atte ue Coney Itt Shc cada KO fiw) hie in SHORES yee HY Wad ye CoN 0% yy a Woe cae “ait an y nt i! M4 "atta 1 i ie ‘ paar Al Mery tet aly vide ayo nyt ty M19, Tita bent Paateney ren Saige SRR ocan pony ase teat HN Hn RA WaT eG ity Vk i we alas att mong ny nf RET ee CG Rare ED Rr wn see pia keer a TTL tema! Fey ath Py iy i AN SCC Th tis ae ‘ , bind ily H ote slut sitar a res alad \amntie fh ee a ae Rr Oe Onna VET code acta Oca EUNe Ha Hye e va ie ‘ 3, iN Wawa ae serial Bich ayy nena SU eon Tt SMO SHAT i int ra ati jt na Ca matey i WV eat d \ ee itanianseeiclit vy asta varieo ay este 1 ate ari Wutnree bebe in a Naveen en bien} see fan pe " Wie ree Sng enna ted ila. A a ny c can Wiceneas Wy Bey ak Teed fe ici Wats hae Wey aaa Horna ane tna wat enti ay Pras 088 on 4d ved be site. eae ue Ayes tah ie fn on Wunbens ey ranted tet a dete sty ee fap tse Deere Tin! tt pte ada On eg haa y Hen BET ee) rae] Pye ge ie TIOwTY ye ane Po ee nauthoneacend tay bra aa. te yet aint ihe i Sobek iRe aa NSIV RST GM Gotth neni Nea hanes 4h f ih binant crate rah Nya) SS Titics ete. a Hats bast So nurnre 8 ana ae wheal oi Rien ase ati nyse H ay ilelea ta ony ih ity sihewateasd ai hotels eAath aa Faved i fase Wome Manan. is" 5y bata aut asa utes Hea Ae Wee br oo Wis Anne nee. Asta nk adh a ny aa he nate gmaucica yaya eC em Ma isthe sh testi BO ern i eit Wert Ne ai ape Lepatistabad ean ico a Ais amnes exits a ay oe Reet c Cray Srey Sh wansh tps bat go i any asie balance Sasa Rea teay, ing Pe (ried >, ratte tase why EAE APR aa (eaten 458 iat bana deer act A ecentie y 7 ieee srr see Ee en Shemale Wok srnidtrnbedcund se atin aed hr oe ei Ce hepenae caPtireae: cine Pita ie Bedutaeeet: brea ate Saeed snatong a Ses gece nate ug tena oA a gis a we Hiqaewie neh aa “aus es sintiien et yse! bined wae tea Henk Spies earn A est non ere War Arab NH pad at Dente rr rene ia ai at! Se freed Dhami gee : d in SNCS CaS wien sea iy Pirie ayant 4 ens pee tote “hat woe Por Tees Wr wn etn tee ‘ irennnn Wd hs) te) ws ihe a wg ; nH vlew Sits be ce Once AoE ibe PT tet, Pr a a wae Pera erent rare i AN A ci * ed ie 07 { if ede cnt rps ; , Hs roy is ett bate * lobia TSIM ras Wi td ttt RONINUTE Wa Dad vor ie rae iia fy ag ee if} rn besa hal ti “yi i a AA ni Sy ae syle Said seine Vite Wa a Das ie Mahal Cane POUR Ie ee Pe a heh betel ' nse : cana vat nwa wining tet eee uy ae ae aN Bie cS inde ay TMs pibtendeet dan tt Fait ota pede ee aie halk ye dg mgt OMe UE aE E RODS NG Cee er racer ne er aan aay sia Uap beasts eat Pulsheg tal tated hominy nmr es Jroviueae rte g in mu Hi tw tokens Po tara arte tataet us iape te Brerhaeech ter rite ssegey eee iy 4 Pek aioe es Sy Sacieute sie et Ga! Peres ora sin iets ata! Sosa ftir the ee eat a eat « ee tw eprint eo Carne urge i ig gevaspens sek Rntincarie te hneay Latin ata alata ae PUP han Cee, ap Rata is resees sid rg gaat PS Coa pe LOH Lae relereshe Ee te ed its ata roe re atts i ieara ety AD BU ebacisitacat Vana ty haga ens seneteat He mae dw ) mo fate Ahn stat Ee Rand ev fet Comair Hind ere ‘ ' Aa oe Fraley ge STON TE PE ad eb aca abe i" ’ mee 10 atk Ase in abet pint tbe ee Prats at} ae i Ha jae ENN eta mo Pre hae teen $34 ‘issues riya re Anan Vii A Se Cin neal itt ink of 4 i! Manan) aH yf a PLR aCe nT if siesta tn etal atta ears ie see ee Se ina Pian tad detent ae i naan Cae Aisin ay ean THe ds SO ia) =< 4 5 , & Sas }YVYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILOLILSNI. NVINOSHLINS S31YVYGI7 LIBRARIES S SMITR SMITE NVINC SMITH SMITH 2 wn = ” > ey 7) = 5 2 +8 e 5 Y a ul = o = oc 4a SSS Cf 4 es = < % a < = % Xs. 7) > ” = = ” = ” ba 7 ud uw oc = ac et load 7 ac = < c » x a < ia < = © z z = : = : = = ro) cad —_ ca = a = = 2 — S pa) S 4 2 IUYVYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3ZIYVYUEIT LIBRARIES | - z - = ss ra = & = 2 z - z a ow -_ mee) — baa ow p= 2 5 a 5 > = m = Pad ey > - P- ress > te 2 ai i i 2 i ~ Gi a wn no fa w 72) i =e eit, Zz Hs = Bo = NOILNLILSNI NVINOSHLINS S3IYVYYGIT_ LIBRARIES SMITHSONIAN INSTITUTION. NOILALILSNI_ w = no z nore wn z =e 7) Fe = = = ee = = GQ. = = a = 4 sl fg = =] z SS a wh hal = x fe) 3: Gif re) au 9 WS = tf > w ” D LY ” on a” Ny n xf ¢ rT) : Ew) 2 = : EY EG = > G = > = _ 2 Ss > G = .— > w Fa 7) > 2 7) » 2 wn J1uvYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI posrlevag LIBRARIES | ud) a > > P= > = >) D ‘a - a i pad — ° am (72) m = Hs e o = 77) LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI_NVINOSHLINS S31YVuaItT Zz 77) z= Ww z n 2 n < = <> = < = < = = 4 Zz Xx = = = z = a a ie AW 65 AG o CR S\N 2 x fe) rT MOS Oo pe 9 =z WA Ss ae. 2 E AS" Z E Z, SENN \E- = Sh OS ae ae 2 = 5 = 8 STITUTION NOILALILSNI_NVINOSHLINS S3INVYEIT_LIBRARI ES SMITHSONIAN _INSTITUTION op NOVLALILSNI wy a uw ra “ty, & w o [a = oc et a bet Ge = < _< a < mal < c = = ce = ce = = = oe 3 ce 5 =) 3 = S at Zz =i z 2 2 aA = J1YVUGIT LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNI NVINOSHLINS SAIUVUGIT LIBRARIES Pe za fs a re poe A — (ED = YN 2G EE OU 2 QW = Ww 2 B® =| SS = “Gp = » 5 Ruus” S ~~ = Yi ly = Noor 5 “Uf S NN a (7p) eo ad g (Mp) 4 IN_INSTITUTION NOILNLILSNI_NVINOSHLIWS $3 1yVUGIT_LIBRARI ES SMITHSONIAN _ INSTITUTION NOILNLIL = w 4 ae uw za wW za ul = wo = = S 7 = v \s ve D ke > ree oP res > SIs ok > J = ee} = a = a) KW = 9) \N INSTITUTION NOILOLILSNI NWINOSHLINS S3luYvUgIT_LIBRARIES SMITHSONIAN INSTITUTION NOILNLIL ze wn Zz a) ay = 4) Zoe 7) = = g&, = = thy», = = <4 = ty Y Mp. rad Zz eel NY $ Y § 2 2256") GF | ; s\n & . YG Gp. Gi; . 2 3 \ XX \ Ye Z = $ Whe? g a= 2 = AN 2 74 = >’ = >" = © = “S my (72) r4 wo Pa (22) 2 a = > IS LIBRARIES SMITHSONIAN INSTITUTION NOILONLILSNI_NVINOSHLINS S3IYVYSIT LIBRARI YF 2 - Zz oe 2 2 (rs) Tr) tu n uJ ” WW o iz 4 i oo aS a = 4 oe = = am c: = Yd oe = Pe = ac “> oc GS WIN « c co = fue] = fae) = SN SS oO = = ro) =" re) as fe) “SS a fe) 3] za ay ed = a a za \N INSTITUTION NOILALILSNI NVINOSHLINS S3IYVYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILOLIL a z = on z = = = z «@ iS o Xe ° o = ow © 2 = Ps) WR a ) iS » = > — PP" : = > te > ra = = 2 a a i 2 res m 2 m n” m ss m n” n z w = (ap) = (72) z NS S3IYVUGIT_ LIBRARIES SMITHSONIAN NOILNLILSNI S3JIMVUSIT LIBRARI g zg « $ z g z g 2 3 AS fx 2 2S 3 = = ‘S SG Ont Se Z WS E LY SE QX F - z S 3 Lo 3 AS |1O Gre; as [e) Y ; whe QA Sam EN & : : é 2 my ESS > = > = > Ss > = : Fa 7) ee a . 3 77) Fa ” IN_INSTITUTION NOILOLILSNI_NVINOSHLIWS S3luvuaiy SMITHSONIAN _INSTITUTION NOILOLIL Fa 2 we ” 2 hs Pas = oc =a Ae a a. = = = < ~ ASN < 4 < = < = a = NN ac c oc = a [e} _ ms 2 sr 3 25 5 2 a WS_S3IYVUdIT LIBRARIES SMITHSONIAN INSTITUTION NOILLNLILSNI Satuyvudit LIBRAR é : 5 mp, 5 g = 6 : ce ies] — 1), [es] — [vs] — , = > = Of > rad > = > 5 x = he = 2 = 2 ” = n° mi no m m = 7) = w z wo ae wn AN INSTITUTION NOILALILSNI NVINOSHLINS S3Iy¥yvH¥aIT INSTITUTION NOIMLALIL za n z no = Q a 7) = = < = < = < = z 4 A ra Si Fa q = = \ 3 x iB z 8 NOG Z x= Oo om re) rT WISN O te ro) S zy = Zz cE NS Zz = zo = >" = >" = aeNS > = >’ ” 4 w az 7p) es = ” Fad NS LIBRARIES SMITHSONIAN INSTITUTION NOILOLILSNI_ us Zz ne > w” S 7) = = no = ow ee a a n” , < = = ‘5 o: = © = ac = oc S oc = a = 4 = 3 2 a z e az AN INSTITUTION NOILNLILSNI NWINOSHLINS S3IYVUGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLtd re . = YSIS [e} SVM = So" A = SY, a WSON 5 ~ SLI S rs «0D Sea 2 wr @ WS 28 CA oD KE a AS 2) Ne COLLIS : : hi mn ‘ : e ; ia wi 7 eo . ve . 7 rr ! it : f eee ee aren as ; , AMERICAN MALACOLOGICAL BULLETIN VOLUME 10 NUMBER 1 Biannual Journal of the American Malacological Union CONTENTS Histology and cell ultrastructure of the mantle and mantle lobes of the Eastern oyster Crassostrea virginica (Gmelin): a summary atlas. CAROL M. MORRISON 3.225 oi cas hctcling «NG ban a oe eee Oh bea es & 1 The histology and ultrastructure of the adductor muscle of the eastern oyster Crassostrea virginica (Gmelin). CAROL M. MORRISON ....................... 25 The Asiatic clam Corbicula fluminea (Miller, 1774) (Bivalvia: Corbiculidae) Genetic relationships among Asian Corbicula: Thai clams are reférable to topotypic Chinese Corbicula fluminea. DAVID S. WOODRUFF, VARAPORN KIJVIRIYA and E. SUCHART UPATHAM......................-00-005 51 Genetic structure and heterozygosity-related fitness effects in the marine snail Littorina littorea. DAVID W. FOLTZ, SANDRA E. SHUMWAY ‘and DENNIS CRISP 32 6 os coe 5 ee a ae 04 cee oe i ee a 55 The arm crown in cephalopod development and evolution: a discussion of morphological and behavioral homologies. SIGURD v. BOLETZKY .................... 61 Reproductive anatomies of Holospira spp. (Gastropoda: Pulmonata: Urocoptidae) from Arizona and Sonora with a new subgenus and a new subspecies. LANCE H. GILBERTSON ................... 0.0.00... 0 00 eae. 71 Life History of the Endangered Fine-Rayed Pigtoe Fusconaia cuneolus (Bivalvia: Unionidae) in the Clinch River, Virginia. SUE A. BRUENDERMAN and RICHARD J. NEVES .................. 00000. e eee 83 Comparative feeding biology of Acteocina canaliculata (Say, 1826) and Haminoea solitaria (Say, 1822) (Opisthobranchia: Cephalaspidea). CHARLES M. CHESTER ............. 0.0... 00. ccc eee 93 Research Note: Comparative Morphology of the Byssi of Dreissena polymorpha and Mytilus edulis. LARRY R. ECKROAT and LOUISE Mi STEELE 0-5 esheets ee eee ere 103 Review: Zoological Catalogue of Australia,-Vol. 8. Non-Marine Mollusca. Winston Ponder ....... 109 Announcement............. ee er ee a eg eR ne lil AMERICAN MALACOLOGICAL BULLETIN BOARD OF EDITORS ROBERT S. PREZANT, Editor-in-Chief Department of Biology Indiana University of Pennsylvania Indiana, Pennsylvania 15705 RONALD B. TOLL, Managing Editor Department of Biology University of the South Sewanee, Tennessee 37375 ASSOCIATE EDITORS MELBOURNE R. CARRIKER College of Marine Studies University of Delaware Lewes, Delaware 19958 GEORGE M. DAVIS Department of Malacology The Academy of Natural Sciences Philadelphia, Pennsylvania 19103 W. D. RUSSELL-HUNTER Department of Biology Syracuse University Syracuse, New York 13210 THOMAS R. WALLER Department of Paleobiology Smithsonian Institution Washington, D. C. 20560 FRED G. THOMPSON, Ex Officio Florida Museum of Natural History University of Florida Gainesville, Florida 32611 R. TUCKER ABBOTT Melbourne, Florida, U.S.A. JOHN A. ALLEN Millport, United Kingdom JOHN M. ARNOLD Honolulu, Hawati, U.S.A. JOSEPH C. BRITTON Fort Worth, Texas, U.S.A. JOHN B. BURCH Ann Arbor, Michigan, U.S.A. EDWIN W. CAKE, JR. Ocean Springs, Mississippi, U.S.A. PETER CALOW Sheffield, United Kingdom JOSEPH G. CARTER Chapel Hill, North Carolina, U.S.A. ARTHUR H. CLARKE Portland, Texas, U.S.A. CLEMENT L. COUNTS, III Princess Anne, Maryland, U.S.A. THOMAS DIETZ Baton Rouge, Louisiana, U.S.A. WILLIAM K. EMERSON New York, New York, U.S.A. DOROTHEA FRANZEN Bloomington, Illinois, U.S.A. VERA FRETTER Berkshire, United Kingdom ROGER HANLON Galveston, Texas BOARD OF REVIEWERS JOSEPH HELLER Jerusalem, Israel ROBERT E. HILLMAN Duxbury, Massachusetts, U.S.A. K. ELAINE HOAGLAND Washington, D.C., U.S.A. RICHARD S. HOUBRICK Washington, D.C., U.S.A. VICTOR S. KENNEDY Cambridge, Maryland, U.S.A. ALAN J. KOHN Seattle, Washington, U.S.A. LOUISE RUSSERT KRAEMER Fayetteville, Arkansas, U.S.A. JOHN N. KRAEUTER Baltimore, Maryland, U.S.A. ALAN M. KUZIRIAN Woods Hole, Massachusetts, U.S.A. RICHARD A. LUTZ Piscataway, New Jersey, U.S.A. GERALD L. MACKIE Guelph, Ontario, Canada EMILE A. MALEK New Orleans, Louisiana, U.S.A. MICHAEL MAZURKIEWICZ Portland, Maine, U.S.A. JAMES H. McLEAN Los Angeles, California, U.S.A. ROBERT F. MCMAHON Arlington, Texas, U.S.A. ANDREW C. MILLER Vicksburg, Mississippi, U.S.A. BRIAN MORTON Hong Kong JAMES J. MURRAY, JR. Charlottesville, Virginia, U.S.A. RICHARD NEVES Blacksburg, Virginia, U.S.A. JAMES W. NYBAKKEN Moss Landing, California, U.S.A. A. RICHARD PALMER Edmonton, Canada WINSTON F- PONDER Sydney, Australia CLYDE F. E. ROPER Washington, D.C., U.S.A. NORMAN W. RUNHAM Bangor, United Kingdom AMELIE SCHELTEMA Woods Hole, Massachusetts, U.S.A. DAVID H. STANSBERY Columbus, Ohio, U.S.A. FRED G. THOMPSON Gainesville, Florida, U.S.A. NORMITSU WATABE Columbia, South Carolina, U.S.A. KARL M. WILBUR Durham, North Carolina, U.S.A. Cover. Io fluvialis (Say, 1825) is the logo of the American Malacological Union. THE AMERICAN MALACOLOGICAL BULLETIN is the official journal publication of the American Malacological Union. AMER. MALAC. BULL. 10(1) ISSN 0740-2783 AMERICAN MALACOLOGICAL BULLETIN VOLUME 10 NUMBER 1 Biannual Journal of the American Malacological Union CONTENTS Histology and cell ultrastructure of the mantle and mantle lobes of the Eastern oyster Crassostrea virginica (Gmelin): a summary atlas. CARO AMOR RISON (oe ang: he ee eae Co ae 1 The histology and ultrastructure of the adductor muscle of the eastern oyster Crassostrea virginica (Gmelin). CAROL M. MORRISON ....................... 25 The Asiatic clam Corbicula fluminea (Miiller, 1774) (Bivalvia: Corbiculidae) in Europe. R. ARAUJO, D. MORENO and M. A. RAMOS .......................... 39 Genetic relationships among Asian Corbicula: Thai clams are referable to topotypic Chinese Corbicula fluminea. DAVID S. WOODRUFF, VARAPORN KIJVIRIYA and E. SUCHART UPATHAM.......................02.05. 51 Genetic structure and heterozygosity-related fitness effects in the marine snail Littorina littorea. DAVID W. FOLTZ, SANDRA E. SHUMWAY? ang: DENNES: CRISP 3 2.045 27 atid wenn ecden fs ond HG 3 ee ee ee 55) The arm crown in cephalopod development and evolution: a discussion of morphological and behavioral homologies. SIGURD v. BOLETZKY .................... 61 Reproductive anatomies of Holospira spp. (Gastropoda: Pulmonata: Urocoptidae) from Arizona and Sonora with a new subgenus and a new subspecies. LANCE H. GILBERTSON ..................... 00.00. c eee eue 71 Life History of the Endangered Fine-Rayed Pigtoe Fusconaia cuneolus (Bivalvia: Unionidae) in the Clinch River, Virginia. SUE A. BRUENDERMAN and RICHARD J. NEVES ...................0..000000005. 83 Comparative feeding biology of Acteocina canaliculata (Say, 1826) and Haminoea solitaria (Say, 1822) (Opisthobranchia: Cephalaspidea). CHARLES M. CHESTER ....................0 0.0 ccc eee 93 Research Note: Comparative Morphology of the Byssi of Dreissena polymorpha and Mytilus edulis. LARRY R. ECKROAT and LOUISE IMS PEER con gacen daar desalil hod Bovis Radin ee os os DEA LA bh ontoe bnew e 4 103 Review: Zoological Catalogue of Australia, Vol. 8. Non-Marine Mollusca. Winston Ponder ....... 109 FATIMOUNGCETIEM Urge wrse area entere Dinning, NR AEs ha ites ott ene eae ie ee et Laat eal, Main lil i! Histology and cell ultrastructure of the mantle and mantle lobes of the Eastern oyster, Crassostrea virginica (Gmelin): a summary atlas Carol M. Morrison Department of Fisheries and Oceans, Box 550, Halifax Fisheries Research Laboratory, Halifax, N.S. B3J 2S7, Canada Abstract. The mantle and mantle lobes of Crassostrea virginica are described using light and scanning and transmission electron microscopy. Also, literature on these structures in bivalves is reviewed. Two types of goblet shaped secretory cells are found in the epithelium surrounding both the mantle and its lobes. There are tracts of ciliated epithelial cells on the inner or pallial surface of the mantle, and on the inner and middle lobes; whereas the outer or shell surface, and the outer lobe have few ciliated cells. Ciliated single celled receptors are present at the distal ends of the tentacles, and probably also on the pallial surface of the mantle. Many of the epithelial cells of the shell surface have extensive basal invaginations, as often found in cells associated with ion transport. The muscle fibres and associated glial cells of the mantle and its lobes have a similar appearance to those of adductor muscles, except that intermuscular junctions are common. The diameters of the thick myofilaments of the radial muscle of the mantle are similar to those of the opaque part of the adductor muscle, whereas the thick myofilaments of the subepithelial muscle fibres and the muscle fibres in the lobes are thinner. The periostracal gland consists of flattened epithelial cells at the base of the periostracal groove, that have extensive apical processes surrounded by periostracum. The latter forms a thin sheet that extends out over the inner surface of the outer lobe, that is covered with secretory granules. This surface is composed of epithelial cells surrounding crypts, and goblet cells full of secretory granules, often with cell bodies extending below the epithelium. The mantle of bivalves is important in the transport 8% magnesium sulphate (Galtsoff, 1964) in a refrigerator at of materials for addition and removal of material to the shell about 4°C. Relaxation took approximately 1.5 hours. Two (Neff, 1972b; Wilbur, 1972, 1985; Machin, 1977; Crenshaw, oysters which were used for preparation for scanning elec- 1980), and as a source of nutritional reserves. Mucous secre- tron microscopy (SEM) were maintained in seawater at 20°C. tions protect the permeable integuement, and ciliary move- One of these was fixed after relaxation with 8% magnesium ment moves the mucus and attached particles to the periphery sulphate at 20°C, which took about 6 hours. (Galtsoff, 1964; Machin, 1977). The mantle lobes at the distal The four oysters used in this study were opened by cut- edge are important in periostracum and shell formation ting through the adductor muscle. For LM and TEM, pieces (Beedham, 1958; Neff, 1972a, b; Bubel, 1973a; Saleuddin and of mantle from each oyster extending from the adductor mus- Petit, 1983) and detection of stimuli (Moir, 1977). cle to the distal edge in the region opposite to the hinge The general anatomy and histology of the mantle and (Fig. 1: area A) were excised and pinned out on dental wax. mantle lobes have been described by Galtsoff (1964) using One sample from each oyster was placed in Karnovsky’s fix- light microscopy, but the features are demonstrated only with ative in phosphate buffer (Karnovsky, 1965), and another was drawings. The purpose of this study is to provide a placed in 1G4F (McDowell, 1978) with seawater replacing preliminary survey of the histology and cell ultrastucture of some of the buffer solution (Howard and Smith, 1983). For the mantle and its lobes, using light micrographs of thin resin SEM, one piece of mantle was excised from each animal as sections as well as transmission and scanning electron for LM and TEM, but was not pinned to a surface, since the microscopy. The study is limited to the region of the mantle surface details of the shell or outer surface were also needed extending from the adductor muscle to the distal edge, op- for study. Another sample of mantle was removed attached posite to the hinge. to the shell, by sawing through both the mantle and shell with a jeweller’s saw. All SEM samples were fixed in 1G4F with METHODS OF PREPARATION seawater. After fixation overnight, the tissues for LM and TEM were cut into strips oriented in the axis from the prox- Two mature oysters that were maintained in ambient imal to the distal edge of the mantle (LS), or across this axis seawater at about 4°C were fixed for light (LM) and transmis- (TS). The samples for LM included the lobes, whereas the sion electron microscopy (TEM). One of these specimens samples for TEM were cut into smaller pieces, and the edge was relaxed before fixation by placing it in seawater containing of the mantle with the lobes was separate. Samples for SEM American Malacological Bulletin, Vol. 10(1) (1993):1-24 1 2 AMER. MALAC. BULL. 10(1) (1993) —__1_i U 5 Fig. 1. Oyster viewed from right side, with right valve removed (redrawn from Galtsoff, 1964). were kept whole. All fixatives and buffers for light and elec- tron microscopy were at pH 7.2. Samples for LM were dehydrated in methanol and embedded in JB4 resin. For TEM, samples were post-fixed in 1% osmium tetroxide, dehydrated in acetone then embed- ded in Taab resin (Marivac Ltd., Halifax, N.S.). LM and TEM samples were embedded so that they would be sectioned in LS or TS. Some LM samples were also embedded so that they would be sectioned parallel to the surface of the mantle. The LM sections were stained with a 1:50 dilution of 1% toluidine blue in 1% sodium borate, or Van Gieson stain (Dougherty, 1981). Taab resin semi-thin sections for light microscopy were also stained with toluidine blue. Sections for TEM were stained with 25% uranyl acetate in methanol (Stempack and Ward, 1964) and lead citrate (Reynolds, 1963), and viewed in an Hitachi HS 9 transmission electron microscope. TEM negatives of transverse sections of the muscles of the mantle and lobes were magnified X10 using a Nikon profile projector, and diameters of the thick myofilaments were measured using the micrometers on the projector. SEM samples were post-fixed in 1.5% osmium. tetroxide, dehydrated in acetone, critical-point dried, mounted with either the mantle surface facing the pallial cavity or facing the shell uppermost, and sputter-coated with gold. They were viewed in an Hitachi S-2400 scanning electron microscope. The following abbreviations are used in the plate cap- tions and elsewhere in this paper (specimen preparation): K,- Karnovsky’s fixative: 1% glutaraldehyde-4% formaldehyde (from paraformaldehyde); 1G4F-1% glutaraldehyde -4% com- mercial formaldehyde in phosphate buffer with sea-water add- ed; MS,-specimen relaxed in magnesium sulphate; JB4,- specimen embedded in JB4 for light microscopy; 0.5um,-0.5um. section of specimen embedded for electron microscopy, stained with toluidine blue; CMB,-chromotrope 2R-methylene blue; MBBF,-methylene blue-basic fuschin; TB,-toluidine blue; VG,-Van Gieson; L.S., longitudinal sec- tion; T.S., transverse section; (cell structure): A,-amoebocyte; AX,-nerve axon; BM,-basement membrane; C,-cilia; CA,- circumpallial artery; CE,-cytoplasmic extension; CF,-collagen fibres; CL,-cross-links; CM,-circular muscle; CN,- circumpallial nerve; CR,-crypt; DB,-dense body; DBA,- diagonal banding; EC,-endothelial cell; F-muscle fibre; G,- granule; GA,-Golgi apparatus; GC,-glial cell; GL,-glycogen; HD,-hemidesmosome; J,-junction; L,-lipid droplet; M,- mucus; MC,-mucous cell; MI,-mitochondrion; ML,-middle lobe; MV,-microvilli; N,-nucleus; NE,-neuromuscular ending; OM,-opaque portion of adductor muscle; P- periostracum; PC,-pallial curtain; PG,-periostracal groove; PGI,-periostracal gland; PS,-surface of mantle proximal to pallial cavity (pallial surface); rudimentary Quenstedt’s mus- cle; RC,-receptor cell; RER,-rough endoplasmic reticulum; RM.,-radial muscle; RN,-radial nerve; S,-sinusoid; SC,- secretory cell; SER,-smooth endoplasmic reticulum; SG,- secretory granule; SGC,-secretory granular cell; SL,-shell lobe; SM,-secretory material; SP,-spicule; SR,-sarcoplasmic reticulum; SS,-surface of mantle proximal to shell (shell sur- face); T,-tentacle; TBA,-transverse banding; TM,-translucent portion of adductor muscle; TNM,-thin myofilaments; TKM,- thick myofilaments; V,-vesicle; VC,-vesicular cell; VG,-vesicle with irregular, electron-dense, granular contents. RESULTS A conspicuous feature of the mantle is the fan-like ar- rangement of the radial muscle, which can be seen in a sec- tion cut parallel to the surface (Fig. 2). This results in more muscle being present peripherally in the mantle. The radial muscle and accompanying nerves are near the surface prox- imal to the shell, and sometimes the nerves are surrounded by muscle fibres (Figs. 3, 4). The mantle consists of a spongy network of vesicular cells, separated by blood vessels and sinusoids, and is covered by a columnar epithelium. Often there are amoebocytes between the vesicular or epithelial cells. Small muscle fibres occur beneath the epithelium, and also run across the tissues of the mantle. Most of the epithelial cells on the inner or pallial sur- MORRISON: MANTLE AND MANTLE LOBES OF THE EASTERN OYSTER 3 Fig. 2. The section is cut parallel to the surface of the mantle. The muscle fibres of the radial muscle radiate out from the region where the mantle is attached to the body, at the left, to the outer edge of mantle, at the right (IG4F, MS, JB4, CMB, scale bar = 0.1 mm). Fig. 3. L.S. mantle. A columnar epithelium covers the upper, pallial surface and the lower, shell surface. The radial nerve and the large fibres of the radial muscle are near the shell surface. There is a layer of small subepithelial muscle fibres near the shell and pallial surfaces of the mantle. Amoebocytes, large vesicular cells and sinusoids are present throughout the mantle (K, MS, JB4, TB, scale bar = 0.1 mm). Fig. 4. T. S. mantle. The radial nerve is surrounded by fibres of the radial muscle. Subepithelial muscle fibres are present, and muscle fibres also extend across the mantle (K, MS, JB4, CMB, scale bar = 20 ym). Fig. 5. Epithelium on PS of mantle. T.S. Goblet shaped secretory cells with distinct granules, one secreting its contents and mucous cells are present among columnar epithelial cells, which have microvilli and sometimes also cilia at the apex. A thick basement membrane supports the epithelium. Subepithelial muscle fibres, amoebocytes and glial cells are present below the epithelium (0.5 wm, K, MS, scale bar = 20 um). Fig. 6. Epithelium on SS of mantle. T.S. The cells are columnar to cuboidal, and many contain darkly staining granules of varying shape and size. Granular secretory cells and mucous cells are present. There are subepithelial muscle fibres, and a radial nerve surrounded by glial cells (0.5 um, K, MS, scale bar = 20 um). 4 AMER. MALAC. BULL. 10(1) (1993) face are columnar, and a thick basement membrane is present (Fig. 5). These cells have microvilli, but there are tracts of cells which also have cilia (Figs. 5, 7 and 8). Some of the latter have several cilia interspersed with microvilli, whereas others have only one or two cilia. The narrow openings of goblet shaped secretory cells can sometimes be seen (Fig. 8). Most of the epithelial cells of the outer or shell surface are shorter than those on the pallial surface, dark granules can often be seen in them, and their basement membrane is thin (Fig. 6). These cells are usually covered with microvilli, although there are a few ciliated cells (Fig. 9). There are granules and sheets of mucus on the shell surface, and many small spicules in grooves or depressions (Figs. 9, 10). There are two types of goblet shaped secretory cells in the mantle epithelium (Figs. 5, 6). One contains distinct, dark- ly staining granules (SGC), the other granules which are more closely packed and tend to coalesce, giving the appearance of a typical mucous cell (MC). The granules of the first type of cell are still distinct when they are discharged. Below the epithelia small muscle fibres run in all directions in close association with elongate glial cells containing dark, round granules. Glial cells also surround the radial nerve (Fig. 6), and accompany the radial muscle fibres (Fig. ll). In a specimen which had not been relaxed with magnesium sulphate, the radial muscle contracted and the epithelium was thrown into folds (Figs. 12, 13). On the shell surface of this specimen, there are subepithelial muscle fibres within the folds (Fig. 13). The ciliated cells of the pallial surface usually have several cilia interspersed with microvilli at the surface (Fig. 14), whereas occasionally a cell was seen with only a few cilia and microvilli at the surface (Fig. 14, insert). The nuclei of the epithelial cells are basal, but many cells do not appear to reach the basement membrane, so appear to form a par- tially stratified epithelium. Mitochondria are situated near the nucleus as well as in the apical cytoplasm, which also contains vesicles, some with dark-staining contents. There is a Golgi apparatus just above the nucleus, and the cytoplasm contains glycogen. In the goblet cells that contain distinct granules, mitochondria can be seen between the granules. The thick basement membrane consists of several fine, filamentous layers. The subepithelial muscle fibres are ac- companied by glial cells and nerve axons. At the shell surface, the epithelial cells often have basal invaginations that form cisternae of smooth endoplasmic reticulum (Fig. 15). The latter are sometimes dilated with dark material (Fig. 15, insert). Between these invaginations are cisternae of rough endoplasmic reticulum, which also ex- tend around the nucleus (Fig. 16). The cytoplasm contains glycogen; round granules which are not membrane-bound, and have the amorphous appearance typical of lipid; and membrane-bound vesicles containing granules with irregular shapes and of varying density that could be secondary lysosomes or phagosomes. Some cells have a darker cytoplasm, containing more densely packed glycogen, than others. Secretory cells containing the membrane-bound granules typical of mucous cells are present (Fig. 15; MC). The membranes are often incomplete where the granules are beginning to coalesce. There are also cells containing distinct, membrane-bound granules (Fig. 16: SG). Below the epithelium is a thin basement membrane. The subepithelial muscle fibres consist of a central myofibril containing thick and thin myofilaments (Fig. 17). Of 98 myofilaments measured near the pallial surface, the widest myofilament was 65 nm in diameter but most were 36-45 nm in diameter. The mean and median width was 42 nm. There are dense bodies in the centre of the myofibril as well as at the sarcolemma, where they form hemides- mosomes or attachment plaques. These muscle fibres are often closely associated with each other, and sometimes form junctions (Fig. 18). The nucleus of the muscle fibre is situated to one side of the myofilaments (Fig. 19). The radial muscle consists of a group of muscle fibres that are larger than those beneath the epithelia (Fig. 20). Some of the fibres that are elongate in transverse section are up to 4.5 um in diameter along the long axis, but most are about 1.4-2.9 wm in diameter. Glial cells and sometimes small blood vessels are found between the muscle fibres. There are pro- files of sarcoplasmic reticulum beneath the sarcolemma, and several dense bodies can be seen at the periphery of each fibre, forming hemidesmosomes, as well as in the centre (Fig. 21). Like the muscle fibres beneath the epithelia, these fibres sometimes form junctions with each other, and with nerve- endings and glial cells. Glycogen rosettes, as well as large dense membrane-bound granules, the gliosomes, can be seen in the glial cells. The nerve-endings are usually accompanied by a glial cell, and sometimes contain small clear vesicles about 70 nm in diameter (Fig. 22), sometimes larger electron- dense vesicles about 160 nm in diameter (Fig. 23). Some end- ings have a mixed population of clear and dense-cored vesicles. The intercellular cleft is narrow, and does not ap- pear to contain collagen or a basement lamina in figure 22, although there appears to be a lamina in figure 23. The nucleus of the muscle fibre is often situated in a peripheral extension of the cytoplasm, which also contains other cell organelles, such as the Golgi apparatus and mitochondria (Fig. 24). The thick myofilaments are cross-striated, and sometimes show diagonal banding like that of the adductor muscles (Fig. 25). Of a hundred thick myofilaments measured, the maximum width was found to be 109 nm, and the mode was in the category of widths of 50-56 nm. The mean width was 62 nm, and the median width 58 nm (Fig. 26). The radial nerve contains a large mass of unmyelinated nerve-cell processes surrounded by a layer of glial cells (Fig. 27). The connective tissue of the mantle is composed of MORRISON: MANTLE AND MANTLE LOBES OF THE EASTERN OYSTER 3) Fig. 7. Surface of PS. Most of the epithelial cells have microvilli at the apex. Tracts of ciliated cells and some cells with only one or two cilia are also present (IG4F, SEM, scale bar = 25 pm). Fig. 8. Surface of PS. Most of the ciliated cells have several cilia interspersed with microvilli, but some have only one or two cilia. The unciliated cells have microvilli of a constant length and width at the apex, and the openings of secretory cells can sometimes be seen between them (IG4F, SEM, scale bar = 5 um). Fig. 9. Surface of SS. Most of the surface is devoid of cilia, although occasional ciliated cells are present. There is mucus and granular material on the surface of the cells (IG4F, SEM, scale bar = 5 ym). Fig. 10. Surface of SS. Microvilli are sometimes evident, but often they are covered by mucus, granules or spicules, especially in crevices (IG4F, SEM, scale bar = 2.5 um). AMER. MALAC. BULL. 10(1) (1993) 4 | Fig. 11. L.S. radial muscle. The muscle fibres are accompanied by long glial cells containing dark granules (K, MS, JB4, VG, scale bar = 20 yum). Fig. 12. PS of mantle. L.S. The epithelial surface, which is supported by a thick basement membrane, is extensively folded. Small subepithelial muscle fibres are present (IG4F, JB4, TB, scale bar = 20 um). Fig. 13. SS of mantle. L.S. This is from the same section shown in figure 12. Small subepithelial muscle fibres are seen in cross-section in folds of the epithelium, and thicker radial muscle fibres are seen in longitudinal section (IG4F, JB4, TB, scale bar = 20 pm). Fig. 14. PS of mantle. L.S. A nucleus can be seen at the base of a columnar epithelial cell with microvilli and cilia at its apex. Mitochondria surround the nucleus and are present in the apical cytoplasm, along with vesicles. Secretory cells contain discrete, darkly staining granules, and there is an amoebocyte between the epithelial cells. There are subepithelial muscle fibres, glial cells, and axons beneath the thick basement membrane (K, MS, TEM, scale bar = 2 pm). Fig. 14: insert. This cell, from the same area as figure 14, has a cilium and a few microvilli at the apex, mitochondria, and vesicles in the apical cytoplasm and a Golgi apparatus above the nucleus (scale bar = | um). MORRISON: MANTLE AND MANTLE LOBES OF THE EASTERN OYSTER 7 cells with many vesicles in the cytoplasm; cells which ap- pear to have amorphous contents, with the cell organelles pushed to the periphery; and cells containing lipid droplets with varying osmiophilia (Fig. 28). With electron microscopy, glycogen can be identified between the vesicles, and some of the cells with amorphous contents are also filled with glycogen (Fig. 29). An electron micrograph of part of a vesicular cell containing lipid droplets of low osmiophilia is shown in Fig. 20, where it occurs among the radial muscle fibres. Numerous small muscle fibres form a network in the mantle, forming junctions with each other. They are similar to those found beneath the epithelium, and consist of a cen- tral bundle of myofilaments, with the nucleus and other cell organelles in a peripheral extension of cytoplasm. Glial cells are also present. As in most bivalves, the outer border of the mantle is divided into three lobes (Fig. 30; Galtsoff, 1964). The radial muscle is thicker at the outer edge of the mantle, so that it extends to the pallial surface (Figs. 30, 31). A band of cir- cular muscle is also seen near the pallial surface at the base of the lobes (Figs. 30, 31). In this region many goblet shaped secretory cells of both types are present in the shell and pallial epithelia, and the bases of these cells often extend below the shell epithelium (Fig. 31). The outer or shell lobe, which secretes the perio- stracum, does not have tentacles; the middle lobe is sensory, and has long, stout tentacles separated from each other by several short, slender tentacles. The inner lobe or pallial cur- tain is muscular, and can be extended into the mantle cavity (Fig. 33), so that the long, stout tentacles of the apposing inner lobes interlock (Galtsoff, 1964). Most of the epithelium covering the lobes is colum- nar, similar to that covering the rest of the mantle (Figs. 34, 35). Many goblet shaped secretory cells are present on the outer (facing the shell) surface near the base of the inner lobe, and on the inner (facing the pallial cavity) surface of the mid- dle and outer lobes. The bodies of the cells on the outer lobe often extend below the epithelium. The epithelial cells in the base of the groove between the inner and middle lobes are more or less cuboidal. The circumpallial nerve is proximal to the base of this groove. The flattened secretory cells of the periostracal gland are at the base of the groove between the outer and middle lobes. Fibres of the radial muscle ex- tend into the lobes, and muscle fibres also run in various directions within each lobe. The epithelial cells of the inner surface of the inner lobe have cilia as well as microvilli (Figs. 36, 37). These cells also have several elements of Golgi apparatus as well as vesicles, some closely associated with the Golgi apparatus, and many mitochondria in the apical cytoplasm (Fig. 37). The epithelial cells of the outer surface contain similar organelles but are less heavily cilated, and have larger vesicles in the apical cytoplasm, some of which are darkly staining (Fig. 38). The epithelial cells on both surfaces have well developed apical junctional complexes, and the adjoining plasmalem- mas are very convoluted. There are longitudinally oriented subepithelial muscle fibres below both surfaces. In the centre of the lobe there are groups of cells similar to the vesicular cells of the mantle, containing glycogen, vesicles and occasionally lipid droplets, as well as muscle fibres oriented in all directions (Fig. 39). There are thick and thin myofilaments and dense bodies in the myofibrils of the muscle fibres, and junctions are sometimes present be- tween them (Fig. 40). Occasionally the dense bodies show oblique orientation for short distances. Out of 47 measure- ments of the diameters of the thick myofilaments of the mus- cle fibres from the inner lobe, the widest diameter was 63 nm, the mean and median widths were 48 nm, and the mode was 45-54 nm. The middle lobe has many goblet shaped secretory cells on its inner epithelial surface and distinct granules can be distinguished in most of them (Figs. 41, 42). The apical cytoplasm of the epithelial cells, some of which are ciliated, is packed with vesicles (Fig. 42). The ciliated cells are in small groups, with the exception of the outer border between the tentacles, which is more heavily ciliated (Figs. 43, 44). This is in contrast to the heavily ciliated inner surface of the inner lobe (Figs. 43, 45). The tentacles of the middle lobe are covered by small tufts of cilia, except near the tip, which is more heavily ciliated (Fig. 46). Besides cells with many cilia, there are some with only one or a few shorter cilia, which often face in different directions (Fig. 47). On the ten- tacles of both the middie and inner lobes some ciliated cells with darkly staining cytoplasm extend below the epithelial surface, connecting with structures that are apparently nerves (Fig. 48). There are few goblet shaped secretory cells on the heavily ciliated outer surface of the middle lobe (Figs. 41, 49), which is covered by a layer of secretory material (Fig. 49). There are many vesicles and sometimes small dense granules in the apical cytoplasm of the epithelial cells, and sometimes blebs of apical cytoplasm containing vesicles ap- pear to join the secretory material on the epithelial surface. The longitudinal musculature is especially well developed near the outer surface of the middle lobe, near the base (Figs. 49, 50, 51). At the base of the groove between the middle and shell or outer lobes there is a recessed region lined by the flattened epithelium of the cells forming the periostracal gland (Fig. 50). There is a transitional region between the ciliated epithelial cells of the outer surface of the middle lobe and these cells; so that cells with long cell extensions surround- ed by secreted periostracum, as well as cilia, are seen on the external surface of the middle lobe near the gland (Fig. 51). The cytoplasm of cells forming the gland is packed with glycogen, there is a Golgi apparatus near the nucleus, and 8 AMER. MALAC. BULL. 10(1) (1993) Fig. 15. SS of mantle. L.S. The epithelial cells have microvilli at the apex. They usually contain membrane bound vesicles with darkly staining inclusions of varying shape and size, and sometimes lipid droplets. There is abundant glycogen in the cytoplasm, which is more densely packed in some cells than others. There are mucous cells between the epithelial cells, and beneath the epithelium subepithelial muscle fibres and an amoebocyte are present (K, MS, TEM, scale bar = 2 wm). Fig. 15: insert. This shows part of the same specimen at a higher magnification. Infoldings of the basement membrane form cisternae of smooth endoplasmic reticulum, some of which are filled with dark material and extend towards the apex of the cell. These infoldings alternate with cisternae of rough endoplasmic reticulum (scale bar = | ym). Fig. 16. L.S. of SS of mantle, near lobes. This portion of the mantle has contracted, in spite of treatment with magnesium sulphate. There are small subepithelial muscle fibres seen in T.S. in folds of the epithelium, as well as secretory cells containing distinct granules. Below the epithelium large radial muscle fibres are visible (K, MS, TEM, scale bar = 2 um). MORRISON: MANTLE AND MANTLE LOBES OF THE EASTERN OYSTER 9 Fig. 17. L.S. of subepithelial muscle fibre near PS of mantle. Thick and thin myofilaments are visible. Collagen fibres are attached to the sarcolemma where there is a hemidesmosome (K, MS, TEM, scale bar = 500 nm). Fig. 18. SS of mantle. Sarcolemmas of two subepithelial muscle cells are closely apposed, forming a junction (K, MS, TEM, scale bar = 300 nm). Fig. 19. PS of mantle. A subepithelial muscle cell is shown in cross section. The nucleus is situated to one side of the cell, and cross sectioned myofilaments, a Golgi apparatus and mitochondria are present in the cytoplasm (K, MS, TEM, scale bar = 1 wn). Fig. 20. T.S. of radial muscle near SS of mantle. The fibres are larger than those of the subepithelial muscle fibres, and glial cells are often closely associated with them. Endothelial cells surround a capillary, and a lipid-containing vesicular cell is present (K, MS, TEM, scale bar = 2 pm). 10 AMER. MALAC. BULL. 10(1) (1993) Fig. 21. T.S. of radial muscle. Thick and thin myofilaments as well as dense bodies can be seen inthe muscle fibre. There are small vesicles of sarcoplasmic reticulum beneath the sarcolemma. There is a junction between two muscle cells, and between one of these cells and a glial cell containing glycogen. Collagen fibres are present between the muscle fibres (K, MS, TEM, scale bar = 500 nm). Fig. 22. Radial muscle. There is a nerve ending containing clear vesicles, accompanied by a glial cell, next to the sarcolemmas of two muscle fibres (K, MS, TEM, scale bar = 500 nm). Fig. 23. This nerve ending contains large, dense vesicles. The accompanying glial cell contains glycogen rosettes (K, MS, TEM, scale bar = 300 nm). Fig. 24. A nucleus is present in the cytoplasm to one side of the myofilaments of a radial muscle cell. There is also a Golgi apparatus, mitochondria and vesicles (K, MS, TEM, scale bar = 500 nm). Fig. 25. L.S. radial muscle. The thick myofilaments show transverse and diagonal banding (K, MS, TEM, scale bar = 500 nm). MORRISON: MANTLE AND MANTLE LOBES OF THE EASTERN OYSTER ll 30 Radial muscle Number of filaments 23 29 35 41 47 53 59 65 71 77 83 69 85 101107 Filament diameter (mid-point) in nm Fig. 26. Graph showing the diameters of the thick myofilaments in the radial muscle of the mantle. The diameters are plotted in categories of 6 nm. The mode is in the category of 50-56 nm, with a mid-point of 53 nm. there are membrane-bound granules with light and dark in- clusions as well as mitochondria (Fig. 52). The apical cytoplasm contains many vacuoles and has many long cytoplasmic processes, each surrounded by a layer of periostracum so that a multilayered, but very thin perio- stracum is secreted (Figs. 51, 52, 53, 54). This extends over the inner surface of the outer lobe, which consists mainly of epithelial cells with microvilli, although a few cilia are present (Figs. 55, 56). Thick and thin myofilaments and dense bodies can be distinguished in the muscle fibres proximal to the gland (Fig. 57). The thick myofilaments are cross-striated, with some indications of diagonal banding, and the dense bodies situated at the sarcolemma form hemidesmosomes (Fig. 58). The epithelial cells of the inner surface form crypts with secretory globules at the surface (Figs. 55, 56, 59). These crypts are lined by cuboidal cells with many vesicles in the apical cytoplasm (Fig. 60). Many goblet shaped secretory cells are present, most of which are of the type containing discrete, dark granules; but others are more like typical mucous cells. The bodies of at least some of these cells ex- tend below the epithelium, and are sometimes closely associated with each other (Fig. 61). The lower surface of the outer or shell lobe has cells with many apical vesicles, and vesicles with electron-dense contents (Fig. 62), but few goblet cells. The base of the outer lobe was contracted in this specimen and, as in the shell surface of the contracted man- tle, muscle fibres are seen in cross section between the bases of the epithelial cells of the inner and outer surfaces (Figs. 59, 62), as well as in all directions within the lobe. They are accompanied by glial cells, nerve axons and neuromuscular endings. The sarcolemmas of adjoining muscle fibres are often parallel and close together, and form junctions (Fig. 63). Sometimes there is an orderly arrangement of one row of thin myofilaments around each thick myofilament (Fig. 64). From 81 measurements of the thick myofilaments, the widest myofilament was 80 nm in diameter, and the thick myofilaments had a mean width of 42 and a median width of 41 nm. The mode was 36-37 nm. DISCUSSION As mentioned above, this study is an overview of the structure of the mantle and its lobes, and many areas need more detailed study. Vesicular cells containing lipid and glycogen are found in the connective tissue of the mantle and its lobes, as described by Galtsoff (1964). According to Galtsoff (1964) and other workers (Bargeton, 1942; Gabbott, 1975), the amount of glycogen is variable, decreasing as the gonads develop, so the cells filled with small vesicles and those filled with glycogen could be different forms of each other, the prevalence of each type depending on the amount of glycogen stored in the mantle. Two types of storage cells have been described in the mantle of Mytilus edulis L. The ‘glycogen cell (=vesicular connective-tissue cell)’’ is similar to the cells described above, whereas the cells containing lipid droplets in the oyster could correspond to adipogranular cells (Lenoir et al., 1989). According to Gabbott (1975), the main energy reserve in the bivalve egg and larva is lipid, which decreases during metamorphosis and early spat growth, whereas the level of glycogen increases as the spat develops. The main energy reserve in the adult is glycogen. Further study of the relationships between the different types of storage cells in the oyster, and their variations at different times of year, is needed. Galtsoff (1964) describes fusiform cells, which appear to be equivalent to the small muscle fibres found crossing the mantle in our study, since they would be difficult to iden- tify as muscle cells without electron microscopy. He also describes longitudinal muscle fibres and elastic fibrils, the latter being more abundant at the free edge and beneath the surface epithelium; these could both be the subepithelial mus- cle fibres described in this account. This arrangement of small muscle cells is similar to that described in Mercenaria mercenaria (Linnaeus, 1758) by Neff (1972b). Small arteries lined by an endothelium and elastic tissue; veins lacking the endothelium, with little elastic tissue, and freely wandering blood cells or ameobocytes were also found in the connec- tive tissue, as described by Galtsoff (1964). Two types of goblet shaped secretory cells were found in the epithelium of the mantle and mantle lobes of Crassostrea virginica, as described by Galtsoff (1964). One, described by Galtsoff as containing eosinophilic granules, has distinct dark granules; the other, described by Galtsoff as a mucous cell, has more closely packed granules that tend to coalesce. Mucins are vitally important for a variety of func- | #2 AMER. MALAC. BULL. 10(1) (1993) Fig. 27. L.S. radial nerve. There is a layer of glial cells at the surface of the nerve, which consists of unmyelinated axons and nerve-endings containing neurofilaments and a variety of vesicles (K, MS, TEM, scale bar = 2 pm). Fig. 28. Mantle. Cells with a vesicular cytoplasm, and cells with large areas of homogenous cytoplasm are present. Dark and light lipid droplets can be seen in some of these. 0.5 um (K, MS, scale bar = 10 um). Fig. 29. Centre of mantle. The vesicular cells contain many vesicles and mitochondria, with glycogen in the cytoplasm between. Other cells are filled with a large mass of glycogen, and most of the cell organelles are pushed to the periphery. Small muscle fibres contain a central core of myofilaments, and are oriented in various directions. Glial cells are also present (K, MS, TEM, scale bar = 2 pm). MORRISON: MANTLE AND MANTLE LOBES OF THE EASTERN OYSTER 13 Fig. 30. L.S. outer edge of mantle and lobes, pallial surface dorsal. The mantle is divided into the inner lobe or pallial curtain, middle or sensory lobe and outer or shell lobe. The circumpallial nerve and artery are at the base of the lobes, and there is a circular muscle above these. The periostracal gland is at the base of the periostracal groove between the middle and shell lobes. Radial muscle fibres extend through the mantle, and some fibres continue into the lobes. Tentacles extend from the middle and inner lobes (IG4F, MS, JB4, CMB, scale bar = 0.3 mm. Fig. 31. T.S. base of lobes. Pallial surface at right, shell surface at left. Secretory cells containing distinct granules and mucous cells can be seen at both surfaces of the mantle. The secretory cells on the shell surface often extend below the epithelium. The radial nerve is surrounded by the radial muscle, which is large and extends across most of the mantle (IG4F, 0.5 wm, scale bar = 50 pm). 14 AMER. MALAC. BULL. 10(1) (1993) Fig. 32. L.S. base of lobes. Radial muscle fibres extend across most of the mantle, and a circular muscle is present near the pallial surface. The radial nerve, circumpallial artery, small subepithelial muscle fibres and muscle fibres passing across the mantle are also visible (IG4F, MS, JB4, CMB, scale bar = 0.1 mm). Fig. 33. Lobes at edge of mantle. The lobes and tentacles are contracted in this specimen. The inner lobe with its tentacles extends up into the mantle cavity. The middle lobe has large tentacles with small, slender tentacles between. The periostracum extends from the periostracal groove over the inner surface of the outer or shell lobe (IG4F, SEM, scale bar = 0.3 mm). Fig. 34. L.S. inner lobe and bases of middle and outer lobes. Muscle fibres extend from the radial muscle into the lobes, and also run obliquely and transversely in the lobes. Goblet shaped secretory cells are concentrated near the base of the inner lobe, and the inner surface of the middle and outer lobes (K, MS, JB4, VG, scale bar = 0.1 mm). MORRISON: MANTLE AND MANTLE LOBES OF THE EASTERN OYSTER 15 tions in bivalves, including feeding, cleansing, protection, shell formation, and dispersal (Prezant, 1985 and 1990). Sim- ple ‘‘goblet’’ cells, the same height as the surrounding epithelium, are common in the epidermis of molluscs. They show great variability in the chemistry of their secretions, but most are mucus-secreting cells (Franc, 1960; Storch and Welsch, 1972; Machin, 1977). Mucus is primarily water, and consists of mucopolysaccharides or glycoproteins with pro- tein and carbohydrate portions (Prezant, 1985). Some goblet cells contain a structureless mass, but some, like those described here, contain dark, membrane-bound granules, which have a denser core in some cells. The granules usual- ly arise in the Golgi area, but can originate from the rough endoplasmic reticulum. The membrane can remain intact until the granule is discharged, or the granules can coalesce before they are discharged (Storch and Welsch, 1972; Machin, 1977). The cells with distinct, dark granules found in C. virginica seem to be of the first type, whereas the cells with lighter granules are of the latter type. Histochemical tests are need- ed to determine the nature of the goblet shaped secretory cells of C. virginica, and would probably reveal more than the two types recognised morphologically. The distribution of the ciliated epithelial cells on the pallial surface is similar to that reported in the fresh-water mussel, Amblema plicata Conrad (Petit et al., 1978) and the tellin, Fabulina nitidula (Dunker) (Kawaguti and Ikemoto, 1962). Beneath the epithelium is an ‘‘elastic basal membrane’’ according to Galtsoff (1964), which is the thick basement membrane shown in this study. In specimens with the radial muscle contracted, this membrane appears to prevent the bases of the epithelial cells of the pallial surface lifting into the folds; whereas the bases of the epithelial cells of the shell surface and of the mantle lobes are drawn up into the folds, which then surround groups of subepithelial muscle fibres. The radiating form of the radial muscle has been shown in drawings by Galtsoff (1964). According to Galtsoff, the high contractile ability of the mantle is related to the presence of the conspicuous radial and concentric musculature, as well as smaller longitudinal and transverse muscles. There appear to be junctions between these muscle fibres, as well as regions where the sarcolemmas are closely apposed. Junctions were not found in the adductor muscle of Crassostrea virginica (Morrison, 1992), but have been described in other molluscan muscles (Twarog et al., 1973; Nicaise and Amsellem, 1983). The muscles of the mantle and its lobes are described as smooth, non-striated by Galtsoff. The thick myofilaments of the radial muscle are similar in width to those of the opaque part of the adductor of Crassostrea virginica, and have similar, though not as clear, cross and diagonal striations, which are typical of paramyosin (Morrison and Odense, 1974; Mor- rison,1992). The radial muscle fibres are similar in diameter to the smaller axis of the flattened fibres of the translucent adductor muscle, but are much smaller than those of the opaque adductor muscle. It would be interesting to know what the differences are between the radial and adductor muscles, since the structure and arrangement of the myofilaments are similar, but the adductor can exhibit ‘‘catch’’. The thick myofilaments of the subepithelial muscles and the muscles in the mantle lobes, although not as wide as those of the radial muscle, are wider than those of the translucent, obliquely striated muscle of the adductor, which have a mean width of 32 nm (Morrison, 1992). The dense bodies of these muscles sometimes show oblique orientation, although the myofibrils are so small that this can only be followed for a short distance. The thin myofilaments of molluscan muscle appear to be similar to the actin filaments in other groups of animals (Heyer et al., 1973). In this study two types of vesicles, about 70 and 160 nm in diameter, were found in neuromuscular junctions of the radial muscle. Possibly this is morphological evidence for the dual innervation indicated by physiological studies on the mantle of the bivalve Spisula solidissima (Dillwyn) (Wilson, 1969); although in the latter study only nerve endings con- taining numerous agranular vesicles were described. A variety of clear and dense vesicles, that could be associated with the presence of different neuromuscular transmitters, have been described in mollusc muscle (Heyer ef al. , 1973; Nicaise and Amsellem, 1983). A similar distribution of vesicles to that found in this study has been reported in Mytilus anterior byssus retractor muscle (McKenna and Rosenbluth, 1973). The nerves and nerve-endings are accompanied by cells of the glio-interstitial network, which have been described in the muscles of various molluscs, and are generally more con- spicuous in marine species (Nicaise and Amsellem, 1983). They have been described in the adductor muscle of Crassostrea virginica (Morrison, 1992). The isolated mantle of the Crassostrea virginica is capable of laying down both the inorganic (mainly calcium carbonate) and organic (conchiolin) components of the shell, and the movement of calcium and bicarbonate ions across it has been studied (Jodrey and Wilbur, 1955; Wilbur, 1972, 1985; Machin, 1977; Crenshaw, 1980; Simkiss and Wilbur, 1989). The shell surface of the mantle is highly permeable to these ions, which are transported through the epithelium of the shell surface into the extrapallial space for shell for- mation. The ion exchange probably takes place in both direc- tions, depending on whether shell is being laid down or whether it is being dissolved. The latter occurs in anaerobic conditions, when acids form as a result of metabolism and dissolve intracellular deposits of calcium carbonate and the inside of the shell. The spicules found on the shell surface of the mantle (using SEM) could result from precipitation of ions from the extrapallial space during specimen preparation. Galtsoff (1964) demonstrated alkaline phosphatase in the epithelium, and stated that both conchiolin-secreting and 16 AMER. MALAC. BULL. 10(1) (1993) Fig. 35. L.S. middle and outer lobes. The periostracum is forming at the base of the periostracal groove. Many muscle fibres are present near the base of the gland as well as in the lobes. Goblet shaped secretory cells are concentrated on the inner surfaces of the lobes, and some extend below the epithelial surface of the outer lobe (K, MS, JB4, CMB, scale bar = 0.1 mm). Fig. 36. L.S. inner lobe. There are small subepithelial muscle fibres next to the pallial or inner surface, which is heavily ciliated, and the outer surface, which has fewer cilia. There is a secretory cell with distinct granules (K, MS, 0.5 ym, scale bar = 30 pm). Fig. 37. Inner surface of inner lobe. The epithelial cells possess cilia and microvilli. Mitochondria, several elements of Golgi apparatus, rough endoplasmic reticulum and vesicles are present in the apical cytoplasm of each cell. Glycogen is more abundant in some cells than others. A layer of subepithelial muscle fibres is present (K, MS, TEM, scale bar = 2 um). Fig. 38. Outer surface of inner lobe. Few cilia are present on the epithelial cells, and none are seen in this micrograph. Large clear vesicles and vesicles with dark contents are present in the apical cytoplasm, and there is a Golgi apparatus and several mitochondria near the nucleus. There is a prominent desmosome in the apical junction, and the adjacent plasmalemmas are convoluted. Subepithelial muscle fibres are present (K, MS, TEM, scale bar = 2 yum). MORRISON: MANTLE AND MANTLE LOBES OF THE EASTERN OYSTER 17 Fig. 39. Centre of inner lobe. Muscle fibres, some being very small, run in all directions. A group of cells containing vesicles, mitochondria and glycogen is present (K, MS, TEM, scale bar = 3 pm). Fig. 40. Part of figure 39 at a higher magnification. Thick and thin myofilaments and dense bodies are present in the muscle fibre (K, MS, TEM, scale bar = | pm). Fig. 41. Middle lobe, outer surface of inner lobe and inner surface of outer lobe. There are many secretory cells on the inner surface of the middle and outer lobes, extending below the epithelium of the outer lobe. Distinct granules can be seen in some of these. The epithelium of the middle lobe is ciliated (0.5 um, K, MS, scale bar = 30 um). Fig. 42. Middle lobe, inner surface. A secretory cell with distinct granules is present among the epithelial cells, which have cilia as well as microvilli, and many vesicles in the apical cytoplasm. Many of these cells do not reach the basement membrane, giving a partially stratified appearance. Subepithelial muscle fibres are present (K, MS, TEM, scale bar = 3 pm). 18 AMER. MALAC. BULL. 10(1) (1993) VANS 4 SRO Ee t Fig. 43. Lobes at edge of mantle. Tentacles extended. The inner surface of the inner lobe or pallial curtain is heavily ciliated. Groups of cilia can be seen on the surface of the middle lobe and both its large and small tentacles, and the outer edge of this lobe is more heavily ciliated. The outer or shell lobe is unciliated (IG4F, MS, SEM, scale bar = 100 um). Fig. 44. Outer edge of middle lobe, inner surface to the right. The inner surface and the tentacle that extends from it have groups of ciliated cells, whereas the outer edge and outer surface of the middle lobe are well-ciliated. The epithelial cells of the tentacle appear in cross-section, because it is sectioned near its surface (0.5 um, K, MS, scale bar = 20 um). Fig. 45. Edge of inner lobe. Many cilia are present on the surface. Cells with an unciliated surface, presumably secretory cells, are also present (IG4F, MS, SEM, scale bar = 20 wm). Fig. 46. Tip of long tentacle from middle lobe. There are cells with only one or a few cilia, and some, especially at the tip, are heavily ciliated (IG4F, MS, SEM, scale bar = 10 pm). Fig. 47. Tip of tentacle from middle lobe. Some cells have only one or two cilia, which face in different directions (IG4F, MS, SEM, scale bar = 5 um). Fig. 48. Tip of tentacle from pallial lobe. A cell with dark cytoplasm and a few cilia, presumably a receptor cell, extends below the epithelial surface (0.5 um, K, MS, scale bar = 20 pm). MORRISON: MANTLE AND MANTLE LOBES OF THE EASTERN OYSTER 19 Fig. 49. Middle lobe, outer surface. The epithelial cells have cilia and microvilli, and are covered by a layer of secreted material. These cells contain varying amounts of glycogen, and some have many vesicles in the apical cytoplasm. Blebs from the apical cytoplasm appear to be joining the secretory material. There are many subepithelial muscle fibres (K, MS, TEM, scale bar = 4 wm). Fig. 50. Base of periostracal groove, between outer and middle lobes. The periostracal gland consists of flattened epithelial cells at the base of the groove, and periostracum can be seen above it. There is secretory material above the epithelial surface of the middle and outer lobes. Muscle fibres are especially well developed near the outer surface of the middle lobe, and around the periostracal gland (0.5 um, K, MS, scale bar = 30 um). Fig. 51. Middle lobe, near base. The epithelial cells become more cuboidal towards the periostracal gland, which is to the left of the micrograph, and long cytoplasmic extensions of the apical cytoplasm surrounded by periostracum extend from some of the cells. There are many muscle fibres below the epithelium (K, MS, TEM, scale bar = 3 pm). Fig. 52. Periostracal gland. The epithelial cells forming the periostracal gland are small and filled with glycogen, and contain vesicles with darkly staining contents. Long cytoplasmic processes surrounded by periostracum extend from these cells. Numerous clear vesicles are present in the apical cytoplasm (K, MS, TEM, scale bar = 2 pum). 20 AMER. MALAC. BULL. 10(1) (1993) calcium-secreting cells are present, although he found no reliable staining methods for identifying them. Goblet shaped secretory cells are abundant on the shell surface at the periphery of the mantle, at the base of the lobes (Fig. 31), where shell formation is most active (Simkiss and Wilbur, 1989), although few of these cells are found on the shell or outer surface of the outer lobe. Basal invaginations of cells are found where cells are involved in ion transport, as in the chloride cells of fish gills, or the cells lining the urinary tract, so the presence of these in the epithelial cells of the shell surface in Crassostrea virginica indicates that these cells could be specialised for the transport of ions. Basal invaginations were also described in the epithelial cells of the shell surface in Mercenaria mercenaria (Neff, 1972b). In both C. virginica and M. mercenaria the thick basement membrane of the pallial surface is lacking, which would facilitate exchange of materials across the epithelium. Most of the calcium in the mantle of bivalves is in a bound form, in spherules or granules, which have been described within intracellular vacuoles and extracellularly (Neff, 1972b; Wilbur, 1972, 1985; Petit et al., 1980), so the dark deposits in the membrane- bound vesicles of the epithelial cells of the shell surface of the oyster mantle and outer lobe could be calcium. The periostracal groove, as described by Galtsoff (1964), is lined by a single-layered, flattened glandular epithelium. This has been termed the ‘‘periostracal gland”’ (Beedham, 1958), although it does not form a compact body, or have a duct. The cells forming the gland seem to be Fig. 53. The periostracum is very thin, and extends across the inner surface of the outer or shell lobe (IG4F, SEM, scale bar = 50 pm). Fig. 54. Periostracal groove. The periostracum extends from the gland, across the inner epithelial surface of the outer or shell lobe (0.5 um, K, MS, scale bar = 20 um). Fig. 55. Inner surface of outer or shell lobe. The surface of most of the epithelial cells is covered by microvilli, although a few cilia are present. The cells are arranged around crypts (IG4F, SEM, scale bar = 10 um). Fig. 56. Inner surface of outer or shell lobe. There are secretory globules at the surface of some of the cells (IG4F, SEM, scale bar = 2.5 um). MORRISON: MANTLE AND MANTLE LOBES OF THE EASTERN OYSTER 21 Fig. 57. Muscle fibres from middle lobe, near gland. Thick and thin myofilaments can be distinguished, as well as dense bodies in the centre of the fibre and hemidesmosomes at the sarcolemma. There are vesicles of sarcoplasmic reticulum next to the sarcolemma (K, MS, TEM, scale bar = | um). Fig. 58. Muscle fibres from middle lobe, near gland. Transverse and some indication of diagonal banding are seen on the thick myofilaments. A hemidesmosome is present at the sarcolemma (K, MS, TEM, scale bar = 200 nm). Fig. 59. Outer lobe, inner surface. The epithelial cells have microvilli at the surface, and there are many vesicles in the apical cytoplasm. They are arranged around crypts filled with secretory material. Many secretory cells containing distinct granules and mucous cells are present. This lobe is contracted, in spite of treatment with magnesium sulphate, so subepithelial muscle fibres are seen in T.S. between the epithelial cells, and in L.S. below this (K, MS, TEM, scale bar = 4 pm). Fig. 60. Outer lobe, inner surface, same region as figure 59, showing small epithelial cells around the crypts. These cells contain many apical vesicles, glycogen and mitochondria (K, MS, TEM, scale bar = 2 pm). Fig. 61. Outer lobe, inner surface, basement membrane of epithelium at upper left. The bodies of goblet shaped secretory cells are sectioned across below the epithelial surface, and are surrounded by subepithelial muscle fibres. In one case, the bodies of a secretory cell containing distinct granules and a mucous cell appear to be closely associated (K, MS, TEM, scale bar = 4 pm). Fig. 62. Outer lobe, outer surface. This epithelium is similar to that of the inner surface, but no goblet shaped secretory cells are present, and there are cell inclusions with dark contents (K, MS, TEM, scale bar = 2 um). 22 AMER. MALAC. BULL. 10(1) (1993) Mee * oe Rd wah Rat 2 ets Fig. 63. Outer lobe, inner surface. TS subepithelial muscle fibres. Thick myofilaments, thin myofilaments and dense bodies are present. The sarcolemmas of two adjacent muscle fibres are closely apposed, forming a junction. K, MS, TEM. Scale bar = 300 nm. Fig 64. Same region as figure 63. As in the adductor muscle, the thick filaments are wider in diameter the farther away they are from the dense bodies and hemidesmosomes. Cross- links can occasionally be seen between the myofilaments. K, MS, TEM. Scale bar = 200 nm. equivalent to the ‘‘basal cells’’ described in other bivalves, which may be limited to a single row of cells connecting the middle and outer lobes (Wada, 1968; Saleuddin, 1974), or can be two to eight cells deep (Bubel, 1973a). Folding in the periostracum and folds or channels in the apical membrane of the basal cells have been described in the pearl oysters Pinctada martensii Dunker and P. fucata (Kawakami and Yasuzumi, 1964; Wada, 1968), and other bivalves (Bevelander and Nakahara, 1967; Neff, 1972a; Bubel, 1973a; Saleuddin, 1974); but none have the extensive array of parallel apical cytoplasmic processes described in the periostracal gland of Crassostrea virginica in the present account. The cells of this gland need to be studied in more detail. There is not as abrupt a transition between the cells of the periostracal gland of Crassostrea virginica and the epithelial cells of the outer surface of the middle lobe as described in other bivalves, and periostracum appears to be secreted in this transitional region. The glandular region is found in all bivalves, but its extent and number of cell types involved is variable (Saleuddin and Petit, 1983). The periostracal gland is restricted to the base of the groove in C. virginica and Ostrea edulis Linnaeus, 1750 (Beedham, 1958), where a thin, hyaline periostracum is formed. The lat- ter is entirely organic and nonmineralised in C. virginica ac- cording to Carriker et al. (1980). In other bivalves, the periostracum starts as a thin layer secreted by the basal cells, termed the ‘‘pellicle’’ by Beverlander and Nakahara (1967), and becomes thicker and altered in composition as it reaches the free edges of the mantle lobes as a result of secretions by one or both epithelial layers lining the periostracal groove (Beedham, 1958; Beverlander and Nakahara, 1967; Neff, 1972a, b; Saleuddin, 1974; Petit et al., 1979; Saleuddin and Petit, 1983). The epithelial cells of the outer surface of the middle lobe of C. virginica contain vesicles and some dense granules. These cells often contain more abundant dense granules in other bivalves (Wada, 1968; Bubel, 1973b; Saleud- din, 1974), but the secretions of these cells do not seem to contribute to the periostracum. In some bivalves the periostracum is closely associated with this epithelium, and it has been suggested that the epithelial cells help to lubricate the periostracum and support it as it moves along the groove (Saleuddin, 1974). Both C. virginica and Pinctada fucata (Wada, 1968) have ciliated cells in this region, although they are not found in other bivalves (Saleuddin, 1974). It is generally accepted that the cells lining the inner surface of the outer lobe of bivalves contribute to the periostracum, but crypts have not been described in other bivalves (Beverlander and Nakahara, 1967; Bubel, 1973b). The cells lining the crypts in Crassostrea virginica are covered with secretory globules, but the periostracum does not ap- pear to become thicker as it passes along the groove. The crypts of C. virginica could result from contraction of the outer lobe, but this would probably result in folds rather than crypts. Goblet shaped secretory cells with bodies extending below the epithelium, like those found in C. virginica, have also been described in Mercenaria mercenaria (Neff, 1972a), and Ostrea edulis (Beedham, 1958). In O. edulis these cells were found not to react with stains for mucin, but to be positive for the mercuric bromophenol blue stain for protein. This could contribute to the periostracum. Relatively high concentrations of alkaline phosphatase were found in the epithelium of the inner surface of the outer lobe of O. edulis by Beedham (1958). The components required for quinone tanning of the periostracal protein have been found in the periostracum of several bivalves (Saleuddin and Petit, 1983). The tentacles of the oyster are very extensible, and are sensitive to light, chemicals, suspended particles and temperature according to Galtsoff (1964). The cells with few cilia and dark cytoplasm found near the tips of the tentacles of the middle and inner lobes appear to be connected with the nerve network beneath the epithelium, and are probably simple single-celled receptors, although Galtsoff states that MORRISON: MANTLE AND MANTLE LOBES OF THE EASTERN OYSTER 23 sensory organs are absent. The similar cells found in the pallial epithelium of the mantle could also be receptors, because these occur over the general body surface of other molluscs (Jones and Saleuddin, 1978). Their distribution and structure in Crassostrea virginica need to be studied in more detail. They did not appear to be on specialised papillae, like those on the tentacles of the giant scallop Placopecten magellanicus (Gmelin) (Moir, 1977); and they seemed to have a variable number of cilia, like those described on the man- tle edge of Helisoma (Planorbella) duryi Wetherby, 1979 (Jones and Saleuddin, 1978). ACKNOWLEDGMENTS Ms. Vivian Marryatt provided expert technical assistance. Ms. Diane Tremblay and Ms. Christine Morrison assisted with preparation of the manuscript and photographs, and Ms. Tremblay prepared figure 26. Dr. Sharon McGladdery provided some of the oyster specimens, and Dr. David Scarratt and Ms. Brenda Bradford maintained and fed the animals. Ms. Joan Kean-Howie and Dr. Peter Beninger reviewed the manuscript, and provid- ed useful suggestions. LITERATURE CITED Bargeton, M. 1942. Les variations saisonniéres du tissu conjonctif vésiculeux de Vhuitre. Bulletin Biologique de la France et de la Belgique. 76:175-191. Beedham, G. E. 1958. Observations on the mantle of the Lamellibranchia. Quarterly Journal of Microscopical Science 99:181-197. Beverlander, G. and H. Nakahara. 1967. An electron microscope study of the formation of the periostracum of Macrocallista maculata. Calcified Tissue Research 1:55-67. Bubel, A. 1973a. An electron-microscope study of periostracum formation in some marine bivalves. I. The origin of the periostracum. Marine Biology 20:213-221. Bubel, A. 1973b. An electron-microscope study of periostracum formation in some marine bivalves. II. The cells lining the periostracal groove. Marine Biology 20:222-234. Carriker, M. R., R. E. Palmer and R. S. Prezant. 1980. Functional ultra- morphology of the dissoconch valves of the oyster Crassostrea virginica. Proceedings of the National Shellfisheries Association 70:139-183. Crenshaw, M. A. 1980. Mechanisms of shell formation and dissolution. Jn: Skeletal Growth of Aquatic Organisms. D. C. Rhoads and R. A. Lutz, eds. pp. 115-132. Plenum Press, New York and London. Dougherty, W. J. 1981. Preparation and staining of semi-thin sections of tissues embedded in water-soluble methacrylate for light microscopy. In: Stain- ing Procedures, fourth ed., G. Clark, ed. pp. 27-38. Williams and Wilkins Co., Baltimore. Franc, A. 1960. Manteau et muscles adducteurs. In: Traité de Zoologie V: P-P. Grasse, ed. pp. 1870-1875. Masson et Cie., Paris. Gabbott, P. A. 1975. Storage cycles in marine bivalve molluscs: a hypothesis concerning the relationship between glycogen metabolism and game- togeneses. In: Proceedings of the 9th European Biological Symposium. H. Barnes, ed., pp. 191-211. Aberdeen University Press. Galtsoff, P. S. 1964. The American Oyster Crassostrea virginica Gmelin. Fishery Bulletin of the Fish and Wildlife Service, Vol. 64. United States Government Printing Office, Washington, D.C. 480 pp. Heyer, C. B., S. B. Kater and U. L. Karlsson. 1973. Neuromuscular systems in molluscs. American Zoologist 13:247-270. Howard, D. W. and C. S. Smith. 1983. Histological techniques for marine bivalve molluscs. National Oceanographic and Atmospheric Ad- ministration Technical Memorandum NMFS/NEC-25. 36 pp. Jones, G. M. and A. S. M. Saleuddin. 1978. Ultrastructural observations on sensory cells and the peripheral nervous system in the mantle edge of Helisoma duryi (Mollusca: Pulmonata). Canadian Journal of Zoology 56:1807-1921. Karnovsky, M. J. 1965. A formaldehyde-glutaraldehyde fixative of high osmolality for use in electron microscopy. Journal of Cell Biology 27:137A-138A. Kawaguti, S. and N. Ikemoto. 1962. Electron microscopy on the mantle of a bivalve, Fabulina nitidula. Biology Journal of Okayama University. 8:21-42. Kawakami, I. K. and G. Yasuzumi. 1964. Electron microscope studies on the mantle of the pearl oyster Pinctada martensii Dunker. Preliminary report. The fine structure of the periostracum fixed with permanganate. Journal of Electron Microscopy 13:\19-123. Lenoir, F., I. Robbins, M. Mathieu, P. Lubet and P. A. Gabbott. 1989. Isola- tion, characterization and glucose matabolism of glycogen cells (=vesicular connective -tissue cells) from the labial palps of the marine mussel Mytilus edulis. Marine Biology 101:495-501. Machin, J. 1977. Role of Integument in Molluscs: In: Transport of Ions and Water in Animals, B. L. Gupta, R. B. Moreton, J. L. Oschman and B. J. Wall, eds. pp. 735-762. McDowell, E. M. 1978. Fixation and processing. In: Diagnostic Electron Microscopy, B. F. Trump and R. T. Jones, eds. Vol. 1, p. 113. John Wiley and Sons, New York, Chichester, Brisbane, Toronto and Songapore. McKenna, O. C. and J. Rosenbluth. 1973. Myoneural and intermuscular junc- tions in a molluscan smooth muscle. Journal of Ultrastructure Research 42:434-450. Moir, A. J. G. 1977. Ultrastructural studies on the ciliated receptors of the long tentacles of the giant scallop, Placopecten magellanicus (Gmelin). Cell and Tissue Research 184:367-380. Morrison, C. M. 1993. The histology and ultrastructure of the adductor muscle of the eastern oyster, Crassostrea virginica (Gmelin). American Malacological Bulletin 10(1):25-38. Morrison, C. M. and P. H. Odense. 1974. Ultrastructure of some pelecypod adductor muscles. Journal of Ultrastructure Research 49:228-251. Neff, J. M. 1972a. Ultrastructural studies of periostracum formation in the hard shelled clam Mercenaria mercenaria (L.). Tissue and Cell 4:311-326. Neff, J. M. 1972b. Ultrastructure of the outer epithelium of the mantle in the clam Mercenaria mercenaria in relation to calcification of the shell. Tissue and Cell 4:591-600. Nicaise, G. and J. Amsellem. 1983. Cytology of muscle and neuromuscular junction. Jn: The Mollusca, A. S. M. Saleuddin and K. M. Wilbur, eds., Volume 4, pp. 1-33. Academic Press, New York and London. Petit, H., W. L. Davis and R. G. Jones. 1978. Morphological studies on the mantle of the fresh-water mussel Amblema (Unionidae): scanning electron microscopy. Tissue and Cell 10:619-627. Petit, H., W. L. Davis and R. G. Jones. 1979. Morphological studies on the periostracum of the fresh-water mussel Amblema (Unionidae): light microscopy, transmission electron microscopy, and scanning electron microscopy. Tissue and Cell 11:633-642. Petit, H., W. L. Davis, R. G. Jones and H. K. Hagler. 1980. Morphological studies on the calcification process in the fresh-water mussel Amblema. Tissue and Cell 12:13-28. Prezant, R. S. 1985. Molluscan mucins: a unifying thread. In: Perspectives in Malacology, R. S. Prezant and C. L. Counts III, eds. American Malacological Bulletin, Special Edition No. 1:35-50. 24 AMER. MALAC. BULL. 10(1) (1993) Prezant, R. S. 1990. Form, function and phylogeny of bivalve mucins. Jn: The Bivalvia-Proceedings of a Memorial Symposium in honour of Sir Charles Maurice Yonge, Edinburgh, 1986, B. Morton, ed. pp. 83-95. Hong Kong University Press, Hong Kong. Reynolds, E. S. 1963. The use of lead citrate at high pH as an electron- Opaque stain in electron microscopy. Journal of Cell Biology 17:208-212. Saleuddin. A. S. M. 1974. An electron microscopic study of the formation and structure of the periostracum in Astarte (Bivalvia). Canadian Journal of Zoology 52:1463-1471. Saleuddin, A. S. M. and H. P. Petit. 1983. The mode of formation and the structure of the periostracum. In: The Mollusca, A. S. M. Saleuddin and K. M. Wilbur, eds., Volume 4, pp. 199-234. Academic Press, New York and London. Simkiss, K. and K. M. Wilbur. 1989. Biomineralization. Academic Press, Inc. Toronto. 337 pp. Stempack, J. G. and R. T. Ward. 1964. An improved staining method for electron microscopy. Journal of Cell Biology 22:697-701. Storch, V. and U. Welsch. 1972. The ultrastructure of epidermal mucous cells in marine invertebrates (Nemertini, Polychaeta, Prosobranchia, Opisthobranchia). Marine Biology 13:167-175. Twarog, B. M., M. M. Dewey and T. Hidaka. 1973. The structure of Mytilus smooth muscle and the electrical constants of the resting muscle. Journal of General Physiology 61:207-221. Wada, K. 1968. Electron microscopic observations of the formation of the periostracum of Pinctada fucata. With English summary, p. 1547. Bulletin of the National Pearl Research Laboratory 13:1540-1560. Wilbur, K. M. 1972. Shell formation in mollusks.. In: Chemical Zoology, M. Florkin and B. T. Scheer, eds., Volume VII, pp. 103-146. Academic Press, New York and London. Wilbur, K. M. 1985. Topics in molluscan mineralization: present status, future directions. American Malacological Bulletin, Special Edition No. 1:51-58. Wilson, D. F. 1969. The basis for double contractions and slow relaxations of non-striated muscles in a pelecypod mantle. Comparative and Biochemical Physiology 29:703-715. Date of manuscript acceptance: 23 October 1992 The histology and ultrastructure of the adductor muscle of the eastern oyster Crassostrea virginica (Gmelin) Carol M. Morrison Department of Fisheries and Oceans, Box 550, Halifax Fisheries Research Laboratory, Halifax, Nova Scotia, B3J 2S7, Canada Abstract. The structure of the translucent and opaque parts of the adductor muscle of the eastern oyster Crassostrea virginica (Gmelin) was studied by light and transmission electron microscopy. The muscle was fixed with the valves held closed and after the muscle had relaxed so that the valves gaped. Several fixatives were used, some with recently developed modifications. The muscle fibres of the translucent and opaque parts of the adductor muscle have a similar organisation, with central myofilaments and peripheral sarcoplasmic reticulum, mitochondria and nucleus. In translucent muscle, the thick myofilaments are thinner than they are in opaque muscle, and the dense bodies are sometimes obliquely oriented. Nerve endings and invaginations of the sarcolemma at hemidesmosomes are shown for the first time in oyster adductor muscle. For the first time, presence of paramyosin is indicated by diagonal as well as transverse periodicities in sections of the thick myofilaments of obliquely striated muscle. These muscles are similar to the anterior byssus retractor muscle of the blue mussel Mytilus edulis Linné, except that no nexal junctions were found between the muscle fibres. Larval oysters develop two adductor muscles, the ‘“‘dimyarian’’ condition, but following attachment the anterior muscle degenerates (Galtsoff, 1964; Elston, 1980), resulting in the ‘‘monomyarian’’ condition. Adult bivalves which bur- row retain two adductors, which are needed for the movements involved in burrowing; but two adductors are not necessary for the sessile oyster. The single adductor is large and more or less centrally placed, facilitating rapid closing in adverse conditions or when clearing the mantle of detritus. It is made up mainly of translucent muscle, which can contract and close the valves quickly. At one side of the adductor muscle is a smaller, crescent-shaped portion of opaque muscle that con- tracts more slowly, but can hold the valves shut against the tension of the hinge ligament for several days (Millman, 1964), a phenomenon known as “‘catch’’. The gross structure (Galtsoff, 1964; Morrison and Odense, 1973), light microscopy and ultrastructure of the myofilaments and isolated paramyosin filaments of the ad- ductor muscle of the eastern oyster Crassostrea virginica (Gmelin) have been described (Philpott er al., 1960; Galtsoff, 1964; Morrison et al., 1970; Cohen et al., 1971; Morrison and Odense, 1974). The most detailed study of the adductor muscle of an oyster is that of the translucent part of the mus- cle of Crassostrea angulata (Lamarck) by Hanson and Lowy (1961). The relationship between thick and thin myofilaments and the structure of isolated thick myofilaments of C. angulata has also been studied (Lowy and Hanson, 1962; Elliott, 1964, 1974 and 1979; Hoyle, 1964; Elliott and Lowy, 1970). Bowden (1958) and Salanki and Zs-Nagy (1966) describe the ap- pearance of the muscle fibres in Ostrea edulis L. using light microscopy only. Bennett and Elliott (1981) and Elliott and Bennett (1982) show the myofilaments of O. edulis in sec- tioned material, but the other cell organelles were poorly fixed. They also show thick myofilaments isolated from the translucent and opaque parts of the adductor of O. edulis, and a striated appearance in tilted cross sections of thick myofilaments from the opaque part. Hanson and Lowy (1960) studied the myofilaments only of an oyster. These studies were focused on the myofilaments because of the interest in the ‘‘catch’’ mechanism. Also, con- sistently good fixation of the other organelles was difficult to achieve. However, recently modifications of fixatives have been developed which seem to preserve tissues more com- pletely. The present study was undertaken with the expecta- tion that the ultrastructure of the myofilaments and the other organelles could be better preserved, to provide a more com- plete description of the oyster adductor muscle. METHODS Fixatives used were 4% glutaraldehyde in seawater (Morrison, 1970), Karnovsky’s fixative (1% glutaraldehyde; 4% formalin prepared from paraformaldehyde, in phosphate buffer; Karnovsky, 1965), IG4F (1% glutaraldehyde; 4% com- mercial formalin in phosphate buffer, McDowell, 1978), 1G4F with seawater replacing some of the distilled water (Howard and Smith, 1983), or 2.5% glutaraldehyde in 0.05 M cac- codylate buffer (David Sims, pers. comm.). Generally, 1G4F with sea-water added gave better results than with phosphate buffer only, so only micrographs of muscle fixed with the American Malacological Bulletin, Vol. 10(1) (1993):25-38 2) 26 AMER. MALAC. BULL. 10(1) (1993) former fixative are shown. All solutions were used at pH 7.2. To obtain fixed preparations of adductor muscles with the valves closed, part of the shell and the tissue around the adductor were removed to permit access of the fixative to the muscle, then the whole oyster was immersed in fixative. To obtain preparations of lengthened adductor muscle, the oyster was placed in 4°C seawater containing 8% MgSO, (Galtsoff, 1964) or 3.75% MgCl, (Hanson and Lowy, 1961), which cause the muscle to relax, until the valves gaped. The time taken for this varied from one to several hours, and the gape varied from 1-6 mm. The gape did not appear to alter when the oyster was touched, but to make sure the valves stayed apart dental wax was inserted between the valves. After removing part of the tissue surrounding the adductor muscle to expose the surface, the oyster was placed in fixative. After 1.5 to 2 hours, part of the exposed surface of the adductor muscle fixed at both lengths was removed and small pieces were placed in fresh fixative. For light microscopy (LM) the specimens were dehydrated in methanol, then embedded in JB4 resin. For transmission electron microscopy (TEM) they were then fixed in 1% osmium tetroxide, dehydrated in acetone then embed- ded in Taab resin. JB4 resin sections were stained with a 1:50 dilution of 1% toluidine blue in 1% sodium borate, methylene blue/basic fuchsin, or Van Gieson stain (Dougherty, 1981). Taab resin semi-thin sections were also stained with toluidine blue for light microscopy. Sections for electron microscopy were stained with 25% uranyl] acetate in methanol (Stempack and Ward, 1964), and lead citrate (Reynolds, 1963), and viewed in an Hitachi transmission electron microscope model MS-9. Paramyosin paracrystals were prepared from the opaque and translucent part of the adductor, using the method described by Philpott et a/. (1960) and Johnson et al. (1959), who worked on the muscles of several bivalves including Crassostrea virginica. RESULTS Better general fixation was obtained than in previous studies of oyster muscle. However, results were variable, often within the same section. Areas of poor fixation were sometimes found near the surface as well as deeper in the tissues. IG4F with seawater was the most dependable fix- ative. There were periodicities in the thick myofilaments, and details of other cell organelles that were not evident in our earlier studies (Morrison, 1970; Morrison and Odense, 1974). TRANSLUCENT MUSCLE The translucent muscle consists of groups of elongate, thin muscle cells or fibres separated by a thin layer of con- nective tissue, the endomysium (Fig. 1). The elongate nuclei are oriented parallel to the long axis of the muscle cell. The muscle fibres are ribbon-like, so in cross section they ap- pear as flattened ovals about 3-4 wm wide and about 17 ym long, which sometimes bifurcate (Fig. 2). The fibres occur in groups, surrounded by a thicker layer of connective tissue, the perimysium. Electron microscopy (Figs. 3, 4) reveals that each mus- cle fibre is composed of densely packed myofilaments and peripheral mitochondria, sarcoplasmic reticulum and a nucleus. The arrangement of thick and thin myofilaments and dense bodies is usually better shown in muscle which was fixed in an extended condition, after relaxation with MgSO, or MgCl, (Fig. 5) than in muscle from oysters fixed with the valves closed. The thin myofilaments are attached to isolated ‘“‘dense bodies’’, which are oriented for short distances in oblique bands. Typically, when viewed with light microscopy no stria- tions are visible (Fig. 1). Oblique striations are occasionally seen, however, in fibres from preparations fixed with the valves closed where the anomalous effect known as ‘‘super- contraction’ has occurred. The thick myofilaments pass in- to the regions at the ends of the sarcomeres where there are normally only dense bodies and thin myofilaments, often becoming folded so that obliquely aligned bands are produced, which are visible using light or electron microscopy (Figs. 6, 7). In a cross section of a sample from an oyster relaxed with magnesium sulphate, fields of thick myofilaments sur- rounded by thin myofilaments, and of dense bodies among thin myofilaments, can be seen (Fig. 8). The thick myofilaments are regularly arranged in approximate hexagonal arrays. At higher magnifications the thick myofilaments are cross-striated with a periodicity of about 5 nm. They also have a diagonal periodicity of about 35 nm and sometimes cross-links to adjacent thin filaments have a similar periodicity (Fig. 9). The transverse periodicity also occurs in isolated paracrystals of paramyosin, with every third bar accentuated (Fig. 10). Measurements of 100 thick myofilaments in cross section revealed a modal peak at about 32 nm (range = 18 to 62 nm). Thin myofilaments are attached to the fusiform dense bodies (Fig. 11). The thick filaments are spindle-shaped, so they appear wider in the centre of a sarcomere and become smaller towards the dense bodies between sarcomeres (Fig. 12). As in mammalian skeletal muscle, where thick and thin filaments overlap the thick myofilaments are surrounded by a ring of thin filaments. This ring is sometimes incomplete around the wider thick myofilaments in the centre of the sarcomere, which would be equivalent to an H zone. Cross links can often be seen between adjacent thick and thin myofilaments and between adjacent thin myofilaments. Some of the dense bodies are attached to the sarcolemma, forming hemidesmosomes (Fig. 13). Filaments from the connective MORRISON: ADDUCTOR MUSCLES OF THE EASTERN OYSTER D4 Fig. 1. Longitudinal section (parallel to the muscle fibres of the adductor extending between the two valves of the oyster) of translucent part of adductor muscle. Specimen was fixed in 2.5% glutaraldehyde in caccodylate buffer, embedded in JB4, and stained with toluidine blue. Long, thin muscle fibres (F) are surrounded by endomysium (E), and groups of muscle fibres are surrounded by perimysium (P) (scale bar = 0.1 mm). Fig. 2. Cross section (across the adductor muscle) of translucent part of adductor muscle. Specimen was fixed in 2.5% glutaraldehyde in caccodylate buffer, embedded in JB4, and stained with Van Gieson. The profiles of the muscle fibres (F) are elongate and sometimes bifurcate; each fibre is surrounded by endomysium (E), and groups of fibres are surrounded by perimysium (P) (scale bar = 30 ym). Fig. 3. Longitudinal section of translucent part of adductor muscle. TEM micrograph of specimen fixed in 1G4F. The muscle fibres have a central core of myofilaments (M) and dense bodies (DB), and hemidesmosomes (HD) are present at the sarcolemma. The peripheral cytoplasm contains elongate mitochondria (MI), sarcoplasmic reticulum (SR) and an elongate nucleus (NU). Between the muscle fibres are glial cells (GC) closely associated with nerve endings (NE) (scale bar = 2 um). 28 AMER. MALAC. BULL. 10(1) (1993) Fig. 4. TEM micrograph of translucent part of adductor muscle of specimen relaxed in Mg SO, and fixed in Karnovsky’s fixative. The same features can be recognised as in figure 3, but the nucleus (NU), mitochondria (ME) and sarcoplasmic reticulum (SR) are more rounded since this is a cross section [(E) endomysium; (GC) glial cell; (M) myofilaments] (scale bar = 2 um) Fig. 5. TEM micrograph of longitudinal section of muscle fibres of translucent part of adductor muscle. Specimen relaxed in Mg SO, and fixed in Karnovsky’s fixative. Thick myofilaments (TKM) are present and thin myofilaments (TNM) are continuous with dense bodies (DB), which are obliquely oriented for short distances. There are profiles of sarcoplasmic reticulum (SR) beneath the sarcolemma (scale bar = | pm). MORRISON: ADDUCTOR MUSCLES OF THE EASTERN OYSTER 29 Fig. 6. Longitudinal section of translucent part of contracted adductor muscle of specimen fixed in 2.5% glutaraldehyde in caccodylate buffer and embedded for electron microscopy, 0.5 wm section stained with toluidine blue. Note oblique contraction bands (CB) in the fibres (F) (scale bar = 20 pm). Fig. 7. TEM micrograph of longitudinal section of contraction bands of translucent part of contracted adductor muscle. Specimen fixed in 2.5% glutaraldehyde in caccodylate buffer. Thick myofilaments (TKM) overlap and become folded in regions where there are dense bodies (DB) (scale bar = | pm). Fig. 8. TEM micrograph of cross section of translucent part of adductor muscle. Specimen relaxed in Mg SO, and fixed in Karnovsky’s fixative. There are fields of thick and thin myofilaments (TKM) and thin myofilaments (TNM) in the muscle fibres, and dense bodies (DB) occur in the fields of thin myofilaments. Hemidesmosomes (HD) present at the sarcolemma. The nucleus (NU) is rounded in cross-section. Vesicles of sarcoplasmic reticulum (SR) as well as mitochondria (MI) occur beneath the sarcolemma. A glial cell (GC) with an accompanying nerve ending (NE) is present between the muscle cells (scale bar = 500 nm). 30 AMER. MALAC. BULL. 10(1) (1993) tissue stroma accumulate next to the hemidesmosome. The sarcolemma is often invaginated where there are hemidesmosomes (Figs. 3, 8). The profiles of sarcoplasmic reticulum are situated just beneath the sarcolemma, and dense material can be seen between the outer membrane of the sarcoplasmic reticulum and the sarcolemma (Fig. 14). The sarcolemmas of adjacent muscle cells are often close to each other (Figs. 3-5, 8), but no well-defined junctions were seen. Many of the mito- chondria in one specimen have filamentous paracrystals and annulated cristae (Fig. 15). Nerve endings, which contain a variety of vesicles, some small and clear, some dense-cored or of varying den- sity, often occur close to the sarcolemma (Figs. 3, 16). Glial cells containing large granules, the gliosomes, are usually closely associated with these nerve endings but do not com- pletely surround them. OPAQUE MUSCLE The muscle fibres of the opaque muscle are more rounded in cross section than those of translucent muscle, so their diameter (10-20 ym) is similar to that of the length but larger than the width of the oval profiles of the translucent muscle fibres (Figs. 17, 18). In the transitional zone between the translucent and opaque muscle the muscle fibres of each type are intermingled (Fig. 19). As in the translucent mus- cle, each muscle fibre is surrounded by endomysium, and the groups of fibres by perimysium. The arrangement of cell organelles is also similar. Central myofilaments are surround- ed by vesicles of sarcoplasmic reticulum and mitochondria, and the sarcolemmas of neighbouring cells are often closely opposed (Figs. 20, 21). The nucleus is situated in cytoplasm to one side of the cell, oriented with its long axis parallel with that of the muscle cell, and dense material can be seen between the sarcolemma and the sarcoplasmic reticulum (Fig. 22). The thick myofilaments are thicker than those of the translucent muscle, having a modal peak at about 61 nm (range = 18 to 123 nm). They exhibit cross striations at 5 nm and often diagonal striations like those of translucent mus- cle, but they are more evident (Figs. 23, 24). Every third cross striation is accentuated. Cross links occur between the thick and thin myofilaments (Figs. 25, 26), and some thin myofilaments have cross links to more than one thick myofila- ment (Fig. 25). As in the thick myofilaments of the translu- cent muscle fibres, the cross links seem to have a similar periodicity to that of the diagonal bands. Sometimes, there is a single ring of thin myofilaments surrounding each thick myofilament in the region of overlap, but more often there are several thin myofilaments between the thick myofilaments. Cross links can sometimes be seen between adjacent thin myofilaments, and occasionally con- nections can also be seen between thick myofilaments (Fig. 26). When the thick myofilaments are cut obliquely, they often appear to be banded (Fig. 27). The thin filaments are attached to dense bodies (Fig. 28) which do not show any special arrangement in the mus- cle fibre (Fig. 29), so this type of muscle is classified as smooth. As in the translucent muscle fibre, some dense bodies are attached to the sarcolemma, forming hemidesmosomes (Fig. 30), and filaments of connective tissue in the endomysium are closely associated with the sarcolemma at these sites. In some places the sarcoplasm surrounding the myofilaments is wide, and sarcolemmal invaginations con- taining filaments of connective tissue extend to the hemidesmosomes (Fig. 31). Glial cells and axons occur be- tween the muscle fibres. Nerve endings, often accompanied by glial cells, are found next to the sarcolemma, sometimes embedded in the sarcoplasm (Fig. 32). They contain a vari- ety of small clear and dense-cored vesicles and larger vesicles containing a varying amount of dense material. DISCUSSION The adductor muscle of oysters, as in most bivalved molluscs, consists of uninucleate muscle cells of small diameter (3-20 wm) that are much longer than those of vertebrate smooth muscle (Twarog, 1976). Each muscle cell contain a single myofibril, and peripheral vesicles of sar- coplasmic reticulum which form couplings with the sarcolem- ma, as described in the adductor muscle of the scallop, Argopecten irradians (Lamarck, 1819) (Nunzi and Franzini- Armstrong, 1981) and the translucent part of the adductor muscle of Crassostrea angulata (Hanson and Lowy, 1961). The invaginations of the sarcolemma at the hemidesmosomes are not as well defined as the transverse tubular system of transversely striated muscle or some unstriated muscle (Dorsett and Roberts, 1980), but may perform a similar transport function. The dense bodies, like the Z-line of striated muscles, anchor the actin filaments. In the translucent part of the ad- ductor of Crassostrea virginica they form oblique bands as in C. angulata, in which the bands are arranged helically around the outer part of the fibre and branch and anastomose in the centre of the muscle fibre (Hanson and Lowy, 1961). These bands were not seen in normal muscle in this study, but can be seen using phase contrast illumination (Hanson and Lowy, 1961). The contraction bands of supercontracted muscle are obvious using regular as well as phase illumina- tion (Bowden, 1958; Hanson and Lowy, 1961; Galtsoff, 1964; Salanki and Zs-Nagy, 1966; Morrison and Odense, 1974). More regular oblique banding occurs in some molluscs such as the octopus and cuttlefish, where the myofibrils surround a central core of mitochondria (Millman, 1967). In this study, the dense bodies, and the actin filaments entering them, were easier to see in specimens fixed in 4% glutaraldehyde in sea MORRISON: ADDUCTOR MUSCLES OF THE EASTERN OYSTER 31 Fig. 9. TEM micrograph of longitudinal section of translucent part of adductor muscle of specimen fixed in 2.5% glutaraldehyde in caccodylate buffer. The thick myofilaments exhibit transverse banding (TBA) at intervals of about 15 nm and diagonal banding (DBA). Cross-links (CL) to the actin filaments can be seen in some areas, at about the same intervals as the diagonal banding (scale bar = 100 nm). Fig. 10. Translucent part of adductor muscle showing paramyosin paracrystal with narrow transverse bands (TBA) about 5 nm apart and accentuated bands (AB) about 15 nm apart (scale bar = 10 nm). Fig. 11. TEM micrograph of longitudinal section of translucent part of adductor muscle of specimen relaxed in Mg SO, and fixed in Karnovsky’s fixative. Cross links (CL) occur between the thick and thin myofilaments, as well as between the thin filaments. The latter enter dense bodies (DB) (scale bar = 200 nm). Fig. 12. TEM micrograph of cross section of translucent part of adductor muscle of specimen relaxed in Mg SO, and fixed in Karnovsky’s fixative. Thick myofilaments (TKM), thin myofilaments (TNM) and dense bodies (DB) are present. Each thick myofilament is surrounded by a ring of thin myofilaments, and cross-links (CL) can often be seen between the two types of myofilament (scale bar = 100 nm). Fig. 13. TEM micrograph of longitudinal section of translucent part of adductor muscle of specimen relaxed in Mg SO, and fixed in Karnovsky’s fixative. There is a hemidesmosome (HD) to one side of the double membrane forming the sarcolemma; on the other side there are connective tissue filaments (scale bar = 200 nm). Fig. 14. TEM micrograph of longitudinal section of translucent part of adductor muscle of specimen fixed in iG4F. A thin layer of dense material is present between a vesicle of sarcoplasmic reticulum (SR) and the sarcolemma (scale bar = 200 nm). 32 AMER. MALAC. BULL. 10(1) (1993) Fig. 15. TEM micrograph of transverse section of translucent part of adductor muscle of specimen relaxed in Mg SO, and fixed in Karnovsky’s fixative. A mitochondrion contains filamentous paracrystals (FP) and annulated cristae (AC) (scale bar = 200 nm). Fig. 16. TEM micrograph of longitudinal section of translucent part of adductor muscle of specimen fixed in IG4F. There is a nerve-ending (NE) containing clear and dense vesicles, and an associated glial cell (GC) near the sarcolemma (scale bar = 1 yan). Fig. 17. Longitudinal section of opaque part of adductor muscle of specimen fixed in 2.5% glutaraldehyde in caccodylate buffer, embedded in JB4 and stained with toluidine blue. The muscle fibres (F) that are wider than those of the translucent muscle, and are separated by endomysium (E) and perimysium (P) (scale bar = 0.1 mm). Fig 18. Cross section of opaque part of adductor muscle of specimen fixed in 2.5% glutaraldehyde in caccodylate buffer, embedded in JB4 and stained with methylene blue/basic fuschin. The muscle fibres (F) are large and have a rounded profile [(E) endomysium; (P) perimysium] (scale bar = 20 ym). MORRISON: ADDUCTOR MUSCLES OF THE EASTERN OYSTER 30 water used in earlier work than in more recently developed fixatives, presumably because less background material in the cell was fixed. Dense bodies also form hemidesmosomes at the sar- colemma in the anterior byssus retractor muscle (ABRM), where it has been suggested that they give structural stabili- ty to the muscle by attaching the muscle cells to the closely associated filaments of connective tissue (Twarog, 1967; Twarog et al., 1973) and propagate tension to neighboring cells via the collagen of the connective tissue (Sobieszek, 1973; Twarog, 1976). Hanson and Lowy (1961) suggested, in their study on the translucent adductor muscle of Crassostrea angulata, that the sarcolemma may be interrupted next to the dense body, but it can be clearly seen that this is not the case in the present study (Fig. 13). In the ABRM nexal or gap junctions about 15 nm across which may be sites of intercellular conductivity have been described between the muscle cells (Twarog et al., 1973), but these were not seen in the present study. However, the sarcolemmas of adjacent cells often run parallel to each other farther apart, but over greater distances than those found in nexal junctions. This has been reported in several other molluscs (Zs.-Nagy and Saldnki, 1970; Nicaise and Amsellem, 1983). The filamentous paracrystals found in the mitochondria of one specimen are similar to those described in the mito- chondria of the myocardial cells of Crassostrea virginica (Hawkins et al., 1980). These workers also found prismatic cristae, and our specimen had annulated cristae. Their specimens, like ours, appeared to be otherwise normal, although in other animals paracrystals are usually associated with disease or states of altered metabolism. Glial cells like those found in both the translucent and opaque parts of the adductor muscle in this study have been described forming a ‘‘glio-interstitial network’’ between tonic muscle cells in a variety of molluscs, including the tonic mus- cle of the adductor of Anodonta, and the ABRM of Mytilus edulis (Amsellem et al. , 1973; Gilloteaux, 1975; Nicaise and Amsellem, 1983). They have not previously been recorded in phasic muscles. These authors suggest that these cells may help to control metabolic exchanges between the interstitial spaces and the nerves and/or muscles. The opaque muscle tissue of the adductor examined in this study is very similar in appearance and physiology to the ABRM of the mussel (Lowy and Hanson, 1962; Millman, 1964; Twarog, 1967; Heumann and Zebe, 1968; and Sobieszek, 1973). However, its thick myofilaments are wider than those of the ABRM, which have a peak width of about 40 nm (Sobieszek, 1973). Unlike the translucent mus- cle, there is no obvious arrangement of dense bodies. Lowy and Hanson (1962) assert that some degree of order is necessary for the sliding-filament mechanism to work effi- ciently, and such an order has been demonstrated in the ABRM using optical diffraction techniques (Sobieszek, 1973). Sobieszek postulates that there are sarcomeres consisting of three to four thick myofilaments which are 25 wm long, thin myofilaments which are 140 wm long and two dense body halves. Such sarcomeres would be difficult to recognise in sectioned material, since they are so long and narrow, so they may be present in the opaque adductor muscle of Crassostrea virginica. Cross-bridges between the thick and thin myofilaments have been shown in other mollusc muscle cells, such as those of the ABRM, in varying degrees of contrac- tion. It has been suggested that direct interaction occurs between the thick myofilaments (Hoyle, 1983), as indicated by the close relationship sometimes seen between them (Fig. 26); but other workers believe that this does not play a part in contraction (Bennett and Elliot, 1989). The thick myofilaments of vertebrate striated muscle are very constant in diameter (about 16 nm) and length (about 1.6 um); whereas the thick myofilaments of most mollusc fibres, including those described in the present study, are wider and vary in width (Levine et al., 1976). This is associated with the presence of varying amounts of paramyosin (Millman, 1967). The proportion of paramyosin in transversely striated muscles such as the fast part of the adductor of Placopecten magellanicus (Gmelin) is low (7% by mass; Chantler, 1991), but is higher in obliquely striated muscle such as that of the translucent part of the adductor of Crassostrea angulata (16-20%; Chantler, 1983) and highest in smooth muscle such as that of the opaque part of the adductor of C. angulata (22-39%; Chantler, 1983). Paramyosin forms the core of the thick myofilament, sur- rounded by myosin (Szent-Gyorgyi et al., 1971) and gives a typical X-ray diffraction pattern originally described by Bear (1944) and Bear and Selby (1956), which subsequently became known as the Bear-Selby net. The transverse striations with a periodicity of about 15 nm, and diagonal striations with a periodicity of about 35 nm described in the thick myofilaments of the translucent and opaque muscle in this study form a ‘‘checkerboard’’ pattern characteristic of the paramyosin core of ‘‘catch’’ muscle (Hanson and Lowy, 1964; Chantler, 1991). The diagonal striations have not been shown previously in a translucent or phasic muscle. These periodicities were first shown by Hall et al. (1945) to cor- respond to the pattern formed by the Bear-Selby net, and result from the tendency of paramyosin molecules to assem- ble with a 14.5 nm intermolecular shift as a result of inter- molecular ionic interactions (Kendrick-Jones et al., 1969: Castellani and Cohen, 1987). The relationship between the patterns seen in paramyosin and myofilament preparations from ‘‘catch’’ muscle using the electron microscope and the Bear-Selby net are considered in detail by Elliott and Lowy (1970), Cohen et al. (1971), and Elliott (1979). The banding patterns seen in thick myofilaments in oblique sections of the opaque muscle in the present study (Fig. 27) have also been 34 AMER. MALAC. BULL. 10(1) (1993) *ehe® é ag Be pete 6, ewig Mee 3. ee By ote a cays? aS? Fig. 19. Transverse section of the transition zone between the opaque and translucent parts of the adductor muscle of specimen fixed in 2.5% glutaraldehyde in caccodylate buffer, embedded in JB4 and stained with methylene blue-basic fuschin. Thick opaque fibres (OF) can be seen to the bottom of the micrograph, and more elongate, thinner translucent ones (TF) can be seen to the top. In the centre fibres of both types are intermingled (scale bar = 20 um). Fig. 20. TEM micrograph of longitudinal section of opaque part of adductor muscle of specimen relaxed in Mg Cl, and fixed in 2.5% glutaraldehyde in caccodylate buffer. A central core of myofilaments (M) and dense bodies (DB) is present in each muscle fibre. There are mitochondria (MI) and vesicles of sarcoplasmic reticulum (SR) next to the sarcolemma. These are not as elongate as in translucent muscle. Endomysium (E) is present between muscle fibres (scale bar = 1 pam). Fig. 21. TEM micrograph of cross section of opaque part of adductor muscle of specimen fixed in Karnovsky’s fixative, showing the dense bodies (D), endomysium (E), myofilaments (M), mitochondria (MI) and sarcoplasmic reticulum (SR) seen in longitudinal section in figure 20 (scale bar = 1 pm). Fig. 22. TEM micrograph of cross section of opaque part of adductor muscle fixed in Karnovsky’s fixative. The nucleus (NU) appears rounded, and is in the cytoplasm to one side of the myofilaments (M) (scale bar = 1 ym). MORRISON: ADDUCTOR MUSCLES OF THE EASTERN OYSTER Se Fig. 23. TEM micrograph of longitudinal section of opaque part of adductor muscle of specimen fixed in 2.5% glutaraidehyde in caccodylate buffer. The thick myofilaments have transverse and diagonal banding (DBA), and sometimes seem to be very close together. Thin filaments (TNM) and glycogen (GL) can also be seen (scale bar = 200 nm). Fig. 24. TEM micrograph of longitudinal section of opaque part of adductor muscle of specimen fixed in 2.5% glutaraldehyde in caccodylate buffer. Transverse banding can be seen, with every third band accentuated (AB). Diagonal banding (DBA) is also present (scale bar = 100 nm). Fig. 25. TEM micrograph of longitudinal section of opaque part of adductor muscle of specimen fixed in Karnovsky’s fixative. There are cross-links (CL) between thick and thin myofilaments, and also between thin myofilaments. CL have a similar periodicity to the diagonal bands (DBA) on the thick myofilaments. One thin myofilament (TNM) has links to two thick myofilaments (scale bar = 100 nm). Fig. 26. TEM micrograph of cross section of opaque part of adductor muscle of specimen relaxed in Mg Cl, and fixed in 2.5% glutaraldehyde in caccodylate buffer. Cross-links (CL) are present between thick and thin myofilaments, and also occasionally between thick myofilaments. A dense body (DB) can be distinguished (scale bar = 100 nm). 36 AMER. MALAC. BULL. 10(1) (1993) Fig. 27. TEM micrograph of cross section of opaque part of adductor muscle of specimen relaxed in Mg Cl, and fixed in2.5% glutaraldehyde in caccodylate buffer. Light and dark bands cross the thick myofilaments (scale bar = 300 nm). Fig. 28. TEM micrograph of longitudinal section of opaque part of adductor muscle fixed in 4% glutaraldehyde in seawater. Thin actin myofilaments enter the dense body (DB) (scale bar = 300 nm). Fig. 29. TEM micrograph of longitudinal section of opaque part of adductor muscle of specimen fixed in 4% glutaraldehyde in seawater. There is no definite pattern in the distribution of the dense bodies (DB) (scale bar = 1 ym). Fig. 30. TEM micrograph of cross section of opaque part of adductor muscle of specimen relaxed in Mg SO, and fixed in Karnovsky’s fixative. The sarcolemma is invaginated next to a hemidesmosome (HD), and connective tissue filaments are concentrated in the adjacent endomysium (scale bar = 100 nm). Fig. 31. TEM micrograph of longitudinal section of opaque part of adductor muscle of specimen fixed in Karnovsky’s fixative. A long invagination of the sarcolemma extends to a hemidesmosome (HD) at the sarcolemma (scale bar = 1 pm). Fig. 32. TEM micrograph of longitudinal section of opaque part of adductor muscle of specimen fixed in 2.5% glutaraldehyde in caccodylate buffer. Nerve endings are embedded in the sarcoplasm and contain small clear (C) and dense-cored vesicles (D) as well as vesicles with varying amounts of granular contents (G), which are usually larger than the other types. An invagination (I) of the sarcolemma (S) extends to the myofibril (scale bar = 500 nm). MORRISON: ADDUCTOR MUSCLES OF THE EASTERN OYSTER 3 shown in Ostrea edulis, C. angulata, C. gigas Thunberg, 1793, Mercenaria mercenaria (Linné, 1758), Pecten maximus (Linné, 1758) and the ABRM of Mytilus edulis when cross sectioned material was tilted (Bennett and Elliott, 1981; Elliott and Bennett, 1982), and these authors show that these pat- terns are caused by viewing the myofilaments down planes of the Bear-Selby net. The thin myofilaments of the sea scallop adductor mus- cle have been shown to be very similar to those in vertebrates (Chantler, 1991). The thin myofilaments of the oyster have a similar appearance and diameter, suggesting that they also have a similar composition. In the translucent adductor mus- cle, there appear to be about 12 thin myofilaments around each thick myofilament, as reported in Crassostrea angulata (Hanson and Lowy, 1961). The muscle fibres of the opaque part of the adductor of Placopecten magellanicus (Cohen et al., 1971) sometimes have obliquely aligned dense bodies, and the thick myofilaments have a similar peak diameter, 36 nm (Morrison and Odense, 1974) as those of the obliquely striated, translu- cent muscle of the oyster. They contain paramyosin and ex- hibit a small degree of ‘‘catch’’ (Hoyle, 1964). The oblique- ly striated muscle of the oyster also has some ability to main- tain ‘‘catch’’ (Hanson and Lowy, 1961), and its X-ray pattern shows the Bear-Selby net typical of paramyosin (Elliott and Bennett, 1982). The presence of paramyosin is also demonstrated by the transverse and diagonal periodicities found in sectioned material in the present study. These two types of muscle are therefore very similar, although one forms the opaque, slow part and the other the fast, translucent part of an adductor muscle. Smooth and obliquely striated muscles in the adduc- tors of bivalves have the same arrangement of organelles, and in this study their thick myofilaments show the same band- ing patterns. There appears to be a spectrum of muscle cell types in bivalve adductor muscles, with slightly varying ultrastructural characteristics, the latter producing a graded series of physiological characteristics. ACKNOWLEDGMENTS I would like to acknowledge the expert technical assistance of Vivian Marryatt. Dr. Sharon McGladdery of D.F.O. in Moncton obtained specimens, and Dr. David Scarratt of the D.FO. Halifax Laboratory maintained and fed the animals. Mr. Ken Freeman, Dr. Ellen Kenchington and Dr. David Scar- ratt kindly reviewed the paper and provided helpful comments. Also, I ex- press my thanks to the two anonymous reviewers who made very thorough and helpful comments on the manuscript. LITERATURE CITED Amsellem, J. G. Nicaise and G. Baux. 1973. Distribution du systéme glio- interstitiel chez les Mollusques: Les parties tonique et phasique du muscle adducteur postérieur chez Anodonta. Experientia 29:1274-1276. Bear, R. S. 1944. X-ray diffraction on protein fibres. Feather rachis, por- cupine quill tip and clam muscle, Vol. 2. Journal of the American Chemical Society 66:2043-2050. Bear, R. S. and C. C. Selby. 1956. The structure of paramosin fibrils accord- ing to X-ray diffraction. Journal of Biophysics Biochemistry Cytology 2:55-70. Bennett, P. M. and A. Elliott. 1981. The structure of the paramyosin core in molluscan thick filaments. Journal of Muscle Research and Cell Motility 2:65-81. Bennett, P. M. and A. Elliott. 1989. The ‘catch’ mechanism in molluscan muscle: an electron microscopy study of freeze-substituted anterior byssus retractor muscle of Mytilus edulis. Journal of Muscle Research and Cell Motility 10:297-311. Bowden, J. 1958. The structure and innervation of lamellibranch muscle. International Review of Cytology 7:295-335. Castellani, L. and C. Cohen. 1987. Rod phosphorylation favors folding in a catch muscle myosin. Proceedings of the National Academy of Sciences 84:4058-4062. Chantler, P. D. 1983. Biochemical and structural aspects of molluscan mus- cle. In: The Mollusca, A.S.M. Saleuddin and K. M. Wilbur, eds. pp. 78-154. Academic Press, San Francisco. Chantler, P. D. 1991. The structure and function of scallop adductor muscles. In: Scallops: Biology, Ecology and Aquaculture, S. E. Shumway, ed. pp. 225-304. Elsevier, Amsterdam. Cohen, C., A. G. Szent-Gyérgyi and J. Kendrick-Jones. 1971. Paramyosin and the filaments of molluscan ‘‘catch’’ muscles. Journal of Molecular Biology 56:223-237. Dorsett, D. A. and J. B. Roberts. 1980. A transverse tubular system and neuromuscular junctions in a molluscan unstriated muscle. Cell and Tissue Research 206:251-260. Dougherty, W. J. 1981. Preparation and staining of semi-thin sections of tissues embedded in water-soluble methacrylate for light microscopy. Jn: Stain- ing Procedures, G. Clark, ed. fourth ed., pp. 27-38. Williams and Wilkins Co., Baltimore. Elliott, A. 1974. The arrangement of myosin on the surface of paramyosin filaments in the white adductor muscle of Crassostrea angulata. Pro- ceedings of the Royal Society of London Series B, 186:53-66. Elliott, A. 1979. Structure of molluscan thick filaments: A common origin for diverse appearances. Journal of Molecular Biology 132:323-341. Elliott, A. and P. M. Bennett. 1982. Structure of the thick filaments in molluscan adductor muscle. In: Basic Biology of Muscles: A Com- parative Approach, B. M. Twarog, R. J. C. Levine and M. M. Dewey, eds. pp. 1I-28. Raven Press, New York. Elliott, A. and J. Lowy. 1970. A model for the coarse structure of paramyosin filaments. Journal of Molecular Biology 53:181-203. Elliott, G. F. 1964. Electron microscope studies of the structure of the filaments in the opaque adductor muscle of the oyster Crassostrea angulata. Journal of Molecular Biology 10:89-104. Elston, R. 1980. Functional anatomy, histology and ultrastructure of the soft tissues of the larval American oyster, Crassostrea vriginica. Pro- ceedings of the National Shellfisheries Association 70:65-93. Galtsoff, P. S. 1964. The American Oyster Crassostrea virginica Gmelin. Fishery Bulletin of the Fish and Wildlife Service Vol. 64. 480 pp. United States Government Printing Office, Washington. Gilloteaux, J. 1975. Innervation of the anterior Byssal Retractor Muscle in Mytilus edulis L. Cell and Tissue Research 161:511-519. Hall, C. E., M. A. Jakus and F. O. Schmitt. 1945. The structure of certain muscle fibrils as revealed by the use of electron stains. Journal of Applied Psychology 16:459-465. Hanson, H. and J. Lowy. 1960. Structure and function of the contractile ap- paratus in the muscles of invertebrate animals. In: The Structure and Function of Muscle, G. H. Bourne, ed. Vol. 1, pp. 265-335. Academic Press, New York and London. 38 AMER. MALAC. Hanson, J. and J. Lowy. 1961. The structure of the muscle fibres in the translu- cent part of the adductor of the oyster, Crassostrea angulata. Pro- ceedings of the Royal Society B 154:173-196. Hanson, J. and J. Lowy. 1964. The structure of molluscan tonic muscles. In: Biochemistry of Muscle Contraction. J. Gergely, ed. Vol. 2. pp. 400-409. Little, Brown and Company, Boston. Hawkins, W. E., H. D. Howse and C. A. Foster. 1980. Prismatic cristae and paracrystalline inclusions in mitochondria of myocardial cells of the oyster Crassostrea virginica Gmelin. Cell and Tissue Research 209: 87-94. Heumann, H. -G. and E. Zebe. 1968. Uber die funktionsweise glatter muskelfasern. Zeitschrift fiir Zellforschung und mikroskopiche Anatomie 85:534-551. Howard, D. W. and C. S. Smith. 1983. Histological techniques for marine bivalve molluscs. National Oceanographic and Atmospheric Ad- ministration (NOAA) Technical Memorandum NMFS/NEC-25. p. 36. Hoyle, G. 1964. Muscle and neuromuscular physiology. In: Physiology of the Mollusca, K. M. Wilbur and C. M. Yonge, eds. Ch. 10, pp. 313-345. Academic Press, New York and London. Hoyle, G. 1983. Muscles and Their Neural Control. John Wiley and Sons, New York. 689 pp. Johnson, W. H., J. S. Kahn and A. G. Szent-Gyorgyi. 1959. Paramyosin and contraction of ‘‘catch-muscles’’. Science 130:160-161. Karnovsky, M. J. 1965. A formaldehyde-glutaraldehyde fixative of high osmolality for use in electron microscopy. Journal of Cell Biology 27:137A-138A. Kendrick-Jones, J., C. Cohen, A. G. Szent-Gy6rgyi and W. Longley. 1969. Paramyosin: Molecular length and assembly. Science 163:1196-1198. Levine, R. J. C., M. Elfvin, M. M. Dewey and B. Walcott. 1976. Paramyosin in invertebrate muscles. II. Content in relation to structure and func- tion. Journal of Cell Biology 71:273-279. Lowy, J. and J. Hanson. 1962. Ultrastructure of invertebrate smooth muscles. Physiological Reviews 42:34-42. McDowell, E. M. 1978. Fixation and processing. In: Diagnostic Electron Microscopy, B. F. Trump and R. T. Jones, eds. Vol. 1. p. 13. John Wiley and Sons, New York, Chichester, Brisbane, Toronto and Singapore. Millman, B. M. 1964. Contraction in the opaque part of the adductor mus- cle of the oyster (Crassostrea angulata). Journal of Physiology 273:238-262. Millman, B. M. 1967. Mechanism of contraction in molluscan muscle. American Zoologist 7:583-591. Morrison, C. M. 1970. A comparison of the adductor muscles of several pelecypods (Phylum Mollusca). Doctoral Dissertation. Dalhousie University, Halifax, Nova Scotia, Canada. 82 pp. + Ill figs. BULL. 10(1) (1993) Morrison, C. M. and P. H. Odense. 1973. Gross structure of the adductor muscle of some pelecypods. Journal of the Fisheries Research Board of Canada 30:1583-1585. Morrison, C. M. and P. H. Odense. 1974. Ultrastructure of some pelecypod adductor muscles. Journal of Ultrastructure Research 49:228-251. Morrison, C. M., M. L. Cameron and P. H. Odense. 1970. Periodicities in the thick filaments of the opaque and translucent parts of the ad- ductor of the oyster, Crassostrea virginica. Canadian Journal of Zoology 48:608-609. Nicaise, G. and J. Amsellem. 1983. Cytology of muscle and neuromuscular junction. In: The Mollusca, A. 8. M. Saleuddin and K. M. Wilbur, eds. pp. 1-33. Volume 4. Academic Press, New York. Nunzi, M. G. and C. Franzini-Armstrong. 1981. The structure of smooth and striated portions of the adductor muscle of the valves in a scallop. Journal of Ultrastructure Research 76:134-148. Philpott, D. E., M. Kahlbrock and A. G. Szent-Gy6rgyi. 1960. Filamen- tous organization of molluscan muscles. Journal of Ultrastructure Research 3:254-269. Reynolds, E. S. 1963. The use of lead citrate at high pH as an electron- opaque stain in electron microscopy. Journal of Cell Biology 17:208-212. Salanki, J. and I. Zs.-Nagy. 1966. Comparative histological investigations on marine lamellibranch adductors in different functional conditions. Acta Biologica Academiae Scientiarum Hungaricae 17:315-328. Sobieszek, A. 1973. The fine structure of the contractile apparatus of the anterior byssus retractor muscle of Mytilus edulis. Journal of Ultrastructure Research 43:313-343. Stempack, J. G. and R. T. Ward. 1964. An improved staining method for electron microscopy. Journal of Cell Biology 22:697-701. Szent-Gyorgyi, A. G., C. Cohen and J. Kendrick-Jones. 1971. Paramyosin and the filaments of molluscan ‘‘catch’’ muscles. II. Native filaments: isolation and characterization. Journal of Molecular Biology 56:239-258. Twarog, B. M. 1967. The regulation of catch in molluscan muscle. Journal of General Physiology 50:157-169. Twarog, B. M. 1976. Aspects of smooth muscle function in molluscan catch muscle. Current Topics in Physiology 56:829-838. Twarog, B. M., M. M. Dewey and T. Hidaka. 1973. The structure of Mytilus smooth muscle and the electrical constants of the resting muscle. Journal of General Physiology 61:207-221. Zs.-Nagy, I. and J. Salanki. 1970. The fine structure of neuromuscular and intermuscular connections in the adductors of Anodonta cygnea L. (Mollusca, Pelecypoda). Annales instituti biologici (Tihany) Hungaricae academiae Scientiarum 37:131-143. Date of manuscript acceptance: 10 June 1992 The Asiatic clam Corbicula fluminea (Miiller, 1774) (Bivalvia: Corbiculidae) in Europe R. Araujo, D. Moreno and M. A. Ramos Museo Nacional de Ciencias Naturales (CSIC), Jos¢ Gutierrez Abascal, 2. 28006, Madrid, Spain Abstract. Two populations of Corbicula fluminea were found in the Iberian Peninsula; one in Spain and the other in Portugal. A detailed description in terms of ecology, shell morphology and microstructure, morphometrics and anatomy is given for the Spanish population from the Mino River. Lectotypes for Tellina fluminea and T. fluminalis, and a neotype for T: fluviatilis are designated and illustrated. Distribution and spread of C. fluminea in Europe are revised. Comparisons among some European populations and the populations from Canton, China, and the Mino River are made. Results suggest that, except for one doubtful population, all records of Corbicula in Europe are attributable to C. fluminea. Corbicula taxonomy begins in 1774 with Miiller who described three species in the genus Te/lina Linne’, 1758: T. fluminalis ‘‘in fluvio Asiae Euphrat’’; 7? fluminea ‘*in arena fluviali Chinae’’; 7. fluviatilis ‘‘in flumine emporium Can- ton Chinae praeterlabente’’. Since then, many living species of Corbicula Mihlfeldt, 1811, have been described in freshwater and estuarine habitats from Southeast Asia, the Indian subcontinent, the Pacific islands, and the easternmost part of Europe and Africa (McMahon, 1983). The fossil record of Corbicula includes Europe, North America and Japan (see Linstow, 1922; Zhadin, 1952; Ellis, 1978; and Britton and Morton, 1979 for a review). The first published record of Corbicula in North America is that of Burch (1944) in 1938. Counts (1981, 1985) cites the presence of the species in 1924 and 1937 in Nanaimo, Vancouver Island, British Columbia, Canada, and in Ray- mond, Pacific County, Washington, respectively. Since then, it spread widely in most lotic and lentic habitats, being a pest with very important economic and ecological effects (Sinclair and Isom, 1963; McMahon, 1983). Many papers have been published with records of new localities and biological data using different species names, mainly C. fluminea, C. manilensis and C. leana (McMahon, 1983 and references therein). Several hypotheses about the importance of the role of human activities in the spread of Corbicula have been treated (e.g. Thompson and Sparks, 1977; McMahon, 1982). The wide geographical and ecological range of Cor- bicula seems to be related to the great variation in shell form and colour. These two features are the most common tax- onomic characters and the only ones used by the early conchologists, suggesting that Corbicula taxonomy probably involves more species names than needed. Thus, Talavera and Faustino (1933) (/n: Britton and Morton, 1979) placed Corbicula manilensis (Philippi, 1844) into synonymy with C. fluminea, Morton (1977) considered C. leana to be a junior synonym of C. fluminea, while C. fluviatilis was previously placed into synonymy with C. fluminea by Prashad (1929). Moreover, a thorough review by Britton and Morton (1979) lead the authors to consider that most Asiatic species previously described could be attributed to two taxa: the freshwater species C. fluminea (Miller, 1774) and the estuarine species C. fluminalis (Miiller, 1774). Studying North American populations of Corbicula on the basis of ecology, functional morphology and reproduc- tive biology, Britton and Morton (1979) concluded that all belonged to the single species, C. fluminea. The results of this paper, and the conclusions of Morton (1982), seem to provide a good discrimination between C. fluminea and C. fluminalis. In the last decade, Corbicula was also introduced into South America (Ituarte, 1981) and Europe. Mouthon (1981) reports the presence of C. fluminalis in France (La Dordogne) and in Portugal (Tajo River estuary). Nagel (1989) cites the species from the Duero River near Oporto (Portugal) and Girardi (1989-1990) indicates the occurrence of C. fluminalis also in France at the Canal du Midi at Grisolles (Tarn and Garonne). We found two Corbicula populations in the Iberian Peninsula, one in Spain and the other in Portugal, apparent- ly corresponding to C. fluminea. These facts suggest that, as occurred in North America, Europe is being currently invaded by this bivalve and that species discrimination is probably not as clear as previously thought (Morton, 1982), because it still seems to allow the use of various species names for morphological variants of the same species concept. American Malacological Bulletin, Vol. 10(1) (1993):39-49 39 40 AMER. MALAC. BULL. 10(1) (1993) Thus, the Iberian populations are described in detail in relation to shell and anatomical characters in order to clarify the taxonomy and distribution of Corbicula. The results are compared with bibliographical data and museums material, including the Miiller collection. MATERIALS AND METHODS The Mino River was sampled from its source to its mouth. The first samplings were carried out in July and October 1989. From June 1990 to June 1991 a monthly sampl- ing was done at Goian, near Vigo in Galicia, Spain, 13 kms from the sea (Fig. 1) (UTM coordinates 29TNG208436). The specimens of the Duero River were collected at Regua (Portugal) in the shore by the Karaman-Chappuis method (Motas, 1962). In the Mino River, the clams were collected by snorkeling, dredging the bottom and sieving the mud along the shore. Temperature, dissolved oxygen, pH and water turbidity were monitored at Goian (Mino River) using an Horiba water checker Model U-7. Conductivity was measured using a Crison conductivimeter Model 523. Values of alkalinity, calcium, total water hardness and carbonate hardness were obtained in situ with Merck Aquamerck. No physicochemical data are available from the Duero River. Some specimens were maintained alive in aquaria at the laboratory for anatomical studies. Prior to the dissec- tions, animals were relaxed with menthol or 1% sodium- pentobarbital (nembutal), extracted from the valves and stained with neutral red. Fig. 1. Map of the distribution of Corbicula fluminea in the Mino River. Shell pores and other microstructural features were observed in the Mino River population with a Jeol JSM T330A scanning electron microscope at accelerating voltages of 15 and 20 kV. Juvenile and adult valves were submerged in 5% sodium hypochlorite (Clorox) for more than 24 hours, cleaned in distilled water with ultrasound, dried at 50°C and then coated (whole or fractured) with a thin layer of gold- palladium in a Balzer SCD 004 Sputter Coating Unit. Shell phenotypes were described. Then, the specimens were compared with descriptions in the literature and specimens in the collections. In addition to the typical measures (length, width and height), and to describe the variability of Corbicula shells, seven measurements shown in figure 2C plus the shell perimeter were determined using a caliper read to the nearest 0.01 mm and camera lucida draw- ings of each left valve. The total shell length (L) was used as a standard size measure for statistical adjustment of the measured variables. The angle (A) inscribed between the in- ner inflection point of the hinge and the ends of the lateral teeth, and the one between the lines d and d’ in figure 2C (Add ’) were also measured on the camera lucida drawings, and transformed in radians for statistical analysis. We also registered the number of sulcations per 5 mm medially measured along the dorsoventral axis and the number of den- ticles per millimeter in the anterior lateral tooth, both in the left valve. A total of 58 European specimens of Corbicula were measured: 45 from the Mino River at Goian (Spain), nine from the Duero River at Regua (Portugal) and four from the Duero (Portugal) (Karl Otto-Nagel, coll.). The mean, standard error, standard deviation and the coefficients of variation from the Mino River population were calculated to evaluate the least variable characters and therefore the most suitable for taxonomic purposes. The Miiller collection material listed in the following section was also measured for comparisons. These were made among Mino, Duero (Regua), Duero (Nagel coll.) and Canton populations. Corbicula lacks well-defined stages of growth and adult size is highly variable. In addition, sampling biases due to the use of previously collected material from archival col- lections could cause artificial heterogeneity across samples (Reist, 1985). Thus for morphometrics to be useful for tax- onomy they must compensate for allometric relationships and be able to estimate differences in shape by removing confounding effects of absolute shell size. As size and shape covary, an analysis of covariance, with L as the covariate (Packard and Boardman, 1987), was performed on the orig- inal data and the among-groups residuals used to de- scribe shape and in subsequent statistical procedures (Reist, 1985, 1986). The residuals of each measurement were then examined by an analysis of variance and Duncan multiple range test to determine significantly different pairs of means. ARAUJO ET AL.: CORBICULA FLUMINEA IN EUROPE 4] AAM LP 16 OG D6 6 H K GP EA Fig. 2. Spanish Corbicula fluminea: A, siphons; B, anatomy (AAM, anterior adductor muscle; LP, labial palps; IG, inner gill; OG, outer gill; DG, digestive gland; G, gonad; H, heart; K, kidney; GP, gonopore; EA, excretory aperture; PRM, posterior pedal retractor muscle; R, rectum; PAM, posterior adductor muscle, M, mantle). C, left valve showing standard measurements (L, length; L 4, length at “4 of height; L *4, the same at 74; Li, length between the ends of the lateral teeth; H, height of the shell; Hi, distance between the upper border of the hinge and the lower end of the shell; W, width of the complete shell; Wh, width of the hinge; P, perimeter of the valve; d, distance between the angle down the anterior cardinal tooth and the lower end of the valve, perpendicular to Li; d’, bisector of the mentioned angle; A, angle; Add, angle between d and d’. TYPE MATERIAL Concerning the three species of Corbicula described by Miiller (1774), it was impossible for us to find any pro- perly designated type material. Counts (1991) says that the holotypes of Corbicula fluminea, C. fluminalis and C. fluviatilis are in the Universitetets Zoologisk Museum of Copenhagen (UZMC), and Morton (1977) and Britton and Morton (1979) illustrate three of these specimens as the types of the three species. However, no data about the origin of the supposed type specimens of C. fluminea and C. fluviatilis are available at the Zoologiske Museum (T. Schigtte, pers. comm.). On the other hand, it was impossible to obtain specimen #152926 of C. fluminea in the Museum of Com- parative Zoology at Harvard cited as the paratype by Counts (1991) and quoted by Johnson (1959) as cotype. Therefore, in this paper we designate lectotypes for Tellina fluminea and T. fluminalis and a neotype for T. fluviatilis as follows. Tellina fluminea. There is no type designated properly. We have studied the following material from the UZMC: two whole shells and two valves from an unknown locality in Miiller’s collection; three whole shells and two valves from Canton, China, unknown collection; 15 whole shells from Canton, China in Spengler’s collection; one whole shell from Canton, China. Probably in Miuller’s collection (T. Schigtte, pers. comm.). This specimen was quoted and figured without label data by Kennard and Woodward (1926) as the ‘‘cotype’’ (meaning syntype). It is designed here as the lectotype; one 42 AMER. MALAC. BULL. 10(1) (1993) whole shell from East India or China in Miiller’s collection. It was quoted and figured by Prashad (1929) who stated that it could be accepted as the “‘cotype’’ of the species. This specimen was figured by Morton (1977) and Britton and Mor- ton (1979) and taken to be the type. It is not recognized here as a syntype because of the doubtful locality and because its measurements do not fit the original description. Lectotype (Fig. 3): The specimen (14.9 x 13.7 x 10.2 mm) from Can- ton, China, and probably in Miiller’s collection (UZMC) agrees with the original description and measurements (Miiller, 1774). Tellina fluminalis. The mention of a ‘‘cotype’’ from Miiller’s collection by Kennard and Woodward (1926) is not a valid lectotype designation. We consider that the mention of ‘‘type’’ by Britton and Morton (1979: Fig. 3) is a valid designation of lectotype according to the Article 74 a of the ICZN. Lectotype: (Fig. 3). One specimen (29.9 x 30 x 21.8 mm) from the Euphrates River, Mesopotamy, in Miiller’s col- lection (UZMC) agrees with the original description, measurements and locality (Muller, 1774). Tellina fluviatilis. Kennard and Woodward’s (1926) plesiotypes have no nomenclatural value, not coming from Miiller’s specimens. Prashad (1929) recorded one specimen as the topotype. This is not a valid designation. The specimen figured by Morton (1977) and Britton and Morton (1979) is from Spengler’s collection. The sole specimen of this species in Miiller’s collection is from Tranquebar, India, therefore we designate here the specimen from Canton, China, figured by Morton (1977) and Britton and Morton (1979) as the neotype of the species. Neotype: One specimen (19.6 x 17 x 10.6 mm) from Canton, China, in Spengler’s col- lection (UZMC) agrees with the original description and locality. The specimen is equivalve, near equilateral and in- tegropaleate, with an external sculpture of concentric sulca- tions. There are eight sulcations in 5 mm of the medial region of the left valve. The colour of the periostracum is dark brown with a pale medial concentric band. The umbones are erod- ed and white. Internally it is violet, darker near the edge and in the area of the siphons. Lateral teeth are white with eight denticles per millimeter measured in the medial region of the anterior lateral tooth of the left valve (Fig. 3). The left valve shows the manuscript letters ‘*‘Sp’’. RESULTS ECOLOGY No Corbicula were observed in the samples collected in July 1989. Nine juvenile specimens were found in October 1989, and in June 1990 a large number of adults and juveniles were found. Several samplings since then show that the number of Corbicula and the area occupied are increasing. In January 1991 the species reached from 8 to 24 km upstream (Fig. 1). The population found at Regua (Portugal) in April 1989 indicates that the species also invaded the Duero River where it lives buried into the gravel in the margins of the river. In the Mino River, Corbicula lives in a section of river 400 m wide that is under tidal influences. Water temperature ranges from 9.2°C in January to 27°C in July. Accordingly, registered values of pH are from 5.9 to 8.2. Values of conductivity at 25°C are low as expected from freshwater habitats in granitic areas. Minimum value is 54 us which correspond to January. It increases during the sum- mer reaching to 450 ys in September. These values are clearly associated with those of total hardness (1 - 3.8 °dH) for the same months. Less variation was found for carbonate hard- ness (1 - 1.8 °dH). Calcium varies from 7 to 18 mg/l and alkalinity from 0.4 to 0.8 mmol/l. Higher values of both parameters were also registered in summer. These yearly variations seem to be related with changes in the marine in- fluence and the water river contribution. The Mino River has well oxygenated waters as shown by the observed values of dissolved oxygen, between 7.8 and 12.4 ppm. Specimens in the shore were sampled at a depth up to 8 cm, which means that at low tide Corbicula lives some hours out of the water. Many more specimens were observed living in the middle of the river (about 6 m depth) in gravel and sandy substrata, than in the sand and mud of the shore. They are also frequent in the mud retained by the vegetation. Flora of the area includes many Elodea canadensis Michx., an invasive species, Potamogeton sp., Ranunculus sp. and Ceratophyllum sp. Other molluscs living in the area are: Musculium lacustre (Miiller, 1774); Pisidium amnicum (Miller, 1774); P. henslowanum (Sheppard, 1823); Potomida littoralis (Cuvier, 1797); Unio pictorum (Linné, 1758); Anodonta cygnea (Linné, 1758); Gyraulus sp.; Bithynia tentaculata (Linné, 1758); Potamopyrgus jenkinsi (Smith, 1889); Valvata piscinalis (Miiller, 1774); Lymnaea sp.; Hippeutis com- planatus (Linné, 1758); Physa acuta Draparnaud, 1805 and Ancylus sp. The abundance of Bithynia tentaculata was spec- tacular. It is also important to mention that Pisidium amnicum was one of the dominant molluscan species in the samples prior to September 1990. In this month only a few specimens were found and since then the population has become extinct. This was coincident with the appearance of a pest of the Cyanophyta Microcystis aeruginosa (Kutz). However, in spite of the fact that the algal bloom receded, in the following month no living specimens of P. amnicum could be found. In the most downstream locality for Corbicula the only ac- companying species was P. jenkinsi whose abundance is considerable. SHELL MORPHOLOGY AND MICROSTRUCTURE The shell of European Corbicula is equivalve, near ARAUJO ET AL.: CORBICULA FLUMINEA IN EUROPE 43 Fig. 3. A. Lectotype of Corbicula fluminalis; Euphrates River, Mesopotamy; U.Z.M.C. (29.9 x 30 x 21.8 mm). B. Neotype of C. fluviatilis; Canton, China; U.Z.M.C. (19.6 x 17 x 10.6 mm). C. Lectotype of C. fluminea; Canton, China; U.Z.M.C. (14.9 x 13.7 x 10.2 mm). equilateral, integropaleate, thick and heavy. The shape is oval in juveniles and tends to be near triangular in adult specimens. The surface has well marked concentric and regular sulca- tions. The number of sulcations in 5 mm of shell appears to be independent of age and varies between four and eight, specimens with five or six sulcations being most frequent (Table 1, Fig. 4A). The hinge is typically heterodont with three cardinal teeth on each valve, and two crenulate lateral teeth, simple in the left valve and double in the right one. In the Mino River population the number of denticles per mm of cardinal tooth is also independent of age. It varies from five to eight, six denticles specimens being most frequent (Table, 1, Fig. 4B). The ligament is exterior and prominent (Fig. 5). The periostracum is yellow-brown in all the specimens. The inner surface of the shell is white with a more or less violet dye (Fig. 5). In the Mino River juveniles, it is frequent to see three strongly pigmented violet areas, one central and triangular and two lateral, just below the lateral teeth, that are also ex- AMER. MALAC. BULL. 10(1) (1993) Table 1. Descriptive statistics of the Mino River population (Note: Var., variable acronym as in Fig. 2, except N°S/Smm=number of sulcations in 5 mm, and N°2D/mm=number of denticles per millimeter of anterior lateral tooth; SD, standard deviation; SE, standard error; Min., minimum; Max., maximum; CV, coefficient of variation; N=45). Var. Mean SD SE L 15.82 2.66 0.39 L4% 14.11 2.15 0.32 L% 15.26 2.63 0.39 Li 12.97 2.12 0.32 H 13.35 2.29 0.34 Hi 12.12 2.00 0.29 Ww 9.49 1.49 0.22 Wh 0.85 0.19 0.03 P 45.94 7.61 1.13 d 11.08 1.83 0.27 d’ 11.23 1.83 0.27 N°S/S5Smm 5.44 0.96 0.14 N°D/mm 6.15 0.71 0.10 A 2.21 0.07 0.01 Add 0.27 0.04 0.006 N° S/Smm N° D/mm 8 10 12 14 16 18 20 22 Fig. 4. Corbicula fluminea from the Mino River. Variation with length in the A. Number of sulcations per 5 mm, and B. Number of denticles per mm of the anterior lateral tooth. Min. Max CV 9.7 20.0 16.81 8.9 16.9 15.25 9.3 19.5 17.21 8.5 16.2 16.34 8.4 17.2 17.16 7.6 15.4 16.53 6.3 11.9 15.72 0.4 1.4 22.91 28.5 58.0 16.56 7 14.3 16.48 7.1 14.3 16.27 4 8 17.75 5 8 11.46 2.02 2.48 3.31 0.16 0.34 16.81 ternally visible. The central one usually spreads and stumps in adults, with variable pigment intensity among specimens. The lateral areas surround two clear yellowish zones at both sides of the umbo, corresponding to the lunula and the scutheon which are more marked in dark specimens (Fig. 5). In the Duero River population, the violet color is more faint, the umbones are eroded and the dark lateral bands are nearly absent (Fig. 5). In scanning micrographs of juvenile shells from the Mino River (2 - 3 mm in length) treated with sodium hypoclorite, the prodissoconch (Fig. 6A) is smooth with irregular granules in a random pattern (Fig. 6B) lacking pores and striation. Sulcations in the early dissoconch are crossed by thin radial ribs in a more or less regular disposi- tion, thus forming a reticulate microsculpture. Downwards, the ribs disappear and they are substituted by a concentric zone with a peculiar design in which parallel and narrow sulcations are grouped forming irregular bands radially disposed (Fig. 6C). No microsculpture is observed in the later dissoconch. Shell pores are present in both surfaces of the dissoconch being more abundant in the inner surface of the early dissoconch, where the estimated density reaches about 200 pores per square millimeter in specimens between 2 and 3 mm in length (Fig. 6D). Pores are circular in cross-section and their diameter is about 2 um at the outer surface (Fig. 6F) and 2.5 um at the inner one, where they are funnel-shaped (Fig. 6E, G, H, I). The pores are openings of narrow tubules that cross perpendicularly the shell (Fig. 6G), although most of them are blinded just before reaching the outer shell sur- face (Fig. 6H). Some of these tubules seem to be filled although the nature of the filling material is uncertain at pre- sent (Fig. 6J). ARAUJO ET AL.: CORBICULA FLUMINEA IN EUROPE 45 Fig. 5. Corbicula fluminea from Iberian Peninsula. A, B, C. Three different sizes from the Mino River (A=24.5 mm; B=17.2 mm; C=8 mm). D. Umbo of a specimen from the Mifio River. F, E. Hinge and umbo of A. G. Specimen from the Douro River (13 mm). SHELL MORPHOMETRY Descriptive statistics for the Mino River population are given in Table |. Variation about the means is substantial, the coefficients of variation being about 17%, except for the width of the hinge that seems to be very variable (23%), the number of teeth (11.5%) and the angle A (3.3%) which is the least variable measurement. A similar situation is found for the raw, pooled data (Table 2). This table also shows that all variables are significantly positively correlated with L (p < 0.0001 except p < 0.03 for the A angle), suggesting that size has an im- portant effect on shape. Most of the relationships are linear or nearly so (r > 0.90) and the variables roughly isometric with the standard size measure for the size range considered 46 AMER. MALAC. BULL. 10(1) (1993) Table 2. Parameters of the untransformed pooled data set for the following populations: Mino, Duero (Regua), Duero (Nagel sample), Canton (China) (Note: Var., variable acronym as in Fig. 2D; SD, standard deviation; SE, standard error; Min., minimum; Max., maximum; CV, coefficient of varia- iion; b, slope; a, intercept; r, correlation coefficient; *, p <0.0001; **, p < 0.03; n=73). Regression statistics Sample statistics Variable on L Log on Log. L Var. Mean SD SE Min. Max. CV b a r b a r L'% 13.71 3.03 0.35 6.6 21.1 22.07 0.777 1.56 0.98* 0.918 0.04 0.99 L% 15.12 3.82 0.44 7.1 21:2 25.32 0.995 -0.43 0.99* 1.025 -0.04 0.99 Li 12.84 3.19 0.37 6.0 22:5 24.86 0.827 -0.08 0.99* 1.004 -0.09 0.99 H 13.51 3.50 0.40 6.4 25.0 25.90 0.896 -0.49 0.98* 1.013 -0.08 0.98 Hi 12.36 3.15 0.36 6.0 22.7 25.50 0.802 -0.18 0.98* 0.983 -0.08 0.98 Ww 9.50 2.48 0.29 4.1 18.2 26.18 0.628 -0.31 0.97* 1.033 -0.25 0.97 Wh 0.96 0.36 0.04 0.4 25.0 37.16 0.076 -0.22 0.81* 1.00 -1.22 0.79 P 45.65 11.26 1.31 21.0 80.0 24.66 2.920 0.01 0.99* 1.00 0.46 0.99 d 11.19 2.84 0.33 5.4 20.7 25.42 0.730 -0.22 0.98* 0.996 -0.14 0.98 d’ 11.36 2.87 0.33 5.5 21.0 25:27 0.735 -0.13 0.98* 0.990 -0.12 0.98 A 2.12 0.13 0.01 1.8 255 6.30 -0.008 2.26 0.25** -0.045 0.38 0.18 Add’ 0.26 0.06 0.007 0.12 0.46 24.60 0.017 0 1* 1 -1.76 1 L 15.63 3.84 0.45 7.5 21.3 24.60 (slope of log-log regression was about 1.0). The high correlation of all variables with L indicate that a univariate approach is reasonable for size adjustment, with L as a standard size variable for it. None of these tests were significant which indicates that there are no measurements that could separate any of the four populations studied. ANATOMY The siphons of the Mino River Corbicula (Fig. 2A) are of ambarine-orange color with black spots and areas. The two siphons are internally surrounded by a black ring. Be- tween the two siphons, and from the exhalant one, there are two well marked black lines. The siphons are surrounded by tentacles, usually with pigmented bases. Those of the inhalant siphon are longer and with black spots in the middle. There is no rule in the disposition of the fused mantle folds papillae down to the inhalant siphon. There are specimens with a single row and others with the papillae in a random pattern (Fig. 2A) and intermediate situations may be observed. The gonadal tissue occupies, in the reproductive season, most of the visceral mass. Each gonad is formed by greenish arborescent follicles branching through the stroma of the visceral mass. The common branch leads to a single gonopore, one on each side of the body just above the ex- cretory aperture in the latero-dorsal edge of the visceral mass (Fig. 2B). One specimen of Goian, captured in July 1990, presented developing larvae in the inner demibranchs (Fig. 7). The larvae are bivalved and D-shaped. DISCUSSION All the European localities described for Corbicula demonstrate the species lives in substrata of sand, mud and gravel. The habitats are lotic areas receiving tidal influences with the exception of the French population cited by Girardi (1989-1990). Physicochemical data of La Dordogne population (Mouthon, 1981) at 21°C are within the range observed in the Mino River. Conductivity values correspond in both cases to freshwaters agreeing with the habitat preferences of Cor- bicula flumina given by Morton (1982) to distinguish this species from C. fluminalis. The Tajo estuary population dif- fers from the others in that the salinity varies between 4 and 18% (=7.400-30.000 »S) (Mouthon, 1981), which indicates values of salinity much higher than expected for C. fluminea (Morton, 1982). No data are available for the populations of the Duero River (Nagel, 1989) or the Canal du Midi (Girardi, 1989-1990). Shell morphology of the Mino River specimens is similar to that of Corbicula fluminea from the UZMC col- lection and differs from the C. fluminalis lectotype (here designated) in that the last is darker, taller and more triangular than C. fluminea. Shell size seems to be the most important difference among the European Corbicula populations. Table 3 shows maximum and minimum values of shell length for all the Table 3. Minimum and Maximum values of shell length in European Cor- bicula populations. Locality Min. Max. Reference Tajo Estuary 25 41 Mouthon, 1981 La Dordogne 16 20 Mouthon, 1981 Duero River 18.1 27.3 Nagel, 1989 Duero River (Regua) TES 13 This study Mino River 9.7 20 This study ARAUJO ET AL.: CORBICULA FLUMINEA IN EUROPE 47 Fig. 6. Corbicula fluminea from Mifio River. A. Prodissoconch. B. Microsculpture of the prodissoconch. C. Microsculpture of the early dissoconch. D. Pores of the inner surface. E. Internal pore. F. External pore. G, H, I. Sections of the shell, showing the pores and tubules. The inner surface above. J. Tubule filled with unknown material. populations. The larger specimens are those of the Tajo Estuary where the smallest shells are greater than the largest ones of other three populations. However, as Britton and Mor- ton (1979) pointed out, ‘‘the Corbicula shell, in general, lacks good taxonomic markers for species discrimination’ ’, so that according to these authors ‘‘giving an excessive value to deter- minate conchological characters, it has lead up in the cur- rent taxonomic confusion’. The results suggest that size has an important effort on shape. For that reason, when shell shape of three populations (Mino River, Duero River at Regua and Duero River- Nagel coll.) and that of Canton, China, is compared after removing the effect of size on the 48 AMER. MALAC. BULL. 10(1) (1993) Fig. 7. Gill of Spanish Corbicula fluminea showing the incubated larvae (Length of the larvae = 0.2 mm). measurements, there are no signficant differences among the european populations nor among them and the one from Can- ton, suggesting that all of them belong to the same species: C. fluminea. Unfortunately we have not enough information about the Tajo population to test whether it also belongs to the same taxonomic unit. Mouthon (1981) cites the presence of many pores in the shell of juvenile specimens of La Dordogne. A detailed microstructural study of Corbicula fluminea from the Mississippi River show the existence of 196 pores per square millimeter (Tan Tiu and Prezant, 1989), which is very close to the density found in the Mino River population (200 pores per square millimeter). In scanning electron micro- graphs of both American and Spanish shells, it is possible to observe that pores are the openings of tubules (Fig. 6G, H), some of them filled by a material identified by the above- mentioned authors as mantle extensions. We have observed similar filling structures in broken shells (Fig. 6J) but the fact that these shells were cleaned with sodium hypoclorite suggest that more studies are needed to determine the com- position of the filling material and therefore the function of the tubules. Concerning the soft parts, Britton and Morton (1979) draw the siphons and describe the differences between Cor- bicula fluminea and C. fluminalis. The first often has a band of pigment in the tentacles of the inhalant siphon that is ab- sent in the last. In the exhalant siphon, C. fluminea shows a ring of pigment internally and is densely pigmented ex- ternally. There are more and bigger sensory tentacles around the exhalant siphon in C. fluminalis. The papillae of the fused mantle folds, dorsal and ventral to the siphons, form a single alternating row in C. fluminea and there are many more papillae arranged in a series of rows in C. fluminalis. The Corbicula of La Dordogne has small brown spots in the base of the papillae of the two siphons and fine dark bands surrounding the mantle portion near and into the holes of the siphons (Mouthon, 1981). This pigmentation is com- pletely absent in the Tajo River Corbicula (Mouthon, 1981) and is very similar to that found in the Mino specimens. The results of this paper suggest that the number and disposition of the mantle papillae is not a useful discriminating character. The Mino River population has a very high variability including all intermediate situations (Fig. 2A). In the rest of the features they are very close to that described for Corbicula fluminea by Britton and Morton (1979). Finally, Britton and Morton (1979) discriminate Corbicula fluminea from C. fluminalis by the fact that the former nourished the fertilized eggs within the inner demibranch while they are not retained in the last. Undoubt- edly, this represents the most important biological difference between both species, though it is difficult to test in many samples and impossible in conchological museum collections. Figure 7 shows one specimen of the Spanish Corbicula population from the Mino River incubating larvae in its in- ner gill demibranch. This evidence, in addition to the previously discussed data of ecology, shell morphology and anatomical features, confirm the hypothesis that the Mino River population belongs to C. fluminea. The available information on the other European Corbicula suggest that there are not enough divergences among them and C. fluminea populations to believe that we are dealing with a different species (except for the Tajo Estuary population). This hypothesis is also supported by the large invasive historical record of C. fluminea (McMahon, 1982, 1983 and references), not shared by any other Corbicula species and the wide ecological range both in natural habitats as in colonized ones. Regarding ecological effects on the native fauna, the disappearance of Pisidium amnicum at Goian was coincident with the Microcystis aeruginosa bloom and also with the demographic increase of Corbicula fluminea. For this reason it is impossible to assess if only one of these phenomena or both were responsible for this extinction. Further research about population dynamics and reproductive strategy must be conducted in order to elucidate the growth and invasive speed of Corbicula fluminea in Europe, but it is a fact that we are witnessing an introduc- tion of this invasive bivalve in the continent. ACKNOWLEDGMENTS We are gratefully indebted to Tom Schi¢tte of the Zoologisk Museum of Copenhagen and Dr. K. O. Nagel for the specimens of Corbicula loaned ARAUJO ET AL.: CORBICULA FLUMINEA IN EUROPE 49 and other useful information. We thank E. Rolan and his wife for providing facilities in our samples trips, and M. A. Alonso-Zarazaga for his help in the discussion of the type material. Thanks also to A. I. Camacho and J. Bedoya for the Corbicula specimens of Regua, Duero River. Rafael Marquez reviewed the English manuscript, hence our gratitude. The SEM photographs are by Miguel Jerez from the Real Jardin Botanico (Madrid). This work received financial support from the Fauna Iberica Projects (DGICYT n° PB87 0397 and PB89 0081). LITERATURE CITED Britton, J. C. and B. Morton. 1979. Corbicula in North America: The evidence reviewed and evaluated. In: Proceedings of the First International Corbicula Symposium 1977, J. C. Britton, ed. pp. 250-287. Texas Christian University. Burch, J. Q. 1944. Checklist of west American mollusks. Family Cor- biculidae. Minutes of the Conchological Club of Southern California 36:18. Counts, C. L. Il. 1981. Corbicula fluminea (Bivalvia: Sphaeriacea) in British Columbia. Nautilus 95(1):12-13. Counts, C. L. II. 1985. Corbicula fluminea (Bivalvia: Corbiculidae) in the state of Washington in 1937, and in Utah in 1975. Nautilus 99(1):18-19. Counts, C. L. II. 1991. Corbicula (Bivalvia: Corbiculidae). Tryonia 21:1-134. Davis, G. M. 1967. The systematic relationship of Pomatiopsis lapidaria and Oncomelania hupensis formosana (Prosobranchia: Hydrobiidae). Malacologia 6(1-2):1-143. Ellis, A. E. 1978. British Freshwater Bivalve Mollusca. Synopses of the British Fauna (New Series) Vol. 11, Linnean Society of London. Academic Press. London. 109 pp. Girardi, H. 1989-1990. Deux bivalves d'eau douce récents pour la faune fran- caise (Mollusca, Bivalvia). Bulletin Societé Et. Sciences naturelles Vaucluse 87-93. International Commission of Zoological Nomenclature. 1985. /nternational Code of Zoological Nomenclature. Third Edition. London. 338 pp. Ituarte, C. F. 1981. Primera noticia de la introduccion de pelecipodos asiaticos en el area Rioplatense (Mollusca, Corbiculidae). Neotropica 27(77):79-82. Johnson, R. I. 1959. The types of Corbiculidae and Sphaeriidae (Mollusca: Pelecypoda) in the Museum of Comparative Zoology, and a bibliographic sketch of Temple Prime, an early specialist of this group. Bulletin of the Museum of Comparative Zoology 120:429-479. Kennard, A. S. and Woodward, B. B. 1926. Note on F. O. Miiller’s Types of Tellina fluminalis, fluminea and fluviatilis. Proceedings of the Malacological Society of London, XV1I:100-101. Linstow, O. v. 1922. Beitrag zur Geschichte und Verbreitung von Corbicula fluminalis. Archiv fur Molluskenkunde 54(4/5):113-144. McMahon, R. F. 1982. The occurrence and spread of the introduced asiatic freshwater clam, Corbicula fluminea (Miiller), in North America: 1924-1982. Nautilus 96(4):134-141. McMahon, R. F. 1983. Ecology of an Invasive Pest Bivalve, Corbicula. In: The Mollusca, Vol. 6, W. D. Russell-Hunter, ed. pp. 505-561. Academic Press, New York. Motas, C. 1962. Procedés des sondages phréatiques- Division du domainesouterran- Classification écologiques des animaux souterrains- Lepsammon. Acta Musei Macedonici Scientiarum Naturalium. Skoplje 8:135-153. Morton, B. 1977. The population dynamics of Corbicula fluminea (Bivalvia: Corbiculacea) in Plover Cove Reservoir, Hong Kong. Journal of Zoology, London 181:21-42. Morton, B. 1982. Some aspects of the population structure and sexual strategy of Corbicula cf. fluminalis (Bivalvia: Corbiculacea) from the Pearl River, People’s Republic of China. Journal of Molluscan Studies 48:1-23. Mouthon, J. 1981. Sur la presence en France et au Portugal de Corbicula (Bivalvia, Corbiculidae) originaire d'Asie. Basteria 45:109-116. Miiller, O. F. 1774. Vermium terrestrium et fluviatilium, sen animalium in- fusoriorum, helminthicorum, et testaceorum, non marinorum, suc- cincta historia, Vol. 2, Testacea. Havnie et Lipsiae. 214 pp. Nagel, K. O. 1989. Ein weiterer Fundort von Corbicula fluminalis (Miiller, 1774) (Mollusca: Bivalvia) in Portugal. Mitteilungen der deutschen malakozoologischen Gesellschaft \7:44-45. Packard, G. C. and T. J. Boardman. 1987. The misuse of ratios to scale physiological data that vary allometrically with body size. In: New directions in ecological physiology. M. E. Feder, A. F. Bennett, W. W. Burggren and R. B. Huey, eds. pp. 217-236. Cambridge University Press. Prashad, B. 1929. Revision of the Asiatic species of the genus Corbicula. III. The species of the genus Corbicula from China, south-eastern Russia, Tibet, Formosa and the Philippine Islands. Memoirs of the Indian Museum 9:49-72. Reist, J. D. 1985. An empirical evaluation of several univariate methods that adjust for size variation in morphometric data. Canadian Journal of Zoology 63:1429-1439. Reist, J. D. 1986. An empirical evaluation of coefficients used in residual and allometric adjustment of size covariation. Canadian Journal of Zoology 64:1363-1368. Sinclair, S.M. and B. G. Isom. 1963. Further studies on the Introduced Asiatic Clam (Corbicula) in Tennessee. Tennessee Stream Pollution Control Board, Tennessee Department of Public Health, Nashville. 75 pp. Tan Tiu, A. and R. S. Prezant. 1989. Shell tubules in Corbicula fluminea (Bivalvia: Heterodonta): Functional morphology and microstructure. Nautilus 103(1):36-39. Thompson, C. M. and R. E. Sparks. 1977. Improbability of dispersal of adult Asiatic clams, Corbicula manilensis, via the intestinal tract of migratory fowl. The American Midland Naturalist 98:219-223. Zhadin, V. I. 1965 (1952). Mollusks of fresh and brackish waters of the U.S.S.R. Keys to the fauna of the U.S.S.R. 46 (Israel Progr. sci. Transl.):I-X VI, 368 pp. Jerusalem. Date of manuscript acceptance: 21 April 1992 ‘ a " a Genetic relationships among Asian Corbicula: Thai clams are referable to topotypic Chinese Corbicula fluminea David S. Woodruff, Varaporn Kijviriya? and E. Suchart Upatham? 'Department of Biology and Center for Molecular Genetics, University of California, San Diego, La Jolla, California 92093-0116, U.S.A. 2Department of Biology, Faculty of Science, Ramkhamhaeng University, Ramkhamhaeng Road, Bangkok 10240, Thailand 3Center for Applied Malacology and Entomology, Department of Biology, Faculty of Science, Mahidol University, Rama 6 Road, Bangkok 10400, Thailand Abstract. A survey of 28 electrophoretically detected allozyme loci revealed that Thai Corbicula are weakly differentiated (Nei’s unbiased genetic distance, D = 0.12-0.16) from Corbicula fluminea from the type locality in southern China, 1800 km away. This finding supports our earlier proposal that 20 nominal Corbicula species from Thailand are junior synonyms of the widespread and conchologically variable C. fluminea. We have shown recently that 21 nominal species of freshwater clams, Corbicula, from Thailand are genetically indistinguishable and most probably referable to the widespread Asian species C. fluminea (Miller, 1774) (Kijviriya ef al., 1991). We found little variation at 24 allozyme loci in Corbicula collected from 40 sites up to 1500 km apart in Thailand. Thirty-five of the samples were found to be genetically identical; clustering at insignificant (D < 0.01) multilocus genetic distances (Nei, 1978). The remain- ing five samples (from northeast Thailand) were weakly dif- ferentiated from the others with D < 0.04. The low levels of genetic differentiation led us to suggest that 20 Thai species were junior synonyms of the earliest named local taxon, C. fluminea, a population of which was recognised from the Chao Phraya River, Bangkok, by Brandt (1974). We argued that our proposed revision would be supported by the demonstration that the Thai Corbicula are genetically similar to Chinese clams from the type locality of C. fluminea. We now present a genetic comparison of the Thai clams with topotypic C. fluminea from south China and show that the proposed synonymy was appropriate. MATERIALS AND METHODS To test the hypothesis that the Thai clams are genetical- ly similar to topotypic Corbicula fluminea we performed a multilocus comparison of two samples of Corbicula from northeast Thailand and one from Hong Kong, 1800 km to the northeast. One of the Thai samples had been shown previously to be genetically identical to putative C. fluminea from Thailand but its relationship to topotypic clams was unknown. The three samples were: Thai 1. Collected from Ubonrat Reservoir, Khon Kaen Province, northeast Thailand. Voucher specimens in Museum, Center for Applied Malacology and Entomology, Faculty of Science, Mahidol University (MUFS-THOO-167) and Los Angeles County Museum of Natural History (LACM 85-366.1). These clams were identified originally using Brandt’s (1974) criteria as Corbicula lydigiana Prime, 1861, by Kijviriya (1990), and subsequently referred to as Sample 23 in Kijviriya et al. (1991), where they were shown to be genetically identical to putative C. fluminea from the Chao Phraya River, Bangkok. Thai 2. Collected at Ban Huikom, Phibun Mangsahan District, Ubon Ratchathani Province, northeast Thailand. Conchologically very similar to Thai 1; specific identity not determined. Voucher specimens: MUFS-THOO-192 and LACM 85-367.1. China. Collected at Ping Long Village, Lam Tseun River Valley, New Territories, Hong Kong. Shell color and morpho- metric variation in this population has been described in detail by Morton (1987). Voucher specimens: LACM 85-368.1. Clams were frozen at -70°C immediately after collec- tion and handcarried to the senior author’s laboratory for genetic analysis. Starch gel electrophoretic separation of allozymes extracted from whole body homogenates was ac- cording to the methods of Kijviriya et al. (1991). For 20 loci the specific methods are given by Kijviriya et al. (1991: see Table | for full names and Enzyme Commission numbers): Aat-I, Aat-2, Es-l, Es-4, Es-5, G6pdh, Gpi, Idh-l, Idh-2, Mdh-l, Mdh-2, Mdhp-I, Mdhp-2, Pep-A-l, Pep-B-2, Pep-B-3, Pgdh, Pgm-l, Pgm-2, Xdh. Three additional loci [Acp (E.C. 3.1.3.2), Cat (1.11.1.6), Ldh (1.1.1.27)] were resolved on the TC 6.0 buffer described therein, and five others were resolved American Malacological Bulletin, Vol. 10(1) (1993):51-53 51 52 AMER. MALAC. BULL. 10(1) (1993) on the TC 6.8 buffer used by Staub et al. (1990); Gapdh (1.2.1.12), G3pdh (1.1.1.8), Mpi (5.3.1.8), Sod-1, Sod-2 (1.15.1.1). Data consisting of genotypes for individual clams scored at all 28 loci were analyzed and Nei’s genetic identi- ties (I) and unbiased genetic distances (D) (Nei, 1972, 1978, respectively) were calculated using the BIOSYS-1 computer programs (Swofford and Selander, 1981). RESULTS Allozymic variation is summarized in Table 1. Twenty- one of the 28 loci showed no detectable variation; seven loci varied geographically. Both Thai samples were isogenic. The Chinese sample showed low levels of genetic variabiltiy, with diallelic systems at four loci and genotypes in panmictic fre- quencies. The Chinese clams are slightly more variable than the most variable samples reported previously from Thailand (Kijviriya et al., 1991) due to the addition of another poly- morphic locus, Mpi, to the survey. The Thai | sample is genetically identical to Sample 23 of Kijviriya et al. (1991), which was collected at the same reservoir. Thai 2 was found to be very similar to Sample 20 of Kijviriya et al. (1991), which was collected from the same district. The small genetic distance (Table 2: D = 0.04) be- tween Thai | and Thai 2 is similar to that reported between Samples 20 and 23 and typical of the maximum genetic dif- ferentiation detected previously among Thai Corbicula. The fixed difference at Es-5 had been detected previously in Thai clams. The Thai Corbicula are very similar to topotypic C. fluminea from Hong Kong. The mean genetic distance, D = 0.14, is largely due to fixed differences at Mdh-l, Mpi, and Pem-/. Table 1. Genetic variability in Asian Corbicula*. Sample: Thai | Thai 2 China Locus Allele frequency Aat-2« 1.00 1.00 0.61 Es-5« 1.00 0.00 0.00 Gpil 1.00 1.00 0.75 Mdh-1« 1.00 1.00 0.00 Mpib 1.00 1.00 0.5/0.5 Pep-A-]« 1.00 1.00 0.64 Pem-]4 1.00 1.00 0.00 Summary statistics N 16 17.5 17.7 A 1.0 1.0 1.1 P 0.00 0.00 0.14 H 0.00 0.00 0.07 *Summary statistics are: N, mean sample size; A, mean no. of alleles per locus; P, proportion of loci polymorphic; H, mean individual heterozygosity. Each variable locus has two alleles except for triallelic Mpi, where alleles a and c co-occur in the Chinese sample. Table 2. Matrix of genetic distance values. Sample Thai | Thai 2 Thai | --- Thai 2 0.04 --- China 0.16 0.12 DISCUSSION The genetic identity among the Asian clams studied here is slightly higher than that reported in an earlier 12-locus comparison of samples from Hong Kong and the United States (Smith et al., 1979): IT = 0.87 vs. 0.84. The Thai Corbicula fluminea are thus more similar to those of China than the latter are to their trans-Pacific derivatives. Although there is no simple relationship between genetic distance and taxonomic level the D = 0.14 value reported here is within the range found among conspecific populations of geographically widespread molluscs and other animals (Davis, 1983; Thorpe, 1983; Woodruff et al. , 1988). Cases of interspecific comparisons with D < 0.14 are known but, in such cases, species status rests on other criteria (especially behavioral traits that could function as reproduc- tive isolating mechanisms or species recognition cues, post- mating sterility barriers, chromosomal reorganization, etc. ) and genetic similarity merely reflects the recency of cladogenesis. We know of no such criteria which might af- fect our interpretation of the genetic similarities discovered among Asian Corbicula. Detailed studies of anatomical varia- tion in clams from China (Morton, 1987) and Thailand (Kijviriya, 1990) failed to reveal any taxonomically signifi- cant variation. Only in the case of conchological features (shell size, shape, sculpture and color), has significant local and geographic variation been documented. Such conspicuous conchological variation, employed by Brandt (1974) to recognize 28 Corbicula species in Thailand, is now seen to be taxonomically irrelevant. Morton’s (1986) opinion, that most or all of Brandt’s taxa are junior synonyms of C. fluminea, is supported. We reaffirm our recommendation (Kijviriya et al., 1991) that 20 nominal Thai taxa for which we have genetic data be synonymized with C. fluminea (Miiller, 1774). Finally, our finding of only moderate differentiation within geographically widespread populations of Corbicula fluminea on the Asian mainland contrasts with the higher levels of differentiation (J = 0.3) reported between Japanese and Philippine clams by Smith et al. (1979). A comprehensive allozymic survey in Asia is now required to establish whether the island populations merit specific recognition. Such a survey could also resolve two issues involving the North American C. fluminea: the identification of the source popula- tion(s) and the number of successful colonizations. Although available genetic data support an Asian mainland origin WOODRUFF ET AL.: GENETIC RELATIONSHIPS AMONG ASIAN CORBICULA 33 hypothesis, they do not preclude an island origin, or multi- ple introductions from either source. Given the apparent genetic uniformity of the Asian mainland populations, it could be difficult, however, to reconstruct the events surrounding their introduction to North America using allozymes. ACKNOWLEDGMENTS We thank Drs. May Yipp and David Dudgeon for assistance in Hong Kong and M. Patricia Carpenter for laboratory assistance in San Diego. This study was supported by the University of California Academic Senate and, indirectly, by grants from the U.S. National Science Foundation and the U.S. Agency for International Development to one of us (DSW). LITERATURE CITED Brandt, R. A. M. 1974. The non-marine aquatic Mollusca of Thailand. Archiv fitr Molluskenkunde 105:1-423. Davis, G. M. 1983. Relative roles of molecular genetics, anatomy, morpho- metrics and ecology in assessing relationships among North American Unionidae (Bivalvia). In: Protein Polymorphism: Adaptive and Tax- onomic Significance, G. S. Oxford and D. Rollinson, eds. pp. 193-222. Academic Press, London. Kijviriya, V. 1990. Studies of the Asiatic clams, (Corbicula, Muhfeld, 1811) in Thailand: Electrophoretic estimates of enzyme variation and the use of anatomy as species indicator. Doctoral dissertation, Mahidol University, Bangkok, Thailand, xii + 184 pp. Kijviriya, V., E. S. Upatham, V. Viyanant and D. S. Woodruff. 1991. Genetic studies of Asiatic clams, Corbicula, in Thailand: allozymes of 21 nominal species are identical. American Malacological Bulletin 8:97-106. Morton, B. 1986. Corbicula in Asia - up-dated synthesis. Proceedings of the Second International Corbicula Symposium. American Malacological Bulletin Special Publication 2:\13-124. Morton, B. 1987. Polymorphism in Corbicula fluminea (Bivalvia: Cor- biculoidea) from Hong Kong and southern China. Malacological Review 20:105-127. Nei, M. 1972. Genetic distances between populations. -American Naturalist. 106:283-292. Nei, M. 1978. Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics 89:583-590. Smith, M. H., J. C. Britton and P. Burke. 1979. Genetic variability in Cor- bicula, an invading species. In: Proceedings of the First International Corbicula Symposium, J. C. Britton, ed. pp. 244-248. Texas Chris- tian University Research Foundation, Fort Worth, Texas. Staub, K. C., D. S. Woodruff, E. S. Upatham and V. Viyanant. 1990. Genetic variation in Neotricula aperta, the intermediate snail host of Schistosoma mekongi: allozyme differences reveal a group of sibling species. American Malacological Bulletin 7:93-103. Swofford, D. L. and R. B. Selander. 1981. BIOSYS-1: A FORTRAN pro- gram for the comprehensive analysis of electrophoretic data in popula- tion genetics and systematics. Journal of Heredity 72:281-283. Thorpe, J. P. 1983. Enzyme variation, genetic distance, and evolutionary divergence in relation to levels of taxonomic separation. /n: Protein Polymorphism: Adaptive and Taxonomic Significance, G. S$. Oxtord and D. Rollinson, eds. pp. 131-152. Academic Press, London. Woodruff, D. S., K. C. Staub, E. S. Upatham, V. Viyanant and H. C. Yuan. 1988. Genetic variation in Oncomelania hupensis: Schistosoma japonicum transmitting snails in China and the Philippines are distinct species. Malacologia 29:347-361. Date of manuscript acceptance: 9 December 1991 ui Genetic structure and heterozygosity-related fitness effects in the marine snail Littorina littorea David W. Foltz,! Sandra E. Shumway? and Dennis Crisp** ‘Department of Zoology and Physiology, Louisiana State University, Baton Rouge, Louisiana 70803-1725, U. S. A. ?Maine Department of Marine Resources and Bigelow Laboratory for Ocean Science, West Boothbay Harbor, Maine 04575, U. S. A. 3Marine Science Laboratories, Menai Bridge, Gwynned LLS9 5TH, U. K. Abstract. The relationships among rate of oxygen consumption under routine and starved conditions, soft tissue dry weight, shell growth rate and heterozygosity at polymorphic allozyme loci were investigated in 86 individuals of the common periwinkle, Littorina littorea (Linnaeus, 1758). Oxygen consumption was measured at the time of collection and after 14 days starvation using a micro-Winkler method. Soft tissue dry weight was estimated from shell height using a regression equation developed from analysis of a set of 34 additional snails. Shell growth rate was estimated as distance between successive lines in thin sections through the growing edge of the shell. Heterozygosity was studied using conventional starch gel electrophoretic techniques to examine enzyme systems controlled by 34 loci; 11 of the loci were polymorphic. The mean observed heterozygosity per locus per individual was 0.073. Continuous variables were adjusted for differences in estimated dry weight by using linear regression to convert all measurements to a standard dry weight. Only for growth rate was there a significant positive association with dry weight. The weight-adjusted values were then tested for an effect of heterozygosity using linear regression. There was no evidence for an association between level of heterozygosity and oxygen consumption under either routine or starved conditions. However, starva- tion caused a significant depression in oxygen consumption, in paired comparisons. There was no detectable association between heterozygosity and shell growth rate. Genetic studies of marine mollusc populations have with more traditional concepts in molluscan physiology. They reported several unexpected findings, including deficiencies argue that the greater metabolic efficiency of more in the numbers of heterozygous individuals and correlations heterozygous individuals results from lower rates of protein between estimates of an individual’s overall level of turnover in such individuals, not from lower rates of energy heterozygosity and various surrogate measures of fitness such metabolism. Further, they suggest that low intensities of pro- as growth or viability (Mitton and Grant, 1984; Zouros and tein turnover reduce the energy required for maintenance, Foltz, 1984, 1987). Among marine molluscs, there is abun- with some of the energy saved being used to support in- dant evidence for age- or size-dependent changes in allele creased feeding rates. A second approach has been to com- frequencies and in the level of observed heterozygosity for pare heterozygosity-fitness correlations from various studies numerous allozyme loci. One common pattern is for the and organisms in a search for general trends. For example, heterozygosity to increase (or the deficiency of heterozygotes Zouros (1987) reported that significant positive correlations to decrease) among older or larger animals, suggesting an between heterozygosity and fitness components were more apparent viability advantage of more heterozygous in- likely to be found in marine molluscs if (1) the allozyme loci dividuals. There is also an extensive literature reporting showed deficiencies in the numbers of heterozygous in- positive correlations between allozyme heterozygosity and dividuals (compared to Hardy-Weinberg expectations), and growth rate in marine molluscs and other organisms; several (2) the sample came from a natural population rather than conclusions have emerged from this work. First, the apparent laboratory populations derived from a small number of correlation between heterozygosity and growth rate can be parents. These two findings suggest an influence of popula- related at the physiological level to either a lower cost of tion genetic structure on the occurrence of heterozygosity- routine metabolism (Koehn and Shumway, 1982; Garton, fitness correlations. The association between degree of heter- 1984), a higher feeding rate (Garton, 1984; Holley and Foltz, zygote deficiency and occurrence of heterozygosity-fitness 1987), or both, in highly heterozygous individuals. Hawkins correlations has not been satisfactorily explained, and seems et al. (1986, 1989) have attempted to integrate these findings counter-intuitive. Although several different hypotheses have Bee es been proposed to explain this association (see below), pre- *Deceased 18 January 1990 sent data are insufficient to decide among them. More data American Malacological Bulletin, Vol. 10(1) (1993):55-60 55 56 AMER. MALAC. BULL. 10(1) (1993) are needed, particularly to determine if the association be- tween heterozygote deficiency and heterozygosity/fitness cor- relations is of wide occurrence. The present study examines genetic structure, heterozygosity, weight, growth rate and oxygen consumption rate in a natural population sample of the marine snail Littorina littorea (Linnaeus, 1758). MATERIALS AND METHODS A total of 86 Littorina littorea (shell height range 16.6—26.2 mm) was collected near West Boothbay Harbor, Maine (during the same tidal cycle and from the same shore height) in the spring of 1987. Immediately after collection, each snail was marked individually with a small plastic tag and its (routine) oxygen consumption rate determined. The snails were then held in tanks supplied with running seawater without food for 14 days, after which rates of oxygen con- sumption were again measured. Seawater temperatures ranged from 8-12°C during the experiments, and all measurements of oxygen consumption were made under ambient conditions. Oxygen consumption rates were measured on individual animals in 70 ml experimental chambers using a micro- Winkler technique (Burke, 1962). The snails were allowed to equilibrate for | hr, during which time the water was aerated. Measurements were carried out over a period of ap- proximately 3 hrs; total oxygen concentration in the ex- perimental chambers never fell below 70% saturation. Con- trols were run using identical vessels with no animals. The snails were then placed in large (2.0 m x 0.5 m x 0.5 m) cages at mid-tide level for four weeks. Ample supplies of U/va and Enteromorpha were maintained in the cages as a food source. The snails were removed from the cage and the shell height of each animal was measured. The free growing edge of each shell was removed with a diamond saw, and the soft tissues were removed and frozen at -80°C for later allozyme analysis. Shells were preserved in 70% alcohol containing a small quantity of borax and air-shipped to Menai Bridge, Wales, for growth measurements in the laboratory of D. J. Crisp. The most recently deposited 2 or 3 mm at the lip of each shell were embedded in resin. Sections were cut parallel to the direction of growth, and the cut surface ground on wet and dry paper and polished on a cloth soaked in ‘*Brasso.’’ The polished surface was etched in 0.01 M HCI for 10-20 min, and acetate peel replicas of the dry surface prepared using the technique described by Ekaratne and Crisp (1982). The peels were examined microscopically and the shell growth rate for each animal estimated from the width of the most recently deposited band. The growth bands were reasonably assumed to correspond to the most recent tidal cycle, because similar bands have been shown to be tidally-produced in Littorina littorea from Menai Strait, Wales (Ekaratne and Crisp, 1982, 1984). Soft tissue dry weight of each animal was estimated by a regression equation obtained from a set of 34 additional snails collected near West Boothbay Harbor. This set had a mean shell height of 18.6 mm (range 6.5—28.2 mm) and a mean soft tissue dry weight of 0.146 g (range 0.002— 0.384 g). The regression equation relating soft tissue dry weight (Y) and shell height (X) was log.(Y) = -12.44 + 3.49 - loge(X), P<0.0001, R? = 0.98. The mean estimated soft tissue dry weight of the 86 snails was 0.153 g (range 0.074—0.314 g). The frozen tissue samples were air-shipped to the laboratory of D. W. Foltz, where they were processed and analyzed electrophoretically by standard methods (e.g. Harris and Hopkinson, 1976). In all, 34 enzyme loci were resolved. Alleles at each polymorphic locus were designated by numbers indicating mobility relative to the most common allele at that locus, which was designated ‘‘100’’; negative numbers indicated cathodally migrating bands. For enzymes with multiple isozymic forms, loci were numbered sequen- tially starting with the most anodally-migrating system. Allele frequencies, fixation indices (F) and observed and expected heterozygosities for each locus were calculated using stan- dard formulas (e.g. Hedrick, 1983). All data were analyzed by the Statistical Analysis System (SAS Institute, Inc., 1985). RESULTS Of the 34 enzyme loci examined, 11 were polymorphic (see Table 1 for allele frequencies and single-locus observed heterozygosities). The fixation index (F) for each polymorphic locus was close to 0, indicating no significant departures from Hardy-Weinberg equilibrium. The average F was 0.003 (range -0.051—0.131). The average number of heterozygous loci per individual was 2.49 (range 0O—5). The frequency of each heterozygosity class was plotted and compared with the ex- pected frequency derived from the II single-locus expected heterozygosities (Fig. 1). The distribution of observed heterozygosity was more platykurtic than the expected distribution, with an excess of both low-heterozygosity and high-heterozygosity individuals. Continuous variables (growth rate, routine and starved oxygen consumption rate) were loge-transformed prior to analysis (to help ensure that each dependent variable had a normal distribution and uniform variance) and were adjusted for differences in estimated soft tissue dry weight by using linear regression to convert all measurements to a standard dry weight (0.146 grams). As expected, there was a signifi- cant positive effect of dry weight on growth rate (P<0.03), but no detectable effect of dry weight on routine (P > 0.07) or starved (P>0.26) VO. The finding of no association be- tween oxygen consumption under routine or starved condi- tions and estimated soft tissue dry weight was unexpected. It differs from previous studies of Littorina littorea (e.g. Newell and Roy, 1973) that have found a positive correlation between oxygen consumption and dry tissue weight. The lack FOLTZ ET AL.: HETEROZYGOSITY IN LITTORINA LITTOREA ay of a correlation is most likely due to the absence of extreme- ly low-weight individuals in the present sample. For exam- ple, in Newell and Roy’s (1973) study the range for estimated soft tissue dry weight was 2—160 mg, whereas the cor- responding range in the present study was 70—300 mg. There was a significant depression in oxygen consumption after star- vation for 14 days. Previous studies of the effect of starvation on oxygen consumption in pulmonate and prosobranch molluscs (summarized by Studier and Pace, 1978) have variously found that oxygen consumption rates increase, decrease or remain constant in starved animals. Factors responsible for such different responses could include length of starvation period, acclimation temperature or sex dif- ferences. Fig. 2 presents growth rate (wm/tidal cycle), adjusted for dry weight differences, and dry weight (in grams) for dif- ferent levels of heterozygosity. Multiple-locus heterozygosi- ty was not a significant source of variation for either adjusted growth rate (P>0.07, R?=0.04) or dry weight (P<0.77), R? <0.01). Single-locus heterozygosity for each of the Il polymorphic loci also was not a significant source of varia- tion for either adjusted growth rate or dry weight, after allow- ing for multiple tests (results not shown). Fig. 3 presents Vo, 30 20) FREQUENCY a PGi ie) 1 2 3 4 5 6-11 NUMBER OF HETEROZYGOUS LOCI Fig. 1. Observed (filled bars) and expected (open bars) numbers of individuals with various numbers of heterozygous loci (out of 34 examined loci) in Littorina littorea. Table 1. Enzyme commission (E.C.) numbers, allele frequencies and observed heterozygosities (Ho), + 1 standard error, at 11 poly- morphic allozyme loci in Littorina littorea. Locus E.C. No. Allele Frequency Acon-1 4.2.1.3 107 .023 + .O11 100 971 + .013 95 .006 + .006 Ho 058 + .025 Acon-2 4.2.1.3 208 .006 + .006 146 122 + .025 100 .872 + .025 Ho .232 + .046 Ck pad peed) 127 052 + .O17 119 012 + .008 100 .936 + .019 Ho .128 + .036 Est-1 Sleds 105 .O11 + .008 100 .989 + .008 Ho .023 + .016 Est-3 3.1.1.1 -119 .273 + .034 100 401 + .038 -67 303 + .035 -59 .023 + .O11 Ho .698 + .050 Glydh 1.1.1.29 116 .064 + .019 100 .936 + .019 Ho 105 + .033 Locus E.C. No. Allele Frequency Pep 3.4.-.- 112 .017 + .010 100 .983 + .O10 Ho .035 + .020 6-Pgd 1.1.1.44 175 029 + .013 147 .238 + .032 133 .047 + .016 100 .686 + .035 Ho 488 + .054 Pgi D319 107 .047 + .016 100 .936 + .019 93 O11 + .008 85 .006 + .006 Ho 128 + .036 Pgm-1 974222 100 913 + .022 71 .087 + .022 Ho IST + .038 Pgm-2 5.4202, 126 .047 + .016 111 106 + .023 100 .694 + .035 83 153 + .028 Ho 447 + .054 Average* Ho .073 + .005 *Average Hy includes the following 23 monomorphic loci: Ada (3.4.4.4); Adh (1.5.1.17); Ald (4.1.2.13); Alkp (3.1.3.1); Ark (2.7.3.3); Est-2 (3.1.1.1); Fum (4.2.1.2); Got (2.6.1.1); G-6-Pdh (1.1.1.49); Gpt (2.6.1.2); Gpdh (1.1.1.8); Idh-I (1.1.1.42); Idh-2 (1.1.1.42); Ipo (1.15.1.1); Lap (3.4.--); Mdh (1.1.1.37); Me (1.1.1.40); Mpi (5.3.1.8); Nsp (2.4.2.1); Odh (1.5.1.11); Pep (LGG) (3.4.--); Sdh (1.1.1.4); Xdh (1.1.1.204). 58 AMER. MALAC. BULL. 10(1) (1993) 30 |e LOGe[Y] = -2.02 + 0.03X ; 2 ® s = e220 |: ; ; ‘ — Q . a g bE e 2 ra é 1) a io) iv 8 3 — : : = 3 8 @ g a a a ) = 40 ' ; O 0} 2 @ = oe i _ 100 O > 7 1O} O : ; | 75 @ . Q g ) g ra ic} ic} o ic} NS @ H a . (o) => oS * a E Lu E = i) : : : : : Fs io) 7 : e 20. 'O LOGelY] = 3.93 - 0.03X 6 0) 1 2 3 4 5 HETEROZYGOSITY Fig. 2. Semi-logarithmic plots of estimated soft tissue dry weight (upper panel) and growth rate (lower panel) verus heterozygosity in Littorina littorea. Regression lines and equations are also shown, but are not statistically significant. (ml O, - hr-') under routine and starved conditions for dif- ferent levels of heterozygosity. The effect of heterozygosity on adjusted oxygen consumption rate was not significant when tested by linear regression, under either routine (P >0.65) or starved (P>0.85) conditions. There was a significant reduction (P< 0.0001 by paired t-test) in adjusted oxygen con- sumption rate under starved conditions (mean = 0.092 ml O, - hr-') when compared to the corresponding mean value (0.113) under initial conditions. DISCUSSION The observed heterozygosity in this study (0.073 + 0.005) is slightly higher than that reported (0.04) in previous studies of Littorina littorea (Fevolden and Garner, 1987; Jan- son, 1987). Fevolden and Garner (1987) found no evidence for heterozygote excesses or deficiencies. They also looked for genetic differences between fast-growing and slow-growing snails at the 6-Pgd locus, with negative results. The present study extends that conclusion to a larger set of heterozygous loci and to oxygen consumption rate comparisons. Plots of VO, versus heterozygosity for both routine and starved con- ditions gave no suggestion of an association for either treat- ment, although the three snails with 0 observed heterozygosity had very low oxygen consumption rates. As seen in Fig. 2, these animals also had very high weights. Although the ox- ygen consumption rates were adjusted for the (linear) effect of log.-transformed dry weight, there could have been some residual effect of weight differences on oxygen consumption. The previous literature on allozyme heterozygosity- fitness correlations has been summarized by Mitton and Grant (1984), Zouros and Foltz (1987) and Zouros (1987). As noted FOLTZ ET AL.: HETEROZYGOSITY IN LITTORINA LITTOREA he Bs 35 : tc . LOGelY] = -2.15 - 0.01X m 225) E15 : = q 8 g a 4 z : " =) i : a nm D 8 8 Oo ° i i 8 = 052 B .15 8 # A ry | : 5 | ia ic) ; I + .10 .- a é a Li | = = a 7 : = 8 a io} | z (o) a 8 @ = ‘ =} 405 Oo 4 : = LOGel[Y] = -2.40 - 0.01X @) 1 2 3 4 5 HETEROZYGOSITY Fig. 3. Semi-logarithmic plots of oxygen consumption rate under fed conditions (upper panel) and under 14-day starvation conditions (lower panel) versus heterozygosity in Littorina littorea. Regression lines and equations are also shown, but are not statistically significant. above, heterozygosity-fitness correlations are more likely to be found when the same set of allozyme loci exhibits a de- ficiency in the number of heterozygous individuals than when they do not. The data obtained for Littorina littorea in this study and by Fevolden and Garner (1987) are consistent with the suggestion that heterozygosity-fitness correlations in natural populations of marine molluscs are largely absent when genotype frequencies closely approximate the Hardy- Weinberg expectations. Two previous studies of heterozygosity and growth rate in marine gastropods (Fujino, 1978; Garton, 1984) reported the co-occurrence of heterozygote deficien- cies and heterozygosity-size (or heterozygosity-growth rate) correlations. Despite the seeming contradiction between de- ficiencies in the numbers of heterozygotes (compared to Hardy-Weinberg expectations) and apparent heterozygote superiority at the same loci, positive associations between these two phenomena have been found in numerous studies. This pattern occurs when different species are compared (Zouros, 1987; Zouros and Mallet, 1989) and also when dif- ferent loci within a single species are compared (Gaffney et al., 1990). As reviewed by Zouros et al. (1988) and Gaffney et al. (1990), at least three explanations can potentially ex- plain the co-occurrence of heterozygote deficiencies and heterozygosity-fitness correlations at the same set of loci. First, molecular imprinting could account for both phenomena (Chakraborty, 1989). Second, a high rate of chromosomal mutation (Thiriot-Quievreux, 1986; Thiriot- Quievreux et al. , 1988) and/or a high rate of single-gene muta- tion, could be responsible for apparent heterozygote de- ficiencies through production of null mutations, where nulls 60 AMER. MALAC. BULL. 10(1) (1993) could either be point mutations resulting in loss of activity at a single allozyme locus or else multi-locus deletions with presumably larger fitness effects. A high mutation rate could also cause apparent heterozygosity-fitness correlations, either through direct reduction in fitness of allozyme active/null heterozygous genotypes (Zouros and Foltz, 1987) or through associative overdominance (Zouros, 1987). Third, partial in- breeding will generate both a deficiency of heterozygotes at individual loci and an inter-locus correlation in the degree of heterozygosity (Haldane, 1949; Bennett and Binet, 1956). These inter-locus correlations can generate apparent heterozygote superiority through inbreeding depression (Strauss, 1986; Bush et al., 1987) even in the absence of any direct fitness effect of the loci examined. As yet, there is in- sufficient knowledge about the genetics and population struc- ture of marine molluscs to allow these possibilities to be tested. ACKNOWLEDGMENTS We thank J. Barter for technical assistance and W. B. Stickle and C. R. Richardson for comments on earlier drafts of the manuscript. Research supported in part by NSF grant BSR 84-07450 to Foltz. LITERATURE CITED Bennett, J. H. and F. E. Binet. 1956. Association between Mendelian fac- tors with mixed selfing and random mating. Heredity 10:51-55. Burke, J. D. 1962. Determination of oxygen in water using a 10-ml syringe. Journal of the Elisha Mitchell Scientific Society 78:145-147. Bush, R. M., P. E. Smouse and F. T. Ledig. 1987. The fitness consequences of multi-locus heterozygosity: the relationship between heterozygosity and growth rate in pitch pine (Pinus rigida Mill.). Evolution 41:787-798. Chakraborty, R. 1989. Can molecular imprinting explain heterozygote de- ficiency and hybrid vigor? Genetics 122:713-717. Ekaratne, S. U. K. and D. J. Crisp. 1982. Tidal micro-growth bands in in- tertidal gastropod shells, with an evaluation of band-dating techniques. Proceedings of the Royal Society of London 214B:305-323. Ekaratne, S. U. K. and D. J. Crisp. 1984. Seasonal growth studies of inter- tidal gastropods from shell micro-growth band measurements, in- cluding a comparison with alternative methods. Journal of the Marine Biological Association of the United Kingdom 64:183-210. Fevolden, S. E. and S. P. Garner. 1987. Environmental stress and allozyme variation in Littorina littorea (Prosobranchia). Marine Ecology Progress Series 39:129-136. Fujino, K. 1978. Genetic studies on the Pacific abalone. II. Excessive homozygosity in deficient animals. Bulletin of the Japanese Society of Scientific Fisheries 44:767-770. Gaffney, P. M., T. M. Scott, R. K. Koehn and W. J. Diehl. 1990. Interrela- tionships of heterozygosity, growth rate and heterozygote deficiencies in the coot clam, Mulinia lateralis. Genetics 124:687-699. Garton, D. W. 1984. Relationship between multiple locus heterozygosity and physiological energetics of growth in the estuarine gastropod Thais haemastoma. Physiological Zoology 57:530-543. Haldane, J. B. S. 1949. The association of characters as a result of inbreeding and linkage. Annals of Eugenics 15:15-23. Harris, H. and H. E. Hopkinson. 1976. Handbook of enzyme electrophoresis in human genetics. Elsevier/North-Holland, New York. Unpaginated. Hawkins, A. J. S., B. L. Bayne and A. J. Day. 1986. Protein turnover, physiological energetics and heterozygosity in the blue mussel, Mytilus edulis: the basis of variable age-specific growth. Proceedings of the Royal Society of London 229B:161-176. Hawkins, A. J. S., B. L. Bayne, A. J. Day, J. Rusin and C. M. Worrall. 1989. Genotype-dependent interrelations between energy metabolism, protein metabolism and fitness. Proceedings of the European Marine Biology Symposium 23:283-292. Hedrick, P. W. 1983. Genetics of Populations. Jones & Bartlett, New York. 629 pp. Holley, M. E. and D. W. Foltz. 1987. Effects of multiple-locus heterozygosity and salinity on clearance rate in a brackish-water clam, Rangia cuneata (Sowerby). Journal of Experimental Marine Biology and Ecology 111:121-131. Janson, K. 1987. Allozyme and shell variation in two marine snails (Littorina, Prosobranchia) with different dispersal abilities. Biological Journal of the Linnean Society 30:245-256. Koehn, R. K. and S. E. Shumway. 1982. A genetic/physiological explana- tion for differential growth rate among individuals of the American oyster, Crassostrea virginica (Gmelin). Marine Biology Letters 3:35-42. Mitton, J. B. and M. C. Grant. 1984. Associations among protein heterozygosity, growth rate, and developmental homeostasis. Annual Review of Ecology and Systematics 15:479-499. Newell, R. C. and A. Roy. 1973. A statistical model relating the oxygen con- sumption of a mollusk (Littorina littorea) to activity, body size, and environmental conditions. Physiological Zoology 46:253-275. SAS Institute, Inc. 1985. SAS User’s Guide: Statistics, Version 5 Edition. SAS Institute, Cary, North Carolina. 956 pp. Strauss, S. H. 1986. Heterosis at allozyme loci under inbreeding and crossbreeding in Pinus attenuata. Genetics 113:115-134. Studier, E. H. and G. L. Pace. 1978. Oxygen consumption in the prosobranch snail Viviparus contectoides (Mollusca: Gastropoda). IV. Effects of dissolved oxygen level, starvation, density, symbiotic algae, substrate composition and osmotic pressure. Comparative Biochemistry and Physiology 59A:199-203. Thiriot-Quievreux, C. 1986. Study of aneuploidy in spats of Ostreidae (Bivalvia). Genetica 70:225-231. Thiriot-Quievreux, C., T. Noel, S. Bougrier and S. Dallot. 1988. Relation- ships between aneuploidy and growth rate in pair matings of the oyster Crassostrea gigas. Aquaculture 75:89-96. Zouros, E. 1987. On the relation between heterozygosity and heterosis: an evaluation of the evidence from marine mollusks. Jsozymes 15:255-270. Zouros, E. and D. W. Foltz. 1984. Possible explanations of heterozygote deficiency in bivalve molluscs. Malacologia 25:583-591. Zouros, E. and D. W. Foltz. 1987. The use of allelic isozyme variation for the study of heterosis. /sozymes 13:1-59. Zouros, E. and A. L. Mallet. 1989. Genetic explanations of the growth/ heterozygosity correlation in marine mollusks. Proceedings of the Euro- pean Marine Biology Symposium 23:317-324. Zouros, E., M. Romero-Dorey and A. L. Mallet. 1988. Heterozygosity and growth in marine bivalves: further data and possible explanations. Evolution 42:1332-1341. Date of manuscript acceptance: 2 March 1992 The arm crown in cephalopod development and evolution: a discussion of morphological and behavioral homologies Sigurd v. Boletzky C.N.R.S., Laboratoire Arago, F-66650 Banyuls-Sur-Mer, France Abstract. Capture and manipulation of prey is the primary function of the arms in cephalopods, but they also take on various other prehensile functions as well as propulsive ones (e.g. ‘walking’ in octopus). In many instances they act as visual signal effectors as well. The entire arm crown can be involved in concerted actions, or individual pairs of arms (or an individual arm in the case of certain hectocotyli) can act in a discrete manner. This paper discusses the problem of arm identity and homology in order to provide a basis for the comparison of arm action patterns and arm postures in different cephalopod groups. Emphasis is placed on data derived from the morphogenetic pathway that leads from a uniform embryonic anlage, through subdivision into distinct arm rudiments, to the differentiation of arms. These data support the homology between the arm crowns of decapods and octopods. Furthermore, there is good evidence for the homology of arm pairs at three positions: dorsal arms, ventral arms, ventrolateral arms (=tentacles of decapods). The study of animal behavior is related closely to the fields of animal physiology and functional morphology. Bullock (1965: 454) stated: ‘‘Behavior is carried out by ef- fectors’’; they have physiological characteristics that are af- fected by their morphological properties, the latter of course include the arrangement of the nerves controlling and co- ordinating the actions of effectors. These intrinsic processes and their underlying mechanisms are studied by neurobiologists who may also be interested in the co- adaptation of receptor and effector systems. Behavioral ecologists, on the other hand, investigate the extrinsic effects of behavior to assess specific behavioral adaptations to an environment. The evolutionary perspective of the whole problem is emphasized by the term adaptation. Here the func- tion of behavior is viewed against the historical background of organismal change through time. From the comparative study of functional morphology one can conclude that structural modifications have occurred only within rather narrow limits in the phylogenetic history of organisms in general (Riedl, 1978). The question of whether the same is true for behavior was raised by ethologists early in this century, and from the comparative study of behavior in more or less closely related species they concluded that certain behavioral features can indeed be considered as discrete heritable characters of the phenotype that must somehow be programmed in the genome, much like morpho- logical characters [see Lorenz (1978) for a review]. If behavior is thus studied in a conceptual framework of evolutionary structure, the basic problem to be approached is the homology of behavior characters. The crucial ques- tion is whether homology of behavior characters can be established only relative to homologous effector systems, or whether homologies exist at higher levels of neural integration. The arms of cephalopods provide an interesting ex- ample of co-ordinated effector systems built into a higher in- tegration structure, the arm crown and brachial lobe system, with more or less clearly recognizable homologies at the lower structural levels. This paper discusses arm identities in coleoid cephalopods, dealing in particular with the significance of homology recognition at different levels of integration. The roles played by the arms in one or several superimposed functional contexts (e.g. feeding, burying, signalling) are thus considered from the viewpoint of ‘multi purpose’ effector systems in a way that is intermediate between the approaches of neurobiologists and behavioral ecologists. THE HOMOLOGY CONCEPT AND ITS APPLICABILITY Homology is a basic concept underlying all com- parative studies in biology [see Roth (1991) for a review]. Homologies can be defined in different ways, either excluding or including the criterion of resemblance in the definition. Dohle (1989: 383) makes a clear distinction between homology and transformation, homology representing ‘‘con- formity based on identical instruction for differentiation, handed on by identical replication’, whereas transformation represents the “‘gradual modification of an initial structure leading to extensive dissimilarity that may even end in an unrecognizable condition’’; he therefore considers re- semblance an essential ingredient of homology. American Malacological Bulletin, Vol. 10(1) (1993):61-69 61 62 AMER. MALAC. BULL. 10(1) (1993) In all instances, the ‘practical’ aim of the search for homology is sorting out non-homologous features, which are characterized by resemblance that is not *‘caused by continuity of information’ (Van Valen, 1982: 305). As phylogenetic rela- tionships between different organisms can only be recognized on the basis of truly homologous features, i.e. ‘‘those that stem phylogenetically from the same feature (or condition) in the immediate common ancestor of these organisms”’ (Bock, 1989: 331), one is faced with the danger of circular- ity; it can only be circumvented by the application of very strict rules in testing the homology hypothesis. As structures that can be subjected to a homology test are generally defined on the basis of recognizable component elements, which can represent ‘standard parts’ at lower in- tegration levels (Riedl, 1978), one ultimately deals with in- creasingly generalized standard parts ending at the cellular and molecular levels. In order to resolve this problem of ‘in- finite regression’, complex patterns of corresponding substruc- tures are required for homology propositions (Dohle, 1989). The use of ontogenetic features requires particular cau- tion, because structures that appear homologous on the basis of pattern conformity are not always achieved along identical differentiation pathways (Dohle, 1989). Wagner (1989: 1163) addresses this problem in the framework of his ‘‘triple agenda of a biological homology concept: conservation, individuality, and uniqueness’’; he states: ‘*To explain these three proper- ties of homologs, not all aspects of development are equally important. The source of cells seems to be irrelevant, as long as they interact in a coherent and conservative manner to pro- duce a certain part of the body”’. THE RELATIVE AGE OF HOMOLOGOUS CHARACTERS The recognition of homologies inevitably raises the question of the phylogenetic age of a character: when did character X first appear? In other words: when did this character become phenotypically stabilized in a population (representing an incipient species) which we can consider the ancestor of the species that still carry this character? On the basis of paleomorphological data derived from the fossil record of cephalopods, we can ‘reconstruct’ more or less realistic paleobiological scenarios, but nothing in the fossil record provides direct evidence for any behavioral character that we observe in the living animals. So we de- pend entirely on the comparative analysis of behavior in liv- ing cephalopods when we deal with the question of what is ‘primitive’ and what is ‘derived’, and we are compelled to use the analytical approach employed by neontologists study- ing phylogenetic systematics based on morphological features. We seek the hierarchical arrangement of monophyletic groups, each defined by shared derived (synapomorphic) characters inherited from the direct common ancestor, and this implies that the homology hypothesis has been tested seriously for each of these characters. In this hierarchical ar- rangement, autapomorphic characters or character states assumed to mark an ancestral species are recognized as syna- pomorphic only in the immediate descendants; such char- acters ‘become’ symplesiomorphic (shared primitive) features as soon as one of these direct descendants is considered as the ancestor of a group of descendants of its own, so that new characters (different from the now symplesiomorphic ones) must be used to define that sub-group. A ‘general pattern’ of behavior can thus represent a symplesiomorphic character when a distinct ‘variant’ of the pattern appears as a synapomorphic character allowing definition of a mono- phyletic group. In other words, the ‘chronological table’ is reset for each character at a precise point of the ‘transforma- tion’ line. It would of course be helpful if we could use the ‘chronological table’ of ontogenetic development as an in- dication of the historical sequence of events. However, ‘recapitulation’ turns out to be rather incomplete due to the relative independence of the different morphogenetic time- tables within organisms which results in heterochronic shifts through evolutionary time. The temporal order of the onto- genetic appearance can be disturbed in particular by size dif- ferences. For example, in young loliginid squids, the lateral spreading of the tentacles during hovering is the ontogenetical- ly earliest arm display (Fig. 1); the fact that it appears earlier (in fact soon after hatching) in Sepioteuthis sepioidea (Blain- ville) than in Loligo vulgaris Lamarck does not mean that one situation is more ‘primitive’ than the other. The difference is apparently related to the size difference of the hatchlings, S. sepioidea being ontogenetically more ‘advanced’ due to the larger size of the egg from which it has hatched. From the early tentacle display, a variety of V-shaped tentacle and arm postures develop during juvenile life, and here we finally observe what could be a truly apomorphic variant in Sepioteuthis sepioidea. In contrast to the other forms of V-shaped postures in which ‘‘the left and right groups of arms are separated’? (Moynihan and Rodaniche, 1982: 77) and ‘‘can point forward, upward, or downward”’, the so-called Split posture is asymmetrical: ‘‘The arms are spearated to left and right as in simpler V’s, but one side is pointed upward while the other side is simultaneously pointed downward’. In contrast, a V-shaped arrangement of the arms pointed forward is so generalized in swimming or hovering cephalopods (including octopodids in slow backward swim- ming) that it seems rather difficult to single out a distinct variant of symmetrical spreading, apart from the question whether the underlying ‘general pattern’ is really homologous. Although the recognition of homologous characters of behavior can be very difficult across higher taxa, an apomor- phic character will still be discrete if it is not the variant of a generalized pattern concerning the whole arm crown. A BOLETZKY: CEPHALOPOD ARM CROWN 63 5 Fig. 1. Loligo vulgaris, one month old, hovering with tentacles spread, seen from the water surface (total length = 15 mm). Fig. 2. Sepietta neglecta, adult, covering itself with sand: a, the animal sits on the substratum, the arms are held under the head; b, the animal is now half-buried due to the action of the funnel (jets of water directed alternatingly forwards and backwards have ‘fluidized’ the sand surface momentarily, so that the animal ‘sinks’ into it); ¢, the animal at the end of the first burying phase, covered nearly entirely; d, e, second phase characterized by ‘sand gathering’ achieved by the dorsolateral arms (arm II). Note the traces of this arm action on the sand surface (total length = 30 mm). Fig. 3. Sepietta obscura, subadult, hovering with arm display. Note dorsolateral arms raised and curled, and ventolateral arms spread downwards with tips curled (total length = 50 mm). Fig. 4. Light micrograph of apical view of a living embryo of Sepia officinalis at stage IX of Naef (1928) showing organ rudiments as dark structures lying on the yolk mass. Around the central palleovisceral complex with the mantle (m) and collar (c) lie the eye (e) and stomodeum (sd) complexes and the paired rudiment of the funnel tube (ft); the latter has just separated from the arm rudiments V and becomes connected with the collar above the statocyst (st). Note the distance between arm rudiments I and II, and the inconspicuousness of arm rudiment IV from which the long tentacles are formed during subsequent stages (field width = 5 mm). Fig. 5. Post-organogenetic stage of Octopus vulgaris (preserved specimen in lateral SEM view , field width = 1.6 mm); the outer yolk sac (below) is only partly visible; the arms are numbered according to the homology hypothesis presented in this paper; the bases of the lateral arms reach around the eye (e) and form the folds (arrows) from which the primary lid is made at subsequent stages. 64 AMER. MALAC. BULL. 10(1) (1993) striking example is the special action of one pair of arms in a precisely defined situation in two sepiolid subfamilies, the Rossiinae and the Sepiolinae. These animals bury themselves in sand during daylight hours, as do other benthic cephalopods such as Sepia. The main work involved in sand covering 1s achieved by blowing up the substratum particles from underneath the body by water jets from the funnel (with vigorous fin movements counteracting the locomotory effect of the funnel jets). In contrast to Sepia, which finally smoothes the sand cover by a few rapid contractions of the dorsal mantle skin, all the benthic sepiolids studied thus far show a very peculiar ‘second phase’ in sand covering: the dorsolateral arms (Fig. 2) are stretched out to about the double normal length and again retracted and partly coiled in a sweeping movement by which sand particles are drawn to the head; this movement is repeated many times in all directions around the head (Boletzky and Boletzky, 1970). This strictly symmetrical pattern of highly sterotyped arm movements clearly represents a synapomorphic character of the Rossiinae and the Sepiolinae. In the third sepiolid subfamily, the pelagic Heteroteuthinae, such a movement pattern proper to the dor- solateral arms is not known. If it is indeed absent, it could a) have been secondarily suppressed; b) have been trans- formed into an autapomorphic variant; or c) have never existed. Evidence for ‘c’ would have to come from other characters indicating that the Heteroteuthinae form an ‘outgroup’ (i.e. stem from an ancestral form older than the common ancestor of Rossiinae and Sepiolinae). Whatever the answer, it must be emphasized that this peculiar action pat- tern is fully ‘compatible’ with the participation in other ac- tions where the dorsolateral arms play a less spectacular role, especially during certain arm displays (Fig. 3; cf. Mauris, 1989). In other words, when the apparently ‘new’ program for the particular actions of the second arm pair in sand cover- ing was assembled in the ancestral form (for which we have no age indication), it had to be integrated in a set of other programs already existing and functioning. BEHAVIORAL HOMOLOGIES AND NEURAL PATHWAYS Packard and Hochberg (1977: 226) present figures of the strikingly similar arm display called ‘flamboyant’ in a young octopus and in a young cuttlefish. They (op. cit.) raise the question: “‘If these patterns are considered homologous one with another, then what is the status of homology? It is not strictly a morphological one’’ (p. 225). They suggest: ‘*Perhaps we should look upon them as behaviourally homo- logous - i.e. that they have the same behavioural origins’’, and they indicate what that implies : ‘‘If the selective advan- tage is in the behaviour itself, then natural selection may be relatively indifferent to the details of the means by which it is produced - so long as the means are effective’ (p. 227). This proposition comes very close to what has been men- tioned earlier concerning the ‘‘source of cells’’that seems to be irrelevant ‘‘as long as they interact in a coherent and con- servative manner to produce a certain part of the body’’ (Wagner, 1989: 1163). The question then is whether behavioral homology can still be defined like a physical structure (‘‘a certain part of the body’’), or whether one could as well be dealing with similar patterns that have evolved independently using homologous effector systems, a special case of convergence called ‘homoiology’ (Lorenz, 1978; Tembrock, 1989). Although Packard and Hochberg (1977: 228) do not address explicitly this question, it is dismissed implicitly by their state- ment that “‘patterns are not a matter of a single set of elements or structures, but of all of them working together - in the skin, in locomotory structures and in modes of action - and given coherence by the brain. It is the repertoires read out from the central nervous programmes that have been selected in evolution’. Ultimately then, behavioral homology should be expressed in the ‘directions for use’ rather than in the ‘cir- cuit diagram’ of the nervous system. But so far we have ac- cess only to the ‘circuit diagrams’ which are so complex that they defy our capacity to sort out structural homologies at that level. Nonetheless, the comparative studies on the cen- tral pathways of the brachial nerves in decapods and octopods (Budelmann and Young, 1985, 1987) show the great poten- tial of morphological analysis of the nervous system building on comparative studies that deal with the general correlations between brain structures and functional adaptations to ecological contexts (Young, 1988). It thus becomes - at least theoretically - possible to map all the neural connections be- tween the periphery of the effector systems and sensory struc- tures and the higher integration centers. Based on the results of centripetal cobalt filling of the brachial nerves of the third arm in Sepia and Loligo, Budelmann and Young (1987: 349) observe ‘‘a general similarity to the organization found by fillings of the brachial nerves of octopods, but with some dif- ferences (Budelmann and Young, 1985)’. As in Octopus also the third arm was used to allow comparison, and consider- ing that in the decapod species “‘differences may exist... for the pathways of the tentacle nerve’? (Budelmann and Young, 1987: 346), the question of the morphological homology between the decapod tentacle and the third arm of octopods (Naef, 1928) becomes interesting. The hypothesis of homology between these arms in decapods and octopods can be tested on the basis of structural properties and posi- tional relationships, but the latter must take account of cer- tain devlopmental features. DEVELOPMENTAL MORPHOLOGY OF THE COLEOID ARM CROWN In morphogenetic terms, the arm crown of cephalopods BOLETZKY: CEPHALOPOD ARM CROWN 65 arises from a paired series of embryonic cell concentrations that lie on either side of the body and head anlage. The situa- tion of an early organogenetic stage of a cephalopod embryo is best visualized when one is looking from the animal pole along the main egg axis (i.e. in apical view); the arm rudiments then appear as a series of knobs arranged in two curved bands (Fig. 4). At their posterior end these bands ap- proach one another close to the sagittal plane, whereas the anterior ends of the anlage lie far apart, forming a roughly U-shaped complex. It is subdivided into the individual arm rudiments (‘knobs’ or ‘buds’), five on either side in decapod embryos, four in octopods. From the inner end of the posterior arm rudiments (close to the bottom of the U), a paired streak splits away to form the folds from which the funnel tube will be made (Naef, 1928). The common morpho- genetic origin of the arms and the funnel tube (the funnel pouch or collar having a different origin) is also demonstrated by the close connection of the ganglionic masses forming the brachial and pedal lobes, from which the cerebral ganglion (forming the supraoesophageal parts of the brain) are clear- ly separated (Marquis, 1989). The further differentiation of the four or five pairs of arm buds is characterized by the formation of small surface folds on the lower side of the arm (facing the outer yolk sac of the embryo) from which suckers differentiate. Along with the increase of arm length, the whole crown progressively contracts so that the open end of the U closes above the in- vaginated buccal mass. The resulting ring shape of the arm crown base forms the virtually radial arrangement of the arms from late embryonic stages. During arm growth at advanced organogenetic stages, two pairs of folds form on the aboral side of the arm crown base (i.e. opposite the suckered arm face). One pair rises from the bases of the arms that are adjacent to the posterior arms (from which the funnel tube rudiments have split), the other pair of folds arises from the bases of the following arms. In the apical view considered up to now, these folds are seen to grow upwards, the first mentioned pair behind the eye, the second in front of the eye. In contrast to this so-called morpholocial (or embryological) orientation, the so-called physiological orientation considers the embryo in an orien- tation resulting from a 90° rotation backwards, so that the arms come to lie anteriorly, the funnel tube below (‘ventral- ly’), and the mantle tip posteriorly. The two pairs of folds described above then appear to grow backwards around the eye complex (Fig. 5). The lower fold is the one arising from the base of the arm which can now be called ventrolateral, the upper fold arises from the lateral or dorsolateral arm of each side. These folds form the ‘primary lid’ which delimits the future orbital cavity surrounding the eye; they are the most important morphological markers for the recognition of arm homologies in coleoid cephalopods! Among the criteria allowing one to recognize homolo- gous structures, positional relationship is crucial (Remane, 1952; Riedl, 1978). It has been mentioned above that clearly homlogous structures are not always differentiated along strictly identical pathways. However, if one observes similar morphogenetic processes leading to the differentiation of organs that appear homologous on the basis of other criteria, and especially if there is not a single exception available to question the morphogenetic similarity, then the hypothesis of homology is greatly reinforced by that similarity. This is exactly what the arm crown development in coleoid embryos tells us about the homology of arms. Unfortunately, the picture is blurred by the conven- tional numeration of arms in decapods and octopods. By ex- cluding the tentacles from the numbered series of arm pairs, one unnecessarily obscures arm homologies. On the basis of the primary lid development, Naef (1928) emphasized that the tentacles of decapods and the ventrolateral arms of octo- pods are homologous (hence also the ventral arms in both groups). For the homologies of the lateral or dorsolateral, and of the dorsal arms, Naef suggested that arms II and III of decapods and arms I and II of octopods are homologous based on the argument that the dorsal arm rudiment in decapod embryos tends to lie removed from the dorsolateral bud II in early organogenesis (Fig. 4). But this hypothesis is ruled out by the positional relationship between the arm rudiments and the so-called meta-brachial vesicles; these are formed from two pairs of ectodermic invaginations, the dor- sal pair lying between the bases of dorsal and dorsolateral arms in decapods and in octopods. Unless positional rela- tionshps between the arms and the vesicles have changed (which is highly unlikely), the position of the dorsal pair of vesicles demonstrates the homology between the second (dorsolateral) octopodan arm pair and either the second or third arm pair of decapods (Boletzky, 1978-79). Loss of the second arm pair in the octopodan ancestor is suggested by the arm crown morphology of Vampyroteuthis, in which the second of a total of five arm pairs is rudimentary (forming the retractile filaments). This peculiar cephalopod probably represents a lineage from which the octopodan line became separated (see Young, 1977; Bandel and Boletzky, 1988). There is no direct morphogenetic evidence that it is actually arm II that is missing in octopods, because theoretically the dorsal fold of the primary lid could have been taken over by the second arm if formation of the arm III had been sup- pressed. In conclusion, there is good evidence of homology for the dorsal, ventrolateral (or tentacle), and ventral arms, respectively; and there is the acceptable hypothesis of absence of arm II in the octopods. What remains to be discussed is the problem of buc- cal arm homology. Among the “‘general cephalopod features’’ Naef (1928: 82) described the division of the brachial com- plex into an ‘‘outer crown of tentacles and an inner one of buccal arms (Buccaltrichter) which surround the mouth as 66 AMER. MALAC. BULL. 10(1) (1993) a third lip; they are more or less rudimentary in the dibranch- lates’? (whereas they are known to form a well-developed complex in Nautilus). Naef emphasized that the buccal arms appear very late in decapod development (his detailed descrip- tion of Sepia development is very conclusive) and interpreted the absence of buccal arms in octopods as the result of com- plete suppression. Given the positional relations between the ‘outer crown’ and the funnel tube, on the one hand, and with the primary lid, on the other (to which we can now add the position of the metabrachial vesicles), it would indeed be dif- ficult to consider the octopodan arm crown as_non- homologous to the ‘outer crown’ of decapods. This non- homology is actually proposed by Zell (1988: 87) who con- cludes from his analysis of adult characters that ‘‘in all pro- bability, the arms of ocotopods are not homologous to the sessile arms and tentacles but to the buccal arms of the decapods ’’. His main arguments are structural resemblances in the muscle arrangement. Positional relationships of arms and buccal arms relative to the mouth are discussed without any reference to the developmental aspects described by Naef (1928) and more recent authors (see Boletzky, 1978-79 for literature). ARM FUNCTIONS AND DEVELOPMENTAL CONSTRAINTS IN EVOLUTION Considering that decapodan and octopodan arm crowns are probably homologous, and that within these arm crowns one can recognize arm identities at five positions in decapods and at four positions in octopods (with well established homologies for three of these positions), it becomes possi- ble to approach the respective roles played by the different arms. To be exhaustive, a list of the functions of individual arm pairs would have to take account of variations due, for example, to the presence or absence of an interbrachial mem- brane, or its varied extension achieved by muscular action. Or when dealing with prehensile function in predation, mating or egg laying of oegopsid squids, the presence or absence of transformed suckers (hooks) would have to be specified. And wherever arm extension, spreading or curling appears to play a visual role in intra- or interspecific interaction, the question of concomitant effects outside the signal function would have to be considered. Needless to say, such a com- plete list of possibilities cannot be provided now. To reach that goal it will be necessary to take account of the developmental morphology of individual arm pairs, because group-specific modifications in the late embryonic ‘programs’ have set the stage for functional specialisations in arm crown evolution. Let us consider the decapod tentacles (modified arms IV) from this point of view. In all decapods, with one excep- tion as far as is known, the tentacle rudiments grow to a size at least equal to the size of other arms in the hatchling, and they immediately serve in prey capture. In the ommastrephid squids (where the arm buds III and V are ‘retarded’ in early development) the tentacle rudiments are drawn together at late embryonic stages, and they become fused to form a ‘proboscis’ carrying suckers at the anterior end which represents the rudimentary tentacle clubs. The subsequent juvenile development culminates in the progressive separa- tion of the two tentacles (see Okutani, 1987 for a review). The transitory fusion can be viewed as the result of an evolu- tionary optimisation process in the line leading to the om- mastrephid ancestor; after all, the separate adult tentacles can be joined at the club base during tentacle ejection in adult ommastrephids! On the other hand, instead of invoking a pro- gressive displacement of an adult character to early onto- genetic stages, one might envisage a developmental constraint as possible cause; it could have been the small embryo size that switched morphogenetic ‘canalization’ over to a new route which proved adaptive. It is interesting that the sole exception among decapods where tentacles are differentiated only during post-embryonic development is found in the pygmy cuttlefish /diosepius, which produces very small eggs. In contrast to the om- mastrephid embryo where the tentacles are fully formed (though fused) while two other arm pairs are retarded, Idiosepius forms all the ‘sessile’ arms during embryonic development, while the tentacle rudiments remain arrested at the early bud stage. This means that in very young Idiosepius prey capture must be achieved by the other arms, in a way similar to the prey capture in planktonic young octo- puses (pouncing attack). When the tentacles of Idiosepius have grown to full length, they apparently serve in prey capture; however, in the adult female they are also used during spawning, to stick the eggs to the substratum (Natsukari, 1970). In all other decapods it is one or several other arms that fulfill this special prehensile function. This observation finally raises the question of tentacle functions other than seizure of prey. Is the /diosepius female really unique in using the tentacles for something other than feeding? Certainly not; it is apparently unique only in hav- ing tentacles participate in spawning. Moreover, this action appears as reverse prehension when compared to prey cap- ture: instead of seizing prey with extended tentacles, then retracting the tentacles and releasing the prey item to the sessile arms (Kier, 1982, 1985), the Idiosepius female receives the egg expelled through the funnel in her arms, then the tentacles seize the egg and finally extend to release it as soon as it sticks to the substratum (Natsukari, 1970) (Fig. 6). Apart from this peculiar behavior pattern, it has been said already that tentacles can participate in various arm displays in squids (Moynihan and Rodaniche, 1982; Hanlon, 1988; Porteiro et al., 1990), even in very young animals (Boletzky, 1987). In these instances, no prehensile action BOLETZKY: CEPHALOPOD ARM CROWN 67 Fig. 6. A semi-schematic presentation of egg laying in /diosepius pygmaeus paradoxus: A, the female is attached with its dorsal adhesive organ to the hard substratum (tank wall), close to eggs previously laid; B, an egg expelled through the funnel is seized with the tentacles; C, the egg is fixed to the substratum by the tentacles; D, the female moves a short distance in prepar- tion for the following egg deposition (from Natsukari, 1970). occurs. What is more intriguing is the fact that prehensile tentacle action may not occur during feeding. Very young Sepia always use their tentacles in prey capture, but older individuals develop an alternative pouncing attack which is generally used for the seizure of slow moving prey animals (Messenger, 1968; Duval et al., 1984). During the position- ing phase (with binocular fixation) the animal makes a deci- sion whether to use the tentacles or make a pouncing attack with the arms. This decision may be influenced by previous failure to catch a particular prey item (Boletzky, 1972). Sepia is special in that the tentacles when not in use are fully retracted in tentacle pouches; nevertheless, arm attacks without use of the tentacles occur also in Loligo (Kier, 1982) and in ommastrephid squids (Bradbury and Aldrich, 1969; Flores, 1983). Considering all the data on tentacle development and function in decapods, it is reasonable to assume that the whole action pattern of ‘tentacle shooting’ in prey capture is in- herited from an ancestor that had arm IV differentiated for rapid extension (Kier, 1982, 1985), the emphasis being on rapid, because slow extension occurs in the arms of both decapods and octopods! The ‘fixation’ of the tentacle pro- gram in early ontogenetic stages has been cancelled apparent- ly only in the idiosepiid ancestor. Although we thus have good reason to consider the tentacle ejection complex (which is coupled with binocular fixation of the target) as a homologous, indeed synapomorphous character of decapods, there is no evidence as yet that the arm attack pattern (which is essentially defined by the ‘negative’ feature of tentacles not being used) of cuttlefish, loliginid and ommastrephid squids is also homologous. Here homology is a likely guess, but homoiology also is conceivable. The inverse situation, with homoiology as the favored hypothesis and actual homology at best conceivable, appears when the octopods are included in the comparative considera- tion of tentacles. The enlargement of the ventrolateral arms in Octopus defilippi Vérany and related species of the ‘Macrotritopus’ group is very unlikely to represent a true homology, because no trace of this feature appears in the cir- rate octopods and in Vampyroteuthis. This incidentally removes also the possibility of deriving the special *prehen- sile’ function of the incirrate hectocotylus from the decapodan tentacle function as it appears in prey capture. CONCLUSIONS In focusing attention on ‘what the arms do’ in cephalopod behavior, the foregoing discussion has somewhat artifically eliminated the integration of arm postures in visual- ly effective “body patterns’ (cf. Fig. 3). However, this aspect has been extensively covered by Packard and Sanders (1969, 1971), Moynihan and Rodaniche (1982), Hanlon and Messenger (1988) and Hanlon (1988). These studies deal with colorful inshore species that can be observed by SCUBA divers, and that can also be kept in the aquarium. Some can even be reared under fully controllable situations. With the development of deep-diving research submer- sibles, a new area has been opened for field observations of living cephalopods, improving the conditions of study com- pared to the earlier investigations using remote control cameras (Roper and Brundage, 1972). In the deep sea, color plays no major role. In the aphotic zone, the arms of cephalopods can take on visually effective signal functions only if they carry light organs. Arm length and movement patterns could thus be adapted to that particular role along with the functional integration of prey capture and swimming 68 AMER. MALAC. BULL. 10(1) (1993) or hovering. To fulfill optimally these latter functions, an arm crown devoid of light organs could be structured differently. The cirrate octopods present a variety of examples emphasiz- ing the integrated function of very long arms with an enormously developed interbrachial membrane, which can become subdivided into an outer web attached to the arms via a so-called intermediate web (see Voss, 1988 for a review). However, the juveniles of these forms start out with very dif- ferent body proportions including especially short arms and very large fins (Boletzky, 1982). Moreover, these short arms are still devoid of the cirri, so that the suckers alone must achieve what suckers and cirri do in the adult. These are just a few aspects to which attention should now be drawn. An equally interesting field is the functional morphology of the outer and inner arm crowns of Nautilus and the related behavioral features. Nautilus uses its numerous tentacles in ways similar to certain arm actions in modern coleoid cephalopods, the ancestor of which must have been characterized by an arm crown comprising only ten arms (Naef, 1928). The guiding thread through comparisons between any of the examples of arm actions mentioned is the attempt to recognize homologies, because the assumption of homology always remains hypothetical. In some cases, the homology hypothesis proves very robust, especially when built on the observation of structural complexity and its ontogenetic for- mation. In other cases, superficial resemblances suggest homology but may not stand rigorous tests. This is the crux of many behavioral features, but when it comes to compar- ing arm actions, it is at least helpful to have an idea of arm homologies. The comparative method provides slow but steady pro- gress in understanding the ‘natural history’ of organisms. Robust hypotheses do not appear out of the blue. As Young (1977: 430) wrote in his ‘phylogenetic considerations’, the con- cluding section of a survey of “Brain, Behaviour and Evolu- tion of Cephalopods’: ‘‘We are gradually learning how to piece the information together and to correlate it with studies of the structure and organization of the brain. Such knowledge, gained inductively, can be used, with caution, for deductions about the probable habits of species not yet investigated at sea’’. ACKNOWLEDGMENTS I thank Drs. John Arnold, Bernd-Ulrich Budelmann and Roger Hanlon for their critical reading of the manuscript. The helpful suggestions of two anonymous reviewers are also gratefully acknowledged. LITERATURE CITED Bandel, K. and S. v. Boletzky. 1988. Features of development and functional morphology required in the reconstruction of early coleoid cephalopods. /n: Cephalopods - Present and Past. J. Wiedmann and J. Kullmann, eds. pp. 229-246. Schweizerbart’sche Verlagsbuch- handlung, Stuttgart. Bock, W. J. 1989. The homology concept: its philosophical foundation and practical methodology. Zoologische Beitrage N. F. 32:327-353. Boletzky, S. v. 1972. A note on aerial prey-capture by Sepia officinalis (Mollusca, Cephalopoda). Vie et Milieu 23(1A):133-140. Boletzky, S. v. 1978-79. Nos connaissances actuelles sur le développement des Octopodes. Vie et Milieu 28-29(1AB):85-120. Boletzky, S. v. 1982. On eggs and embryos of cirromorph octopods. Malacologia 22:197-204. Boletzky, S. v. 1987. Juvenile behaviour. In: Cephalopod Life Cycles, Vol. II. P. R. Boyle, ed. pp. 45-60. Academic Press, London. Boletzky, S. v. and M. V. v. Boletzky. 1970. Das Eingraben in Sand bei Sepiola und Sepietta (Mollusca, Cephalopoda). Revue suisse de Zoologie 77:536-548. Budelmann, B. U. and J. Z. Young. 1985. Central pathways of the nerves of the arms and mantle of Octopus. Philosophical Transactions of the Royal Society of London B 310:109-122. Budelmann, B. U. and J. Z. Young. 1987. Brain pathways of the brachial nerves of Sepia and Loligo. Philosophical Transactions of the Royal Society of London B 315:345-352. Bullock, T. H. 1965. Physiological Bases of Behavior. In: Ideas in Modern Biology. J. A. Moore, ed.pp. 449-482. The Natural History Press, Garden City, New York. Bradbury, H. E. and F. A. Aldrich. 1969. Observations on feeding in the squid Illex illecebrosus illecebrosus (LeSueur, 1821) in captivity. Canadian Journal of Zoology 47:913-915. Dohle, W. 1989. Zur Frage der Homologie ontogenetischer Muster. Zoologische Beitrage N. F. 32:355-389. Duval, P., M.-P. Chichery and R. Chichery. 1984. Prey capture by the cut- tlefish (Sepia officinalis L.): an experimental study of two strategies. Behavioural Processes 9:\3-21. Flores, E. E. C. 1983. Laboratory observations on the visual attack of the squid 7odarodes pacificus. Memoirs of the National Museum of Vic- toria 44:205-227. Hanlon, R. T. 1988. Behavioral and body patterning characters useful in taxonomy and field identification of cephalopods. Malacologia 29:247-264. Hanlon, R. T. and J. B. Messenger. 1988. Adaptive coloration in young cut- tlefish (Sepia officinalis L.): the morphology and development of body patterns and their relation to behaviour. Philosophical Transactions of the Royal Society of London B 320:437-487. Kier, W. M. 1982. The functional morphology of the musculature of squid (Loliginidae) arms and tentacles. Journal of Morphology \72:179-192. Kier, W. M. 1985. The musculature of squid arms and tentacles: ultrastructural evidence for functional differences. Journal of Morphology 185:223-239. Lorenz, K. 1978. Vergleichende Verhaltensforschung - Grundlagen der Ethologie. Springer Verlag, Wien. 307 pp. Marquis, F. 1989. Die Embryonalentwicklung des Nervensystems von Octopus vulgaris Lam. (Cephalopoda, Octopoda), eine histologische Analyse. Verhandlungen der Naturforschenden Gesellschaft in Basel 99:23-7S. Mauris, E. 1989. Colour patterns and body postures related to prey capture in Sepiola affinis. Marine Behaviour and Physiology 14:189-200. Messenger, J. B. 1968. The visual attack of the cuttlefish, Sepia officinalis. Animal Behaviour 16:342-357. Moynihan, M. and A. F. Rodaniche. 1982. The behaviour and natural history of the Caribbean Squid Sepioteuthis sepioidea. Advances in Ethology, Supplements to Journal of Comparative Ethology 25:1-151. Naet, A. 1928. Die Cephalopoden, 1. Teil, 2. Band. Fauna e Flora del Golfo di Napoli, 35:1-357. Natsukari, Y. 1970. Egg-laying behavior, embryonic development and hatched larva of the Pygmy Cuttlefish, /diosepius pygmaeus paradoxus BOLETZKY: CEPHALOPOD ARM CROWN 69 Ortmann. Bulletin of the Faculty of Fisheries, Nagasaki University, 30:15-29. Okutani, T. 1987. Juvenile morphology. Jn: Cephalopod Life Cycle, Vol. Il. P. R. Boyle, ed. pp. 33-44. Academic Press, London. Packard, A. and G. D. Sanders. 1969. What the octopus shows to the world. Endeavour 28:92-99. Packard, A. and G. D. Sanders. 1971. Body patterns of Octopus vulgaris and maturation of the response to disturbance. Animal Behaviour 19:780-790. Packard, A. and F. G. Hochberg. 1977. Skin patterning in Octopus and other genera. In: The Biology of Cephalopods. M. Nixon and J. B. Messenger, eds. Symposia of the Zoological Society of London 38:191-231. Porteiro, F. M., H. R. Martins, and R. T. Hanlon. 1990. Some observa- tions on the behaviour of adult squids, Loligo forbesi, in captivity. Journal of the Marine Biological Association of the United Kingdom 70:459-472. Remane, A. 1952. Die Grundlagen des Natiirlichen Systems, der vergleichenden Anatomie und der Phylogenetik. Geest und Portig, Leipzig. 400 pp. Riedl, R. 1978. Order in Living Organisms. John Wiley & Sons, Chichester, New York. 313 pp. Roper, C. F. E. and W. L. Brundage. 1972. Cirrate octopods with associated deep sea organisms: new biological data based on deep benthic photographs (Cephalopoda). Smithsonian Contributions to Zoology 121:1-46. Roth, V. L. 1991. Homology and hierarchies: Problems solved and unre- solved. Journal of Evolutionary Biology 4:167-194. Tembrock, G. 1989. Homologisieren in der Ethologie. Zoologische Beitrage N. F. 32:425-436. Van Valen, L. M. 1982. Homology and causes. Journal of Morphology 173:305-312. Voss, G. L. 1988. Evolution and phylogenetic relationships of deep-sea octopods (Cirrata and Incirrata). In: The Mollusca, Vol. 12 Paleon- tology and Neontology of Cephalopods. M. R. Clarke and E. R. Trueman, eds. pp. 253-276. Academic Press, New York. Wagner, G. P. 1989. The origin of morphological characters and the biological basis of homology. Evolution 43:1157-1171. Young, J. Z. 1977. Brain behaviour and evolution of cephalopods. Jn: The Biology of Cephalopods. M. Nixon and J. B. Messenger, eds. Sym- posia of the Zoological Society of London 38:377-434. Young, J. Z. 1988. Evolution of the Cephalopod Brain. In: The Mollusca, Vol. 12 Paleontology and Neontology of Cephalopods. M. R. Clarke and E. R. Trueman, eds. pp. 215-228. Academic Press, San Diego, New York, London. Zell, H. 1988. Zur Histologie und Homologie von Cephalopodenarmen. Archiv fir Molluskenkunde 119:87-106. Date of manuscript acceptance: 3 September 1991 Reproductive anatomies of Holospira spp. (Gastropoda: Pulmonata: Urocoptidae) from Arizona and Sonora with a new subgenus and a new subspecies* Lance H. Gilbertson Department of Biology, Orange Coast College, P. O. Box 5005, Costa Mesa, California 92628, U. S. A. Abstract. The reproductive systems of seven species of Holospira from Arizona and Sonora, Mexico are illustrated and compared. They exhibit four different morphologies based on the size, shape and arrangement of the penial complex and the spermathecal duct. These four morphological groups represent lineages that do not completely correspond to the three presently recognized subgenera from this region. Sonoraloa Gilbertson subg. nov. and Holospira dentaxis alamellata Gilbertson, ssp. nov. are described. H. remondi (Gabb, 1865), H. dentaxis Pilsbry, 1953 and H. mazatlanica (Bartsch, 1943) are transferred from subgenus Allocoryphe to Sonoraloa. Land snails comprising the urocoptid genus Holospira von Martens, 1860, exhibit moderate-sized (ca. 8-20 mm in length), elongated, turriform shells that retain their spire. They inhabit three southwestern states (Arizona, New Mexico, and Texas) and range southward through most of Mexico. Most species are very xerophytic and calcicolous. Three of the six subgenera recognized presently [Allocoryphe Pilsbry, 1946; Eudistemma (Dall, 1896); and Holospira s.s. von Martens, 1860] are represented in Arizona and Sonora, Mexico, the northwestern corner of the Holospira range. The taxonomy of the genus is based entirely on shell characteristics such as overall size, whorl size, shape of the aperture, sculpture and aspects of the internal column. However, these characters tend to be somewhat unreliable for systematic studies due to their variability. Thompson (1964) suggested that true interspecific relationships within the genus are best determined by the soft anatomy. Unfortunately, holospiras are poorly known in this regard. Descriptions of the reproductive systems of only a few species (11) have been published (Pilsbry, 1903; Gilbertson, 1989a, b; Thompson, 1964). Five of these are from Arizona and Sonora, Mexico (Gilbertson, 1989a, b). The present paper describes the reproductive anat- omies of three additional Holospira species from Arizona and two from Sonora. The anatomical configuration of the soft parts of the two Sonoran species, along with aspects of the embryonic whorls of their shells, indicates a separate lineage which requires the description of Sonoraloa subg. nov. MATERIALS AND METHODS Estivating snails were collected from under surface *This paper is part of the 1989 AMU symposium ‘‘Systematics, Anatomy, and Evolution of western North American land snails in honor of Walter B. Miller’’ [see AMB 8(2)]. rocks and/or dead plant material (especially agaves and yuccas) usually in areas of limestone deposits. The Mexican species were collected at sites in central Sonora, between Hermosillo and Sahuaripa (Fig. 1), including the Sierra Batamote. In Arizona, collections were made in the Dragoon and Huachuca Mountains. Mature specimens were immersed individually in a small vial of water for three to five days until drowned. The shell of each specimen was broken and removed carefully. Then the reproductive system of the animal was dissected free from the other internal organs and fixed in 70% ethanol. The reproductive organs were stained with Delafield Haematoxy- lin, destained with 3% acid alcohol, counter-stained with Eosin Y, and slide-mounted (see Gregg, 1959; Naranjo- Garcia, 1989). The reproductive systems of two to four specimens of each species were dissected and mounted. Institutions cited in this article are abbreviated as follows: ANSP, Academy of Natural Sciences of Philadelphia; LACM, Los Angeles County Museum of Natural History; SBMNH, Santa Barbara Museum of Natural History; UNAM, Universidad Nacional Autonoma de Mexico; USNM, United States National Museum-Smithsonian Institution. TAXONOMIC DESCRIPTIONS Genus Holospira von Martens, 1860 Subgenus Allocoryphe Pilsbry, 1946 (Figures 2, 3a) Type species: Holospira (Allocoryphe) minima von Martens, 1897, by original designation. The subgenus is monotypic. Type locality: Cerro de la Campana, Hermosillo, Sonora, Mexico. Diagnosis: Shell exhibits a rounded aperture, expanded American Malacological Bulletin, Vol. 10(1) (1993):71-81 71 72 AMER. MALAC. BULL. 10(1) (1993) GULF OF CALIFORNIA + BAJA CALIFORNIA Approx. limit of Holospira range AYN. Sierra Madre Occidental (A)—@)_ Holospira sites @ Cities 50 1 = —— KILOMETERS 150 ° 110 Tucson @ SONORA Fig. 1. Map of Arizona and Sonora, Mexico showing approximate locations of collecting sites of Holospira spp. discussed in this paper. A. H. (Eudistemma) tantalus campestris; B. H. (E.) danielsi; C. H. (E.) ferrissi; D. H. (Allocoryphe) minima; E. H. (H.) milleri; F. H. (Sonoraloa) remondi laevior; G. G. (S.) dentaxis alamellata spp. nov. peristome, hollow ribs and a somewhat enlarged, alamellate, internal column. Embryonic whorls flat-sided and angular at the upper, outer margin. Male anatomy with tubular epiphallus inserting laterally into penis producing a penial caecum. Penial retractor muscle broad, inserting on apex of penial caecum. Material examined: ANSP 166366, Cerro de la Campana, Hermosillo, Sonora, 55+ shells. SBMNH 35519, Cerro de la Mona, ca. 21 km east of Hermosillo, Sonora, N side of Hwy 15, 29°02.9°N, 110°39.4’W, elevation ca. 350 m, shell and slide-mounted reproductive system (illustrated specimens). SBMNH 35626, Cerro de la Mona (as above), 8 shells. LACM 88-354.1, Cerro de la Mona (as above), 7 shells. Remarks: Pilsbry (1946) erected the subgenus Allocoryphe (allo, different; coryph, top) in a footnote for the ‘‘special group” of holospiras inhabiting northwestern Mexico, be- tween the Sierra Madre Occidental and the Gulf of California (Fig. 1). This subgenus is restricted herein to Holospira minima, which is the only Holospira species known to ex- hibit angular, embryonic whorls (i.e. a ‘‘different top’’). These whorls are sculptured with ‘‘regularly spaced ribs that are overlayed by an open mesh of granular reticulations’’ (Thompson, 1988:92). Subgenus Sonoraloa Gilbertson subg. nov. (Figs. 3c, 4, 5, 6, 7) Type species: Holospira remondi (Gabb, 1865), designated herein. H. remondi, H. dentaxis Pilsbry, 1953, and H. mazatlanica (Bartsch, 1943) are transferred from subgenus Allocoryphe. Type locality: 1.5 leagues (ca. 6 km) from Arivechi, Sonora, Mexico. Distribution: Sonora and Sinaloa, Mexico. Diagnosis: Shell generally similar to that of subgenus Allocoryphe with regard to aperture, peristome, and whorl size but differing by having one lamella (the axial), or none, GILBERTSON: HOLOSPIRA REPRODUCTIVE ANATOMIES ve) Fig. 2. Holospira (Allocoryphe) minima. A. Apertural view of shell. B. Reproductive system (except albumen gland and ovotestis) (scale bars = 1 mm). on the internal column and having rounded, embryonic whorls. Male anatomy with a tubular epiphallus inserting apically on an ovate penis, adjacent to insertion of penial retractor muscle. Material examined: ANSP 25046, Holospira remondi, 1.5 leagues (ca. 6 km) from Arivechi, Sahuaripe Valley, Sonora, one lectotype. ANSP 166484, H. dentaxis, drift at ford of Rio Yaqui about 17.5 km N of Ciudad Obregon, Sonora, holotype, 3 paratypes. USNM 381625, H. mazatlanica, drift at Mazatlan, Sinaloa, Mexico, 4 paratypes. Etymology: Sonoraloa is a contraction of the names of the two northwestern Mexican states which comprise the known range of this subgenus, Sonora and Sinaloa. Remarks: Holospira dentaxis, H. mazatlanica and possibly H. remondi (including numerous subspecies) were described on the basis of river drift shells. Because of this, it is very difficult to locate living populations. However, populations referrable subspecifically to H. remondi and H. dentaxis have been located and their reproductive anatomies are discussed herein. Holospira kinonis Baily and Baily, 1940, is the only other species that Pilsbry (1953) placed in the subgenus Allo- coryphe. However, its shell is very slender and turriform, giv- ing it the external appearance of an Epirobia, and its em- bryonic whorls are rounded, but differ from those of the other species by not enlarging in diameter. Because of these shell features, and lack of anatomical data, its taxonomic status is uncertain. Holospira (Sonoraloa) remondi laevior Pilsbry, 1953 (Figs. 3c, 7) Synonym: Holospira remondi (Gabb, 1865) Material examined: ANSP 188316, Holospira remondi laevior, drift of Laguna Preza Rodriguez, Hermosillo, Sonora, holotype, 3 paratypes. SBMNH 35520, H. remondi laevior. Sierra Batamote, east-central Sonora (near El Milagro Mine), ca. 1,070 m, 20.2 km E of Rio Yaqui Bridge along Hwy 15 from La Estrella to Bacanora, 28°57.7’N, 109°30.5’W, Sonora, shell and slide-mounted reproductive system (il- lustrated specimens). SBMNH 35603, H. remondi laevior, Sierra Batamote (as above), 5 shells. LACM 88-355.1, H. remondi laevior, Sierra Batamote (as above), 6 shells. Remarks: The cylindroconic shells of the present popula- tion from the Sierra Batamote are slightly longer (mean length = 14.6 mm for Il shells) than that of holotype H. remondi laevior (length = 13.8 mm) but overlap occurs (range = 13.4 - 17.0 mm). Their somewhat inflated internal columns (ca. 0.18-0.20 of shell width) suggest a close relationship with genus Coelostemma Dall, 1895. Bequaert (1973) synonomyzed Holospira remondi laevior (and all other subspecies) with the nominate subspecies without comment. It is considered valid herein due to the characteristically elongated, cylindric portion of its shell. Holospira (Sonoraloa) dentaxis alamellata Gilbertson ssp. nov. (Figs. 4-6) Diagnosis: Shell comparable to Holospira dentaxis dentaxis Pilsbry, 1953, in overall shape, shape of the aperture, whorl size and sculpture but slightly larger and alamellate (holotype H. d. dentaxis is 11.0 mm in length, 3.5 mm in diameter, 13 whorls and has an axial lamella). Shell of holotype: Shell medium-sized for genus, moderately cylindroconic in shape, convex with ninth whorl of greatest diameter, tan in color. Whorls very convex, 12.1 in number. Embryonic whorls 2.4 in number, rounded, minutely granular, increasing in diameter. All post-embryonic whorls sculptured with moderate-sized, solid, axial ribs which are slightly retractively slanted and arched, having intercostal spaces ca. three times width of ribs. Penultimate whorl with 31 ribs. Aperture slightly auriculate; peristome extended somewhat from body whorl and moderately expanded except on upper-outer margin. Umbilicus narrowly perforate. Internal column smooth, slightly expanded, increasing in diameter apically. Maximum length 11.7 mm, diameter 4.0 mm. Paratypes: Seven paratypes vary from turriform to cylindro- conic in shape. They range from 10.9 to 11.9 mm in length (mean 11.5 mm) and 3.7 to 4.0 mm in diameter (mean 3.9). Radula: Radula typical for genus. Formula 16-6-1-6-16 (= 45 74 AMER. MALAC. BULL. 10(1) (1993) Fig. 3. Basal reproductive organs of Holospira spp. representing the four lineages discussed herein. A. H. (Allocoryphe) minima. B. H. (Holospira) milleri. C. Holospira (Sonoraloa) remondi laevior. D. H. (Eudistemma) danielsi. Abbreviations: EP, epiphallus; PC, penial caecum; PE, penis; PR, penial retractor muscle; SD, spermathecal duct; VD, vas deferens; VE, verge (scale bar = | mm). GILBERTSON: HOLOSPIRA REPRODUCTIVE ANATOMIES 75 Fig. 4. Holospira (Sonoraloa) dentaxis alamellata ssp. nov. A. Apertural view of holotype. B. Dorsal view of paratype opened to show internal column (scale bar = 1 mm). teeth per row). Reproductive system: Penis small and slightly ovate due to a constriction at junction with genital atrium. Vas deferens enlarging into tubular epiphallus which inserts apically into penis. Spermathecal duct rounded basally with internal folds, proceeding apically into short constricted section, and then enlarging with scalloped-appearing interior; spermatheca tapering into duct. Spermathecal diverticulum lacking. Type locality: Riperian gorge, S side of Hwy 15, 7.5 road km W of Sahuaripa, 29°01.5’N, 109°18.7’W, elevation ca. 700 m, Sonora Mexico. Disposition of types: Holotype: SBMNH 35514. Paratypes: ANSP 389889; LACM 2246; UNAM-IB 190; USNM 860570. SBMNH 35515, slide-mounted reproductive system (illustrated specimen). SBMNH 35593, 35594, two slide- mounted reproductive systems. Etymology: This subspecies is named for the absence of lamellae on the internal column. Remarks: Holospira dentaxis dentaxis Pilsbry, 1953, H. d. lamellaxis Pilsbry, 1953, H. d. striatella Pilsbry, 1953 and H. d. pomatia Pilsbry, 1953 were described on the basis of drift shells from the mouth of the Rio Yaqui. Bequaert (1973) synonomyzed all subspecies with the nominate subspecies. DESCRIPTIONS OF REPRODUCTIVE ANATOMIES Holospira species inhabiting Arizona and Sonora, Mexico exhibit at least four distinctly different arrangements (= lineages) of the reproductive system based on variations of the male genitalia and the female spermathecal duct. Other female organs such as the free oviduct, uterus and albumen gland are morphologically similar, varying only in length. In all cases, the spermathecal duct is completely separate from the free oviduct; hence, a vagina is lacking. A spermathecal diverticulum is present in the Arizona species (lineage 1) and is lacking in all three of the Sonoran groups (lineages 2-4). LINEAGE |: Subgenus Eudistemma (Dall, 1896) Holospira (Eudistemma) danielsi Pilsbry and Ferriss, 1915 (Figs. 8, 9) Material examined: SBMNH 35516, foothills on N side of mouth of Cochise Stronghold Canyon West, Dragoon Mts., 31°56.0°N, 110°00.1’W, elev. ca. 1615 m, Cochise Co., Arizona (at or near type locality), one slide-mounted reproductive system (illustrated specimen). SBMNH 35595, 35596 (data as above), two slides. Reproductive system: The male anatomy is characterized by Fig. 5. Reproductive system of Holospira (Sonoraloa) dentaxis alamellata spp. nov. Abbreviations: AG, albumen gland; EP, epiphallus; FO, free oviduct; GA, genital atrium; HD, hermaphroditic duct; OT, ovotestis; PE, penis; PR, penial retractor muscle; SD, spermathecal duct; SP, spermatheca; VD, vas deferens (scale bar = | mm). 76 AMER. MALAC. BULL. 10(1) (1993) Fig. 6. Basal reproductive organs of Holospira (Sonoraloa) dentaxis alamellata spp. nov. Abbreviations: EP, epiphallus; PE, penis; SD, spermathecal duct (scale bar = 1 mm). an elongate penis which contains slight internal folds of tissue. The apical portion of the penis is a caecum due to the lateral insertion of the epiphallus. The epiphallus is an elongate (Appendix 1), tubular enlargement of the vas deferens which contains a serrate-appearing, glandular endothelium in its proximal portion. The penial retractor muscle is rather slender, and inserts on the rounded (dome-shaped) apex of the penial caecum. The spermathecal duct is characterized by a diverticulum and an elongate, gradually tapering spermatheca. A retractor muscle inserts on the basal portions of the oviduct and spermathecal duct. Holospira (Eudistemma) tantalus campestris Pilsbry and Fer- riss, 1915 (Figs. 8, 10) Synonym: Holospira campestris Pilsbry and Ferriss, 1915 Material examined: SBMNH 35517, Wood Canyon near a rock quarry, NW end of Dragoon Mountains, 32°00.5’°N, Fig. 7. Holospira (Sonoraloa) remondi laevior. A. Apertural view of shell. B. Reproductive system (excluding albumen gland and ovotestis) (scale bars = 1 mm). Fig. 8. Apertural view of three Arizona Holospira species. A. Holospira (Eudistemma) danielsi. B. H. (E.) tantalus campestris. C. H. (E.) ferrissi (scale bar = | mm). 110°00.2’W, elevation ca. 1640 m, Cochise Co., Arizona, one slide-mounted reproductive system (illustrated specimen). SBMNH 35597 (data as above), one slide. Reproductive system: Similar to H. danielsi except: 1) the spermathecal duct is longer and more slender (including the base); 2) the epiphallus is longer and more slender; 3) the GILBERTSON: HOLOSPIRA REPRODUCTIVE ANATOMIES ri) apex of the penial caecum is tapered; 4) the penis lacks in- ternal folds. Holospira (Eudistemma) ferrissi Pilsbry, 1905 (Figs. 8, 11) Material examined: SBMNH 35518, foothills immediately SE of Manila Mine, NW end of Huachuca Mts., 31°33.2’N, 110°26.2’W, elev. ca . 1620 m, Cochise Co., Arizona (type locality), one slide-mounted reproductive system (illustrated specimen). SBMNH 35598 (data as above), one slide. Reproductive system: Similar to Holospira (Eudistemma) danielsi except that: 1) the system is smaller overall (ca. 0.70); 2) the basal portion of the spermathecal duct is distinctly rounded; 3) the apex of the penial caecum is tapered and; 4) the penis lacks internal folds. Remarks: Eudistemma (see Bequaert and Miller, 1973:138) is the primary subgenus of Holospira in Arizona. It inhabits a small southeastern corner of the state and extends eastward through southern New Mexico into northwestern Texas. One population referrable to H. (E.) ferrissi Pilsbry, 1905 has been located in extreme northern Sonora (Bequaert and Miller, 1973). Holospira (Eudistemma) danielsi, H. (E.) ferrissi and H. (E.) tantalus campestris, exhibit similar reproductive systems. The comparable systems of two other species in subgenus Eudistemma, namely H. (E.) arizonensis and H. (E.) chiricahuana Pilsbry, 1905, have been published (Gilbert- son, 1989b). H. arizonensis has a slender (including the base), elongated (14.8 mm in the described specimen) spermathecal duct which closely resembles that of H. tantalus campestris. H. chiricahuana exhibits distinct undulated internal folds in the penis and lower spermathecal duct that are not apparent in the presently described species (nor in H. arizonensis). A retractor muscle inserting on the basal female structures is present in all species observed. The majority of the mus- cle was eliminated during dissection of the photographed specimens of H. tantalus campestris and H. ferrissi. Holospira (Holospira) sherbrookei Gilbertson, 1989, from the Chiricahua Mountains, exhibits a reproductive system that corresponds with those of Eudistemma spp. It is placed in Holospira s.s. because of its quadrilamellate shell condition. The Eudistemma reproductive system is somewhat comparable to that of snails in subgenus Bostricocentrum Strebel, 1880, from southern Mexico (Thompson, 1964). Snails in both subgenera exhibit a spermathecal diverticulum, an elongate spermatheca and a long, slender, tubular epiphallus. However, in Bostricocentrum the epiphallus in- serts apically into the penis and the penis exhibits a terminal knob for the attachment of the retractor muscle. LINEAGE 2: Subgenus Allocoryphe Pilsbry, 1946 Holospira (Allocoryphe) minima (Figs. 2, 3a) ip SPD Fig. 9. Reproductive system of Holospira (Eudistemma) danielsi (except ovotestis). Abbreviations: EP, epiphallus; PC, penial caecum; PE, penis; PT, prostate gland; RM, retractor muscle; SD, spermathecal duct; SP, spermatheca; SPD, spermathecal diverticulum; UT, uterus (scale bar = 1 mm). Material examined: SBMNH 35519, Cerro de la Mona, ca. 21 km east of Hermosillo, Sonora, N side of Hwy 15, 29°02.9°N, 110°39.4’W, elevation ca. 350 m, one slide- mounted reproductive system (illustrated specimen). SBMNH 78 AMER. MALAC. BULL. 10(1) (1993) 35599, 35600 (data as above), two slides. Reproductive system: The male anatomy is characterized by the presence of a stout, tubular epiphallus which inserts laterally into the penis producing an apical penial caecum. This caecum is ca. two-thirds of the total penis length and (along with the penis proper) contains internal folds of tissue. Fig. 10. Reproductive system of Holospira (Eudistemma) tantalus campestris (except ovotestis) (scale bar = 1 mm). The penial retractor muscle attaches to the apex of the penial caecum and is unusually broad. The spermathecal duct is in- flated basally and its spermatheca tapers into the duct. Remarks: Holospira minima is the only species known to exhibit this morphology of the reproductive system. It ap- pears to correlate with the presence of its distinctive, em- bryonic whorls. LINEAGE 3: Subgenus Sonoraloa Gilbertson, subg. nov. Holospira (Sonoraloa) remondi laevior Pilsbry, 1953 (Figs. 3c, 7) Material examined: SBMNH 35520, Sierra Batamote, east- central Sonora (near El Milagro Mine), ca. 1,070 m, 20.2 km E of the Rio Yaqui Bridge along Hwy 15 from La Estrella to Bacanora, 28°57.5’N, 109°30.5’W, Sonora, Mexico, one slide-mounted reproductive system (illustrated specimen). SBMNH 35601, 35602 (data as above), two slides. Reproductive system: The male anatomy is characterized by a vas deferens that enlarges abruptly into a tubular epiphallus. It continues to enlarge distally before inserting apically into the penis. The penis is ovate, constricting at its junction with the genital atrium and exhibits a diagnostic in- ternal outline. The penial retractor muscle is moderate in size, inserting apically onto the penis, adjacent to the epiphallus. The spermathecal duct is expanded basally (ca. same size as penis), and proceeds apically into an internally convoluted section; the spermatheca is oval-shaped. Remarks: The penial complex of Holospira (Sonoraloa) remondi laevior is similar to that of H. (S.) dentaxis alamellata ssp. nov. They differ from the male anatomy of H. (Allocoryphe) minima by: 1) the apical insertion of the epiphallus into the penis, 2) the oval shape of the penis and 3) the narrower (ca. one-half to one-third width) penial retrac- Fig. 11. Reproductive system of Holospira (Eudistemma) ferrissi (except ovotestis. A. Entire system. B. Basal anatomy (scale bars = 1 mm). GILBERTSON: NN SRERaaedi ha gnncens ATR RIA ener Fig. 12. Holospira (H.) milleri. A. Apertural view of shell. B. Male reproductive system and female spermathecal duct (scale bar = 1 mm). tor muscle. These differences confirm the validity of Sonoraloa subg. nov. LINEAGE 4: Holospira s.s. von Martens, 1860 Holospira (Holospira) milleri Gilbertson, 1989 Figs. 3b, 12) Material examined: SBMNH 35521, E side of Rio Yaqui in a ravine near mouth of El Alamo Wash, ca. 1.5 km S of the military footbridge at El Novillo; 28°58.1’N, 109°37.5’W, elev. ca. 260 m, Sonora, Mexico (type locality), one slide-mounted reproductive system (illustrated specimen). SBMNH 35604, 35605, 35606 (data as above), three slides (two specimens possibly immature). Reproductive system: See Gilbertson (1989a) for original description. The male system exhibits a small penial sac which contains a verge. The vas deferens inserts into a short, conic epiphallus atop the penial sac. The basal portion of the spermathecal duct typically contains an undulated, internal lumen although it is not seen on the illustrated specimen. The spermatheca is indented medially (reniform) and constricted basally before joining the spermathecal duct. Remarks: The male anatomy of Holospira (H.) milleri is dramatically different from that of all other described Holospira spp. (for which the anatomy is known). It is the HOLOSPIRA REPRODUCTIVE ANATOMIES 79 only species that exhibits a verge in the penial sac and whose vas deferens does not enlarge into a tubular epiphallus. An undescribed ‘‘form’’ with a trilamellate shell was discovered at the same site as Holospira (Sonoraloa) remondi laevior in the Sierra Batamote. Its anatomy is very similar to that of H. (H.) milleri (including the presence of a verge). However, because of the trilamellate shell condition, this ‘*form’’ cannot be included in Holospira s.s. (along with H. milleri) as the subgenus is defined at present (quadrilamellate only). The reproductive systems of three additional species assigned to subgenus Holospira s.s. have been published (Pilsbry, 1903; Gilbertson, 1989b). They are H. (H.) sher- brookei Gilbertson, 1989 from Arizona, H. (H.) goldfussi (Menke, 1847) from Texas and H. (H.) nelsoni Pilsbry, 1903 from Coahilla, Mexico. They exhibit arrangements of the reproductive system that are significantly different from H. (H.) milleri and from each other. Hence, it is apparent that Holospira s.s. is polyphyletic and in need of revision. CONCLUDING REMARKS Studies of the reproductive system indicate a greater than anticipated number of Holospira lineages from the north- western corner of its range. Additional field work, especial- ly in Mexico, is necessary for continued systematic evalua- tion of the genus. Examination of the soft anatomies of liv- ing populations is essential for an adequate understanding of their phylogeny. Eventually, the taxonomy of the entire genus should be revised to include anatomical data. ACKNOWLEDGMENTS I sincerely thank Edna Naranjo-Garcia, James E. Hoffman and Walter B. Miller for accompanying me on collecting expeditions including one to Sonora at which time we returned to sites of Holospira populations discovered many years earlier by Dr. Miller. These populations represent the Sonoran species whose anatomies are described herein. I also thank M. Andria Garback at the Philadelphia Academy of Sciences and Paul R. Greenhall at the National Museum of Natural History; Smithsonian Institution for the loan of type material of Sonoran Holospira spp. and Dwayne L. Moses for assistance with the illustrations. Dr. Miller and three anonymous reviewers offered many helpful suggestions. In addition, I thank the Professional Development Institute of Orange Coast College for awarding me partial release time from teaching duties for the preparation of this manuscript. LITERATURE CITED Baily, J. L. and R. I. Baily. 1940. A new urocoptid mollusc from the State of Sonora, Mexico. Nautilus 53(3):94-95; Pl. 12, Fig. 1. Bartsch, P. 1943. Notes on Mexican urocoptid mollusks. Journal of the Washington (D.C.) Academy of Sciences 33(2):54-59. Bequaert, J. C. and W. B. Miller. 1973. The Mollusks of the arid Southwest with an Arizona check list. University of Arizona Press, Tucson, Arizona. i-xvi + 271 pp. 80 AMER. MALAC. BULL. 10(1) (1993) Dall, W. H. 1895. Synopsis of the subdivisions of Holospira and some related genera. Nautilus 9(5):50-S1. Dall, W. H. 1896. Diagnoses of new mollusks from the Survey of the Mex- ican Boundary. Proceedings of the United States National Museum 18(1033):1-6. Gabb, W. M. 1865. Descriptions of three new species of Mexican land shells. American Journal of Conchology 1(3):208-209, Pl. 19. Gilbertson, L. H. 1989a. A new species of Holospira (Gastropoda: Pulmonata) from Sonora, with the reproductive anatomy of Holospira minima. Veliger 32(1):91-94. Gilbertson, L. H. 1989b. A new species of Holospira (Gastropoda: Pulmonata) from Arizona, with the reproductive anatomies of H. arizonensis and H. chiricahuana. Veliger 32(3):308-312. Gregg, W. O. 1959. A technique for preparing in-toto mounts of molluscan anatomical dissections. Annual Report of the American Malacological Union for 1958 25:39. Naranjo-Garcia, E. 1989. Four additional species of Sonorella (Gastropoda: Pulmonata: Helminthoglyptidae) from Sonora, Mexico. Veliger 32(1):84-90. Martens, E. von. 1860. Die Heliceen nach Naturlicher Verwandschaft systematisch geordnet von Joh. Christ. Albers. 2nd edition. Berlin. 359 pp. Martens, E. von. 1890-1901. Biologia Centrali Americana. Terrestrial and Fluviatile Mollusca. London. i-xxviii + 1-706 (1897:249-288), Pls. 1-44. Menke, K. T. 1847. Vier neue Arten der Gattung Cylindrella Pfr. Zeitschrift fur Malakozoologie 4:1-3. Pilsbry, H. A. 1903. Manual of Conchology (2) 15, Urocoptidae. Philadelphia i-vili + 323 pp. Pilsbry, H. A. 1905. Mollusca of the southwestern States. I. Urocoptidae; Helicidae of Arizona and New Mexico. Proceedings of the Academy of Natural Sciences of Philadelphia 57:211-290, pls. 1-27. Pilsbry, H. A. 1946. Land Mollusca of North America. Monographs of the Academy of Natural Sciences of Philadelphia 3,2(1):i-iv + 520 pp. Pilsbry, H. A. 1953. Inland Mollusca of Northern Mexico. II. Urocoptidae, Pupillidae, Strobilopsidae, Valloniidae, and Cionellidae. Proceedings of the Academy of Natural Sciences of Philadelphia 105:133-167, pls. 3-10. Pilsbry, H. A. and J. H. Ferriss. 1915. Mollusca of the southwestern States. VII. The Dragoon, Mule, Santa Rita, Baboquivari and Tucson Ranges, Arizona. Proceedings of the Academy of Natural Sciences of Philadelphia 67:363-418, pls. 8-15. Strebel, H. and G. Pfeiffer. 1880. Beitrag zur Kenntniss der Fauna Mex- ikanischer Land und Susswasser-Conchylien, pt. IV. Hamburg. 1-112 pp. 15 pls. Thompson, F. G. 1964. Systematic studies on Mexican land snails of the genus Holospira, subgenus Bostricocentrum. Malacologia 2(1):131-143. Thompson, F. G. 1988. The hollow-ribbed land snails of the genus Coelostemma of the southwestern United States and Mexico. Bulletin of the Florida State Museum: Biological Sciences 33(2):87-111. Date of manuscript acceptance: 13 February 1992 GILBERTSON: HOLOSPIRA REPRODUCTIVE ANATOMIES APPENDIX 1. Measurements (mm) of Holospira reproductive anatomies figured herein. Species H.. danielsi H. dentaxis alamellata H. ferrissi H. milleri H. minima H. remondi laevior H. tantalus campestris APPENDIX 2. Measurements (mm) of Holospira shells figured herein. Penis 3.0 0.6 2:7 0.6 2.0 1.5 3:9 Penial Retr. Muscle 2-5 3.3 1.7 1.9 3.0 2.0 5.0 Epi- phall. 11.0 5.5 11.0 0.3 8.2 7.4 19.0 Sperma- Sperma- thecal thecal Sperma- Free Duct Divert. theca Ovid. Species Height H. danielsi 11.7 H. dentaxis alamellata 11.7 H. ferrissi 8.1 H. milleri 13.1 H. minima 14.5 H. remondi laevior 14.5 H. tantalus campestris 10.0 Width 3.9 4.0 3.4 3.7 4.5 4.7 3.8 12.6 6.0 2.4 3.8 13.4 : 1:2 4.8 8.2 4.0 1.8 2:3 10.9 - 1.5 4.1 12.1 - 1.6 3.8 16.8 - 1:3 37 15.3 7 36 6.3 APPENDIX 3. Revised taxonomy of the genus Holospira from Arizona, Sonora and Sinaloa (*indicates those species for which the reproductive system has been described). Subgenus Holospira H. cyclostoma H. milleri* H. sherbrookei* Subgenus Allocoryphe Hi. minima* Subgenus Sonoraloa subg. nov. H. remondi* H. dentaxis* H. mazatlanica Subgenus Eudistemma . arizonensis* . chiricahuana* . danielsi* . ferrissi* . millistriata . tantalus* . Whetstonensis Incertae sedis H. kinonis ZrrrTrtits 81 Life History of the Endangered Fine-Rayed Pigtoe Fusconaia cuneolus (Bivalvia: Unionidae) in the Clinch River, Virginia Sue A. Bruenderman! and Richard J. Neves U. S. Fish and Wildlife Service, Virginia Cooperative Fish and Wildlife Research Unit?, Department of Fisheries and Wildlife Sciences, Virginia Polytechnic Institute and State University, Blacksburg, Virginia, 24061-0321 U.S.A. Abstract. The reproductive cycle, fish hosts, and selected population statistics of the endangered fine-rayed pigtoe Fusconaia cuneolus (Lea, 1840) were investigated during 1986-1987 in the Clinch River, Virginia. Examination of gravid females and drift samples indicated that the summer brooder is gravid from mid-May to late July and releases most glochidia in mid-June. Diel samples of stream drift contained a peak of glochidia in early morning. Of 1619 collected and examined fishes of 39 species for glochidial attachment, infestation by amblemine glochidia was highest on cyprinids (27 - 46%). Six species were identified as likely hosts for glochidia of the fine-rayed pigtoe. Induced infestations of fishes with glochidia in the laboratory confirmed eight host species: fathead minnow (Pimephales promelas); river chub (Nocomis micropogon); stoneroller (Campostoma anomalum); telescope shiner (Notropis telescopus), Tennessee shiner (N. leuciodus); white shiner (Luxilus albeolus); whitetail shiner (Cyprinella galactura); mottled sculpin (Cottus bairdi). Age class and growth rate characteristics, determined by sectioning of valves collected in muskrat middens, were similar between two demes in the river. The fine-rayed pigtoe reaches a maximum length of about 90 mm and age of 35 yr. Annual growth averaged 5 mm/yr through age 10 and decreased to about 2 mm/yr thereafter until senescence. Specimens less than age 10 were uncommon, and no individuals under age 6 were collected. Cohort structure of live specimens and collections of valves indicated that the fine-rayed pigtoe population is declining in the Clinch River, Virginia. The upper Tennessee River watershed in southwestern all four gills served as marsupia for its glochidia, which are Virginia harbors a diverse assemblage of freshwater mussels subelliptical in shape and of similar length and height (0.16 (Unionidae) that seems to be in significant decline. Of the mm). Other aspects of its biology and life history were roughly 45 reported species from this region, 13 are listed unknown. The objectives of this study, therefore, were to as endangered by the federal government and another 20 are describe the chronology of the reproductive cycle, identify proposed for protection by Virginia (Neves, 1991). Declines fish hosts, and determine the demographics of two demes of in abundance and diversity of mussels are due largely to the fine-rayed pigtoe in the Clinch River, Virginia. anthropogenic activities (Bates and Dennis, 1978; Ahlstedt, 1984). However, a consortium of state and federal agencies STUDY SITE now are involved actively in the protection and recovery of biodiversity in these rivers. The fine-rayed pigtoe pearly mussel Fusconaia cun- eolus (Lea, 1840) was listed as endangered on 14 June 1976 (Federal Register 41:24062-24067). Historically this species was widespread in tributaries of the Tennessee River in Virginia, Tennessee, and Alabama, but now is restricted to reaches of seven rivers in the upper and lower Tennessee River system. The largest population occurs upstream of Norris Reservoir in the Clinch River, Virginia and Tennessee (Fig. 1). Ortmann (1921) noted that this species was a short-term brooder (tachytictic), breeding from May to July, and that The principal study site was at Slant, Clinch River Mile (CRM) 223.3 in Scott County, Virginia (Fig. 2). From an ‘Present address: Virginia Department of Game and Inland Fisheries, 2206 South Main Street, Blacksburg, Virginia 24060 in 0 | 2 >The Unit is jointly supported by the United States Fish and Wildlife Service, cm me aa ae ae ae the Virginia Department of Game and Inland Fisheries, the Wildlife Management Institute, and Virginia Polytechnic Institute and State University Fig. 1. Specimen of Fusconaia cuneolus from the Clinch River, Virginia. American Malacological Bulletin, Vol. 10(1) (1993):83-91 83 84 AMER. MALAC. BULL. 10(1) (1993) SLANT CLINCHPORT — ar Fig. 2. Location of study sites on the Clinch River, Virginia. assessment of muskrat predation on endangered mussels in the Clinch River, Neves and Odom (1989) recorded 32 mussel species and a large population of the fine-rayed pigtoe at this site. The river reach with Fusconaia cuneolus was approx- imately 325 m long and 50 m wide, with substrata of coarse gravel, cobbles, and bedrock. One additional site at Pendleton Island (CRM 226.3), a preserve of The Nature Conservancy, provided gravid females for glochidia and fresh-dead valves for demographic analysis. MATERIALS AND METHODS FIELD STUDIES During the summers of 1986 and 1987, we monitored the incidence of gravid females at the study sites. Specimens of Fusconaia cuneolus were opened carefully with modified o-ring expanders to examine gills. Because this species is not sexually dimorphic in shell characters, gender could be deter- mined only during the period of gravidity. Gravid females were identified by intensely pink and enlarged marsupial gills (Ortmann, 1921). Observations of weekly changes in colora- tion of female gills (marsupia) in 1986 prompted a more detailed examination in 1987. At Slant, examined females were placed under hardware mesh cloth in 1986 to prevent depreda- tion by muskrats. In 1987, these stockpiled females were marked with 5 mm x 3 mm numbered disc tags (Floy Tag Co., Seattle, Washington?) to allow weekly monitoring of in- dividuals. We recorded date, color of gills, and percent gravid females. Mid-day water temperatures were recorded at Slant with a max-min thermometer. Fecundity was computed from the number of conglutinates in a sacrified female (70 mm length) and the mean number of embryos per conglutinate. Because the glochidia of Fusconaia cuneolus were similar in appearance to three other species of the subfamily Ambleminae at the study site, we collected and statistically compared glochidia from each of five specimens of these 3Reference to trade names does not imply government endorsement of commercial products. other amblemine mussel species: long-solid F subrotunda (Lea, 1831); shiny pigtoe F cor (Conrad, 1834); Tennessee clubshell Pleurobema oviforme (Conrad, 1834). Glochidia of mussel species in the subfamily Ambleminae are distinguish- able by their characteristic size and shape (Neves and Widlak, 1988). Sample sizes were 50 glochidia for the Fusconaia spp. and 42 glochidia for P. oviforme. Expelled glochidia from gravid specimens were measured in length, width, and hinge lengths (Table 1). A Kruskal-Wallis test was used to deter- mine whether valve dimensions differed among species. An LSD procedure with ranks identified those dimensions that were significantly different. To determine the period of release of glochidia, we sampled stream drift weekly with three square-framed drift nets (0.045 m?) of 130 um mesh nylon netting with removable cod-ends. Drift nets were staked into the river bottom down- stream of a mussel bed on a line transect for roughly 1 hr between 1000 and 1500 hr on each sampling date. Drift samples were collected from 16 June to 13 Aug 1986, and 29 May to 30 July 1987. Samples were preserved in 10% formalin buffered with sodium borate to prevent dissolution of glochidial valves. In the laboratory, rose bengal was add- ed to samples to facilitate recognition of glochidia. A 0.5 mm mesh sieve was used to remove large particulate matter, and finer debris and glochidia were collected in a 130 um mesh sieve. The latter was examined in a gridded petri dish with a dissecting microscope (25-40X) and glochidia were removed and counted. Densities of glochidia were computed from measurements of water depth and water velocity at the net opening and from counts of glochidia per sample. The diel periodicity of glochidial release of Fusconaia cuneolus was determined at Slant on 16-17 June 1987, the recorded time of peak release in 1986. Every 4 hr during a 24 hr period, we positioned one drift net in the stream where the greatest densities of glochidia were collected in 1986. Numbers of glochidia were tabulated for each of the six time periods to compare densities in a 24 hr daily cycle. The fish fauna at Slant was sampled for roughly 2 hr weekly in a 100 m section of the river with a backpack elec- troshocker and dip nets. Stunned fishes were collected, sorted and preserved in 10% buffered formalin. In the laboratory, opercular flaps were removed to examine gills with a dis- secting microscope (20-40X). No amblemine glochidia were observed externally on the fishes. We tabulated frequency of occurrence and number of glochidia per fish. Fish species with encysted amblemine glochidia were considered possi- ble hosts for the fine-rayed pigtoe. LABORATORY INFESTATIONS We confirmed fish hosts of the fine-rayed pigtoe in the laboratory by induced infestations of putative fish hosts with glochidia. Dechlorinated tap water, oxygenated by Venturi- type diffuser aerators, was recirculated through a central BRUENDERMAN AND NEVES: FINE-RAYED PIGTOE LIFE HISTORY 85 Table. 1. Mean dimensions (mm) + SD and morphometric comparisons of glochidia of selected amblemine species from the Clinch River, Virginia. Glochidial Pleurobema Fusconaia F. cor F. subrotunda dimension oviforme cuneolus Length 0.185 + 0.007 0.181 + 0.008 0.181 + 0.006 0.181 + 0.009 Width 0.193 + 0.004 0.193 + 0.005 0.146 + 0.006 0.150 + 0.028 Hinge length 0.150 + 0.005 0.165 + .009 0.122 + 0.005 0.118 + 0.005 Underscore indicates no difference between species (p = 0.05). gravel biofilter and 96 L rectangular tanks. Fishes for induced infestations were collected from river reaches in Virginia, where mussels were absent or rare, to eliminate the possibility of acquired immunity from prior exposure to glochidia (Neves et al., 1985). Target fish species were electroshocked, scooped with a bucket, and transported in coolers contain- ing aerated river water and 2% salt solution. Ice packs were added when necessary to maintain water temperatures (20° -26°C) comparable to those in laboratory holding tanks. Prior to infestations, collected fishes were acclimated for approximately five days. A 4% benzocaine solution was used to anesthetize fishes and facilitate the identification of cyprinids. Fishes were sorted by species into separate tanks, and the tops of standpipes were covered with 5 mm mesh screen to prevent escape. Mature glochidia were obtained from gravid females collected at Pendleton Island and Slant, Virginia. Mussels were transported to the laboratory and induced to abort con- glutinates by placing them in individual 400 ml beakers without substratum. Beakers were set in a Living Stream (Frigid Units, Inc., Toledo, OH) at a mean temperature of 22°C until glochidia were expelled. If females failed to release glochidia after five days, the beakers were removed from the stream and held at room temperature (23°C). The increased temperature and static water were usually effective in pro- moting expulsion of glochidia. Conglutinates were retrieved with a large-bore pipette, and glochidia were extricated from the conglutinate by repeated agitation in a finger bowl. In- festations were performed with three containers to hold, anesthetize, and recover infested fishes. Procedures for in- festation, examination of fishes, and recovery of juveniles were performed according to Zale and Neves (1982). Juveniles were identified by their yellowish color, thickened ventral margins, and movement of the protruding foot. A sample of juveniles was relaxed in propylene phenoxitol, preserved in 10% buffered formalin, and later photographed. A fish species was considered a host of the fine-rayed pigtoe if encystment and metamorphosis to the juvenile stage occurred. Most laboratory tests that confirmed host fish species in 1986 were repeated in 1987. AGE AND GROWTH Valves of the fine-rayed pigtoe, collected at Pendleton Island and Slant, were measured and aged by thin-sectioning with a Buehler Isomet low-speed diamond-tipped saw (Buehler Ltd., Evanston, Illinois). Five to ten valves per length group (10 mm intervals) from each site were thin- sectioned (Neves and Moyer, 1988). Age of each specimen was determined by counting internal growth lines on thin- sections using a compound microscope (50X). The point where the annulus reached the valve margin was marked with a black felt-tipped pen. The marked thin-sections then were held facing the cross-section from which the cut was pro- duced. Internal growth lines were compared with external growth lines, and false annuli were identified (Neves and Moyer, 1988). Because specimens younger than age nine were rare, we obtained back-measured lengths of younger age groups. On five specimens from each site, external annuli on their cross-sections were marked. By overlaying this valve section onto the matching uncut valve, we measured lengths of the growth lines at previous ages. For these back-measured lengths, we used valves with minimal erosion and easily distinguished annuli. Total lengths by age were obtained from thin sections and fitted to a modified version of the von Bertalanffy growth equation (Gallucci and Quinn, 1979): L, = (w/k) (l-ex (to) where t is a given age in years, t, is the theoretical time when length is zero, k is a growth constant,and w describes growth rate near t,. The parameter w is the product of K and Loo and replaces Loo in the original von Bertalanffy equation. Gallucci and Quinn (1979) concluded that this new parameter (w), because of its statistical robustness, allows comparison of growth rates among populations. Non-linear procedures were applied to derive estimates of w, k, and t,. Based on these values, estimates of L, were computed (SAS Institute, 1982). Confidence intervals of parameter estimates were compared for overlap to determine differences in growth of Fusconaia cuneolus between sites. The equation of predicted lengths at each age, derived from thin-sectioned valves, was used to predict ages from total lengths of unaged 86 AMER. MALAC. BULL. 10(1) (1993) specimens. We obtained lengths from valves of Fiusconaia cuneolus collected in muskrat middens at Slant from 1980 to 1986. Ages of 353 mussels in this collection were computed by age-length relationships, and the age-class distribution of the deme at this site was determined. Lengths of 43 live specimens, ran- domly collected at Slant in 1987, also were converted to ages to determine age class structure. The two age distributions were compared with the G-test (R X C test of independence) to determine whether collection method (human vs muskrat) influenced interpretation of sampling results. Nomenclature for mussels is according to Turgeon et al. (1988). Binomials for fishes follow Robins et al. (1991). RESULTS REPRODUCTIVE CYCLE Gravid females of the fine-rayed pigtoe were collected first on 16 June 1986. Four of eight specimens were gravid. Between 16 July and 6 Aug J986, percent gravidity was high (50-100% ), but sample size on each date was small (N = 4). However, a random sample of 33 fine-rayed pigtoes collected at Pendleton Island on 25 July also indicated that many females (61%) were gravid during this time. One female was gravid at Slant on 6 Aug. No gravid specimens were collected thereafter. In 1987, two gravid females were collected on 10 June (Fig. 3). Most of the stockpiled females at Slant were col- lected again on 18 June 1987. Gravid specimens were collected as late as 6 Aug 1987 at Pendleton Island, which corroborated our observations in 1986. Females began incubating eggs in 100 a 80 nia) al’) @ PENDLETON ISLAND 2S SLANT & wad) ~ 60 a (7) BY ~ Ss (36) Q 50 g > (20) 6 <= 40 j a4 o 30 } (24) 3 20) Ge (19) ——_ 10 71) A(l2) (14) (73) Ets) 0 7) _ _ Ss a eee 3 23 29 10 18 24 1 8 15 20 30 6 MAY JUNE JULY AUG. SAMPLE DATE Fig. 3. Frequency of gravid females in the Clinch River, Virginia, during summers, 1987 and 1988 combined (sample sizes in parentheses). Table 2. Water temperatures and densities of glochidia of the fine- rayed pigtoe in weekly drift samples from the Clinch River at Slant, Virginia, 1986 and 1987. Sample date Temperature Weekly median Glochidial range (°C) temperatures _—_ densities (°C) (no/100m3) 1986 6/16 — 25.4 43.0 6/25 21.4-27.4 24.4 1.9 7/01 22.4-27.9 25.1 6.6 7/09 23.4-27.9 25.6 27.0 7/16 24.4-28.4 26.4 39.7 7/25 25.4-29.9 27.6 8.9 7/30 24.4-29.4 26.9 9.4 8/05 21.4-28.4 24.9 2.9 8/13 22.9-27.4 25.1 3.9 1987 5/29 — 25.4 1.7 6/10 -- 23.0 54.7 6/18 22.9-24.9 23.0 126.6 6/24 23.4-25.1 24.0 38.3 7/01 23.0-25.6 23.5 20.8 7/08 22.0-24.3 23.0 60.3 7/16 21.6-24.2 23.5 3.9 7/20 21.9-24.0 23:3 7.9 7/30 25.0-27.4 26.0 0.8 8/05 26.2-28.0 26.8 = early to mid-May, as judged by the additional collection and examination of females in 1988. Females were not gravid at Pendleton Island (N = 21) or Slant (N = 27) on 3 May, but by 23 May, one of 12 gravid females examined at Slant and nine of 20 at Pendleton Island were gravid. Mid-day water temperatures at Slant increased from 16° to 21°C during this time period. The marsupia of all gravid females collected on 10 June and 4 July contained peach-colored conglutinates. By frequent examination of females at Slant and those brought to the laboratory, we noted a change in marsupial color with maturi- ty - from pink to orange to light peach. In the laboratory, females with light peach marsupia consistently released mature glochidia, held together in a loose gelatinous matrix. Females released mature glochidia within seven to 14 days after the pink color was observed, at water temperatures between 20° and 24°C. Development from embryo to glochidium required roughly 2 wk within this temperature range. Mature glochidia were light orange to peach in color and free of vitelline membranes. Unfertilized eggs, embryos, and immature glochidia were distinctively pink, but unfer- tilized eggs lacked vitelline membranes. When aborted, im- mature glochidia were enclosed in tightly organized, pink conglutinates. Marsupia of females, immediately following release of glochidia, exhibited a bubbly appearance and were less turgid than the gills of gravid females. BRUENDERMAN AND NEVES: FINE-RAYED PIGTOE LIFE HISTORY 87 Table 3. Incidence of glochidial infestations on cyprinid species from the Clinch River, Virginia, 17 June to 30 July, 1987. No. No. infested % infested Species examined Ambleminae Other Ambleminae Other Notropis amblops (Rafinesque, 1820) 265 1 26 0.4 9.8 N. volucellus (Cope, 1865) 199 54 44 27.1 22.1 Campostoma anomalum (Rafinesque, 1820) 88 ah 5 8.0 5.7 Luxilus chrysocephalus Rafinesque, 1820 59 4 8 6.8 13.6 N. sp. 48 22 12 45.8 25.0 N. ariommus (Cope, 1868) 46 1 2 22 4.3 Nocomis micropogon (Cope, 1865) 45 14 3 31.1 6.7 Erimystax dissimilis (Kirtland, 1840) 41 0 0 0.0 0.0 Cyprinella galactura (Cope, 1868) 33 13 8 39.4 24.2 Notropis telescopus (Cope, 1868) 31 2 1 6.7 3.2 L. coccogenis (Cope, 1868) 30 2 1 6.7 3.3 N. leuciodus (Cope, 1868) 20 6 2 30.0 10.0 E. insignis (Hubbs and Crowe, 1956) 18 0 ) 0.0 0.0 C. spiloptera (Cope, 1868) 13 4 yA 30.8 15.4 Pimephales notatus (Rafinesque, 1820) 8 1 0 12.5 0.0 Phenacobius uranops Cope, 1867 0 0 0.0 0.0 N. photogenis (Cope, 1865) i 0 0 0.0 0.0 C. whipplei Girard, 1856 4 0 2 0.0 5.0 N. rubellus (Agassiz, 1850) 2 0 0 0.0 0.0 Total 965 131 116 For host fish experiments, only females with peach- colored marsupia were transported to the laboratory. The 41 gravid females held in the laboratory during 1987 released glochidia from 14 June to Il Aug at water temperatures between 20° and 24°C. Thirty-three females released con- glutinates with nearly 99% mature, viable glochidia. Eight females released conglutinates with 3 to 5% unfertilized eggs. Immature glochidia were released by only one female on 19 July. Conglutinates were subcylindrical in shape and approx- imately 6 mm long and 1.5 mm wide with two layers of tightly aggregated glochidia about 0.8 mm deep. One conglutinate from each of five females contained a mean of 236 + 38.1 embryos or glochidia. Fecundity was about 113,000 embryos for the sacrificed female. Hinge lengths differed among each of the four amblemine species (Table 1; p = 0.0001, n = 42-50 glochidia per species). Glochidia of Fusconaia cuneolus were greater in width than those of F cor and F. subrotunda, but were not different from those of Pleurobema oviforme. Total lengths of glochidia of all species did not differ. The glochidia of the fine-rayed pigtoe most closely resembled those of P. oviforme, but were visually distinguished by a longer hinge length and more rotund appearance. We observed two peaks in the presence of glochidia in 1986 and 1987 (Table 2). In 1986, peak densities of glochidia occurred in mid-June (43.0/100 m3) and mid-July (39.7/100 m3). A similar trend in densities was observed in 1987, although the second peak occurred one week earlier than in 1986. Multiple linear regression analyses indicated that densities of glochidia were not correlated with date of sample, weekly median temperature prior to sample date, or water velocity at the drift net. From 16 to 17 July 1987, we collected glochidia of Fusconaia cuneolus in stream drift throughout the 24 hr period, at water temperatures ranging from 21.2° to 23.0°C. Densities of glochidia were lowest at 2300 and 0300 hr (1/100 m*) and peaked at 0700 hr (118/100 m?). Water velocity at the drift net was constant at 0.4 m/sec for the period. FISH HOSTS From 17 June to 30 July 1987, we collected and ex- amined 1,150 fishes representing 39 species at Slant for at- tached glochidia. Thirteen fish species, all cyprinids, were infested with glochidia of the Ambleminae (Table 3). The highest prevalence of infestation on cyprinids occurred on 17 June (17%) and 8 July (25%) 1987 (Table 4), dates corre- sponding to the highest densities of glochidia of Fusconaia cuneolus in stream drift. Minnow species with the highest incidence of infestation were the sawfin shiner Notropis sp. (45.8%), whitetail shiner Cyprinella galactura (Cope, 1868) (39.4%), river chub Nocomis micropogon (Cope, 1865) (31.1%), spotfin shiner C. spiloptera (Cope, 1868) (30.8%), Tennessee shiner Notropis leuciodus (Cope, 1868) (30%) and mimic shiner N. volucellus (Cope, 1865) (27.1%). These fish species were identified as the most probable hosts for the fine- rayed pigtoe. On various sample dates in 1987, glochidia identified as Fusconaia cuneolus were found encysted in the gills of four species of cyprinids; river chub, whitetail shiner, stoneroller, 88 AMER. MALAC. BULL. 10(1) (1993) 2005 179 Zz fy 150 - 2 } 1S) 5.0 mm in shell length fed increasingly on bivalves, probably reflecting an inability of small snails to ingest large prey. The snails appear to locate prey with the aid of Hancock’s organ and sensory patches (here reported for the first time) situated along the anterior edge of the cephalic shield and foot. Individual prey are grasped with the radula and pulled whole into the buccal cavity. Sediment particles appear to be removed passively by the jaws. Haminoea solitaria (Say, 1822) indiscriminately ingests sediment containing diatoms, unidentified algae, and detritus. Han- cock’s organ and sensory palps in H. solitaria probably do not aid in selection of specific food particles, but could be involved in location of areas where diatoms and algae are abundant. Cephalaspidean opisthobranchs are common members of the shallow water marine benthos. Previous studies revealed that cephalaspids feed on a variety of food items, including bivalve and gastropod mollusks (Hurst, 1965; Berry, 1988; Berry and Thompson, 1990; Morton and Chiu, 1990), foraminiferans (Burn and Bell, 1974a, b; Shonman and Nybakken, 1978; Berry, 1988; Berry and Thompson, 1990), polychaete worms (Hurst, 1965; Rudman, 1972a, b; Yonow, 1989), and algae (Fretter, 1939; Rudman, 1972a). Few studies have dealt with the feeding habits of cephalaspids in detail. Two species, Acteocina canaliculata (Say, 1826) and Haminoea solitaria (Say, 1822), occur commonly along the New England coast (Abbott, 1974) and offer an opportunity to investigate and contrast the feeding biology of two genera in detail. Acteocina canaliculata is a small (4-6 mm) opistho- branch found commonly in both oceanic and estuarine en- vironments in bare or vegetated, sandy to muddy subtidal habitats (Franz, 1971; Gosliner, 1979; Mikkelsen and Mik- kelsen, 1984). The only species of Acteocina studied previous- ly with regard to feeding, A. culcitella (Gould, 1852), is a selective predator on foraminiferans (Shonman and Nybak- ken, 1978). Haminoea solitaria, another small (10-12 mm) opistho- branch, occurs in environments similar to those of Acteocina canaliculata. Except for H. brevis (Quoy and Gaimard, 1833), which is reported to feed on bivalve mollusks (MacPherson and Gabriel, 1962), members of the genus Haminoea are ‘Present address: Department of Zoology, University of New Hampshire, Durham, New Hampshire 03824, U.S.A. reportedly herbivorous (Fretter, 1939; Rudman, 197la,b; Gib- son and Chia, 1989). Diet and feeding behavior have been described for H. hydatis (Linnaeus, 1758) and H. zelandiae (Gray, 1843) (Fretter, 1939; Rudman, 197la, b). Both species feed on filamentous green algae using the jaws and radula to break pieces of algae and the gizzard plates to grind them. This paper presents information on the dietary composition, changes in diet, and behaviors associated with the location and consumption of food by A. canaliculata and H. solitaria. MATERIALS AND METHODS Acteocina canaliculata and Haminoea solitaria were collected in summer 1989 and spring 1990 from a sandy-mud flat in Bluff Hill Cove, Point Judith Salt Pond, Galilee, Rhode Island, USA (41° 23’N, 71° 30’W) (Fig. 1). Qualitative field observations on the relative abundance of both snails were made approximately every month from April 1989 to September 1990. Individuals for diet analysis were collected by pass- ing the flocculent top | cm of substratum through a 1.0 mm sieve. The animals were fixed in 5% buffered formalin im- mediately after collection and transferred to 75% ethanol after 24 to 48 hours. The digestive tract of Acteocina canaliculata was removed with the aid of a dissecting microscope, and the contents spread on a microscope slide. Food items were transferred with the aid of a small camel hair brush to double- stick tape mounted on a microscopic slide. Items were iden- tified to the lowest possible taxonomic level utilizing Loosanoff et al. (1966), Abbott (1974), Cushman (1976), and American Malacological Bulletin, Vol. 10(1) (1993):93-101 93 94 AMER. MALAC. BULL. 10(1) (1993) ey 419 24° Great Island Salt Pond Fig. 1. Location of study site where Acteocina canaliculata and Haminoea solitaria were collected. Specific collecting sites are marked with an asterisk (scale bar = 200 m). Todd and Low (1981). Preliminary observations revealed that Haminoea solitaria feeds on diatoms and detritus. Therefore, the con- tents of the digestive tract were determined following the methods of Kesler ef al. (1986). The digestive tract was removed, the contents mixed with a small amount of water, and evenly distributed on a glass slide. A grid containing 50 evenly spaced fields was viewed at 100x and scored for the presence of diatoms, filamentous algae, animal parts, sand, and detritus. The presence of each food type was quantified as the proportion of total fields containing that food item out of all fields that contained one or more food items. This pro- cedure assumes that the proportion of fields containing an item reflects the amount of the food item in the diet. Items were identified utilizing Kennett and Hargraves (1988). Ad- ditionally, fecal remains from ten individuals were spread on a Slide and viewed at 100x to determine which of the ingested material was being digested. Multivariate analysis of variance (MANOVA) was per- formed on dietary proportions with respect to shell length and sample date (Sokal and Rohlf, 1981). Shell length of each animal was measured to the nearest 0.1 mm. Three size groups were chosen to give equal sample sizes between groups. Statistical analyses were performed using Statistical Analysis System (SAS Institute, Cary, NC). The results are presented as an F-ratio (F), with the appropriate degrees of freedom (d.f.), and level of significance (P). Snails used for feeding observations were collected as above and maintained in small aquaria. Algal and sediment samples were taken at the same time and location as the snails in order to present the same potential food sources to the snail as are found in their natural habitat. Feeding behavior was recorded with a video camera, mounted so as to view the ventral region of the snail, and a video cassette recorder that allowed frame-by-frame analysis of behavioral events. Potential food items of Acteocina canaliculata (foraminiferans and bivalves) were selected carefully from substratum samples and placed in small petri dishes. These prey organisms were acclimated to laboratory conditions for at least 24 hr prior to observations. Attempts were made to remove sediment particles clinging to foraminiferans to im- prove the clarity of viewing feeding activity. A variety of potential food items, including a thin layer of sand contain- ing micro-organisms and various algal species found in the sediments at the mudflat, was presented to Haminoea solitaria in a small glass chamber (10 x 10 x 12 cm). Voucher specimens of Acteocina canaliculata (MCZ 302559) and Haminoea solitaria (MCZ 302560), collected from the study site, are located in the collections of the Museum of Comparative Zoology (MCZ), Harvard University. RESULTS ACTEOCINA CANALICULATA Acteocina canaliculata was present on the Bluff Hill Cove mud flat throughout 1989 and 1990, with the highest abundances occurring during July and August 1989. Larger snails (shell length > 3.9 mm) were more abundant during these two months (Fig. 2). Juveniles (shell length > 1.0 mm) first appeared in March and April 1990. 1. DIET Sixty-eight percent of the snails (n = 97, 2.0-6.8 mm shell length) contained food particles (Table 1). Food remains, consisting of foraminiferans and bivalves, were found in the esophagus, gizzard, stomach, and intestine. Ninety-eight per- cent of the food remains in the stomach and intestine were crushed yet remained as discrete particles. Food remains found posterior to the stomach consisted mainly of shell or test fragments and fecal material. Sediment grains were found in only nine individuals. The gizzard plates of all dissected individuals showed no obvious sign of physical damage. Foraminiferans found in the digestive tract belonged to three suborders: Textulariina, Miliolina, and Rotaliina. The majority of miliolid foraminiferans recovered from the digestive tract was crushed. Some fragments did allow generic identification. The remaining fragments were classified as ‘‘miliolid spp.’’ due to the porcelaneous appearance of the CHESTER: FEEDING BIOLOGY OF ACTEOCINA AND HAMINOEA 95 100 [] Small, < 3.8 mm Medium, 3.9-5.4 mm f4 Large, >5.5mm Oe SS rrey *,, a ESOS Frequency (%) oO So RSS RSS SASS Ee 20 Y 1 0 CaN fe ee 4 July 1989 August 1989 March 1990 April 1990 Fig. 2. Percentages of three sizes of Acteocina canaliculata collected over a four-month period from July 1989 to March 1990. test (Todd and Low, 1981). Other foraminiferan fragments could not be identified and were classified as ‘‘unidentified forams.’’ Eighty-three percent of Textulariina and Rotaliina found in the esophagus and gizzard were whole and showed no obvious signs of chemical or physical damage. No bivalves were found in the esophagus, and all bivalves found in the gizzard, stomach, and intestine were crushed, making identification impossible in some cases. However, Gemma gemma (Totten, 1834) was usually identi- fiable based on the purplish coloration of its shell (Abbott, 1974). Snail size was correlated with the type of food item in the gut (Fig. 3). The proportion of miliolid foraminiferans was signficantly lower in medium and large snails (MANOVA: F = 4.07, d.f. = 1, 61, P = 0.0481). In addi- tion, there were fewer unidentified foraminiferan fragments in larger snails, probably an artifact because ‘‘fragments’’ taken from larger cephalaspids were larger and therefore more readily identifiable. A significantly higher abundance of bivalves was found in the guts of medium and large animals (MANOVA: F = 10.61, d.f. = 1, 61, P = 0.0018). In fact, only one animal < 5.0 mm had a crushed bivalve in its intestine. The proportion of bivalves in the diet was significant- ly higher in July and August than in March and April (MANOVA: F = 12.44, d.f. = 1, 61, P = 0.0008) (Fig. 4). Unidentified foraminiferans were significantly higher in March than in other months (MANOVA: F = 4.33, d.f. = 1, 61, P = 0.0417). Again, this was probably due to the relative ease of identifying the larger foraminiferan fragments taken from larger animals. 2. BEHAVIORAL OBSERVATIONS: MOVEMENT Movement was observed in 18 individuals (size = 3.6-5.2 mm). Acteocina canaliculata burrows into the top few millimeters of the substratum using the combined cephalic shield and tentacles as an efficient plow (Fig. 5). Sediment particles are carried by ciliary action along the dorsal sur- face of the cephalic shield and the posteriorly directed tentacles, and are sloughed off along either side of the shell. Typically, when moving, the head-foot extends forward, gliding across the substratum. With the foot extended anterior- ly, the shell is retracted towards the head and the cycle repeated. Moving in this manner, the average speed for A. canaliculata was 17.6 mm per min on glass (n=3). 3. BEHAVIORAL OBSERVATIONS: FEEDING A number of sense organs are located along the leading edge of the cephalic shield and foot. Two slightly darker sen- sory patches are located on either side of the mouth along the anterior edge of the cephalic shield (Fig. 6). Two other similar patches exist in corresponding positions on the foot. This area is highly active and was observed repeatedly touching the substratum and potential food items. Hancock’s organs are located on either side of the head in antero- posterior grooves created by the cephalic shield and the foot. The animal characteristically swings the foremost part of the Table 1. List of recovered food remains (including fragments) from 66 Acteocina canaliculata arranged by taxonomic group. % in Gut refers to the frequency of a food item in the gut. % Snails refers to the percent of snails in which a food item was present. Food Item % in Gut % Snails — # of Particles Suborder Textulariina Trochammina 23L 57.4 77 Suborder Miliolina Miliommina 9.2 23.5 30 Quinqueloculina 8.3 21.9 2] Miliolid spp. 26.2 50.0 85 Suborder Rotaliina Pseudononion 2:2 7.4 7 Astrononion 1.9 7.4 6 Haynesina 0.6 2.9 2 Nonionella 0.3 1.5 1 Elphidium 0.3 1.5 1 Suborder Unknown Unidentified Forams 6.8 22.1 22 Class Bivalvia Gemma gemma 14.2 26.5 46 Unidentified bivalves 6.5 17.7 21 9 nN 100 C1 Small, < 3.8 mm Medium, 3.9 - 5.4 mm (4 Large, >5.5 mm Occurrence in the gut (%) on oO RR Qe SS RAR 4 Rotaliina Miliolina Textulariina 3s Unid. forams Bivalves Fig. 3. Percentage of food items in the gut of three sizes of Acteocina canaliculata (bar = standard error of the mean). 100 O July 1989 90 Z] August 1989 ~ March 1990 > 80 D April 1990 a 70 <= = 60 | @ 20 Y c g 7 rT) 40 VEY = AY (S) x § 20 Aye 10 nA AY 5 ; Aa atte Rotaliina Miliolina 3°) Cc < © =) ~ x ® | Unid. forams Bivalves Fig. 4. Percentage of food items in the gut of Acteocina canaliculata col- lected in a four-month period from July 1989 to April 1990 (bar = standard error of the mean). AMER. MALAC. BULL. 10(1) (1993) head back and forth, apparently searching for food. On six separate occasions, Acteocina canaliculata turned towards a prey items from a distance of 1 to 3 mm. Feeding was observed in five individuals of Acteocina canaliculata (size = 4.5-5.2 mm). Upon contact, a poten- tial food item is touched repeatedly by the sensory patches. The anterior edge of the cephalic shield and foot is com- pressed into a funnel-shaped cavity with the mouth in the center and sensory patches on either side. The mouth is then pressed against the food item, the buccal mass contracted, the odontophore protruded through the mouth, and the item gripped with the radula. The food item is then pulled into the buccal cavity. The jaws, located near the anterior end of the buccal cavity, appear to scrape off most of the sand grains clinging to the food item as it is drawn past the jaws into the mouth. All foraminiferans ingested by Acteocina canaliculata were alive when eaten. Living foraminiferans are differen- tiated easily from empty tests because they are surrounded by detritus and sediment grains gathered by the reticulopodia (Cushman, 1976). Empty tests were ignored consistently by A. canaliculata, and they were passed up and over the cephalic shield along with sediment particles. Capture and ingestion of a living bivalve (Gemma gemma) was observed only once. In this instance, a 5.2 mm snail turned towards the bivalve from a distance of 1.6 mm, and repeatedly touched it with its sensory patches. The bivalve was approximately 350 um in height. It was gripped by the radula along the anterior edge, just anterior to the umbo, with the ventral opening facing down. Several attempts were made to draw the bivalve through the mouth and, after approximate- ly two minutes, it was ingested. HAMINOEA SOLITARIA Haminoea solitaria was abundant seasonally, and it first appeared in mid-June 1989. The smallest individual col- lected during that summer was 3.0 mm in shell length. Field observations revealed that H. solitaria was extremely abun- dant during summer 1989, with large individuals more prevalent than small individuals in July and August. None were found after 17 November 1989. H. solitaria was first collected in 1990 on 13 July. The smallest individual collected during 1990 was 3.7 mm. 1. DIET Sixty-seven percent of the snails (n = 106, 3.8-12.0 mm shell length) contained food items in their digestive system. Diatoms accounted for 36% and detritus accounted for 34% of the material ingested. Diatoms found included Gram- matophora, Thalassiosira, Coscinodiscus, Peridinium, Navicula, Gyrosigma, Skeletonema, and Chaetoceros. Sand also constituted a large portion (26%) of the material found in the gut. Algae composed 3% of the diet. No attempt was CHESTER: FEEDING BIOLOGY OF ACTEOCINA AND HAMINOEA 97 Dorsal View Lateral View shell head extension Fig. 5. Dorsal and lateral view of Acteocina canaliculata movement: cs, cephalic shield; ct, cephalic tentacle; e, eyespot; ft, foot; oh, Hancock’s organ (scale bar = 1 mm). made to identify algal species because of the small size of the fragments recovered. Additionally, a small fraction of the diet consisted of animals parts (1%). Animal parts included nematodes, copepods, nauplii larvae, ostracods, polychaete setae, oligochaetes, bivalves, and foraminiferans. The bivalves and foraminiferans showed no obvious signs of physical or chemical damage. Qualitative observations made of the fecal remains revealed that nearly all diatoms and animal parts con- sisted of empty tests suggesting that a large percentage of the diatoms and animals were digested. The gizzard plates in 56 of the 106 Haminoea solitaria examined showed signs of damage. The outer, horny layer was usually torn away, creating pits in the plates. This was particularly apparent in larger individuals. The proportion of sand and algae in the diet of Haminoea solitaria increased significantly with increasing snail size (MANOVA: sand, F = 10.43, d.f. = 1, 69, P = 0.0002; algae, F = 6.02, d.f. = 1, 69, P = 0.0166) (Fig. 7). Conversely, the amount of diatoms and detritus in the diet was significantly less in large animals (MANOVA: diatoms, F = 871, d.f. = 1, 69, P = 0.0043; detritus, F = 15.64, d.f. = 1, 69, P = 0.0002). There was no significant change in the proportion of animal remains in the diet of small and large snails. The diet was highly variable among sample dates, even when the dates were only a few days apart. Significant dif- ferences existed between sample dates in the summer of 1989 for detritus, sand, and algae (MANOVA: detritus, F = 3.73, d.f. = 7, 69, P = 0.0019; sand, F = 4.53, d.f. = 7, 69, P = 0.0004; algae, F = 3.58, d.f. = 7, 69, P = 0.0026) but not for diatoms or animals parts (Fig. 8). The proportions of diatoms in the diet decreased over time. There was no ap- parent trend in detritus and sand, but the percent of algae did exhibit an increase through August, a drop at the end of August and at the beginning of September, followed by a subsequent increase during September. 2. BEHAVIORAL OBSERVATIONS: MOVEMENT Movement was observed in 18 individuals of Haminoea solitaria (size = 3.3-11.2 mm). The foot of H. solitaria is 98 AMER. MALAC. BULL. 10(1) (1993) ft m sp Fig. 6. Anterior view of Acteocina canaliculata: ct, cephalic tentacle; ft, foot; m, mouth; oh, Hancock’s organ; sp, sensory patches (scale bar = 0.5 mm). relatively larger than that of Acteocina canaliculata and has two parapodial lobes which curve up and around the anterior of the shell, often meeting mid-dorsally (Fig. 9a). Posterior to the foot, the floor of the mantle cavity extends postero- ventrally beyond the shell, forming the posterior mantle lobe. When moving, H. solitaria glides smoothly through the substratum. The foot secretes an almost complete tube of mucus which helps prevent sediment particles from clogging the mantle cavity. Sediment particles are moved along the ciliated cephalic shield where they are entrapped in the mucous secretions of the foot and sloughed off posteriorly. H. solitaria moves much faster than A. canaliculata, travel- ing an average of 97 mm per min on glass (n=3). 3. BEHAVIORAL OBSERVATIONS: FEEDING Hancock’s organs are located on either side of the head in antero-posterior grooves created by the cephalic shield and foot (Fig. 9b). They are smaller than in Acteocina canali- culata, and consist of slightly wavy ridges. In addition to Han- cock’s organ, two sensory palps are located on either side of the mouth, also in the antero-posterior groove. Sensory patches were observed in a few individuals, but could not be consistently found in all animals that were observed. Haminoea solitaria swings its head back and forth while mov- ing forward. The leading edge of the cephalic shield and foot is highly active and repeatedly touches anything with which the animal comes in contact. Feeding was observed in seven individuals (size = 4.6-10.9 mm). The feeding behavior of Haminoea solitaria is similar to that of Acteocina canaliculata. The buccal mass contracts, the odontophore protrudes through the mouth, and the radula grasps a mouthful of sediment. The odontophore is then withdrawn and the buccal mass relaxes. All individuals observed appeared to ingest sediment at random with no selection of food particles. However, on several occasions H. solitaria rejected ingested particles by expectorating them. DISCUSSION Among opisthobranch gastropods, cephalaspids show the greatest diversity in feeding strategies, ranging from herbivores that graze algal mats or consume pieces of algae to carnivores that engulf whole prey animals (Kohn, 1983). The two species studied here have very different diets and exhibit two different feeding modes. Acteocina canaliculata seeks out and ingests whole foraminiferans and bivalves, while Haminoea solitaria consumes mouthfuls of sediment contain- ing diatoms, detritus, and algae. Hard-shelled prey pass through H. solitaria apparently undigested. In general, foraminiferans are ingested by two types of predators, unselective predators that ingest sediments con- taining foraminiferans, and selective predators that preferen- tially ingest foraminiferans (Lipps and Valentine, 1970; Murray, 1973). The diet and feeding behavior observed demonstrate that Acteocina canaliculata is a selective predator of foraminiferans in the sense that the snails are not selec- ting random sediment samples. Selective predation of forams 70 CJ) Small, < 6.2 mm se eC Medium, 6.3 - 8.6 mm = 50 {4 Large, > 8.7 mm .o)) = 40 & » 30 Oo © = 20 Oo oO O 10 0 3 : n nH a ® £ 2 © fe) = £ = E o ¢ * a ao < Fig. 7. Percentages of food items in the gut of three sizes of Haminoea solitaria (bar = standard error of the mean). CHESTER: FEEDING BIOLOGY OF ACTEOCINA AND HAMINOEA 99 Occurrence In the gut (%) = s Le)) @ = £& 8 £ 8 10 ze se Animal aD ma 2 6 & g 4 @ g Soe erent Geen oc te.) n nn aoe Qe a a 32223 3 3 8 40 Detritus Occurrence In the gut (%) Occurrence In the gut (%) Fig. 8. Percentages of food items in the gut of Haminoea solitaria for eight sample dates in 1989 (bar = standard error of the mean). has also been observed in A. culcitella (Shonman and Nybak- ken, 1978) and other cephalaspids (Hurst, 1965; Burn and Bell, 1974a,b; Berry, 1988; Berry and Thomson, 1990). These observations further suggest that A. canaliculata locates food items before they are physically contacted using Hancock’s organ and sensory patches. Hancock’s organ functions as a chemosensory organ in other cephalaspids (Lemche, 1956; Hurst, 1965; Edlinger, 1980; Kohn, 1983). Sensory patches have been observed in the genus Philine (Hurst, 1965; Rud- man, 1972c). In P. aperta Linnaeus, 1766 they are believed to aid in prey location and capture (Hurst, 1965). These organs, in A. canaliculata, appear to play a specific role in food acquisition, that of locating individual foraminiferans. In A. canaliculata, sensory patches appear to play a similar role. Ontogenetic changes in diet have been observed in other gastropods, as well as in seastars, chitons, and fish (Hurst, 1965; Hughes, 1980; Taylor, 1980; Berry, 1988; Paine, 1988; Berry and Thomson, 1990). The diet of Acteocina canaliculata changed with shell size. As size increased, the 100 AMER. MALAC. BULL. 10(1) (1993) Dorsal Ventral Fig. 9. External anatomy of Haminoea solitaria. A. Dorsal and ventral views (scale bar = 2 mm). B. Lateral view of head-foot with lateral parapodia reflected ventro-laterally (scale bar = 1 mm): cs, cephalic shield; e, eyespot; ft, foot; gs, external seminal groove; ml, posterior mantle lobe; oh, Han- cock’s organ; p, parapodial lobes; ps, sensory palps. abundance of miliolid foraminiferans in the gut decreased. Additionally, there was an increase in the proportion of bivalves in the guts of snails larger than 5.0 mm. This dietary pattern could represent an inability of small A. canaliculata to ingest large prey items. Foraminiferan size is generally be- tween 50 um and 300 pm (Cushman, 1976). Bivalves are typically larger than 200 »m when they settle. For example, Mya arenaria Linnaeus, 1758, has a minimum shell size at metamorphosis of 225 x 209 um (Loosanoff et al., 1966), thus representing the smallest bivalve species that is poten- tially available to A. canaliculata at Galilee. Snails smaller than 5.0 mm in length could be restricted to ingesting only foraminiferans due to the size of their digestive tract. From about 5.0 mm shell length the diameter of the digestive tract or size of the radula would be large enough to allow A. canaliculata to incorporate bivalves, as well as larger foraminiferans, into its diet. Unfortunately, the number of measurable food remains was too small to adequately test this hypothesis. A shell length of 5.0 mm could also represent a prac- tical limit to the size at which the energy gained by eating foraminiferans is greater than the energy expended in searching for and ingesting them. Assuming that there is more energy to be gained by eating bivalves compared to foramini- ferans, and that Acteocina canaliculata is maximizing its energy intake, then bivalves may be more profitable to snails larger than 5.0 mm (see Hughes, 1980). A significantly higher proportion of bivalves was found in the diet of Acteocina canaliculata in July and August when larger snails were more common. This suggests that seasonali- ty of prey species may be another factor in determining diet. This has been demonstrated in Retusa obtusa (Berry, 1988; Berry and Thomson, 1990) and other gastropod predators, such as Conus, Thais, and Polinices (Hughes, 1980, 1986). Unfortunately, no data concerning natural abundances and seasonality of the prey species at Galilee were made during this study. In contrast with Acteocina canaliculata, the diet and feeding behavior of Haminoea solitaria revealed that it in- discriminately ingests sediments containing diatoms, algae, and detritus. No apparent selection of food items was observed, indicating that the sensory organs do not play as specific a role in food selection as they do in A. canaliculata. However, these organs could be more generally involved in locating patches of sand containing a high abundance of diatoms or other organic material, or not involved in feeding at all. The diet of Haminoea solitaria changed with respect to both sample date and snail size. The indiscriminate feeding behavior observed in H. solitaria suggests that the abundances of specific food items in the diet should parallel their relative abundances in the habitat. As with planktonic diatoms, benthic diatoms show a definite seasonal cycle in abundance and species composition (Kennett and Hargraves, 1988). Typically, there is a spring bloom characterizing certain species, a low abundance during July, August or September, followed by a bloom in October characterizing other species. The trends observed in the diet of H. solitaria with respect to predator size and sample date might be the result of seasonal changes in abundance of prey species. Future studies on these animals should concentrate on a number of areas. The diets of Acteocina canaliculata and Haminoea solitaria reveal the importance of food abundance and seasonal distribution in fulfilling dietary requirements. Additionally, the size of ingested food items appeared to be important in the diet of A. canaliculata. Future studies should examine not only the abundance of prey organisms in the en- CHESTER: FEEDING BIOLOGY OF ACTEOCINA AND HAMINOEA 101 vironment, but the life cycle and size of ingested prey species to understand fully the relationships between predator and prey. Sensory organs appear to have a specific role in food acquisition in A. canaliculata, but the role of such organs in H. solitaria is unclear. Further research is necessary to determine more specifically the role sensory organs play in food detection and selection in these cephalaspids. ACKNOWLEDGMENTS I am grateful to R. Bullock for providing to me the opportunity to conduct this research and for his enthusiastic support throughout the study. I would also like to thank H. Bibb, T. Napora, and R. Turner for their con- structive comments and valuable insights. L. Harris, W. Lambert, M. Lit- vaitis, W. Watson, J. Berman, P. M. Mikkelsen and one anonymous reviewer provided comments on the manuscript. J. Heltshe and R. Hanumara pro- vided advice on statistical analysis. K. Davignon provided illustrative ad- vice. B. Peterson and K. Thomas assisted with field work and provided moral support. Funding was provided by the University of Rhode Island, Depart- ment of Zoology. This paper formed part of a master’s thesis at the University of Rhode Island. LITERATURE CITED Abbott, R. T. 1974. American Seashells, 2nd ed. Van Nostrand Reinhold, New York. 663 pp. Berry, A. J. 1988. Annual cycle in Retusa obtusa (Montagu) (Gastropoda, Opisthobranchia) of reproduction, growth, and predation upon Hydrobia ulvae (Pennant). Journal of Experimental Marine Biology and Ecology 117:197-209. Berry, A. J. and D. R. Thomson. 1990. Changing prey size preferences in the annual cycle of Retusa obtusa (Montagu) (Opisthobranchia) feeding on Hydrobia ulvae (Pennant) (Prosobranchia). Journal of Experimental Marine Biology and Ecology 121:145-158. Burn, R. and K. N. Bell. 1974a. Description of Retusa chrysoma Burn sp. nov. (Opisthobranchia) and its food resources from Corner Inlet, Vic- toria. Memoirs of the National Museum of Victoria 35:115-119. Burn, R. and K. N. Bell. 1974b. Description of Retusa pelyx Burn sp. nov. and its food resources from Swan Bay, Victoria. Journal of the Malacological Society of Australia 3:37-42. Cushman, J. A. 1976. Foraminifera, Their Classification and Economic Use, 4th ed. Harvard University Press, Cambridge. 605 pp. Edlinger, K. 1980. Beitrage zur Anatomie, Histologie, Ultrastructur und Physiologie der chemischen Sinnesorgane einiger Cephalaspidea (Mollusca, Opisthobranchia). Zoologischer Anzeiger 205:90-112. Franz, D. R. 1971. Development and metamorphosis of the gastropod Acteocina canaliculata (Say). Transactions of ihe American Microscopical Society 90:174-182. Fretter, V. 1939. The structure and functioning of the alimentary canal of some tectibranch molluscs, with a note on excretion. Transactions of the Royal Society of Edinburgh 59:599-646. Gibson, G. D. and F-S. Chia. 1989. Description of a new species of Haminoea, Haminoea callidegenita (Mollusca: Opisthobranchia), with a comparison with two other Haminoea species found in the northwest Pacific. Canadian Journal of Zoology 67:914-922. Gosliner, T. M. 1979. A review of the systematics of Cylichnella Gabb (Opisthobranchia: Scaphandridae). Nautilus 93:85-92. Hughes, R. N. 1980. Optimal foraging in the marine context. Oceanography and Marine Biology Annual Review 18:423-481. Hughes, R. N. 1986. A Functional Biology of Marine Gastropods. Croom Helm Publishers, London. 245 pp. Hurst, A. 1965. Studies on the structure and function of the feeding apparatus of Philine aperta with a comparative consideration of some other opisthobranchs. Malacologia 2:281-347. Kennett, D. M. and P. E. Hargraves. 1988. Benthic marine diatoms. Jn: Freshwater and Marine Plants of Rhode Island, R. G. Sheath and M. M. Harlin, eds. pp. 127-135. Kendall/Hunt publishers, Dubuque. Kesler, D. H., E. H. Jokinen, and W. R. Munns. 1986. Trophic preferences and feeding morphology of two pulmonate snail species from a small New England pond, U.S.A. Canadian Journal of Zoology 64:2570-2575. Kohn, A. J. 1983. Feeding biology of gastropods. In: The Mollusca, Vol. 5. Physiology, Part 2, A. S. M. Saleuddin and K. M. Wilbur, eds. pp. 2-53. Academic Press, London. Lemche, H. 1956. The anatomy and histology of Cylichna (Gastropoda: Tectibranchia). Spolia Zoologica Musei Hauniensis 16:1-278. Lipps, J. H. and J. W. Valentine. 1970. The role of Foraminifera in the trophic structure of marine communities. Lethaia 3:279-286. Loosanoff, V. L., H. C. Davis and P. E. Chanley. 1966. Dimensions and shapes of larvae of some marine bivalve mollusks. Malacologia 4:351-435. MacPherson, J. H. and C. J. Gabriel. 1962. Marine Molllusks of Victoria. Melbourne University Press, Melbourne. 475 pp. Mikkelsen, P. S. and P. M. Mikkelsen. 1984. Comparison of Acteocina canaliculata (Say, 1826), A. candei (d’Orbigny, 1841), and A. atrata spec. nov. (Gastropoda: Cephalaspidea). Veliger 27:164-192. Morton, B. and S. T. Chiu. 1990. The diet, prey size and consumption of Philine orientalis (Opisthobranchia: Philinidae) in Hong Kong. Journal of Molluscan Studies 56:289-299. Murray, J. W. 1973. Distribution and Ecology of Living Benthic Foraminiferids. Crane, Russak and Co., New York. 274 pp. Paine, R. T. 1988. Food webs: road maps of interactions or grist for theoretical development? Ecology 69:1648-1654. Rudman, W. B. 197la. Structure and functioning of the gut in the Bullo- morpha (Opisthobranchia), Part I, Herbivores. Journal of Natural History 5:647-675. Rudman, W. B. 1971b. On the opisthobranch genus Haminoea Turton and Kingston. Pacific Science 25:545-559. Rudman, W. B. 1972a. Structure and functioning of the gut in the Bullo- morpha (Opisthobranchia), Part II. Acteonidae. Journal of Natural History 6:311-324. Rudman, W. B. 1972b. Structure and functioning of the gut in the Bullomorpha (Opisthobranchia), Part IV. Aglajidae. Journal of Natural History 6:547-560. Rudman, W. B. 1972c. Structure and functioning of the gut in the Bullomorpha (Opisthobranchia), Part III. Philinidae. Journal of Natural History 6:459-474, Sokal, R. R. and F. J. Rohlf. 1981. Biometry, 2nd ed. W. H. Freeman and Co., New York. 859 pp. Shonman, D. and J. W. Nybakken. 1978. Food preference, food availability, and food resource partitioning in two sympatric species of cephalaspi- dean opisthobranchs. Veliger 21:120-126. Taylor, J. D. 1980. Diets of sublittoral predatory gastropods of Hong Kong. In: The Marine Fauna and Flora of Hong Kong and Southern China. B. C. Morton and C. K. Tseng, eds. pp. 907-920. Hong Kong Univer- sity Press, Hong Kong. Todd, R. and D. Low. 1981. Marine fauna and flora of the northeastern United States. Protozoa: Sarcodina: benthic Foraminifera. National Oceanographic and Atmospheric Administration (NOAA) Technical Report (National Marine Fisheries Service) Circular 439. 51 pp. Yonow, N. 1989. Feeding observations on Acteon tornatilis (Linnaeus) (Opisthobranchia: Acteonidae). Journal of Molluscan Studies 55:97-102. Date of manuscript acceptance: 4 September 1991 ~ pris : { ‘ (i F i a J Research Note Comparative Morphology of the Byssi of Dreissena polymorpha and Mytilus edulis Larry R. Eckroat and Louise M. Steele* The Pennsylvania State University at Erie, The Behrend College, Erie, Pennsylvania 16563-0800, U.S.A. Abstract. Scanning electron micrographs reveal differences in byssal stems, threads and plaques of two common biofouling mussels, the zebra mussel Dreissena polymorpha (Pallas), and the blue mussel Myrilus edulis (Linnaeus). These differences include anchorage of the stem, the pattern in which threads branch from the growing stem, thread surface topography, plaque orientation, and the morphology of the region of the thread that extends into the plaque. Morphological dissimilarities are likely to be related to differences in the mechanics of byssus formation and to differences in the surface texture and length of the ventral groove. Morphological differences between byssi of various biofouling mussels should be considered as researchers develop and adapt mechanisms of control. The post-larvae of most bivalve molluscs possess byssal structures by which they are secured to the substratum as they undergo metamorphosis (Yonge, 1962). Some bivalves retain a byssal apparatus into the adult stage. For example, the marine blue mussel Mytilus edulis (Linnaeus), which is well known for its commercial importance, and the freshwater zebra mussel Dreissena polymorpha (Pallas), a recent invader of the Great Lakes (see Mackie ef al., 1989 for a review of the zoogeography of D. polymorpha), are two species that retain a byssus as juveniles. Thus, the byssus plays a major role early in the lives of most bivalves and is important throughout the lives of some species. Purchon (1977) lists two major advantages to the presence of a byssal attachment. First, by forming a byssal attachment, molluscs do not expend energy unnecessarily to maintain position on a substratum. In addition, when the tide ebbs or the water level falls, a byssate mollusc can withdraw its foot and close its shell valves to prevent desiccation. Mussels are, therefore, regarded as being among the most difficult fouling organisms to control because they are capable of withstanding conditions that eliminate most other species. For instance, chlorination has been used in attempts to con- trol both Mytilus edulis and Dreissena polymorpha. Holmes (1970) reports that in M. edulis, one of the most serious of- fenders in fouling marine intake systems (Clapp, 1950), the principal effect of chlorination was a depression in the ac- tivity of the foot, leading to a reduction in the number of threads formed. In addition, chlorination interfered with the *Current address: Case Western Reserve University, School of Medicine, Cleveland, Ohio 44106, U.S.A. quinone tanning process of thread formation, so that the threads formed were weaker than those of unchlorinated animals (Holmes, 1970). Therefore, although chlorination created a flow dependent distribution of mussels due to weakening of their byssus attachment systems, chlorination failed to suppress fouling by M. edulis in areas of low water flow. In D. polymorpha, chlorination is effective in control- ling zebra mussel larvae, but adult mussels are able to pro- tect themselves from intermittent doses of chlorine by ad- duction (Anon., 1991). Since zebra mussels were first identified in the Great Lakes as a two-year-old age class in 1988 (Hebert ef al. , 1989), they have been causing fouling problems for users of raw water by accumulating on exposed surfaces and physically blocking intake pipes by attaching to one another to form layers as thick as 0.3 m (Clarke, 1952; Griffiths et al. , 1989). Because the zebra mussel contributes to the biofouling of water supplies primarily by forming byssal attachments, the study of byssus morphology could lead to the development of a method of controlling this mollusc. It has been reported that the byssus of Dreissena poly- morpha is similar to the byssal apparatus found in Mytilus edulis and other marine bivalves (Moore, 1991). In previous studies, however, the authors (Steele, 1991; Eckroat ef al., 1992) observed that the structure of the byssus of D. poly- morpha differed from the byssal structure of M. edulis that had been described in the literature (Brown, 1952; Smeathers and Vincent, 1979). Therefore, the current study was under- taken to clearly delineate differences between the byssi of D. polymorpha and M. edulis by making morphological American Malacological Bulletin, Vol. 10(1) (1993):103-108 103 104 AMER. MALAC. BULL. 10(1) (1993) comparisons of their stems, threads, and plaques. Accurate comparisons are especially important because morphological differences between the byssi of various biofouling mussels should be recognized as researchers develop and adapt con- trol mechanisms that affect the byssus. MATERIALS AND METHODS Live specimens of Dreissena polymorpha, 2.1-2.5 cm in length, were collected at Lampe Marina on Presque Isle, Erie, Pennsylvania, during October 1990 and maintained in a 15°C freshwater aquarium for approximately one month, where they were fed 200 ml of an alga mixture (3 g of Chlorella spp. blended in | L of water) daily. Additional live specimens of D. polymorpha, 1.5-2.3 cm in length, were col- lected in Thompson’s Bay on Presque Isle, Erie, Penn- sylvania, during May 1991 and maintained in a 15°C freshwater aquarium overnight before their byssal threads were prepared for examination. Live specimens of Mytilus edulis measuring approximatelyt 1.0 cm and 5.0-6.5 cm were collected at Wachapreague, Virginia, and at Rye, New Hamp- shire, respectively, and maintained in a 24°C saltwater aquarium (31 ppt) for 14 days and fed marine Chlorella spp. daily as described above. To prepare specimens for scanning electron micro- scopy, either the byssi were removed, or the mussels’ shells were opened so that the byssi could be studied intact. Specimens were fixed in 5% glutaraldehyde in sodium phos- phate buffer (pH 7.2) for 24-48 hours at 5°C, washed in buf- fer alone for 15 minutes, and dehydrated in a graded ethanol series. After specimens were critical point dried using car- bon dioxide as the transitional fluid, they were mounted on aluminum stubs and sputter coated with gold-palladium. Specimens were examined and photographed with a Hitachi S-570 scanning electron microscope. RESULTS Scanning electron micrographs provided more detailed views of byssal structures that have been diagrammatically represented by other workers (Brown, 1952; Smeathers and Vincent, 1979). Comparisons between the byssi of Dreissena polymorpha and Mytilus edulis were limited to noting morph- ological differences because age differences of specimens make size comparisons of structures meaningless. In intact specimens of Dreissena polymorpha, the byssal stem was concealed inside a collar-like structure that was continuous with the foot, and the region where the byssal threads branched from the stem was not visible (Fig. 1). The stem of Mytilus edulis, which emerged from a raised area at the base of the foot was, however, visible in intact specimens, and the locations at which the threads branched from the stem were exposed (Fig. 2). Additional observations showed that the ventral groove extended to the distal tip of the foot of M. edulis, but extended only about half the length of the foot in D. polymorpha. Removal of the byssus from specimens of Dreissena polymorpha revealed that most of their byssal threads branched from one linear location of the stem and from all around the stem’s circumference (Fig. 3). Cuffs or sheaths, formed by the outer laminae, were observed at the bases of the threads in some specimens (Eckroat et al., 1992). Ex- amination of intact specimens of Mytilus edulis showed that threads, attached to the stem by overlapping rings, branched from progressively more distal locations along two opposite edges of the stem (Fig. 4). In the proximal region, threads of Dreissena poly- morpha were cylindrical and smooth (Fig. 5), but the prox- imal portions of threads of Mytilus edulis were flattened with a crimped edge and had corrugations or folds around their circumferences (Fig. 6). The distal half of threads of D. polymorpha became increasingly rough with longitudinal ridges that were most pronounced at the distal ends of the threads (Fig. 7). Threads of M. edulis, however, were smooth in the distal region (Fig. 8). Threads of Dreissena polymorpha and Mytilus edulis formed oblique angles with the substratum and ended distal- ly in thin, oval plaques (bottom portions of Figs. 9 and 10). In D. polymorpha, threads became broader, flattened, and bifurcated as they expanded along the plaque’s wide axis (Fig. 9, near top), which was oriented perpendicular to the longitudinal axis of the thread (bottom portion of Fig. 9). Threads of M. edulis, however, trifurcated and formed thin, root-like extensions that continued into the plaque (top por- tion of Fig. 10). The narrow axis of the plaque was oriented perpendicular to the longitudinal axis of the thread (Fig. 10). DISCUSSION Byssus morphology is related to the mechanical events that occur during byssus formation. One of the first mechanical events involves the movement of the foot tightly against the substratum so that the mussel can secrete a plaque (Waite, 1983; Eckroat et al., 1992), which is molded by the distal depression of the ventral groove of the foot (Brown, 1952; Waite, 1983). After a plaque has been formed, an in- dividual thread, which is of a fluid consistency, is extruded from the ventral groove (Smyth, 1954; Waite, 1983). The plaques of Mytilus edulis and Dreissena polymorpha are very thin at their peripheral regions, but their orientations on the substratum differ. In addition, the structure of the portion of the thread that continues into the plaque is dissimilar in the two species. These differences in plaque orientation and distal thread morphology could result because the ventral groove does not extend to the distal tip of the foot in D. polymorpha as it does in M. edulis, which could affect the ECKROAT AND STEELE: COMPARISON OF BYSSI 105 Fig. 1. Intact byssus of Dreissena polymorpha with threads emerging from a collar-like structure that conceals the stem (scale bar = 1.00 mm). Fig. 2. Intact byssus of Mytilus edulis with the stem emerging from a raised structure near the base of the foot (scale bar = 0.30 mm). Fig. 3. Stem of D. polymorpha. Threads branch from around the circumference of one proximal-distal location of the stem (scale bar = 0.60 mm). Fig. 4. Stem of M. edulis. Threads branch from progressively distal regions along two opposite sides of the stem and are attached by overlapping rings (scale bar = 0.50 mm). 106 AMER. MALAC. BULL. 10(1) (1993) ii i | 4 * ty ae Se MCN oe EK: SOOKE Fig. 5. Smooth, proximal portion of thread of Dreissena polymorpha (scale bar = 10.0 um). Fig. 6. Flattened proximal portion of thread of Mytilus edulis with crimped edge and folds around circumference (scale bar = 43.0 ym). Fig. 7. Distal portion of thread of D. polymorpha with longitudinal ridges (scale bar = 23.1 pm). Fig. 8. Smooth, distal portion of thread of M. edulis (scale bar = 38.0 um). ECKROAT AND STEELE: COMPARISON OF BYSSI 107 shape of the distal depression as the foot is pressed against a substratum. In addition, differences could arise in the shape of the distal depression that could affect the morphology of the distal ends of the threads if the muscle action of the foot differs in the two species. As threads are secreted as a liquid, they are molded to the walls of the ventral groove (Brown, 1952; Smyth, 1954; Bairati and Vitellaro-Zuccarello, 1974; Waite, 1983). Our results, which agree with those of other workers (Brown, 1952; Smeathers and Vincent, 1979), show that the threads of Mytilus edulis are corrugated in the proximal third and smooth in the distal two-thirds. The threads of Dreissena polymorpha, which are smooth in the proximal half and rough with longitudinal ridges in the distal half, therefore, differ from those of M. edulis. These thread surface differences sug- gest that the walls of the ventral groove of D. polymorpha have a different surface topography than those of M. edulis. Once a thread has been completed, specimens of Mytilus edulis change position and repeat the process to form another thread (Clapp, 1950). The changes in position made by Dreissena polymorpha and M. edulis could differ because in the two species, the patterns in which the threads branch from the stem differ. In D. polymorpha, threads branch from around the circumference of the stem, but in M. edulis, threads branch from two opposite sides. The ways in which the threads of the two species are attached to the stem are dissimilar. Our observations agree with those of Brown (1952), who reported that the threads of Mytilus edulis are attached to the byssus with overlapping rings that are fused with the fibrous laminae of the byssal root, while the threads of some Dreissena polymorpha specimens have cuffs at their bases (Eckroat et al., 1992). Unlike the threads of M. edulis, which branch from pro- gressively distal locations along the stem, threads of D. polymorpha branch from the stem at one linear location. Brown (1952) explained that in M. edulis, as the fibrous laminae of the root increase in length, the stem lengthens and older threads are carried farther from the body of the organism. The thread branching pattern observed in D. polymorpha, however, suggests that as threads are added to the byssus, the stem does not lengthen in the region where the threads branch. New threads are attached to the outside of the stem, which increases stem diameter but not stem length. The byssi of Dreissena polymorpha and Mytilus edulis differ in the anchorage of the stem, the pattern in which threads branch from the growing stem, thread surface topography, plaque orientation, and the morphology of the region of the thread that extends into the plaque. Therefore, although previous reports (Moore, 1991) stated that the byssi of these molluscs were similar, marked structural differences exist. Because byssal attachment is fundamental to the suc- cess of mussels that colonize hard substrata, information concerning structural characteristics of the byssus could be used together with knowledge of thread formation and attach- ment to develop or adapt mechanisms of controlling biofoul- ing mussels. It is therefore important to realize that among different molluscan species, the byssal apparatus has diverse designs. As control mechanisms are proposed, researchers should recognize such morphological dissimilarities and the possible differences in the mechanics of byssus formation that they could reflect. ACKNOWLEDGMENTS The authors thank Dr. Bruce Barber of The Virginia Institute of Marine Sciences and Dr. Nadine Folino of Philips Exeter Academy for collecting specimens of Mytilus edulis, William S. Barbour for collecting specimens of Dreissena polymorpha, and Kathy Mauro for typing the manuscript. LITERATURE CITED Anon. 1991. Industries Learn to Control Zebra Mussels. Seiche 1991 (Spring):7. Minnesota Sea Grant, University of Minnesota. Bairati, A. and L. Vitellaro-Zuccarello. 1974. The ultrastructure of the byssal apparatus of Mytilus galloprovincialis, 11. Observations by microdissec- tion and scanning electron microscopy. Marine Biology 28:145-158. Brown, C. H. 1952. Some structural proteins of Mytilus edulis. Quarterly Journal of Microscopic Science, Third Series 93:487-502. Clapp, W. F. 1950. Some biological fundamentals of marine fouling. Trans- actions of the American Society of Mechanical Engineers 72:101-107. Clarke, K. B. 1952. The infestation of waterworks by Dreissena polymorpha, a freshwater mussel. Journal of the Institute of Water Engineers 6:370-379. Eckroat, L. R., E. C. Masteller, J. C. Shaffer, and L. M. Steele. 1992. The byssus of the zebra mussel: Morphology, byssal thread formation, and detachment. In: Zebra Mussels: Biology, Impacts, and Control. T. F. Nalepa and D. W. Schloesser, eds. pp. 239-263. Lewis Publishers, Inc., Chelsea, Michigan. Griffiths, R. W., W. P. Kovalak, and D. W. Schloesser. 1989. The zebra mussel, Dreissena polymorpha (Pallas, 1771), in North America: Im- pact on raw water users. Jn: Proceedings of the Service Water Reliability Improvement Seminar. pp. 1-26. Electric Power Research Institute, Palo Alto, California. Hebert, P. D. N., B. W. Muncaster, and G. L. Mackie. 1989. Ecological and genetic studies on Dreissena polymorpha (Pallas): A new mollusc in the Great Lakes. Canadian Journal of Fisheries and Aquatic Science 46:1587-1591. Holmes, N. 1970. Marine fouling in power stations. Marine Pollution Bulletin 1:105-106. Mackie, G. L., W. N. Gibbons, B. W. Muncaster, and I. M. Gray. 1989. The zebra mussel, Dreissena polymorpha: A synthesis of European experiences and a preview for North America. Ontario, Queen’s Printer for Ontario. 76 pp. Moore, S. G. 1991. Structure and formation of the byssus. Dreissena poly- morpha: Information Review 2(1):4-5. New York Sea Grant Extension. Purcheon, R.D. 1977. The Biology of the Mollusca. Pergamon Press, New York. 596 pp. Smeathers, J. E. and J. F. V. Vincent. 1979. Mechanical properties of mussel byssus threads. Journal of Molluscan Studies 45:219-230. Smyth, J. D. 1954. A technique for the histochemical demonstration of poly- phenoloxidase and its application to eggshell formation in helminths and byssus formation in Mytilus. Quarterly Journal of Microscopical Science 95, part 2:139-152. 108 AMER. MALAC. BULL. 10(1) (1993) Fig. 9. Plaque of Dreissena polymorpha oriented with its wide axis perpendicular to the longitudinal axis of the thread in perpendicular vertical planes (bottom - scale bar = 0.43 mm). Broad, flattened portion of thread expanding along wide axis of plaque (top - scale bar = 86.0 nm). Fig. 10. Plaques of Mytilus edulis. The narrow axis of each plaque is perpendicular to the longitudinal axis of the thread to which it is attached (bottom - scale bar = 1.0 mm). Thin, root-like extensions spreading into plaque (top - scale bar = 0.20 mm). Steele, L. M. 1991. Structural characteristics of the byssus of the zebra mussel, Yonge, C. M. 1962. On the primitive significance of the byssus in the bivalvia Dreissena polymorpha (Pallas), as seen using scanning electron and its effects in evolution. Journal of the Marine Biological Associa- microscopy. Proceedings of The Fifth National Conference on tion of the United Kingdom 42:113-125. Undergraduate Research, Vol. 2:871-876. Waite, J. H. 1983. Adhesion in byssally attached bivalves. Biological Review Date of manuscript acceptance: 15 November 1991 58:209-231. BOOK REVIEW Zoological Catalogue of Australia, Vol. 8. Non-Marine Mollusca Brian J. Smith, Australian Government Publishing Service. 1991. 405 pp. Austr. $49.95. A new checklist of Australian non-marine molluscs has been published, almost 50 years after Iredale produced the last checklists dealing with these animals. This work has been the product of Brian Smith’s labours over many years - he admits to 10 years in the introduction - during which time Australian malacologists eagerly awaited its completion. There could be few more tedious occupations than pro- ducing a checklist, especially one with the detail required in the Zoological Catalogue of Australia, of which this work is volume 8. The work lists over 2,000 available species names, over 1,000 of these recognised as valid species names, placed in about 400 genera in 57 families. These figures also include taxa restricted to brackish water - marginal marine groups (e.g. Ellobiidae, Assimineidae). Smith estimates that 30-40% of the Australian non-marine fauna remains undescribed. The work itself suffers somewhat from being forced into an editorial straight jacket for which the author is not responsible. It was designed for terrestrial vertebrate checklists and the limitations of what apparently was an anti- quated data base system. For example all references are spelled out in full - yes, really in full - title as well as a journal reference and the usual pagination, etc. every time a reference is cited - even if it appears several times on the one page (as is often the case). Needless to say this curious policy makes the work more voluminous than it otherwise might have been. In addition the rigid format does not allow remarks under 109 genera or species - only under families, with the result that important comments relating to particular genera or species are easily overlooked. It is also unfortunate that available names that can without doubt be placed in a genus, but can- not be readily assigned to ‘‘valid’’ species as synonyms, or reliably treated as a ‘“‘valid’’ species in their own right, are not listed under the genus but in a ragbag of ‘‘incertae cedis’’ at the end of the family. This work is important and timely, not just because it’s a useful reference to the fauna, but because it’s the only reference that we have that gives us anything like a real pic- ture of the diversity of non-marine molluscs in Australia. Given Smith’s estimate of 1,000 valid species, with 30-40% undescribed, gives a total fauna of 1,300-1,400 species (some estimates suggest 2,000), dispelling the myth that the Australian non-marine molluscan fauna is_ relatively depauperate. Because many genera are in a chaotic state tax- onomically, a number of taxonomic decisions have been made in this work, some in conjunction with other workers. This work will be a valuable addition to the library of anyone with an interest in non-marine molluscs - and essential to anyone working with the Australian fauna. —Winston Ponder Australian Museum Sydney NSW 2000 Australia L 59th ANNUAL MEETING THE AMERICAN MALACOLOGICAL UNION ABOARD THE NORDIC EMPRESS WITH STOPS AT FREEPORT, NASSAU, AND COCOCAY BAHAMAS JUNE 21- 25, 1993 Following the traditional format, the meeting will include a symposium, featuring pat- terns of speciation in molluscs, contributed papers sessions on diverse aspects of molluscan biology, field trips to various islands, and an auction of publications and other items of malacological interest. The Bahamian Islands provide unique opportunities for biologists to study and compare tropical insular faunas with the temperate faunas of North America, as well as providing an exciting and unforgettable workshop for family members and conference attendees alike. This is a very unusual and innovative program which we were fortunate to arrange at an unbelievably low cost. The offer is available to members PROVIDING WE MAKE COMMITMENTS TO RESERVING SPACE. We need to receive your deposit of $100.00 per person to secure cabin and meeting spaces. Rates are for double occupancy per person. You may specify your choice for a room- mate. Discounted airfares are available for those who need to fly to Miami. Outside cabin: $675.00 Inside cabin: $598.50 Space is limited. Early deposit is required to reserve your place. For further information please contact: Fred G. Thompson President, AMU Florida Museum of Natural History University of Florida Gainesville, Florida 32611 (904) 392-1721 ll CONTRIBUTOR INFORMATION The American Malacological Bulletin serves as an out- let for reporting notable contributions in malacological re- search. Manuscripts concerning any aspect of original, unpublished research, important short reports, and detailed reviews dealing with molluscs will be considered for publication. Each original manuscript and accompanying illustra- tions must be submitted with two additional copies for review purposes. Text must be typed on one side of 8% x Il inch bond paper, double-spaced, and all pages numbered con- secutively with numbers appearing in the upper right hand corner of each page. Leave ample margins on all sides. Form of the manuscript should follow that outlined in the Council of Biology Editors Style Manual (fifth edition, 1983). This can be purchased from the CBE, 9650 Rockville Pike, Bethesda, Maryland 20814, U.S.A. Text, when appropriate, should be arranged in sections as follows: 1. Cover page with title, author(s) and address(es), and suggested running title of no more than 50 characters and spaces. Authors should also supply five key words, placed at the base of this page, for indexing purposes. 2. Abstract (less than 5% of manuscript length) 3. Text of manuscript starting with a brief introduc- tion followed by methodology, results, and discus- sion. Separate sections of text with centered sub- titles in capital letters. 4. Acknowledgments 5. Literature cited 6. Figure captions All binomens must include the author attributed to that taxon the first time the name appears in the manuscript [e.g. Crassostrea virginica (Gmelin)]. This includes non-molluscan taxa. The full generic name along with specific epithet should be written out the first time that taxon is referred to in each paragragh. The generic name can be abbreviated in the re- mainder of the paragraph as follows: C. virginica. References should be cited within text as follows: Hillis (1989) or (Hillis, 1989). Dual authorship should be cited as follows: Yonge and Thompson (1976) or (Yonge and Thomp- son, 1976). Multiple authors of a single article should be cited as follows: Beattie et al. (1980) or (Beattie et al., 1980). In the literature cited section of the manuscript refer- ences must also be typed double spaced. All authors must be fully identified, listed alphabetically and journal titles must be unabbreviated. Citations should appear as follows: Beattie, J.H., K.K. Chew, and W.K. Hershberger. 1980. Dif- ferential survival of selected strains of Pacific oysters (Crassostrea gigas) during summer mortality. Pro- ceedings of the National Shellfisheries Association 70(2):184-189. Hillis, D.M. 1989. Genetic consequences of partial self- fertilization on population of Liguus fasciatus (Mollusca: Pulmonata: Bulimulidae). American Malacological Bulletin 7(1):7-12. Seed, R. 1980. Shell growth and form in the Bivalvia. Jn: Skeletal Growth of Aquatic Organisms, D.C. Rhoads and R.A. Lutz, eds. pp. 23-67. Plenum Press, New York. Yonge, C.M. and T.E. Thompson. 1976. Living Marine Molluscs. William Collins Son & Co., Ltd., London. 288 pp. Illustrations should be clearly detailed and readily reproducible. All line drawings should be in black, high quali- ty ink. Photographs must be on glossy, high contrast paper. All diagrams must be numbered in the lower right hand cor- ners and adequately labeled with sufficiently large labels to prevent obscurance with reduction by one half. Magnifica- tion bars must appear on the figure, or the caption must read Horizontal field width = xmm or xpm. All measurements must be in metric units. All illustrations submitted for publica- tion must be fully cropped, mounted on a firm white back- ing ready for reproduction, and have author’s name, paper title, and figure number on the back. All figures in plates must be nearly contiguous. Additional figures submitted for review purposes must be of high quality reproduction. Xerographic reproduction of photomicrographs or detailed photographs will not be acceptable for review. Abbreviations used in figures should occur in the figure caption. Indicate in text margins the appropriate location in which figures should appear. Color illustrations can be included at extra cost to the author. Original illustrations will be returned to author if requested. Any manuscript not conforming to AMB format will be returned to the author for revision. New Taxa. The Bulletin welcomes complete descriptions of new molluscan taxa. Establishment of new taxa must con- form with the International Code of Zoological Nomenclature (1985). Descriptions of new species-level taxa must include the following information in the order as given: higher taxon designation as needed for clarity; family name with author and date; generic name with author and date; Genus species author sp. nov. followed by numeration of all figures and tables; complete synonymy (if any); listing of type material with holotype and any other type material clearly designated along with complete museum catalogue or accession infor- mation; listing of all additional non-type material also with full museum deposition information; type locality; diagnosis and full description of material done in telegraphic style in- cluding measurements and zoogeographic distribution as necessary; discussion; etymology. Descriptions of new supra- specific taxa should include type species (for new genus) or type genus (for new family), diagnosis and full description done in telegraphic style, and list of included taxa. Proofs. Page proofs will be sent to the author and must be checked for printer’s errors and returned to the printer within a three day period. Significant changes in text, other than printer errors, will produce publishing charges that will be billed to the author. Mailing. All overseas mailing must be done via airmail. The American Malacological Union will not be responsible for deferred publication of manuscripts delayed in surface mail. Charges. There are no mandatory page costs to authors lack- ing financial support. Authors with institutional, grant or other research support will be billed for page charges. The current rate is $35.00 per printed page. Acceptance and ulti- mate publication is in no way based on ability to pay page costs. Reprints. Order forms and reprint cost information will be sent with page proofs. The author receiving the order form is responsible for insuring that orders for any coauthors are also placed at that time. Submission. Submit all manuscripts to Dr. Robert S. Prezant, Editor-in-Chief, American Malacological Bulletin, Depart- ment of Biology, Indiana University of Pennsylvania, Indiana, Pennsylvania, 15705-1090, U.S.A. Subscription Costs. Institutional subscriptions are available at a cost of $32.00 per volume. [Volumes 1 and 2 are available for $18.00 per volume]. Membership in the American Malacological Union, which includes personal subscriptions to the Bulletin, is available for $25.00 ($15.00 for students). All prices quoted are in U.S. funds. Outside the U.S. postal zones, add $5.00 seamail and $10.00 airmail per volume or membership. For subscription or membership information contact AMU Secretary-Treasurer, David Hargreave, Depart- ment of Science Studies, Western Michigan University, Kalamazoo, Michigan, 49008, U.S.A. AMERICAN MALACOLOGICAL BULLETIN VOLUME 10 NUMBER 2 Biannual Journal of the American Malacological Union s CONTENTS Editorial: COmmMents saz jc ccs sees ccstcs Giseete tec cune avis ctsnenlcvtooes ber asbevetis ox subiosseviae Metos ies iereseedeavbelbexricatas So eaeaeee: iJ TROY CWS 11S ge csicci sa coiss aa vey site nes tees a eases cece tncd naca bates pst aecan aces aetna vena esdunatypeainadeiniete estan uevauntnaiatecnehee iV Anatomy and systematics of a new species from China, Aegista laoyelingensis (Stylommatophora; Bradybaenidae). LU ZHANG... cece cece eeeeeeceeeceeseeeeaeenetenesseseaeeneeeaeeaees 113 Annual gonadal cycle of the land snail Scutalus tupacii (Pulmonata: Bulimulidae). MARIA GABRIELA CUEZZO | xaciccecisscssaveisesesbedaessecesdeas}saynesseseeiceassiasssestaredetsatevonseteseraveasaeeancanenes 121 Functional anatomy of Fossula fossiculifera (D’ Orbigny, 1843) (Bivalvia: Mycetopodidae). WAGNER E. P. AVELAR...w..0.. eee ccc eeeeneesetaeeeeeeneeneeeneaes 129 ( Functional morphology of Heterodonax bimaculatus (Linné, 1758) (Bivalvia: J Psammobiidae). WALTER NARCHI and OSMAR DOMANESCHI......000. ee f roe 139 A Origin and decline of the estuarine clam Rangia cuneata in the Neches River, ; Et Texas. RICHARD C. HARREL...0000.0.. ccc ceceeseeeeseeeeeeeeseeseeaeseeeeeeeseeeesaeeatens Ls cijcemmeakes aaue 153 Research Note: Two common European viviparid species hybridize. ANDRZEJ FALNIOWSKI, ANDRZEJ KOZIK and MAGDALENA SZAROWSKA .....000o cece 161 Sententia: The Archaeogastropoda: a clade, a grade or what else? GERHARD HASZPRUNAR 2. scsicseces ccaccssesguaiesileitessnpevvastzadgeistecessosexssesvedusvencvtsesaveceis macsacdaecapeenieaaics 165 1992 AMU SYMPOSIUM ON BIOLOGY OF CARIBBEAN MOLLUSKS Giving and receiving: the tropical Atlantic as donor and recipient region for invading species. GEERAT J. VERMEIJ and GARY ROSENBERG ......0000.. ees 181 The evolution of “Chlamys” (Mollusca: Bivalvia: Pectinidae) in the tropical western Atlantic and eastern Pacific. THOMAS R. WALLER .......................... Stee G catia vetoes 195 The zoogeographic implications of the prosobranch gastropods of the Moin Formation of Costa Rica. DAVID G. ROBINSON .............ccccccccsssssscssscescseseeseesssseecstscsesacseees 251 A database approach to studies of molluscan taxonomy, biogeography and diversity, with examples from western Atlantic marine gastropods. SAIC ROUSEEIN BE IRG ic ss lcs cetera sa cate ene ag utitewsctsc tien Sop be avanasaceld de ewdonancahcaa nes rte aed nioteandinnens 257 A bibliography of Caribbean malacology 1826 - 1993. PAULA M. MIKKELSEN, RUDIGER BIELER and RICHARD E. PETIT ..00..00.0...cccccccccccccsccsecseceeeseeeaeceesesecesessssesssecsesees 267 — continued on back cover — AMERICAN MALACOLOGICAL BULLETIN ROBERT S. PREZANT, Editor-in-Chief Department of Biology BOARD OF EDITORS Indiana University of Pennsylvania Indiana, Pennsylvania 15705 ASSOCIATE EDITORS MELBOURNE R. CARRIKER College of Marine Studies University of Delaware Lewes, Delaware 19958 GEORGE M. DAVIS Department of Malacology The Academy of Natural Sciences Philadelphia, Pennsylvania 19103 R. TUCKER ABBOTT Melbourne, Florida, U.S.A. JOHN A. ALLEN Millport, United Kingdom JOHN M. ARNOLD Honolulu, Hawaii, U.S.A. JOSEPH C. BRITTON Fort Worth, Texas, U.S.A. JOHN B. BURCH Ann Arbor, Michigan, U.S.A. EDWIN W. CAKE, JR. Ocean Springs, Mississippi, U.S.A. PETER CALOW Sheffield, United Kingdom JOSEPH G. CARTER Chapel Hill, North Carolina, U.S.A. ARTHUR L. CLARKE Portland, Texas, U.S.A. CLEMENT L. COUNTS, III Princess Anne, Maryland, U.S.A. THOMAS DIETZ Baton Rouge, Louisiana, U.S.A. WILLIAM K. EMERSON New York, New York, U.S.A. DOROTHEA FRANZEN Bloomington, Illinois, U.S.A. ROGER HANLON Galveston, Texas, U.S.A. CONSTANCE E. BOONE, Ex Officio Malacology Department Houston Museum of Natural Science Houston, Texas 77030 BOARD OF REVIEWERS JOSEPH HELLER Jerusalem, Israel ROBERT E. HILLMAN Duxbury, Massachusetts, U.S.A. K. ELAINE HOAGLAND Washington, D.C., U.S.A. VICTOR S. KENNEDY Cambridge, Maryland, U.S.A. ALAN J. KOHN Seattle, Washington, U.S.A. LOUISE RUSSERT KRAEMER Fayetteville, Arkansas, U.S.A. JOHN N. KRAEUTER Baltimore, Maryland, U.S.A. ALAN M. KUZIRIAN Woods Hole, Massachusetts, U.S.A. RICHARD A. LUTZ Piscataway, New Jersey, U.S.A. GERALD L. MACKIE Guelph, Ontario, Canada EMILE A. MALEK New Orleans, Louisiana, U.S.A. MICHAEL MAZURKIEWICZ Portland, Maine, U.S.A. JAMES H. McLEAN Los Angeles, California, U.S.A. ROBERT F. MCMAHON Arlington, Texas, U.S.A. RONALD B. TOLL, Managing Editor Department of Biology Wesleyan College Macon, Georgia 31297-4299 W. D. RUSSELL-HUNTER Department of Biology Syracuse University Syracuse, New York 13210 THOMAS R. WALLER Department of Paleobiology Smithsonian Institution Washington, D. C. 29560 ANDREW C. MILLER Vicksburg, Mississippi, U.S.A. BRIAN MORTON Hong Kong JAMES J. MURRAY, JR. Charlottesville, Virginia, U.S.A. RICHARD NEVES Blacksburg, Virginia, U.S.A. JAMES W. NYBAKKEN Moss Landing, California, U.S.A. A. RICHARD PALMER Edmonton, Canada WINSTON F. PONDER Sydney, Australia CLYDE F. E. ROPER Washington, D.C., U.S.A. NORMAN W. RUNHAM Bangor, United Kingdom AMELIE SCHELTEMA Woods Hole, Massachusetts, U.S.A. DAVID H. STANSBERY Columbus, Ohio, U.S.A. FRED G. THOMPSON Gainesville, Florida, U.S.A. NORMITSU WATABE Columbia, South Carolina, U.S.A. KARL M. WILBUR Durham, North Carolina, U.S.A. Cover. Io fluvialis (Say, 1825) is the logo of the American Malacological Union. THE AMERICAN MALACOLOGICAL BULLETIN is the official journal publication of the American Malacological Union. AMER. MALAC. BULL. 10(2) ISSN 0740-2783 AMERICAN MALACOLOGICAL BULLETIN VOLUME 10 NUMBER 2 Biannual Journal of the American Malacological Union CONTENTS Ecitomialk@@rimien tse cxseseenes Sse it eaten ps satech adecnertnerce ee ee bee bs donaeks Re Vie WEL EIS Ue resree ata fers un vie cipne eat viadi stan eta eomenncti ate tenn eomtc ttre ean, es $2 Ram aria Av Anatomy and systematics of a new species from China, Aegista laoyelingensis (Stylommatophora; Bradybaenidae). LU ZHANG..........c.ccccscssssssesesssseseseesesesescscsesesesesssetesesesscseessessl 13 Annual gonadal cycle of the land snail Scutalus tupacii (Pulmonata: Bulimulidae). IVEAIRTATGAB REE ACO ULL ago eo rsac cess ccs ears atteeneeue etre hc ceeds ace aege a eusieen es tee ase 121 Functional anatomy of Fossula fossiculifera (D’ Orbigny, 1843) (Bivalvia: Mycetopodidae). WAGNER E. P. AVELAR....0... cc cieccceneceececeeeeeeeeeeeeeeseestenseaes 129 Functional morphology of Heterodonax bimaculatus (Linné, 1758) (Bivalvia: Psammobiidae). WALTER NARCHI and OSMAR DOMANESCHI ......000. ccc eeeeeeeeteeeees 139 Origin and decline of the estuarine clam Rangia cuneata in the Neches River, TeX aS: CAR DG BRAID oo oc ets ae ccdee tstactoeic oes soak pesavers aetna tea eas oe eataronnede Teas 153 Research Note: Two common European viviparid species hybridize. ANDRZEJ FALNIOWSKI, ANDRZEJ KOZIK and MAGDALENA SZAROWSKA ......00000cceceeceeeeeees 161 Sententia: The Archaeogastropoda: a clade, a grade or what else? GERARD THASZ PRU INA RR olisytre rere ime nese. Dare asia uk tu anime tm nniotn deena lan poaeauuecesanrao eae rains 165 1992 AMU SYMPOSIUM ON BIOLOGY OF CARIBBEAN MOLLUSKS Giving and receiving: the tropical Atlantic as donor and recipient region for invading species. GEERAT J. VERMEIJ and GARY ROSENBERG .....00000..o eee 181 The evolution of “Chlamys” (Mollusca: Bivalvia: Pectinidae) in the tropical western Atlantic and eastern Pacific. THOMAS R. WALLER .....0.... ee ceccecceeteeeeeeeeteeeeseeeeeas 195 The zoogeographic implications of the prosobranch gastropods of the Moin Formation of Costa Rica. DAVID G. ROBINSON |... cece ceeceeeeeeeeseeeeeeesetseeeseeeseeeaees Zol A database approach to studies of molluscan taxonomy, biogeography and diversity, with examples from western Atlantic marine gastropods. GARYG ROSENBERG x52 oo osso occa leeae acts asec rsa eee Heusen carcanece senvecusdssuassoSaeesodasSesortatsc sata vspeeeusgadtuecaadsy 25 A bibliography of Caribbean malacology 1826 - 1993. PAULA M. MIKKELSEN, RUDIGER BIELER and RICHARD E, PETIT ..............ccccccccscsscssssesssscsscscsseacsccecsvsvcecsvsueeteveeeessvees 267 — continued — Review: Field Guide to Freshwater Mussels of the MidWest. .0...........cceceececceccccccccccscccesssessesssceecececceeseeeeeeeas 291 Piramal Report ..0ssiiacatssessieanteneer caste Sicteosaeseiavacansn coe tacars eee csveeetecceccnetar et aniceeet tie tate cj mmen en ees eee 292 JATIN OWUNCEMEN Gai ccc berccecscecccccscesce abe cckee vil ceasGiek eae taco eae e Doe tae eee ee eens ee ee 293 TEREM GINO TIGIN od iowa bess ease ues eas ace eee Se ee 294 EDITORIAL COMMENT Volume 10(2) of the American Malacological Bulletin will measure ten years of my efforts as Editor of the official AMU journal publication. During this time I have had the distinct pleasure and honor of working with some of the best mala- cologists and biologists that our times have to offer. The joint efforts of our Board of Editors, Board of Reviewers, innumer- able reviewers, authors, various AMU Councils and Publication Committees, and uncounted AMU members, have led to a journal that has grown in quality and distribution. With this issue I will terminate my Editorship of the Bulletin knowing that the dedication of these numerous individuals and groups will keep our AMU journal thriving through years to come. I would like, however, to personally acknowledge some individuals that put in more than their fair share of time and consistently helped navigate the success of our journal. The Editorial Board for Volume 1 of the American Malacological Bulletin, published in May 1983 by Braum Brumfield Press, was intimately involved with the development of the journal. That Board, composed of M. R. Carriker, G. M. Davis, A. J. Kohn (ex officio), R. Robertson, and W. D. Russell-Hunter, set the tone and pace for the journal. George Davis kindly took a great deal of his valuable time to meet with me at the Academy of Natural Sciences to give me “an inside look” at how a successful journal is run. I owe Louise Kraemer a special note of thanks for her early encouragement and unending patience. George Davis, Louise Kraemer and Mel Carriker offered tremendous support to my fledgling editorship. Gus Russell-Hunter and Robert Robertson were a steady source of good advice and support. Mel Carriker served as Chair of the Editorial Board and made sure our annual meetings ran smoothly and efficiently. Through the years others served on the Editorial Board, in ex officio capacity as President of the AMU, and I thank each for their time and energy. In 1988 Thomas Waller joined the Editorial Board and brought with him his enthusiasm, characteristic good sense and dedication to malacolo- gy. Tom now serves as Chair of the Editorial Board. Up through volume 4(2) I acted as sole Editor of the Bulletin. At that time, however, the journal had grown to a point where I could no longer handle day to day operation of the review process, copy editing, and finances of the journal, and maintain a faculty position, without assistance. The finances were confidently turned over to the Treasurer of AMU, Anne Joffe. The role of copy editing and acting as general liaison with our printer fell to our Managing Editor, Ronald B. Toll. To Ron I owe a great deal of thanks for his willingness to serve, and serve with distinction. As the newly elected Editor-in-Chief, Ron brings his experience to insure the progress and growth of our journal. In 1985 we moved to Shaughnessy Printers and have remained with them ever since. Denny Shaughnessy has provid- ed many kind suggestions for our “book” and helped insure a quality “product”. Carol Underwood's spirited production of the Bulletin through all phases of publication and her efforts, especially to meet last week’s deadline, have been greatly appreci- ated. On the institutional homefront, many thanks for the very professional aide of Deena Kelly and especially Linda Andrew who managed a continuous flow of correspondence and helped me maintain my own deadlines. Indiana University of Pennsylvania and the University of Southern Mississippi supported our AMU journal by making time, supplies, phone lines, mailing, and most importantly, secretarial assistance continually available. I have, for ten years, been able to continue with the Editorial responsibilities of the Bulletin, and carry on some sem- blance of an academic career, only because of the understanding, caring, encouragement, and positive outlook of my wife, Fran P. Prezant. Additional special thanks are extended to Connie Boone, Paula Mikkelsen, Bill Lyons, Dick Petit, Anne Joffe, Ray Neck, and the late Alan Solem and Dee Dundee for their support and encouragement, and for their dedication to and hopes for AMU and its journal. To all reviewers of manuscripts, and they number over 250 as listed in this issue, thank you for your time and stamina. The continued support of the American Malacological Union in building a professional jour- nal with international credentials, speaks for itself as we close a decade of Bulletin publication. We can look forward with confidence to the continued growth of the American Malacological Union and our AMU journal. Robert S. Prezant July 1993 lll REVIEWERS FOR THE AMERICAN MALACOLOGICAL BULLETIN VOLUMES 1 - 10 1983 - 1993 Publication of volume 10(2) of the American Malacological Bulletin represents the tenth anniversary of the official journal publication of the American Malacological Union. During this decade the journal expanded from one to two issues per year, and from the national to the international realm. The success of the Bulletin is directly attributable to the quality control exerted by the numerous reviewers we have used over the years. All of these individuals contributed to the growth of the AMU journal. The editors of the Bulletin would like to thank each reviewer for their dedication and professionalism through the years. AMB REVIEWERS: 1983 - 1993 Abbott, R. T. Conover, R. J. Jensen, K. R. Miller, W. D. Scheltema, R. S. Ahlstedt, S. Counts, C. L. Jokinen, E. Minichev, Y. S. Schikeyko, A. Aldridge, D. W. Crenshaw, M. Jolopainen, I. J. Moore, D. Schmekel, L. Allen, J. A. D'Asaro, C. N. Jones, A. Morse, M. P. Scott, P. Allmon, W. D. Dassart, G. B. J. Jones, D. Morton, B. Seapy, R. Anderson, G. Davis, G. M. Joosse, J. Morton, J. E. Sheetz, R. Andrews, J. D. Dietz, T. H. Kaas, P. Murawski, S. Shumay, S. Arnold, J. M. Dillon, R. T. Kay, E. A. Murray, H. Siddall, S. E. Aronson, R. B. Dockery, D. T. Keen, R. Murray, J. J. Sickel, J. Audesirk, G. Dundee, D. Kennedy, V. S. Neck, R. Slack-Smith, S. Audersirk, T. Earhart, G. Kesler, D. Neves, R. Slotta, L. S. Babrakzai, N. Edmunds, M. Kier, W. M. Newell, R. C. Smith, D. Baker, F. T. Eernissee, D. J. King, C. Newell, S. Smith, J. T. Baxter, J. M. Emberton, K. C. Koehn, R. K. Nixon, M. Solem, A. Bayne, C. J. Emerson, W. K. Kohn, A. J. Noore, F. Stansbery, D. H. Bieler, R. Eversole, A. G. Kool, S. Nybakken, J. Steneck, R. Boag, D. A. Fabry, V. Kraemer, L. R. O'Dor, R. K. Stickle, W. Bogan, A. E. Fairbanks, H. L. Kraeuter, J. N. 6) Foighil, D. Strathmann, R. R. Boicourt, B. Faulkner, J. Krantz, G. Quievreux, C. Tan Tiu, A. Boletzky, S. Fischer, F. Krenter, J. Packard, A. Taylor, R. Bonar, D. Franzen, D. Kuo, A. Y. Palmer, A. R. Theler, J. L. Boss, K. Fretter, V. Kupferman, I. Panley, G. Thiriot-Quievreux, G. Bouchet, P. Fritz, L. W. Kuzirian, A. M. Parmalee, P. W. Thompson, F. G. Bourne, G. Garcia, J. C. Lalli, C. M. Payne, B. S. Thompson, T. E. Bourne, N. Garton, D. W. Langton, C. Pechenik, J. Tillier, S. Brahme, S. B. Ghiselin, M. T. Larsen, J. Perron, F. Todd, C. D. Britton, J. C. Gilmer, R. W. Lauritsen, D. D. Petit, H. Toll, R. B. Brocco, S. Gold, J. R. Leise, E. Ponder, W. Tompa, A. Brock, V. Gordon, M. E. Leonard, J. Porter, H. Trdan, R. J. Brown, K. Gosliner, T. M. Lindberg, D. R. Pratt, W. Tripp, M. R. Brown, S. C. Gosling, E. M. Little, C. Prezant, R. S. Trueman, E. R. Buckley, D. E. Hadfield, M. G. Livingston, D. Priest, W. Twaroy, B. M. Bullock, R. C. Hamilton, P. Luciano, C. S. Pritchard, D. Van Der Spoel, S. Burch, J. B. Hanlon, R. T. Lunz, J. Prush, R. Vecchione, M. Burky, A. J. Hansen, T. A. Lutz, R. A. Reeder, R. Vokes, E. Butler, W. B. Harasewych, M. G. Lyons, W. G. Reynolds, P. Voltzow, J. Cake, E. W. Hargis, W. MacKenzie, C. Richter, G. Voss, G. Calloway, B. Harris, L. G. Mackie, G. L. Rittschoff, D. Voss, N. Calow, P. Hartfield, P. Malek, E. Rivest, B. Waite, H. Cameron, A. Hartwick, E. B. Mann, R. Robertson, R. Ward, P. Carefoot, T. H. Haskin, H. H. Martin, I. D. Rofft, D. A. Waller, T. R. Carriker, M. R. Haven, D. Mather, J. A. Rollins, H. Waren, A. Carter, J. G. Heard, W. H. Mattice, J. Rollinson, D. Watabe, N. Carter, N. Heller, J. A. Maxwell, P. Roper, C. F. E. Way, C. M. Castagna, M. Herbich, J. B. Mazurkiewicz, M. Ropes, J. Wen-Lung Wu, W. Chalermwat, K. Hickman, C. S. McClead, M. Rosenberg, G. D. Wilbur, K. Chambers, S. Hillman, R. E. McLeod, M. J. Rosewater, J. R. Willan, R. C. Cherry, D. Hirota, J. McLean, J. H. Roth, B. Woodrff, D. S. Christensen, C. C. Hoagland, K. E. McMahon, R. F. Rudman, W. B. Wright, C. Clark II, G. R. Hochberg, F. G. Menzel, R. W. Runham, N. W. Wu, S. K. Clark, K. B. Holopainen, I. J. Messenger, J. Russell-Hunter, W. D. Yochelson, E. E. Clarke, A. H. Hornbach, D. J. Metcalf, A. L. Saleuddin, A. S. M. Young, R. E. Clarke, D. Houbrick, R. S. Michel, H. B. Salvini-Plawan, L. von Coan, E. V. Hughs, R. N. Mikkelsen, P. M. Saul, L. E. Coney, C. C. Isom, B. G. Miller, A. Scheltema, A. iV Anatomy and systematics of a new species from China, Aegista laoyelingensis (Stylommatophora; Bradybaenidae) Lu Zhang* Department of Biology, Northeast Normal University, Changchun, Jilin, The People’s Republic of China Abstract: This paper describes a new species of pulmonate land snail, Aegista laoyelingensis, from Jilin Province, Northeastern China. The shell of this gastropod is small, light brown, unbanded, and has a rough surface. Whorls number about 5.5 + 0.1; the last whorl is never keeled. Aegista laoyelingen- sis has a short epiphallus, but lacks a flagellum. This species has a unique pattern of fusion of the central nervous system, and there is variation in mantle pigmentation. The Bradybaenidae are a large, widespread family of pulmonate land snails with a distribution centered in Southeast Asia but ranging to Europe and North America. The Bradybaenidae comprise the following subfamilies: Aegistinae Kuroda and Habe, 1951; Bradybaeninae Pilsbry, 1924; Euhadrinae Minato, 1988; Helicostylinae Ihering, 1909; and Monadeniinae Nordsieck, 1987 (see Miller and Naranjo-Garcia, 1991). Over 80 genera and subgenera are involved (Richardson, 1983). The species of Aegista are common snails distributed in China, Japan and nearby Islands. There are about 33 Chinese species of Aegista described by Heude (1882), Tryon (1888), Pilsbry (1895), Wiegmann (1900), Yen (1939), Chen and Gao (1987). Unfortunately, these authors based their descriptions solely on shell characters; they did not supply anatomical data. The purposes of the paper are: (1) to describe a new species from Northeastern China, Aegista laoyelingensis, and pre- sent the detailed anatomy of the new species; (2) to docu- ment the variation in mantle pigmentation; (3) to discuss the classification of the Aegistinae. MATERIALS AND METHODS Specimens were collected from a small area of Laoyeling (Mountains), Jilin Province, the People’s Re- public of China, kept in jars filled with fresh leaves, and brought back to the laboratory of Northeast Normal Uni- versity. After 5 hr, the living snails were drowned in dis- tilled water for about 24 hr and allowed to fully extend. The specimens were fixed in 95% ethanol for about | hr, and *Present Address: Department of Malacology, Academy of Natural Sciences, 1900 Benjamin Franklin Parkway, Philadelphia, Pennsylvania 19103-1195, U.S.A. then transferred to, and preserved in, 70% ethanol. Five adults of the alcohol-preserved specimens were dissected, and organs measured at the Academy of Natural Sciences of Philadelphia, using a Wild MSA dissecting microscope. Scanning electron microscope studies on the radula were done at Northeast Normal University, and photographed with a Hitachi scanning electron microscope. The internal structure of the penis, ganglia and mucous gland were observed by cutting and opening the outside wall surround- ing these organs. Institutional abbreviations are as follows: ANSP, Academy of Natural Sciences of Philadelphia; NNUC, Northeast Normal University of China. SYSTEMATICS Family: Bradybaenidae Pilsbry, 1934 Genus: Aegista Albers, 1850 Aegista laoyelingensis Zhang sp. nov. Type locality. Laoyeling (Mountains), Tiangang commune, Panshi County, Jilin Province, the People’s Republic of China, 43.20° N, 127.40° E, on the bottom of a hill close to a railway station (about 2.5 km to the North of the station) elevation between 700 and 800 m. Specimens were collected by the author on 16 July 1987. Habitat for the snails was a narrow area between a small stream and a trail. The main vegetation of the habitat was composed of the following trees: Quercus, Populus, Juglans and Acer; and bushes: Corylus, Eleutherococcus and Lespedeza. Aegista laoyelingensis preferred to crawl on the tops of bushes, with most snails collected from the leaves of Corylus and Juglans growing beside the stream. Etymology. Named for the Mountains. American Malacological Bulletin, Vol. 10(2) (1993):113-119 113 114 AMER. MALAC. BULL. 10(2) (1993) MORPHOMETRICS Holotype: shell height, 6.6 mm; shell diameter, 13.1 mm; whorls 5.6. (Fig. 1, NNUC 87001, preserved at the Institute of Zoology, Academia Sinica, Beijing, China.) Fig. 1. Three views of holotype of Aegista laoyelingensis. Paratype 1: shell height, 7.1 mm; shell diameter 14.3 mm; whorls 5.6, NNUC 87002. Paratype 2: shell height, 6.3 mm; shell diameter 12.5 mm; whorls 5.5, ANSP 391682. Paratype 3: shell height, 6.5 mm; shell diameter 12.9 mm; whorls 5.6, (in author’s collection). Paratype 4: shell height, 5.6 mm; shell diameter 11.1 mm; whorls 5.6, ANSP 391683. DESCRIPTION Shell. Shell small, depressed, with an open umbilicus, about 1/5 of shell diameter; shell light, somewhat thin, but solid; semitranslucent, uniformly light brown, not banded, whorls 5.6, surface often roughened by irregular growth lines. Spire low, with horn-colored apex, and with distinctly impressed suture. Body whorl not keeled or angulated, quite large in proportion; descending anteriorly. All whorls visible in the umbilicus and internal sutures deep. Aperture lunar, the peristome thin, colorless and toothless; narrowly reflexed at base. Mantle. Mantle cavity length 26.9 mm. Mantle colors of holotype and paratype 1, 3 and 4 are lightly colored form, there is no pigmentation from the collar to the epithelium covering the uppermost part of the kidney (Fig. 2A). Variation in mantle pigmentation was found in paratype 2. The mantle is dark, the speckled melanin pattern can extend to the epithelium covering the uppermost part of the kidney (Fig. 2B). Radula. Radula formula (from two individuals) is: 23-10-1-10-23 . (1 central tooth; 10 lateral teeth; 68 23 marginal teeth; 68 rows) Each central tooth consists of two distinctly pointed cusps (Fig. 3b: 1). There is one cusp on the outside edge of Fig. 2. Variation in mantle pigmentation. A, light form; B, dark form (speckled form). ZHANG: NEW AEGISTA FROM CHINA 115 Fig. 3. Radula of Aegista laoyelingensis. a, left marginal teeth; b, (1) a central tooth; (2) a left lateral tooth; c, (3) a right marginal tooth; (4) a right lateral tooth. each lateral tooth, but the inside edge is evenly rounded (Fig. 3b: 2). There is a gradual change in tooth form from the lateral teeth to marginal teeth. The distinction of mar- ginal teeth from the lateral teeth can be described by the following characters: (1) the marginal teeth are longer than the laterals (Fig. 3c, 4), (2) the cusp on the outside edge of each marginal tooth (Figs. 3a and 3c: 3) has a sharper point than each lateral tooth does. Digestive system. Major features are shown in figure 4: (1) sD 2mm Fig. 4. Digestive tract of Aegista laoyelingensis. BM, buccal mass; C, crop; I, intestine; OE, oesophagus; R, rectum; S, stomach; SD, duct of salivary gland; SG, salivary gland. the oesophagus (OE) is distinctly separated from the crop (C); (2) cylindrical crop widens from the oesophagus to the stomach (S); (3) intestinal loop is long, reaching the level of the distal limit of the crop; (4) internal morphology shows 6-7 unciliated ridges run from the upper end of the oesophagus into the stomach. Reproductive system. The reproductive tract is shown in figure 5. Major features are: (1) two mucous glands (M) connect to an accessory sac (AC) along with the dart sac (DS). There is a sharp, calcified dart (dotted line, Fig. 5, DS), about 1.2-1.3 mm long (Fig. 7); (2) each mucous gland is hollow, and the gland cavity (GC) is surrounded by a thick and wrinkled wall. (Fig. 6: W); (3) flagellum absent; (4) a short epiphallus [Fig. 5: E, (3.4 mm long)] cannot be distinguished from the penis externally, the differences only Fig. 5. Reproductive system of Aegista laoyelingensis. AC, accessory sac; AG, albumen gland; AT, atrium; B, bursa copulatrix sac; DS, dart sac; E, epiphallus; G, gonad (ovotestis); H, hermaphroditic duct; M, mucous gland; P, penis; PRM, penial retractor muscle; SO, spermoviduct; SR, sem- inal receptacles; ST, stalk of bursa copulatrix; V, vas deferens. 116 AMER. MALAC. BULL. 10(2) (1993) Fig. 6. Cross section of a mucous gland of Aegista laoyelingensis. GC, gland cavity; W, wall of the cavity. 1mm Fig. 7. Dart of Aegista laoyelingensis. can be seen by cutting and opening the penial wall; (5) internal morphology of the epiphallus shows 7-8 longitudi- nal ridges (Fig. 8: E); (6) penis (P) has longitudinal radiat- ing rows of pustules, lower portion of the penis has anasto- mosing longitudinal ridges, but without verge and penial pilaster (Fig. 8); (7) spermoviduct (SO) length is about 21.8 mm; duct of bursa copulatrix (ST) length is 14.4 mm. Central nervous system. The central nervous system is shown in Figure 9. The features are: (1) the length of cere- bral ganglia (CG) are about 2.0 mm; (2) the length of the cerebral commissure (CC) is about 1.4 mm; (3) the length of the cerebral-pedal connective (CPC) is about 6.1 mm; (4) the length of the cerebral-pleural connective (CPLC) is about 7.6 mm; (5) the right pleural (RPL) and the right parietal ganglia (RPA) are completely fused, the fused Fig. 8. Internal morphology of the penis. E, epiphallus; P, penis; PS, penis sheath. Fig. 9. Central nervous system of Aegista laoyelingensis. CC, cerebral commissure; CG, cerebral ganglia; CPC, cerebrol-pedal connective; CPLC, cerebral-pleural connective; LPA+LPL+V, left parietal ganglion and left pleural ganglion plus visceral ganglion completely fused; LPG, left pedal ganglion; RPA+RPL, right parietal and right pleural ganglia fused; RPG, right pedal ganglion. ganglion is about 1.6 mm long; (6) the left pleural (LPL) and the left parietal ganglia (LPA) plus the visceral gan- glion (V) are completely fused, the fused ganglion is about ZHANG: NEW AEGISTA FROM CHINA 117 species with some more recently described species of Aegista from Japan (Kuroda and Habe, 1951; Habe, 1957; Azuma, 1970; Sorita, 1980; Ogaito and Sorita, 1981; Azuma and Azuma, 1982; Sakurai and Sorita, 1982; Minato, 1983; Minato, 1988). I list the following characters to distinguish Aegista laoyelingensis from all other known 1.6 mm long. Comparisons. I compared Aegista laoyelingensis with the published species from China (Table 1) and the collections of Aegista at the ANSP, National Museum of Natural History, Washington D. C., and Institute of Zoology of Beijing (Heude Museum) (China). I also compared this Table 1. Comparisons among Aegista laoyelingensis and the published species from China. Species #of Diam. Diagnostic characters Surface Peristome (Author) Whorls (mm) Aegista oculus (Pfeiffer) 8 25 Periphery subangular, white banded Rugosely plicate Subangulately reflected, white and thin A. chinensis (Philippi) 8 25 Last whorl slowly increasing Shining White, sublabiate A. vermes (Reeve) 8 32: Last whorl scarcely deflected in Shining, with a white line Whitish front, periphery subangular A. pseudochinensis 8 27 Last whorl a little defected Shining Reddish widely reflected (Mollendorff) A. herpestes (Heude) 9 22 Last whorl obsoletely angulated Whitish band White, narrowly reflected A. furtiva (Heude) 6 19 Last whorl narrow, with obsoletely Slightly impressed suture White, expanded angulated periphery A. aubryana (Heude) qi 21 Periphery obtusely angulated, Brownish white White and thin deflected in front A. accrescens (Heude) - 16 Last whorl obtusely angulated With a whitish band White, narrowly reflected A. alphonsi (Deshayes) 7 9 Last whorl a little deflected, Irregularly punctate Thick, sinuous, reflected obtusely angulated above A. platyomphala 7 17.5 Last whorl obtusely angulated above Well impressed suture Thick and white (Mollendorff) A. subchinensis 7 17 Last whorl scarcely descending in Shining, with a narrow Thick and white (Mollendorff) front, sugangulated on the periphery white band A. initialis (Heude) 6 12 Last whorl with obtuse white banded Minutely striate White, sinuous and thick peripheral angle A. serpestes (Heude) 75 16 Last whorl with obsoletely angulated Shining White and thick periphery A. hupeana (Gredler) 6.5 13-17 Last whorl with angulated periphery Somewhat shining Broadly expanded, thick A. permellita (Heude) 75 19 Last whorl keeled Slightly impressed suture White A. tenerrima (MOllendorff) 6 18 Last whorl with subangulated periphery Slightly impressed sture White and thick A. virilis (Gredler) 7 15 Last whorl with obtusely angulated Shining Little expanded periphery A. laurentii (Gredler) 4.5 11 Last whorl large Shining Not reflexed A. radulella (Heude) 6 13 Last whorl with obtusely angulated Wart-like scales Acute periphery A. araneaetela (Heude) 5 9 Last whorl widely umbilicated Strongly costate Acute A. puberosula (Heude) 8 10 Last whorl with obtusely angulated Minute granulous scales Little reflected periphery Plectotropis cathcartae 6 2] Periphery compressly carinated Reddish corneous White (Reeve) P. mackensii 6.5 30 Periphery acutely carinated Hairy Expanded above, thin (Adams & Reeve) P. gerlachi (MOllendorff) 6 19-21 Last whorl with hairy carina Suture superficial Simple, a little reflected below P. laciniosula (Heude) 6 30 Periphery acutely carinate Hairy White, slightly expanded P. trichotropis (Pffeiffer) 6.5 17 Last whorl with acute ciliated periphery Slightly impressed spiral Simple, a little expanded above lines P. mellea (Pffeiffer) 5 22 Periphery carinated Shining White, narrowly expanded above P. tapeina (Benson) 6 15.5 Last whorl descending in front Thread-like suture White, dextrally a little expanded P. hupensis (Gredler) 6.5 20 Last whorl not descending in front, Brownish above, whitish- — White base convex, subangulated, around rayed below the umbilicus P. shanghaienwsis 7 13 Last whorl not descending, periphery Shining Subreflected, simple (Pffeiffer) carinated P. barbosella (Heude) 6.5 15 Last whorl obsoletely angulated Shaggy epidermis Little thick P. Diploblepharis 5.5 11.5 Last whorl keeled Shining White and thick (Mollendorf) P. submissa (Deshayes) 4.5 11 Periphery angulated Smooth Thick and white 118 AMER. MALAC. BULL. 10(2) (1993) species in the genus: 1. Shells small (diameter = 13.0 + 1.1 mm) in size for the genus (diameter range for the larger congeners is 15-32). 2. Whorls number 5.5 + 0.1; last whorl never keeled or angulated, shell surface rough. 3. Flagellum absent (an important character to dis- tinguish the new species from these published species in Japan). Other congeners have either a long or short flagel- lum (A. trochula Adams, 1868; Aegista kandai Azuma, 1970; A. tokyoensis Sorita, 1980; A. nunobikiensis Ogaito and Sorita, 1981; A. hakusanensis Azuma and Azuma, 1982; A. itoi Kuroda and Azuma, 1982; A. kanmuriyamen- sis Azuma and Azuma, 1982; A. tokyoensis choshiensis Sakurai and Sorita, 1982; A. tadai Minato, 1983). The shell of Aegista laoyelingensis most closely resembles in shape that of A. furtiva Heude, 1885, from Southwestern China; but the latter shell is larger (diameter, 19 mm), the surface smoother, the peristome white, thicker and more widely reflected, and the last whorl has an obso- letely angulated periphery. DISCUSSION Minato (1988) listed more than 70 species and sub- species of Aegista from Japan and nearby islands. Kuroda and Habe (1951) placed the genus Aegista in the subfamily Aegistinae, based on different characters of shells and geni- tal organs (such as, snails with more or less strongly devel- oped flagellum). Their classification follows (see Habe, 1955); Family: Bradybaenidae Subfamily: Aegistinae Genus: Aegista Albers, 1850 Subgenera: Aegista Albers, 1850. Type species: Aegista chinen- sis (Philippi, 1845). Plectotropis Martens, 1860. Type species: Plecto- tropis elegantissima (Pfeiffer, 1849). Coelorous Pilsbry, 1900 (in Coccoglypta Pilsbry 1900, see Richardson, 1983). Type species: Coccoglypta cavicollis (Pilsbry, 1900). Buliminopsis Heude, 1890 (in Pseudobuliminus Gredler, 1886, see Richardson, 1983). Type species: Pseudobuliminus buliminoides (Heude, 1882). Trishoplita Jacobi, 1898 (placed in Bradybaeninae, see Richardson, 1983). Type species: Trishoplita pallens (Jacobi, 1898). Nesiohelix Kuroda and Emura, 1943 (placed in Bradybaeninae, see Richardson, 1983). Type species: Nesiohelix caspari (MG6llendorff, 1884). Bradybaenid shells are variable in form (ranging from globose to depressed or lens-shaped), and in color pat- tern (from uniformity to many banded). For this reason Pilsbry (1895) listed some anatomical characters (but in his paper, the external anatomy and genitalia for Aegista were unknown) for distinguishing different genera of the Brady- baenidae. He concluded that Plectotropis, Mastigeulota and Euhadra possess a flagellum. The other genera lack it, probably by degeneration. It is likely that the degeneration had occurred in some Asian species grouped in Aegista too, such as the species reported by Wiegmann (1900): Plectotropis diploblepharis Mollendorf, 1902; P. submissa Deshayes, and Aegista laoyelingensis. The flagellum of these species is almost always lacking. Accordingly, it appears that the classification of the Aegistinae by Kuroda and Habe cannot be accepted prior to a complete revision in Aegista. I agree with Nordsieck (1987) that the Asian Bradybaenidae (with stimulatory organ consisting of a dart sac, usually with an accessory dart sac and one or two dart glands) represent (in contrast to the American or the European Helicoidea) a strikingly consistent group, that cannot be arranged into subfamilies. For the Bradybaeninae species (distributed in Eastern Asia), they have the stimula- tory organ with two dart glands that are usually divided. Since the different groups from Japan do not differ in the structure of the stimulatory organ (see Azuma, 1970; Sorita, 1980; Ogaito and Sorita, 1981; Azuma and Azuma, 1982; Kuroda and Azuma, 1982; Sakura and Sorita, 1982; Minato, 1983), there is no reason to divide Bradybaeninae into more subfamilies. Only the groups distributed in the Philippines can be separated based on the structure of its stimulatory organ (stimulatory organ with one simple dart gland), as the subfamily Helicostylinae. Mantle pigment patterns are variable in Aegista laoyelingensis. Cain (1971) gave a detailed description of mantle polymorphisms in Monacha cantiana (Montagu) and demonstrated that mantle color is genetically con- trolled. Because the specimens used in my research were collected from a population in a habitat with the same envi- ronmental factors (temperatures and vegetation), the varia- tion of mantle pigmentation could be caused by genetic mutations. ACKNOWLEDGMENTS This study was supported by a Jessup Award from the ANSP. Part of the field work was supported by a grant from Northeast Normal University. I am indebted to Drs. G. M. Davis, K. C. Emberton, R. Robertson, G. Rosenberg and A. Schuyler, and two anonymous reviewers for reading and ZHANG: NEW AEGISTA FROM CHINA bee) criticizing the manuscript. I am thankful to the following museums and curators for the use of their collections: Dr. G. M. Davis (Department of Malacology, ANSP); Dr. R. S. Houbrick (Division of Mollusks, National Museum of Natural History, Washington D. C.); Drs. Y. Y. Liu and D. N. Chen (Heude Museum, Institute of Zoology, Academia Sinica, Beijing). I hereby express my thanks to the follow- ing people for their helpfulness in my research: M. A. Garback, C. Hesterman, A. Bogan, H. Robertson, A. L. Hu and D. Z. Jing. LITERATURE CITED Azuma, M. 1970. Description of a new species of subgenus Coelorus (Pilsbry, 1906) from Oita Pref., Japan. Venus 29 (2): 59-64. Azuma, M. and Azuma, Y. 1982. Descriptions of three new species of the Genus Aegista Albers, 1850 (Bradybaenidae) from Japan. Venus 41 (3): 167-174. Cain, A. J. 1971. Undescribed polymorphisms in two British snails. Journal of Conchology 26: 410-416. Chen, D. N. and Gao, J. X. 1987. Economic fauna sinica of China. Science Press (4):55-98. Habe, T. 1955. Anatomical studies on Japanese land snails (3). Japanese Journal of Malacology 18:4. Habe, T. 1957. Anatomical studies on the Japanese land snails (9) Aegista (Neoaegista) trochula (A. Adams) and Aegista (Lepidopisum n. subg.) verrucosa (Reinhardt). Japanese Journal of Malacology 19 (3.4): 165-168. Heude, P. M. 1882. Notes sur les mollusques terrestres de la vallée du Fleuve Bleu. Memoires concernant |’Histoire naturelle I’Empire Chinois (2): 87-110. Kuroda, T. and Habe, T. 1951. Two new species and two new subspecies of Aegista from Chugoku District. Japanese Journal of Malacology 16 (5-8):79-83. Kuroda, T. and Azuma, M. 1982. Description of three new species and a new subspecies of land snails from Japan. Venus 41 (1): 10-19. Miller, B. W. and Naranjo-Garcia, E. 1991. Familial relationships and bio- geography of the Western American and Caribbean Helicoidea (Mollusca: Gastropoda: Pulmonate). American Malacological Bulletin 8 (2): 147-153. Minato, H. 1983. A new Aegista (Bradybaenidae) from Hiroshima-Ken Japan. Venus 41(4):247-250. Minato, H. 1988. A systematic and bibliographic list of the Japanese land snails. Shirahama 8:153-174. Nordsieck, H. 1987. Revision des systems der Helicoidea (Gastropoda: Stylommatophora). Archiv Fiir Molluskende 118 (113): 9-50. Ogaito, H. and Sorita, E. 1981. A new species of the genus Aegista Albers, 1850 (Subgenus Plectotropis Martens, 1860) from Kobe City, Japan. Venus 39(4): 205-211. Pilsbry, H. A. 1895. Guide to the study of Helices. Manual of Conchology (2) 9: 200-215. Richardson, C. L. 1983. Catalog of species. (Bradybaenidae). Tryonia 9 (i, ii): 1-253. Sakurai, K. and Sorita, E. 1982. A new subspecies of the genus Aegista (s. s.) from Choshi City, Chiba Prefecture, Honshu. Venus 40(4): 195-199, Sorita, E. 1980. A new species of the genus Aegista (s. s.) from the Koishikawa Botanical Garden, Tokyo. Venus 39(3): 142-147. Tryon, G. W. 1888. Family Helicidae. Manual of Conchology. Second series, (4): 50-62. Wiegmann, Fritz. 1900. Binnen-Mollusken aus Westchina und Centralasien. Annuaire Du Musee Zoologique. St. Petersbourg Imprimerie De L’Academie Imperiale Des Sciences 5: 1-50. Yen, T. C. 1939. Die Chinesischen land-und SiiBwasser-Gastropoden des Natur-Museums Senckenberg. Abhandlungen Der Sencken- bergischen Naturforschenden Gesellschaft 444: 1-233. Date of manuscript acceptance: 2 February 1993 Annual gonadal cycle of the land snail Scutalus tupacti (Pulmonata: Bulimulidae) Maria Gabriela Cuezzo Facultad de Ciencias Naturales e Instituto Miguel Lillo, Universidad Nacional de Tucuman-CONICET, Miguel Lillo 205, 4000 S.M. de Tucuman, Argentina Abstract: The annual gonadal cycle and its relation to the seasons were investigated in the land snail Scutalus tupacii (d'Orbigny). Collections of adult snails were carried out monthly over two consecutive years. The animals were processed for anatomical and histological studies. The ovotestis is composed of a number of tubules or elongated acini, each one containing both male and female sexual cells as well as Sertoli and follicular cells. Spermatogenesis and oogenesis occur in all the acini. The annual cycle of the ovotestis was divided into three phases: 1) Pre-breeding (September-November: Spring), characterized by the reactivation of spermiogenesis and the subsequent increase in number of mature sperm; 2) Breeding (November-February: Spring-Summer), characterized by the great number of spermatozoa in the gonad, and the highest proportions of mature oocytes; 3) Post Breeding (March -September: Autumn-Winter, with abundance of spermatogonia, oogonia and primary spermatocytes, and the interruption of spermiogenesis. Although the proliferation of male and female germinal cells is practically coincident, male gametes attained maturity at least one month before the female gametes. The post-breeding period occurs during hibernation which is coincident with the dry season. Based on the data obtained, the annual cycle of reproductive activity in this species depends principally upon the wet or rainy season. The annual gonadal cycle of many species has been tupacii were carried out from September 1988 to March previously investigated [e.g. Lymnaea stagnalis (Linné) 1991 in “Reserva Aguas Chiquitas” (El Cadillal, Tucuman, (Berrie, 1966); Helix pomatia (Lind, 1973); Semperula Argentina), which is part of the subtropical tucuman-boli- maculata (Tompleton) (Nanaware and Varute, 1975); Helix vian forest. The samples were divided into two groups. One aspersa Miller (Gomot and Griffond, 1987)]. Related group used for histological studies was fixed in Bouin's or works include research on the maturation of the reproduc- in Baker's fixative (Humanson, 1979). A second group was tive tract of various species (see Lusis, 1966; Smith, 1966; used for anatomical analysis, after being relaxed in cooled Runham and Laryea, 1968; Sokolove and McCrone, 1978; boiled water. These were then immersed in Baker's fixative. Runham and Hogg, 1979; Cuezzo, 1990) as well as the life The ovotestis and hermaphroditic duct were dissected out, cycles of a number of groups (see Van Der Laan, 1975, dehydrated in an ascending alcohol series, embedded in 1980; Hodasi, 1979; Bailey, 1980; Solem, 1984; Roth, Paraplast and sectioned at 6 um. Sections were stained with 1986). More recently, there have been a growing number of Ehrlich haematoxylin-eosin and Mallory (Azan) Heiden- studies that analyse the influence of different factors on hain (Humanson, 1979). gametogenesis under laboratory conditions (Gomot and Monthly, ovotestis and hermaphroditic ducts were Gomot, 1988; Gomot, Gomot and Griffond, 1989; Griffond fixed in Karnovsky's solution (Karnovsky, 1965) with 0.1 and Medina, 1989). M phosphate buffer (pH 7.2), post-fixed in 2% osmium The genus Scutalus Albers, 1850, is well represent- tetroxide and embedded in Epon-araldite. Semi-thin sec- ed in the western region of Tucuman Province (27°S, tions of | um were stained with toluidine blue and exam- 66°W), Argentina. Despite the large and widespread popu- ined by light microscope. Voucher specimens have been lations, very few data are available regarding the biology of deposited in the Fundacién Miguel Lillo's malacological these bulimulid land snails. In the hope of filling part of collection (Catalogue No. FML 001000). this gap, a population of S. tupacii (d'Orbigny) was sam- pled over two consecutive years to find out the characteris- tics of the gonadal cycle and the possible changes that RESULTS occur during the different seasons. As in other pulmonate gastropods, the ovotestis or hermaphroditic gonad of adult Scutalus tupacii is found MATERIALS AND METHODS embedded in the digestive gland which is located in the upper whorls of the shell. The activity and the size of the Monthly collections of adult land snails Scutalus ovotestis depend upon the age and size of the animal as American Malacological Bulletin, Vol. 10(2) (1993):121-127 12] 122 AMER. MALAC. BULL. 10(2) (1993) well as the season. The structure of the ovotestis was found to be similar to most pulmonates (Joosse and Reitz, 1969; Luchtel, 1972a, b; de Jong-Brink et al., 1977). From November to May (end of Spring to early Autumn), the acini are whitish and clearly separated from each other. At the beginning of April, a progressive reduc- tion in the size of the ovotestis occurs that peaks between June and July (Winter). The decrease in length and diame- ter of the acini are accompanied by a change in colouration from white to light brown. The pigmentation is more evi- dent in the distal part of the acini. In October the ovotestis progressively begins to recover its size and color. The annual cycle of the ovotestis can be divided into three phases of activity: 1) Post-breeding phase (March-September), Fall-Winter; 2) Pre-breeding phase (September-October), Spring; 3) Breeding phase (Novem- ber-February), Spring-Summer. 1) Post-breeding phase (Figs. 1-4): During this phase the acini are small in diameter ee * S27 with abundant interacinar space (Fig. 1). The post-breeding part of the cycle is characterized by an abundance of sper- matogonia and primary spermatocyte (Figs. 2, 3). Oogonia are present isolated or in clusters of two to four cells and are always located inside the epithelium lining the acini, remaining in contact with the basal lamina. These cells are recognizable by a poorly contrasted cytoplasm (Fig. 4). Spermatids are scarce and are present only in early stages of differentiation. Few morphologically mature spermato- zoa are present in the lumen of the acini. In the distal region of the acini the presence of mature Sertoli cells is remarkable. Their abundant cytoplasm is highly vacuolated and extends into the lumen, contributing to the “packed” appearance of the acini. Male germ cells at different stages of differentiation are clearly embedded within the same Sertoli cell. 2) Pre-breeding period (Figs. 5-10): Externally the ovotestis maintains the same appear- ance as in winter. The acini have few or lack spermatozoa Figs. 1-4: Post-breeding period. Fig. 1. General appearance of the acini during the Post-breeding period (Ac, acinar wall; scale bar = 20 pm). Fig. 2. Group of spermatogonia (Sg) in close contact with a Sertoli cell (S) with dark nuclei (scale bar = 10 ym). Fig. 3. Group of spermatocytes (Sc) embedded in the cy- toplasm of a Sertoli cell (S). (scale bar = 10 pm). Fig. 4. Three young oocytes in contact with the acinar wall (O, oocyte; Ac; acinar wall, scale bar = 10 ym). CUEZZO: GONADAL CYCLE OF SCUTALUS TUPACII 123 Figs. 5-10: Pre-breeding period. Fig. 5. General aspect of an acinus during the pre-breeding period (Ac, acinar wall; Sc, spermatocytes; Sz, spermatozoa; Sp, spermatids; S, Sertoli cells; scale bar = 40 pm). Figs. 6-9. Different stages of spermatid maturation. Fig. 6. Early spermatid stage. These cells possess a round nucleus with anterior and posterior plaque already differentiated (N, nucleus; scale bar = 10 um). Figs. 7-9. Mid spermatid stage. The nucleus becomes indented at the posterior plaque. An elongating axonome is visible. (N, nucleus; A. axoneme; G, Golgi apparatus; scale bar = 10 pm). Fig. 10. Late sper- matid stage. Elongated nucleus with condensed chromatin. The heads are embedded in the cytoplasm of a Sertoli cell (S). (scale bar = 10 um). 124 AMER. MALAC. BULL. 10(2) (1993) (Fig. 5). Toward the end of September and the beginning of October spermatids proliferate and different stages start to become apparent (Figs. 6-10). Spermatids become more abundant than spermatocytes. Also a growing number of spermatozoa can be observed in the lumen of the acinus. At the bottom of the acini vitellogenic oocytes are present but never more than one per acinus. 3) Breeding period (Figs. 11-14): During this part of the cycle all the spermatogenic stages, from spermatogonia to spermatozoa are present in the ovotestis (Figs. 11, 12). The lumen of the acini are filled with spermatozoa. Concerning the female line, previtel- logenic and vitellogenic oocytes (Figs. 13, 14) (Griffond and Bolzoni-Sungur, 1986) are distributed along the acini with a tendency towards maturation near the bottom of the acinus. In longitudinal sections of ovotestis (Figs. 11, 12), no interacinar space is observed and the layers of the adja- cent acini are in contact. The ovotestis of individuals that had just copulated were almost empty, lacking mature male gametes. The male phase in the ovotestis of Scutalus tupacii starts during the post-breeding period with an active prolif- eration of spermatogonia and spermatocytes, coincident with hibernation (end of autumn and during winter). Although mature spermatozoa are present in the ovotestis in every month of the year it is important to remark that spermiogenesis is inactive during winter. Therefore the mature gametes are less abundant in the acini while the snails are hibernating. As soon as spermiogenesis begins, toward the end of October and beginning of November, the lumen of the acini fills with spermatozoa. Mature oocytes are present throughout the year being especially abundant during the breeding period. “Ps Os, a yO: Figs. 11-14: Breeding-period. Fig. 11. Longitudinal section of the proximal part of several acini. The lumens are saturated with spermatozoa (Sz) (Ac, aci- nar wall; scale bar = 40 pm). Fig. 12. Longitudinal section of the distal portion of several acini. The more advanced stages, spermatocytes (Sc), spermatids (Sp) and mature oocytes (O) are observed in this region (scale bar - 50 pm). Fig. 13. Young premeiotic oocyte resting on the acinar wall (Ac) (O, oocyte; scale bar = 10 pm). Fig. 14. Vitellogenic oocyte. Note the presence of follicular cells (Fc) that envelope the oocyte (O, oocyte; Ac, acinar wall; scale bar = 20 um). CUEZZO: GONADAL CYCLE OF SCUTALUS TUPACII 125 Female cell proliferation begins in autumn, during May and June, but the number of mature oocytes grow between November and December in the pre-breeding and breeding periods. The seminal vesicle stores spermatozoa throughout the year. In the summer months, however, a noticeable increase in both the number of stored male gametes and the diameter of the hermaphroditic duct was noted. Although rather constant patterns were recorded in the three repro- ductive periods, individual variation in the lengths of the periods existed. SEASONALITY IN THE REPRODUCTION According to W. Koppen's climatic classification (Torres Bruchman, 1976), the region that this population inhabits has a mesothermal climate with a dry winter (tem- perature of the warmest month higher than 22°C and of the coldest month lower than 18°C). The vegetation corre- sponds to a basal subtropical forest which is characterized by the presence of big trees [Phoebe porphyria (Grisebach) Mez, Jacaranda mimosifolia D. Don, Juglans australis Grisebach and Tipuana tipu (Benth.) O. Kuntze] associated with epiphyt plants like Phlebodium aureum (L.). Abundant ferns with wide front fronds cover the ground. The pres- ence of a wet season in the region is the key to understand the pattern of activity of these land snails. The length of this period is approximately 6 months with the heaviest rains between January and March (Fig. 15). Winter is the dry season when the snails are most of the time dormant. In1989, the dry season was longer than in 1990 and conse- quently the hibernation period was prolonged. At the begin- ning of the dry period most of the animals bury themselves vertically with the aperture of the body whorl below the soil and the spire above. Other snails simply protect them- Precipitation (mm) Temperature (°C) 30 Sept Dec March Jun Sept Dec March Jun Sept Dec March ———? —————“- MONTHS —— Precipitation —— Temperature 15 —Hibernating Period Fig. 15. Dry and wet seasons from 1989 to 1991. selves under a thick layer of dry leaves and secrete a thin, transparent epiphragm. No mucus tracks were observed during these months (temperature range: 0 - 20°C). Another noticeable change is the retraction of the animal's body into the shell as a consequence of weight and water loss. Reproductive activity begins between October and November when the first rains can deeply moisten the soil. From this moment it is possible to observe copulating snails, especially nocturnally, but also during certain days, in particular the rainy ones. Mating can last several hours. The first clutch of eggs, both in 1988 and 1989, were found in November, generally buried in moist soil at the base of large trees. The eggs have an opaque-white albuminous substance within a nearly transparent membrane. The num- ber of eggs per clutch ranged from 70 to 120 with diame- ters from 2.5 to 3.0 mm. Under laboratory conditions the duration of development ranged from 15 to 18 days. DISCUSSION Three different regions in gonadal acini of Helix aspersa juveniles (Griffond and Bride, 1987) and in Biomphalaria glabrata (Say) (de Jong Brink, 1976) have been described. In adult Scutalus tupacii these regions are not clearly identifiable. A trend toward a compartmentaliza- tion in male medullar and female cortical regions occurs. Also there is a higher frequency of vitellogenic oocytes, mature follicular and Sertoli cells at the bottom of the aci- nus. Morphologically mature spermatozoa are found in the acinar lumen with their nuclei oriented toward the bottom and the flagellae toward the neck of the acinus. Despite the lack of strict regionalization, most of the mature gametes are typically found at the bottom of each acinus. Young germinal cells are distributed randomly along the acini. Coincidently, Runham and Hogg (1979) have found in Deroceras reticulatus (Miiller) that there is a gradient of oocyte size being largest at the acinar base. However, they also noted that the great enlargement of the acini could, by itself, completely explain the observed oocyte distribution: largest oocytes arise first from the base, the oldest part of the acinus and remain there without any active movement. Although proliferation of male and female germinal cells is practically coincident, male gametes reach maturity at least one month before the female ones. A number of studies, carried out in several species (Lusis, 1966; Runham and Laryea, 1968; Luchtel, 1972; Runham, 1978) support the hypothesis of a clear separation in time of male and female phases without superposition in the time of prolifer- ation of the germinal cells. During the breeding period of Scutalus tupacii both mature spermatozoa and oocytes are present in the ovotestis, coincidently with the summer 126 AMER. MALAC. months and wet season. During winter, the dry season, hibernation takes place and gonadal activity is minimal with predominating male and female juvenile germinal cells. Luchtel (1972) suggests that Arion circumscriptus L. would be defined as a protandrous hermaphrodite con- sidering the time at which the male and female gametes reach maturity. However, in terms of time of differentiation of the spermatogonia and oogonia the species could be con- sidered a simultaneous hermaphrodite and finally in terms of appearance of primary spermatocytes and oocytes Arion could be a protogynous hermaphrodite. Scutalus tupacii could be considered a protandrous hermaphrodite if we define “protandry” as the maturation of male gametes earlier than the female. However the pres- ence of mature oocytes and spermatozoa is coincident dur- ing the breeding period without a great separation in time of both phases. Among the pulmonates already studied, most are true protandric hermaphrodites (Lusis, 1966; Smith, 1966; Runham and Hunter, 1970; Parivar, 1978) except for the Achatinellidae which are registered as pro- togynous. Based upon the analysis of the climatic data (Fig. 15) and mainly upon observations in the field, the annual cycle of activity in Scutalus tupacii depends principally upon precipitation. With the beginning of the wet season in the summer the snails start their reproductive activity, copu- lating and laying eggs during these months. The duration of hibernation is intimately related to the duration of the dry season. The other factors such as the temperature and pho- toperiod, are not as important as precipitation and humidity of the habitat in the actvitity cycle of S. tupacii. ACKNOWLEDGMENTS My sincere thanks are due to Drs. Marcelo O. Cabada and Ernestina Teisaire for their permanent advice and encouragement through- out this work. I am indebted to Dr. A. Medina of Universidad de Cadiz, Spain, who reviewed the manuscript. I also thank Dr. Robert Prezant, Indiana University of Pennsylvania, USA, for revising the language. This research was partially supported by Grants PID 3055500/88 Consejo Nacional de Investigaciones Cientificas y Técnicas (CONICET, Argentina) to Marcelo O. Cabada and Project 224/90 Consejo de Investigacion de la Universidad Nacional de Tucuman (CIUNT) to Ernestina Teisaire. LITERATURE CITED Bailey, S. E. R. 1981. Circannual and circadian rhythms in the snail Helix aspera Miiller and the photoperiodic control of annual activity and reproduction. Journal of Comparative Physiology 142:89-94. Berrie, A. D. 1966. Growth and seasonal changes in the reproductive organs of Lymnae stagnalis (L.) Proceedings of the Malacological Society of London 38:199-212. Cuezzo, M. G. 1990. Maturation of the reproductive tract of Neohelix BULL. 10(2) (1993) major (Binney) (Gastropoda: Polygyridae). American Malacolog- ical Bulletin 8(1):19-24. de Jong-Brink, M., H. Boer, T. Hommes and A. Kodde. 1977. Spermato- genesis and the role of Sertoli Cells in the freshwater snail Biomphalaria glabrata. Cell and Tissue Research 181:37-58. Gomot, P.and B. Griffond. 1987. Repercussion de la duree d'eclairement journalier sur l'evolution des cellules nourricieres et de la lignee male dans l'ovotestis Helix aspera. Reproduction, Nutrition and Development 27 (1A):95-108. Gomot, P. and L. Gomot. 1989. Etude exploratoire de la spermatogenese induite par la chaleur chez l'escargot Helix aspersa en hibernation Role du cerveau. Invertebrate Reproduction and Development 16:23-32. Gomot, P., L. Gomot and B. Griffond. 1989. Evidence for a light compen- sation of the inhibition of reproduction by low temperatures in the snail Helix aspersa. Ovotestis and Albumen Gland responsiveness to different conditions of photoperiods and temperatures. Biology of Reproduction 40:1237-1245. Griffond, B. and J. Bride. 1987. Germinal and non-germinal lines in the ovotestis of Helix aspersa: a survey. Roux's Archives of Develop- mental Biology 196:113-118. Gniffond, B. and A. Medina. 1989. Timing of spermatogenesis and sper- miation in snails Helix aspersa bred under short photoperiods: A histologic and quantitative autoradiographic study. Journal of Experimental Zoology 250:87-92. Griffond, B. and J. Bride. 1981. Etude histologique et ultrastructurale de la gonade d'Helix aspersa Miiller a l'eclosion. Reproduction, Nutrition and Development 21:149-161. Hodasi, J. K. 1979. Life-history studies of Achatina achatina (Linne). Journal of Molluscan Studies 45:328-339. Joosse, J. and D. Reitz. 1969. Functional anatomical aspects of the ovotestis of Lymnaea stagnalis. Malacologia 9(1):101-109. Karnovsky, M. J. 1965. A formaldehyde-glutaraldehyde fixative of high osmolality for use in electron microscopy. Journal of Cell Biology 27:137-138. Luchtel, D. 1972a. Gonadal development and sex determination in Pulmonate Molluscs: I Arion circumscriptus. Zeitschrift fiir Zellforschung und Mikroskopische Anatomie 130:279-301. Luchtel, D. 1972b. Gonadal development and sex determination in Pulmonate Molluscs II: Arion ater rufus and Deroceras reticulatum. Zeitschrift fiir Zellforschung und Mikroskopische Anatomie 130:302-311. Lind, H. 1973. The functional significance of the spermatophore and the fate of spermatozoa in the genital tract of Helix pomatia (Gastropoda: Stylommatophora). Journal of Zoology, London 169:39-64. Lusis, O. 1961. Postembryonic changes in the reproductive system of the slug Arion ater rufus L. Proceedings of the Malacological Society of London 137:433-468. Nanaware, S. and A. Varute. 1975. Histochemical studies on the mucosub- stances in the ovotestis of a hermaphrodite land pulmonate Semperula maculata in seasonal breeding-aestivation cycle. Acta Histochemistry 54:107-124. Roth, B. 1986. Observations on the range and natural history of Mona- denia setosa (Gastropoda: Pulmonata) in the Klamath Mountains, California, and the taxonomy of some related species. Veliger 29(2):169-182. Runham, N. 1978. Reproduction and its control in Deroceras reticulatum. Malacologia 17(2):341-350. Runham, N. M. and A. Laryea. 1968. Studies on the maturation of the reproductive system of Agriolimax reticulatus (Pulmonata: Limacidae). Malacologia 7:93-107. Runham, N. M. and N. Hogg. 1979. The gonad and its development in CUEZZO: GONADAL CYCLE OF SCUTALUS TUPACII 127 Deroceras reticulatum (Pulmonata: Limacidea) Malacologia 18:391-399. Smith, B. J. 1966. Maturation of the reproductive tract of Arion ater (Pulmonata: Arionidae). Malacologia 4(2):325-351. Sokolove, P. G. and E. McCrone. 1978. Reproductive maturation in the slug Limax maximus and the effects of artificial photoperiod. Journal of Comparative Physiology 125:317-325. Solem, A. and C. C. Christensen. 1984. Camaenid land snail reproductive cycle and growth patterns in semiarid areas of north-western Australia. Australian Journal of Zoology 32(4):471-491. Solem, A. 1991. The next challenge: life styles and evolution. American Malacological Bulletin 8(2):173-175. Torres Bruchmann, E. 1976. Atlast agroclimatico de Tucuman. Uni- versidad Nacional de Tucuman. Facultad de Agronomia - Nro. 7:1-80. Van Der Laan, K. L. 1975. Aestivation in the land snail. Helminthoglypta arrosa (Binney). Veliger 17(4):36. Van Der Laan, K. L. 1980. Terrestrial pulmonate reproduction: seasonal and annual variations and environmental factors in Helminthoglypta arrosa (Binney) (Pulmonata: Helicidae). Veliger 23:48-54. Date of manuscript acceptance: 20 January 1993 Functional anatomy of Fossula fossiculifera (D'Orbigny, 1843) (Bivalvia: Mycetopodidae) Waener E. P. Avelar Department of Biology, Faculty of Philosophy, Sciences and Letters of Ribeirao Preto, University of Sao Paulo, 14040-901 Ribeirao Preto, SAo Paulo, Brazil Abstract: Fossula fossiculifera (D'Orbigny, 1843) is the type species of the genus Fossula, a member of the family Mycetopodidae. The genus is restricted to South America. The species is found in the Parana River and its tributaries from Argentina to the state of Sao Paulo, Brazil, and in the Paragacu River on the Atlantic ridge of the state of Bahia, Brazil. These medium-sized bivalve molluscs live buried in muddy substrata. The incurrent aperture is fringed along the inner fold while the excurrent aperture is smooth. The ctenidia are of type D (Atkins, 1937a). The stomach is a type IV structure (Purchon, 1958) with morphological differences on the right wall. The posterior gut is voluminous; the typhlosole runs throughout its length. The animals are hermaph- rodites and incubate their eggs in the marsupium of the inner demibranchs. In South America the family Mycetopodidae is represented by six genera: Fossula, Iheringella, Mono- condylaea, Leila, Anodontites, and Mycetopoda (Parodiz and Bonetto, 1963). According to Ortmann (1921), F: fossi- culifera (D'Orbigny, 1843) occurs in the Parana River and its tributaries from Argentina to the state of Sado Paulo, Brazil. Bonetto (1961), when discussing the geographic distribution of unionoideans in the Argentine Republic, extended the distribution of F. fossiculifera to the Para- guacu River, on the Atlantic ridge of the state of Bahia, Brazil. Hass (1930) recognized three subspecies of Fossula: F. fossiculifera fossiculifera, F. fossiculifera balzani (Ihering, 1893), and F. fossiculifera braziliensis (Ihering, 1910). According to this author, F) fossiculifera fossiculi- fera is distributed as described by Ortmann (1921), whereas F, fossiculifera balzani occurs in the Paraguai River, state of Mato Grosso do Sul, Brazil, and F: Fossiculifera brazili- ensis occurs in the Paraguacgu River, state of Bahia, Brazil. Ortmann (1921) and Bonetto (1961) recorded a single species for the genus. Lange de Morretes (1949) reported two species, F. fossiculifera and F. braziliensis, without mentioning the species reported by Hass (1930). In the present report, I shall follow the systematic treatment proposed by Bonetto (1961). Systematic studies on the functional anatomy of limnic bivalves in Brazil, include those conducted by Mansur (1972, 1973), Hebling and Penteado (1974), Hebling (1976), Veitenheimer and Mansur (1978), Mansur and Anflor (1981), Mansur and Silva (1990) and Avelar and Santos (1991). Additional studies of the functional anatomy of the Mycetopodidae have been reported by Hebling (1976), who studied the comparative anatomy of Anodontites trapesialis (Lamarck, 1819) and A. trapezeus (Spix, 1827). Mansur (1974) exam- ined the shell and the morphology of the digestive system of Monocondylaea minuana (D’ Orbigny, 1835). Mansur and Silva (1990) studied the comparative morphology and microanatomy of Bartlettia stefanensis (Moricand, 1856) and A. tenebricosus (Lea, 1834). Veitenheimer and Mansur (1978) studied the morphology, histology and ecology of Mycetopoda legumen (Ortmann, 1888). The objective of the present investigation was to study the structure, ciliary feeding currents and other func- tional adaptations of Fossula fossiculifera as a contribution to the understanding of the biology of the species, and also to provide the basis for future research related to South American unionideans. MATERIALS AND METHODS Live specimens of Fossula fossiculifera specimens were collected from the Pardo River, municipality of Ribeirao Preto (21° 7'S, 47° 45'W). Five specimens were captured at three-month intervals (15 Feb, 15 May, 17 Aug and 21 Nov). A total of 20 animals were kept alive in the laboratory in aquaria at 25°C. Some animals were anes- thetized with magnesium chloride and fixed in 10% buffered formalin for 24 hr. Others were preserved in 70% alcohol for morphological examination. To complete the anatomatical studies, detailed drawings of the animals, and of the arrangements of their internal organs, were performed using anesthetized ani- American Malacological Bulletin, Vol. 10(2) (1993):129-138 129 130 AMER. MALAC. BULL. 10(2) (1993) mals. Ciliary currents of the mantle, ctenidia, labial palps and stomach were observed under a stereomicroscope using carmine or carborundum as indicators. Some structures (palps, ctenidia, mantle and visceral mass) were fixed in aqueous Bouin's, cut into 7-10 um sections, and stained with Ehrlich's hematoxylin and eosin, for anatomical examination. HABITAT According to Bonetto (1961), Fossula fossiculifera occurs in the Paraguai River and in the middle and upper Parana River, reaching the Atlantic coast of the state of Bahia in the Paraguagu river. The animals preferentially live buried in muddy substrata under calm waters and can be captured at depths of 0.7 to 1.0 m. The bivalves burrow almost completely into the substratum, leaving only their posterior end exposed (Fig. 1). They can be captured by probing the river bottom with one's hands or feet. Several species of bivalve were captured at the collection site, among them Anodontites trapesialis, Diplodon rotundus gratus (Wagner, 1827), D. fontaineanus (D'Orbigny, 1835), D. delodontus expansus (Kiister, 1856), D. martensis (Ihering, 1892), and Castalia undosa undosa (Martens, [pl 10 mm Fig. 1. Fossula fossiculifera. External view of the left side showing extended foot and aperture (ex, excurrent aperture; is, incurrent aperture). 1827). The most abundant species were D. rotundus gratus and A. trapesialis, followed by D. fontaineanus and F. fos- siculifera, with the rare occurrence of D. delodontus expan- sus and D. martensi. It should be pointed out that bivalves of the families Hyriidae or Mycetopodidae, except Anodontites trapesialis, that occur in the Pardo River measuring less than 2 cm in length are difficult to identify because the juvenile shells are very similar and all characterized by the presence of six to 12 ribs in the umbonal region. FUNCTIONAL MORPHOLOGY SHELL. The shell of Fossula fossiculifera (Figs. 2 and 3) is subcircular in contour, equivalve and inequilateral, with a winged posterior border rising above the umbo. The perio- stracum in species smaller than 2.0 cm is yellowish to olive-brown in color in specimens ranging in size from 2.0 to 7.0 cm in length. In specimens larger than 7.0 cm, the periostracum is brownish green. These observations are similar to those reported by Ihering (1910) and Ortmann (1921). The umbo (u) is prosogyrate, with a worn perio- stracum. The lunule (lu) is small, dark in color and with oval contours. The outer opisthodetic ligament (1) starts below the umbo and extends posterodorsally to the valve isthmus. Two minor ribs originate in the umbo and extend to the ventral region, the most dorsal one ending in the region of the diaphragm that separates the apertures. In young and adult specimens, the growth lines anastomose, especially in the posterior region where they are elevated and have a lamellar aspect. The above observations are sim- Fig. 2. Fossula fossiculifera. External view of the left valve showing the lines of growth. AVELAR: ANATOMY OF FOSSULA FOSSICULIFERA 13] prms pams 10 mm Fig. 3. Fossula fossiculifera. Internal view of the right valve, showing the muscle scar (aams, anterior adductor muscle scar; arms, anterior retractor muscle scar; 1, ligament; lu, lunule; pams, protractor muscle scar; prms, posterior retractor muscle scar; t, tooth; u, umbo. ilar to those reported by Hebling (1976) for Anodontites trapezeus. According to Ortmann (1921), the valve mea- surements of Fossula fossiculifera varied from 80 to 85%, in relation to length. In the present study, height was 76 to 84% of length in a lot of 20 animals. The hinge is well developed, with one pseudocardi- nal tooth in the left valve and two pseudocardinal teeth in the right valve. In specimens larger than 7.0 cm, the spac- ing between the pseudocardinal teeth in the right and left valves becomes increasingly wider and the teeth simply become a projection of the hinge, eventually even becom- ing obsolete. The inner surface of the valve is white or bluish white, with the submarginal region yellowish cream in color. In recently sacrificed specimens with soft parts removed, the inner surface stains greenish in color. This results from pallial mucus that polymerizes when in contact with the external medium, and forms a thin film. After a few weeks, this film detaches from the shell revealing the white pearly color of the nacre. The umbonal cavity is shallow and the scars of the dorsal mantle muscles were not observed. Hebling (1976) reported the presenceof these muscles in Anodontites trape- sialis and their absence in A. trapezeus. The scars of the anterior adductor muscle (aams) and of the anterior retrac- tor muscle of the foot (arms) (Fig. 3) are continuous and present an oval contour. The anterior retractor muscle of the foot forms a deep scar in a position dorsal to the anterior adductor muscle. The scar of the protractor muscle of the foot (pms) has a rounded contour and is located in the region of the posterolateral third of the anterior adductor muscle. The scar of the posterior retractor muscle of the foot (prms) is narrow and dorsally located in relation to the scar of the posterior adductor muscle. The scar of the poste- rior adductor muscle (prms) is approximately of the same size and shape as that of the anterior adductor muscle. In specimens measuring 40 and 80 mm in length, the size of the scars reached six and 12 mm, respectively. The pallial line (pls) starts below the scar of the anterior adductor muscle and continues parallel to the mar- gin of the shell, ending at the base of the posterior adductor muscle scar. The margin of the mantle leaves a well defined scar that starts dorsally to the anterior adductor muscle and runs along the entire margin of the shell parallel to the scar of the pallial line and ending dorsally in the shell isthmus. MANTLE. When the valve and mantle are removed from the left side of the animals, the pallial cavity is exposed. The outer (secretory) and median (sensory) folds of the mantle are quite close to each other. In the region of the incurrent aperture, the inner fold (muscular) has a fringed border appearing to be formed by small papillae, as defined by Ortmann (1921), that extend far forward, decreasing in size and gradually disappearing. Posterior to the foot, approximately 1 cm dorsad from the shell margin, the mantle presents a muscular ele- vation forming an evagination that I term the posterior fold of the mantle (pfm) (Fig. 4 and elsewhere) and which ends at the point where the ctenidia join the mantle. Anterior to this fold, the mantle is not thickened, as shown in figure 5. Fig. 4. Fossula fossiculifera. Frontal section of the posterior region of the edge of mantle showing the posterior fold of mantle and the pallial muscle on the mantle lobes (if, inner fold; mf, middle fold; of, outer fold; pm, pal- lial muscle; pfm, posterior fold of mantle). 132 AMER. MALAC. BULL. 10(2) (1993) if I\ of \ mf Fig. 5. Fossula fossiculifera. Frontal section of the anterior region of the edge of mantle and the pallial muscle (if, inner fold; mf, middle fold; of, outer fold; pm, pallial muscle). The mantle margin on both the left and right sides presents two joining regions located dorsoventrally to the anal open- ing, defining the excurrent aperture. The incurrent aperture is characterized by the presence of fringes of the inner mar- gin of the mantle. The mantle is yellowish in color, and the color of the inner, middle and outer lobes tends to be salmon. Between the inner and middle lobes, in the posteri- or region, there is black pigmentation that delimits the incurrent and excurrent apertures. The pedal opening is wide and starts in the region located posteroventrally to the anterior adductor muscle and is limited in the posterior region by the ventral region of the incurrent aperture, where the first fringes of the inner margin of the mantle arise. APERTURES. The apertures of Fossula fossiculifera (Fig. 6) are of the A II type (Yonge, 1957). The excurrent aper- ture corresponds to the anal opening as defined by Ortmann (1921). It is wide with a smooth free border and becomes crenulated depending on the state of contraction. In living animals, the length of the excurrent aperture does not exceed | cm beyond the shell margin. The inhalant aperture 12.5 mm Fig. 6. Fossula fossiculifera. Frontal view of incurrent and excurrent aper- tures showing the posterior fold of mantle (ex, excurrent aperture; if, inner fold; is, incurrent aperture; mf, middle fold; of, outer fold; pa, posterior adductor; pfm, posterior fold of mantle; rpm, retractor posterior muscle). is separated from the inner lobes of the mantle by a solid connection. The incurrent aperture has a fringed margin, and forms a continuous structure with the pedal opening. The length of the extended incurrent aperture does not exceed | cm from the shell margin in living animals buried in the substratum. MUSCLE AND FOOT. The musculature of Fossula fossi- culifera (Fig. 7) is similar to that of members of Hyriidae ————1 10 mm Fig. 7. Fossula fossiculifera. Lateral view of musculature after removal of the valves, mantle, palps and visceral mass (aam, anterior adductor mus- cle; arm, anterior retractor muscle; pam, posterior adductor muscle; pm, protractor muscle; prm, posterior retractor muscle). AVELAR: ANATOMY OF FOSSULA FOSSICULIFERA 133 and Mycetopodidae described by Mansur (1972), Hebling and Penteado (1974), Hebling (1976), and Avelar and Santos (1991). Diplodon rotundus gratus, Anodontites trapesialis, A. trapezeus, Castalia undosa undosa and F. fossiculifera all lack ciliation on the foot. MANTLE CAVITY Topography. The position of the main organs of the mantle cavity is indicated in figure 8. The visceral mass is yellow and a heavily ciliated region is responsible for rejectory currents that carry particles to the posterior region of the animal. The foot is salmon colored. The ctenidia are yel- lowish in color and extend posteriorly from the umbonal region to the base of the exhalant aperture. The mantle mar- gins are free, leaving a wide pedal opening. The labial palps (Ip) which are large and suboval in shape. The ventral margins of the labial palps are plain and without “sculpture”. Fig. 8. Fossula fossiculifera. Organs and ciliary currents of mantle cavity after removal of left shell valve and mantle lobe (a, anus; aam, anterior adductor muscle; arm, anterior retractor muscle; ex, excurrent aperture; f, foot; id, inner demibranch; is, incurrent aperture; k, kidney; Ip, labial palps; od, outer demibranch; pam, posterior adductor muscle; pfm, posteri- or fold mantle; pm, protractor muscle; prm, posterior retractor muscle; tc, pseudocardinal tooth; u, umbo; v, ventricle). Labial Palps. The palps of Fossula fossiculifera (Figs. 9 and 10) are yellowish in color and symmetrical, with folded inner surfaces and smooth outer surfaces. An area with no folds exists on the inner surfaces of both the anterior and posterior region. The anterior region is connected to the proximal oral groove (pog) and the posterior region to the anterior channel (ac). The food particles that reach the labi- al palps arrive from the marginal food groove (g) of the inner demibranch, either from the anterior channel that directs particles from the outer demibranchs, or from the Fig. 9. Fossula fossiculifera. Labial palps of left side; arrows show direc- tion of ciliary currents (ac, anterior channel; ag, anterior groove of mantle; g, food groove of inner demibranch; ilp, inner labial palps; m, mantle; olp, outer labial palps; pog, proximal oral groove). Fig. 10. Fossula fossiculifera. Diagrammatic representation of the ciliary mechanisms on the folded inner surface of the labial palps to show the various ciliary tracts (a, anterior; p, posterior). anterior mantle groove (ag) that sends the particles from the mantle to the dorsum of the outer labial palps. This last path was observed by Mansur (1972) and Avelar and Santos (1991) in some Hyriidae. The mechanisms of parti- cle screening and acceptance on the part of the labial palps are similar to those observed by Hebling (1976) and Avelar and Santos (1991). Ctenidia. The ctenidia of Fossula fossiculifera are of type D (Atkins, 1937a), i.e. characterized by the presence of a marginal food groove along the inner demibranchs only. According to Atkins (1937a), this type of ctenidium is char- 134 AMER. MALAC. BULL. 10(2) (1993) acteristic of the Unionidae except Etheria. The ctenidia of Fossula fossiculifera are arranged diagonally in relation to the visceral mass. The anterior fila- ments of the outer demibranchs are smaller than those of the inner demibranch and gradually increase in width pos- teriorly. The inner demibranch (id, Fig. 8) has approximate- ly six more folds than the outer demibranch, which are clearly visible in the anterior region. The inner demibranch (Fig. 9) grows anteriorly in relation to the mantle and vis- ceral mass, forming a wide and easily visible anterior chan- nel (ac) that ends in the dorsal region of labial palps. The deimbranchs (Fig. 11) are plicate, with filaments varying in number from 17 to 29 in the ascending lamella (alod) and descending lamella (dlod) of the outer demibranch (od). In the inner demibranch (id), the number of filaments in both lamellae (dlid and dlod) is the same as in the outer demi- branch. alod UUW” Lan npn Pa ; LY OY 8) DQ Ny, 4 : anh ps Mn, ) Byes od oS) ; EPO NS : _ ; D AY @ ) AY IWZ0 Aa Z WE me p LS -) 9 1 « f x ees , / s a. | | B Fig. 4. Heterodonax bimaculatus. A, appearance of fully extended exha- lent siphons, showing movements of the tentacles around their orifices. B, detail of the distal end of the inhalent siphon. The orifice of the inhalent siphon is fringed with twelve tentacles, six of them more developed and situated at the end of the longitudinal rows. The tips of the tentacles are flattened and at its basal region there is a milk-white spot. Completely buried clams extend their siphons as long as their own shell length. The inhalent siphon extends straight through the substratum keeping its opening a little above the sediment surface. Surrounding suspended materi- al and dense particles lifted from the bottom reach the pal- lial cavity by the inhalent current. The tentacles can be directed inward to the siphonal aperture creating a feeble barrier against the entrance of large particles; when in clean water they bend outwards. 142 AMER. MALAC. BULL. 10(2) (1993) The exhalent siphon curves dorsally as it extrudes from the substrate; its opening is maintained out of the bot- tom surface and far from the inhalent. When observed in the laboratory, living specimens protracted both siphons, the exhalent twice or more distance of the inhalent. This condition is maintained even in well relaxed and preserved specimens. The behaviour of the inhalent siphon during feeding was used, along with other features, by Pohlo (1969) and Reid and Reid (1969) as a criterion for classification of feeding types in bivalves. Later, both Reid (1971) and Pohlo (1982) dedicated more attention to this problem and proposed a more useful criterion for categorizing feeding types. MANTLE The two mantle lobes are completely separated, leaving a large pedal gape that extends from the anterior adductor to the cruciform muscle. The middle fold is mod- erately developed and bears only one row of cylindrical, cup shaped tip tentacles, similar to the ones described by Narchi (1978) for Donax hanleyanus Philippi, 1847. The middle folds along with the protraction of the foot and the siphons are extended well beyond the limits of the shell exposing an external surface covered by a thin layer of periostracum. The inner surface of the mantle is ciliated and its ciliary currents are shown in figure 5. Over a wide dorsal area of the mantle surface the ciliary currents are antero- ventrally directed; they carry particles to an anterior con- Fig. 5. Heterodonax bimaculatus. Inner surface of right mantle lobe, showing ciliary cleansing currents; the arrows show the minor currents and the main rejection tract. vergent area immediately beneath the external labial palp; that area also receives particles from the anterior part of the anterior adductor muscle. Rejection currents are directed ventrally, but a major, powerful rejection tract is situated near the ventral margin, running posteriorly from a point near the anterior adductor muscle to the base of the inhalent siphon. MUSCULATURE AND FOOT The adductor muscles and pedal musculature are shown in figure 6. The anterior adductor muscle (aam) is elongated and more developed than the posterior adductor (pam), that is transversely oval. The foot is axe-shaped without a flattened sole.There is no aperture of the duct of the byssus gland as Graham (1934a) described for Gari tellinella. The foot of Heterodonax bimaculatus emerges from the elongated anterior region of the shell to probe the sand and quickly burrow into the substrate, in a similar fashion to that of Donax hanleyanus (Narchi, 1978) and of Mesodesma mactroides (Narchi, 1981). The well developed protractor muscles (pm) are attached to the shell juxtaposed to the postero-ventral sur- face of the anterior adductor muscle. From its insertion on the shell, the protractors pass backwards, twist abruptly as they enter the foot, and spread fanwise to form the outer- most muscular layer of the foot. The two posterior retractor muscles (prm) are in- serted on the shell antero-dorsally to the posterior adductor muscle. They pass towards the midline where their bundles of fibers cross each other and split again; the ones coming from the right pass deeply into the left side of the foot, and vice versa, lying internal to the layer formed by the protrac- tor muscle. The two anterior retractor muscles (arm) are insert- ed on the shell postero-dorsally to the anterior adductor muscle. Thence they pass ventrally converging so as to form a thick median bundle; without crossing each other, the right and left muscles pass deeply into the foot where they form its innermost muscular layer. In addition to these muscles, the visceral and ventral parts of the foot present transversely-directed fibres which form a large number of strands crossing from one side to the other. These isolated bundles were considered (Graham, 1934a) to form the true intrinsic pedal muscles. Despite careful searching no pedal elevator muscles were found in the specimens studied. The cruciform muscle, characteristic of all Tellinacea (Ihering, 1900), lies at the ventral side of the base of the inhalent siphon. Its morphological features and functions were discussed by Graham (1934), Yonge (1949), Moiieza and Frenkiel (1974, 1976, 1977). NARCHI AND DOMANESCHI: HETERODONAX BIMACULATUS 143 Fig. 6. Heterodonax bimaculatus. Diagrammatic representation of the musculature, as seen from the left side (aam, anterior adductor muscle; arm, anterior retractor muscle; f, foot; pm, protractor muscle; pam, posteri- or adductor muscle; prm, posterior retractor muscle). CTENIDIA The arrangement of the organs in the mantle cavity after removal of the left valve and mantle lobe is shown in figure 7. The form of the ctenidia and the general course of the ciliary currents are shown diagrammatically in figure 8. The inner demibranch hangs much lower than the outer and aA sph the ascending lamella of the latter rises considerably above the level of the gill-axis, forming a supra-axial extension as Ridewood (1903) described for the Psammobiidae. The inner demibranch is wider than the outer, particularly ante- riorly as in Asaphis dichotoma (Anton, 1839) [= A. viola- cens (Forsskal, 1775)] (Narchi, 1980). The ctenidia of Heterodonax bimaculatus (Figs. 7, 8, 9) are heterorhabdic with a groove along the free ventral margin of the inner demibranch only. They seem to be like the type C (la) described by Atkins (1937a). The posterior half of the infra-axial extension and the posterior end of the supra-axial extension of the outer demibranch are shallowly plicate while the rest of the demi- branch is smooth. There is no interruption of the cilia at the free edge of the outer demibranch, the appearance is that of a simple bending of the filaments. On the ascending lamella of the outer demibranch ciliary currents on the smooth area and on the crests and in the troughs between the plicae convey particles toward the free margin. Material is driven around the free margin and toward the gill axis on the descending lamella. Some parti- cles are transferred onto the descending lamella of the inner demibranch. Unlike the situation in Asaphis dichotoma (Narchi, 1980), there is an incipient oralward current along the ungrooved free edge of the outer demibranch. This is due to pam rors: yay $3 Maaco ye QS Ware’ nN Th 3 Wa \ee apes Noe cm od m Fig. 7. Heterodonax bimaculatus. Animal viewed from the left side after removal of the left shell valve and mantle lobe (the siphons and foot are shown somewhat contracted). Arrows show the direction of ciliary currents (a, anus; aam, anterior adductor muscle; arm, anterior retractor muscle; cm, cruciform muscle; dd, digestive diverticula; ex, exhalent siphon; f, foot; id, inner demibranch; ilp, inner labial palp; in, inhalent siphon; od, outer demibranch; olp, outer labial palp; pam, posterior adductor muscle; prm, posterior retractor muscle; sae, supra-axial extension of outer demibranch; u, umbo). 144 AMER. MALAC. BULL. 10(2) (1993) the presence at the margin of coarse terminal cilia on the posterior half of the frontal surface of each filament; they beat toward the edge and forward, creating a slight longitu- dinal current or frequently sending particles off the demi- branch onto the inner demibranch. Except for a small smooth anterior portion, the rest of the inner demibranch is plicate. The frontal currents on the descending lamella are directed ventralward on the crest and sides of the plica and dorsalward on the troughs. They are only ventralward on the ascending lamella (Fig. 8). The presence of a dorsalward current makes these gills different from the Type C (la) of Atkins (1937a) where all the cur- rents are ventralward. Atkins (1937a) included Gari ferven- sis and G. tellinella of the Psammobiidae in her Type C (1a) gill ciliation. The number of filaments per plica ranges from 12 to a maximum of 14 at the inner demibranch and 7 to 8 in the outer demibranch. On each side of the body there are two oralward currents, one in the ctenidial axis and the other in the groove along the free ventral margin of the inner demi- branch. Two additional, incipient oralward currents exist; one at the free margin of the outer demibranch and the other at the point where the ascending lamella of the inner ie Fig. 8. Heterodonax bimaculatus. Diagrammatic transverse section show- ing the form and ciliary currents on one ctenidium. Arrows indicate the directions of major currents; solid circles, oralward currents and hollow circles, incipient oralward currents. demibranch fixes on the visceral mass. The 70-116 um wide filaments (Fig. 9) are separat- ed by 100 um long laterofrontal cilia (lfc). The lateral cilia (Ic) produce a powerful respiratory and feeding current throughout the ctenidia and attain a length of 58 to 68 um. The 12 um long frontal cilia (fc) give place to increasingly longer (up to 200 um) terminal cilia (tc) in the distal ends of the filaments which are placed around the food groove. A simple row of 170 um long, stout cilia (rc), is found on the anterior half of the frontal surface of the fila- ments on both lamellae of the inner demibranch. Each row extends from the free edge of the filament toward the cteni- dial axis, attaining a varied length. In some specimens these cilia were present over the lower quarter to a half of the middle part of the inner demibranch, while in the anterior and posterior thirds of the latter they extended but a short distance from the free edge. Sparse, long (190 um) frontal cilia (rc;) were ob- served at the inner demibranch of Heterodonax bimacu- latus, similar to those recorded by Narchi (1972a) from the outer demibranch of the venerid Tivela mactroides (Born, 1778). The specialized large frontal cilia (rc, rcj) were not observed on the outer demibranch of living specimens; in preserved specimens it was impossible to recognize them on either demibranch. Along the marginal groove of the inner demibranch a fan-shaped group of long (to 170 um) guard cilia (gc) are clearly visible. Stout (70-110 um), cir- rus-like cilia (c) are present on the ends of the filaments bordering the marginal groove, mainly at the posterior part of both demibranchs. They beat obliquely forward. Gen- erally there is only one at the tip of each filament. LABIAL PALPS The labial palps (Fig. 10A) have the same basic structure and muscular activity as those of Gari tellinella (Graham, 1934a) and Asaphis dichotoma (Narchi, 1980). The inner face of the labial palps of Heterodonax bimacula- tus has almost invariably eight folds, whereas there are 12 in G. tellinella (Graham, 1934a). The particles collected on the outer face of the palps are carried to the dorsal margin and then passed to the inner face. The internal, smooth, dorsal margin is relatively wide and its ciliary currents convey particles to the median region of the plicate area of the palp. The inner demibranch of the ctenidium projects deeply between the palps. Par- ticles coming from the inner demibranch and ctenidial axis as well as those collected by the palps from the mantle and visceral epithelium pass to the inner folded surface of the palps. The sorting mechanisms on the folded surface of the palps (Fig. 10B) carry particles in three major currents, NARCHI AND DOMANESCHI: HETERODONAX BIMACULATUS 145 N LS ty, la = “4 : i ‘ | . + S : 2 AW. ‘ aus om 4 td “(a On LDL en ZE Ghar ce / AGA ff (MM. Fig. 9. Heterodonax bimaculatus. A, distal end of two living filaments of the inner demibranch, showing the arrangement of cilia. Short arrows indicate ven- tralward currents; long arrows, oralward currents. B, transverse section of some filaments, showing the arrangement of the different groups of cilia (c, cirrus- like cilia; fc, frontal cilia; gc, guard cilia; Ic, lateral cilia; lfc, latero-frontal cilia; rc, long stout cilia; rc}, long frontal cilia; tc, terminal cilia). 146 AMER. MALAC. BULL. 10(2) (1993) namely; a powerful current "a", on the aboral side of each plica; particles traveling dorsad on this current are caught by a tract of cilia and throw up to the crest of the plica. Here cilia convey fine particles from crest to crest in an oralward current "c" or throw large ones into the troughs between the folds. In the troughs, rejection currents "b" remove unwanted material onto the smooth ventral margin of the palp, where a vigorous rejection tract carries it to the distal end and on into the mantle cavity. Rejection currents "b" were more intensive on the proximal half of the palps; they were not observed all through the dorsal region of the organ. These ciliary tracts in Heterodonax bimaculatus are oral aboral Fig. 10. Heterodonax bimaculatus. A, relationship between the inner and the outer labial palps, showing ciliary currents and acceptance tracts (ilp, inner labial palp; olp, outer labial palp). B, diagrammatic representation of the ciliary currents on the folds of the inner surface of the labial palp (a, dorsalward current on the aboral side of a plica; b, rejection current at the troughs between the folds; c, oralward current). similar to those described by Narchi (1980) for Asaphis dichotoma, differing only by the absence in the first species of currents at the oral face of the folds. The labial palps have an intense muscular activity of their own. They can roll up very easily or contract rhythmi- cally without changing their shape. The inner palps often bend their tip to touch the visceral mass and capture parti- cles on it. The outer labial palps do the same, touching the mantle epithelium. ALIMENTARY CANAL The inner demibranch of the ctenidia projects for- ward between the bases of the inner and outer labial palps. Anteriorly the palps are continuous with the smooth dorsal and ventral lips of the mouth, which have conspicuous mus- cular movements. Living, dissected specimens were observed to swallow pieces removed from their own cteni- dia, muscles or mantle edge. Pieces of tissue as large as the diameter of the mouth were sucked up promptly. Arrival of material between the lips stimulates the mouth to expand and then to close pushing the material backward into the DE Fig. 11. Heterodonax bimaculatus. A, diagrammatic representation of the alimentary canal, as seen from the left side (ax, appendix; dd), digestive diverticula from the left pouch; dd2, digestive diverticula from the right caecum; dh, dorsal hood; h, heart; hg, hindgut; k, kidney region, mg, midgut; 0, oesophagus; r, rectum; ss, style-sac). B, four different midgut tracts found in different specimens. NARCHI AND DOMANESCHI: HETERODONAX BIMACULATUS 147 expanding oesophagus. After this, the two anterior retractor muscles contract and pull the foot slightly back. The pres- sure exerted by the retractors and foot upon the oesophagus forces material into the stomach. The oesophagus opens into the globular stomach in its anterior dorsal wall. The conjoined style-sac and intes- tine open into the postero-ventral region of the stomach and pass ventrally toward the foot (Fig. 11A). The midgut arises from the distal end of this wider tube, and ascends parallel to the style-sac and posterior side of the stomach toward the pericardial cavity. In Gari tellinella, Graham (1934a) illus- trated the midgut separating from the style-sac near its dis- tal end, bending back on itself and, exhibiting two or three closely packed coils, passing along the posterior part of the style-sac towards the hinder-end of the stomach. Yonge (1949) observed in the Tellinacea that there is considerable variation in the degree of coiling, i.e. in the total length of the midgut. He observed that for Donax vit- tatus and Gari tellinella the midgut is quite short. Narchi (1978, 1980) observed the same in D. hanleyanus and Asaphis dichotoma. Heterodonax bimaculatus has a short midgut which exhibits considerable intraspecific variation (Fig. 11B): straight, or with one to four loose coils before it enters the pericardial cavity via its anterior wall. The hindgut passes through the pericardial cavity, traverses the ventricle and terminates in the anal papilla on the anterior face of the posterior adductor muscle. STOMACH The anatomy of the stomach of the genus Hetero- donax (Fig. 12) is generally similar to that of other mem- bers of the Psammobiidae, but it is useful to describe and compare it with Gari togata (Deshayes, 1854) and Asaphis deflorata (Linné, 1758), adequately studied by Purchon (1960) and with A. dichotoma described by Narchi (1980). The stomach lies dorsally and is surrounded lateral- ly and ventrally by a thick layer of digestive diverticula and gonads. The oesophagus (0) enters the stomach antero-dor- sally, its orifice into the stomach being marked ventrally by a transverse, lobed ridge (rm), which represents the swollen extremities of numerous, well defined longitudinal folds. The combined style-sac (ss) and midgut (mg) leave the pos- tero-ventral wall of the stomach. The minor intestinal typhlosole (mt) terminates at the mouth of the midgut. The major typhlosole (ty) consists of two distinct parts. There is a stiff, raised semicircular elevation (e), which curves to the left over the floor of the stomach, the upper surface of which is lined by the gastric shield (gs) and shows no cil- lary activity. Under the margin of this shelf, cilia beat strongly towards the orifice of the left caecum (lc); this ridge is not accompanied by an extension of the intestinal groove and cannot be regarded as the major typhlosole as Purchon (1960) defined when describing the stomach of Gari togata. The major typhlosole properly arises in the mouth of the midgut and passes forward over the floor of the stomach, accompanied throughout its course by the intestinal groove (ig). Contrary to what was observed with G. togata, Asaphis deflorata (Purchon, 1960) and A. dichotoma (Narchi, 1980), the major typhlosole does not send a semi-circular tongue into the right caecum (rc). Near to the mouth of the right caecum the major typhlosole forms a semicircular loop, runs transversely across the anterior floor of the stomach, and enters the left caecum (Ic). Due to this feature, the stomach of the Heterodonax bimaculatus could not be classified as Type V of Purchon (1960). There are five to six ducts of the digestive diverticu- la opening into the right caecum. At the left caecum there are only three, and one of these ducts receives a flare from the major typhlosole. The major typhlosole extends to the apex of the left caecum, turns back and terminates at the orifice of the left caecum after describing a loose and incomplete spiral of about one and a half turn. The dorsal hood (dh) is large and well defined, the lower border of its orifice being protected by a very stiff, saddle-shaped sheath of the gastric shield. A broad ridge (r) enters the dorsal hood on the anterior side of its roof, while two parallel slender ridges (rj, r2) lie on the posterior side of the roof. Between the ridges r and rj there is a ciliated area on which material is driven downward, to the apex of the dorsal hood. The appendix (ax) is a large and well defined, dis- tensible, conical chamber. It opens into the stomach by a large orifice surrounded by irregularly parallel folds that penetrate deeply into it. The majority of these folds convey particles from the mouth of the dorsal hood inward to the appendix; a few of them, running on the right wall of the stomach remove particles from the appendix onto the style- sac. No sand grains were found inside the appendix. The function of this organ was discussed by Yonge (1949), Purchon (1960) and Reid (1965). The left pouch (Ip) lies antero-ventrally to the open- ing of the dorsal hood; the posterior border of its mouth is protected by a stiff, saddle-shaped flare of the gastric shield. The left pouch passes backward under the dorsal hood. About nine ducts enter the left pouch from the diges- tive diverticula on the left side of the body. A sorting area of extremely fine striations (sag), divided into two bands, lies on the left anterior floor of the stomach and penetrates deeply into the left pouch. One band runs close to the wing of the gastric shield; at the distal end of the left pouch its striation invades the aperture of ducts of the digestive diver- ticula. The other band runs at the anterior wall of the left pouch where remaining ducts open. The ciliary currents on 148 AMER. MALAC. BULL. 10(2) (1993) Fig. 12. Heterodonax bimaculatus. Interior of the stomach after 1ts opening by an incision in the dorsal wall (ax, appendix; dh, dorsal hood; e, raised semicir- cular elevation; el, long forwardly projecting tongue; gs, gastric shield; ig, intestinal groove; Ic, left caecum; Ip, left pouch; mg, midgut; mt, minor typhlosole; 0, oesophagus; r, broad ridge; r, r2, ridges on the posterior wall of the dorsal hood; rc, right caecum; rm, rim to the oesophageal orifice; s, swelling of the left anterior wall of the stomach; ss, style-sac). these two regions of sac always convey particles dorsal- ward; at the apertures of the ducts, cilia beat outward. A broad swelling (s) on the left anterior wall of the stomach isolates the apertures of the left pouch and dorsal hood; its ciliary currents remove particles from the mouth of the left pouch into the dorsal hood. Together with the broad ridge "r" it forms an efficient barrier against the free passage of material coming from the oesophagus to the cav- ity of the stomach. Reaching the ridge "r", isolated particles from the oesophagus are driven to the left, then into the dorsal hood via a ciliary tract on the dorsal side of the broad swelling (s). The parallel ridges "r;" and "ry" emerging from the dorsal hood pass downward over the roof and right wall of the stomach ending up near the opening of the midgut. A "U"-shaped ventral tongue (el), similar to those described in Gari togata, Asaphis deflorata (Purchon, 1960) and A. dichotoma (Narchi, 1980), lies to the right and ventrally to ridges "r;" and "ra". The particles on its dorsal margin are driven to the gutter between "el" and "r;" and then to the dorsal hood. Elsewhere, the cilia on the tongue pass parti- cles to the intestinal groove (ig) or over the major typhlo- NARCHI AND DOMANESCHI: HETERODONAX BIMACULATUS 149 sole and then either to the right or left caecum. The crystalline style projects into the stomach and is brown in colour. In living dissected animals with stomach, appendix and dorsal hood empty, the crystalline style could be seen through the stomach wall. A few seconds after adding carmine suspension to the pallial organs, a red mucous strand reached the stom- ach. Rotation of the style causes the material to be carried downward, toward the floor of the stomach. The crystalline style rotates in a clockwise direction when viewed from the anterior end, and the observed speed varied from 13 to 16 r.p.m. at 20°C and attained 26 r.p.m. at 21.5°C. Five min- utes later, the red particles were driven by cilia from the floor and right wall of the stomach to the dorsal hood. Next, the appendix started to become red, giving clear evidence that it received particles from the dorsal hood. Thirty min- utes after the beginning of the experiment, the foregut was stained red. An extremely viscous mass with amorphous materi- al was observed within the stomach of living dissected specimens and in fixed animals. The viscous mass that filled the stomach extended to all corners of this organ and few mineral grains (up to 0.2 mm) and entire diatoms (0.08 mm) were found. DISCUSSION AND CONCLUSIONS Heterodonax bimaculatus is an intertidal species living shallowly buried in sand or gravelly sand, in protect- ed bays. A similar habitat is occupied by H. ludwigii (Boss, 1969) and H. pacificus (Keen, 1971; Coan, 1973). Broek- huysen and Taylor (1959 cited by Boss, 1969) obtained H. ludwigii in a sand bar in the estuary at Kosi Bay. According to Boss (1969), this latter species has "a preference for estuarine conditions with a considerable amount of sus- pended matter". At Cod6 Beach, the circulating water inside Flamengo Creek often keeps a large amount of fine organic matter in suspension. As in the species of Gari, representative of the British seas (Yonge, 1949), Heterodonax bimaculatus seems to be specially adapted to the sandy or sandy gravel substrata, because it occurs only intertidally on Cod6 Beach. Its presence on the ripple sand bars formed on the sublittoral zone could be accidental. Specimens buried on the border between the sand beach and clay sand zone, or those less efficient in their ability to burrow quickly in the sand zone, could be scoured out of their natural habitat and transported into the sublittoral. Heterodonax bimaculatus presents some features typically found in clams that inhabit unstable sediments. As with Donax hanleyanus (Narchi, 1978) and Mesodesma mactroides (Narchi, 1981), it has a wedge-shaped smooth shell and a well developed and active foot. The elevator pedis muscle, characteristic of these high speed burrowing bivalves and correlated with the habit of burrowing in firm sand (Yonge, 1949; Narchi, 1978), is lacking in H. bimacu- latus. Atkins (1937a) classified Gari tellinella as a deposit feeder; Yonge (1949) observed the inhalent siphon of this species and G. fervensis opening widely for the pas- sive intake of loose surface deposit and suspended material. Conflicting statements concerning the nature of feeding in the Tellinacea received exhaustive consideration by Pohlo (1969, 1982), Reid and Reid (1969) and Reid (1971). According to the criteria for feeding types in bi- valves, Heterodonax bimaculatus is a suspension feeder, more precisely a non-selective suspension feeder in Pohlo's (1982) classification. The clam possesses large ctenidia rel- ative to the labial palps, which are small and characteristi- cally with few folds; the outer demibranchs are not reflect- ed although they are provided with a wide supra-axial extension; ctenidia type C (la) (Atkins, 1937a) (slightly modified) with a well-developed marginal groove on the inner demibranch; a waste canal (Kellogg, 1915) absent; a vertical orientation in the burrow and only finger-like non straining tentacles around the inhalent aperture. As in the donacids Egeria radiata (Purchon, 1963), Iphigenia brasiliensis (Narchi, 1972) and Donax itself (Yonge, 1949), the siphons of Heterodonax bimaculatus are relatively shorter and wider compared with most extant tel- linaceans (particularly the deposit feeders Tellinidae, Scro- biculariidae and some Semelidae). In the species with shorter and wider siphons the inhalent current caused by the beating of the lateral cilia on the ctenidia is less concen- trated and correspondly less powerful (Yonge, 1949). H. bimaculatus is unable to tear off bottom material and suck it in actively as do the specialized deposit feeders, hence the clam deals with less pseudofaeces and these are ex- pelled by the usual route, through the inhalent siphon, with- out the protection of a pair of folds that prevent the pseudo- faeces from being washed forward by the very concentrated inhalent current. In addition to the loss of ramified straining tenta- cles, the passive behaviour of the inhalent siphon which is kept well clear from the sediment surface seems to corrobo- rate the suspension feeding habit of Heterodonax bimacula- tus. Despite this, the clam may derive part of its nutritional requirements from loose surface deposits as well as from the dense deposited material. This latter is achieved pas- sively when the inhalent aperture is kept widely open, or at slightly below, the sediment surface. At these times sur- rounding benthic organisms and sand grains may fall into the pallial cavity. The presence of specialized large frontal cilia (rc, 150 AMER. MALAC. BULL. 10(2) (1993) rc,) on the frontal surface of the filaments of the inner demibranchs is evidence that Heterodonax bimaculatus deals with large particles inside its pallial cavity. These stout frontal cilia are undoubtedly a specialization for removing unwanted material, such as sand grains from the lamellae (Atkins, 1937; Narchi, 1972). The bivalves described as possessing them are all sand dwellers, silty sand dwellers or borers in rock or wood (Atkins, 1937). These cilia in H. bimaculatus seem to be efficient in their function since only few and minute (up to 0.2 mm) mineral grains were found inside the stomach of one specimen examined for this purpose. In Heterodonax bimaculatus the marginal groove on the inner demibranchs possesses fine guard cilia in com- mon with Gari tellinella and G. fervensis (Yonge, 1949) and a wide variety of other Filibranchia and Eulamelli- branchia listed by Atkins (1937), most of which inhabiting silty or muddy substrata. According to this author, guard cilia are presumably efficient in dealing with the particles of a muddy soil, but not sufficiently robust when coarse material has to be dealt with. At Codo Beach not only the layer subjacent to the superficial clean sand contains a large amount of silt but so does the circulating water, which may sporadically remove and transport mud particles from the sublittoral zone. Having both large frontal and guard cilia, Heterodonax bimaculatus is adapted to deal simultaneously with fine and coarse material present in its environment. The observations made in the laboratory with the clam ingesting pieces of their own tissues, suggests that it is able to regulate the sorting mechanisms in the pallial cavity. Submitted to a dense inhalent current the animal can reject the excess of material. If the surrounding water is scanty in food material Heterodonax bimaculatus can retain and ingest even those large particles usually rejected. The small size of the palps and their reduced number of sorting folds could represent a simplification to facilitate an acceptance of large particles. The large ctenidia of Heterodonax bimaculatus are responsible for the principal selection before material is passed to the palps; so, the palps could be maintained as small as possible with little contribution to the sorting devices inside the pallial cavity. The intense muscular activ- ity of the palps, which roll up or bend at the sides to touch the visceral or mantle epithelium, contributes respectively to discharge the excess of material or to capture useful par- ticles travelling ventralward on those epithelia. Some species of the Psammobiidae were classified as deposit feeders by Hunt (1925) and Atkins (1937a), non- selective by Pohlo (1982) or even suspension feeders by Yonge (1949). According to Pohlo, the non-selective feed- ing type as observed in the family Solecurtidae and Psammobiidae and in some species of the Donacidae, rep- resents an intermediate stage of the evolution of the telli- naceans from a selective suspension feeding ancestral to a final stage represented by the specialized deposit feeders. The alimentary canal of Heterodonax bimaculatus is similar to that of Gari tellinella (Graham, 1934a), and the stomach resembles that of the Tellinidae and Semelidae (Yonge, 1949) in relation to the presence of a straight style- sac with an associated intestinal groove and in the presence of an appendix. The internal anatomy of the stomach resembles that of G. togata and Asaphis deflorata studied by Purchon (1960) and A. dichotoma described by Narchi (1980), differing only in a few aspects. The large and distensible appendix of Heterodonax bimaculatus differs from that of Gari togata, described by Purchon (1960) as a pear-shaped chamber opening into the stomach by a slender, smooth walled neck. Purchon (1960) and Narchi (1980) referred to the appendix of Asaphis deflorata and A. dichotoma, respectively, as a capacious sac, with the opening into the stomach guarded by two large flashy pads and flanked by two sorting areas (SA 11) of irregularly parallel folds, which do not penetrate into the appendix. Ciliary currents in that sorting area beat inward, passing material into the appendix; no outward ciliary cur- rents were found by these authors. H. bimaculatus presents all around the mouth of the appendix only parallel folds penetrating deeply into the organ. Outward ciliary activity was detected on some of those parallel folds running on the right wall of the stomach. As_ stated by Purchon (1960) and confirmed by Reid (1965), the real function of the appendix is for tempo- rary storage of exceptionally large particles which have escaped rejection by the sorting mechanisms of the pallial cavity. In Heterodonax bimaculatus, very fine particles of carmine were found in the appendix. Its emptying is carried out by ciliary activity and it is possibly aided by muscular contraction of its walls, as proposed by Purchon (1960) for Asaphis deflorata. In the stomach of Heterodonax bimaculatus there are not bead-like swellings in front of the broad ridge "r" which extends up to the right wall of the stomach into the dorsal hood, as well as any series of shallow blind pockets close to the mouth of the style-sac and midgut, present in Gari (Purchon, 1960). Purchon (1960) stated that the type IV stomach is more primitive and possibly ancestral to type III and type V stomachs. Through a process of "juvenilization", i.e. by a reduction in size of individuals accompanied by a general simplification of parts of the organs, which if including a withdrawal of the tongues of the major typhlosole from association with the ducts of the digestive diverticula, stom- ach type V could revert to the ancestral type IV. According NARCHI AND DOMANESCHI: HETERODONAX BIMACULATUS es) to Purchon (1960), this theory could be advanced to account for the occurrence of a type IV stomach in the eulamellibranch families Sphaeriidae, Thyasiridae and in Chama multisquamosa (Chamidae). The type IV stomach of Heterodonax bimaculatus could represent a similar, and so far the only known, such reversion in the Psam- mobiidae. In a revision of the basic form and adaptations of the Lucinacea, Allen (1958) admitted the occurrence of a simplification in the stomach of Thyasiridae and Lucinidae, accomplished by the loss of sorting areas and reduction of the number of apertures leading from the stomach to the digestive diverticula and an increase in the size of these apertures. This simplification occurred to facilitate the acceptance of large particles, since the animals live in an environment where the food supply is so low that all avail- able particulated food has to be accepted. Thus the findings of Allen (1958) for the Lucinacea fit in well with the views expressed by Purchon (1960). The simplicity of the stomach of Heterodonax bimaculatus expressed by the presence of only a few and poor sorting areas, in addition to the simplicity of the labial palps and the voracity exhibited in the laboratory, suggest a convergence with the Lucinacea. Purchon (1960) consid- ered the paucity and the simplicity of the sorting areas of the stomach of Gari togata as being probably correlated with the unusually high viscosity of the stomach contents, which prevents individual particles from being brushed against the ciliary sorting area. According to Pohlo (1982), the type IV stomach is also the ancestral condition from which the type V stomach evolved. This point of view is supported by the presence in the extant Donacidae (Donax), which he considered as retainers of more ancestral features of a type IV stomach. From a Donax-like stage with a selective mode of suspen- sion feeding, evolution then proceeded to less selective sus- pension feeding as seen in some Donacidae and also in the Solecurtidae and Psammobiidae. The stomach also shows an advance to type V (Pohlo, 1982), a stage present in some of the Donacidae, e.g. Donax trunculus (Moiieza and Frenkiel, 1976), Egeria radiata (Purchon, 1963), Iphigenia brasiliensis (Narchi, 1972) and D. hanleyanus (Narchi, 1978). This stomach type then remains throughout the rest of the superfamily (Pohlo, 1982). From Pohlo's (1982) point of view, Heterodonax bimaculatus can be regarded as being near the ancesteral condition in the Tellinacea and so, at the base of the Psam- mobiidae lineage, since it has retained a type IV stomach and the majority of its anatomical features relate to a selec- tive suspension feeding behaviour except in the absence of straining tentacles around the inhalent aperture. LITERATURE CITED Allen, J. A. 1958. On the basic form and adaptations to habitat in the Lucinacea (Eulamellibranchia). Philosophical Transactions of the Royal Society of London. Series B 241:421-484. Abbott, R. T. 1974. American Sea Shells. Van Nostrand Reinhold Co., New York. 663 pp. 24 pls. Atkins, D. 1937. On the ciliary mechanisms and interrelationship of Lamellibranchs. Part II. Sorting devices on the gills. Quarterly Journal of Microscopical Science 79:339-373. Atkins, D. 1937a. On the ciliary mechanisms and interrelationships of Lamellibranchs. Part III. Types of lamellibranch gills and their food currents. Quarterly Journal of Microscopical Science 79:375-421. Boss, K. J. 1969. The subfamily Tellininae in South African Waters (Bivalvia, Mollusca). Bulletin Museum of Comparative Zoology 138(2):81-162. Coan, E. V. 1973. The Northwest American Psammobiidae. Veliger 16(1):40-57. Cosel, R. von. 1989. Taxonomy of Tropical West African Bivalves II. Psammobiidae. Bulletin du Muséum National d'Histoire Naturelle, Paris, 4th Series II, Section A, N° 4:693-731. Graham, A. 1934. The Cruciform Muscle of Lamellibranchs. Proceedings of the Royal Society of Edinburg 54:17-30. Graham, A. 1934a. The structure and relationships of Lamellibranchs pos- sessing a cruciform muscle. Proceedings of the Royal Society of Edinburg 54:158-187. Hunt, O. D. 1925. The food of the bottom fauna of the Plymouth fishing grounds. Journal of the Marine Biological Association of the United Kingdom. New Series 13:560-599. Ihering, H. von. 1897. A Ilha de Sao Sebastido. Revista do Museu Paulista 2:129-216, 2 pls. Ihering, H. von. 1900. The musculus cruciform in the order Tellinacea. Proceedings of the Academy of Natural Science of Philadelphia 480-481. Keen, A. M. 1969. Superfamily Tellinacea. In: Treatise on Invertebrate Paleontology, (N), Mollusca 6, R. C. Moore, ed. University of Kansas and the Geological Society of America, Lawrence, Kansas. pp. 613-643. Keen, A. M. 1971. Sea Shells of Tropical West America. Stanford Uni- versity Press, California. 1064 pp., 22 pls. Kellogg, J. L. 1915. Ciliary mechanisms of Lamellibranchs with descrip- tions of anatomy. Journal of Morphology 26(4):625-701. Lange de Morretes, F. 1949. Ensaio de catalogo dos moluscos do Brasil. Arquivos do Museu paranaense 7(1):5-216. McLean, R. A. 1951. The Pelecypoda or bivalve mollusks of Porto Rico and the Virgin Islands. Scientific Survey of Porto Rico and the Virgin Islands 17(1):1-183. Moiieza, M. and L. Frenkiel. 1974. Contribution a l'étude des structures palleales des Tellinacea. Morphologie et structures du manteau de Donax trunculus L. Proceedings of the Malacological Society of London 41(1):1-19. Moiieza, M. and L. Frenkiel. 1976. Premiéres données sur I'ultrastructure de l'organe sensoriel du muscle cruciforme de Donax trunculus L. (Mollusque Lamellibranche Tellinacea). Archives de Zoologie Experimentale et Générale 117(4):485-492. Moiieza, M. and L. Frenkiel. 1977. Le fonctionnement du muscle cruci- forme des Tellinacea. Journal of Molluscan Studies 43(2):189-191. Narchi, W. 1972. On the biology of Iphigenia brasiliensis Lamarck, 1818 - Bivalvia, Donacidae. Proceedings of the Malacological Society of London 40(2):79-91. Narchi, W. 1972a. Comparative study of the functional morphology of [52 AMER. MALAC. BULL. 10(2) (1993) Anomalocardia brasiliana (Gmelin, 1791) and Tivela mactroides (Born, 1778) (Bivalvia, Veneridae). Bulletin of Marine Science 22:643-670. Narchi, W. 1978. Functional anatomy of Donax hanleyanus Philippi, 1847 (Donacidae-Bivalvia). Boletim de Zoologia, Sao Paulo 3:121-142. Narchi, W. 1980. A comparative study of the functional morphology of Caecella chinensis Deshayes, 1855 and Asaphis dichotoma (Anton, 1839) from Ma Shi Chau, H ong Kong. Jn: Proceedings of the First International Workshop on the Malacofauna of Hong Kong and Southern China, B. S. Morton, ed. pp. 253-276. Hong Kong University Press, Hong Kong. Narchi, W. 1981. Aspects of the adaptive morphology of Mesodesma mac- troides (Bivalvia: Mesodesmatidae). Malacologia 21(1-2):95-110. Pohlo, R. 1969. Confusion concerning deposit feeding in the Tellinacea. Proceedings of the Malacological Society of London 38:361-364. Pohlo, R. 1982. Evolution of the Tellinacea (Bivalvia). Journal of Molluscan Studies 48:245-256. Purchon, R. D. 1960. The stomach in the Eulamellibranchia, stomach types IV and V. Proceedings of the Zoological Society of London 135:431-489. Purchon, R. D. 1963. A note on the biology of Egeria radiata Lam. (Bivalvia, Donacidae). Proceedings of the Malacological Society of London 35(6):251-271. Reid, R. G. B. 1965. The structure and function of the stomach in bivalves molluscs. Journal of Zoology 147:156-184. Reid, R. G. B. 1971. Criteria for categorizing feeding types in bivalves. Veliger 13(4):358-359. Reid, R. G. B. and A. Reid. 1969. Feeding processes of members of the genus Macoma (Mollusca: Bivalvia). Canadian Journal of Zoology 47(4):649-657. Ridewood, W. G. 1903. On the structure of the gills of the Lamel- libranchia. Philosophical Transactions of the Royal Society of London. Series B 195:147-284. Rios, E. C. 1985. Seashells of Brazil. Fundacgao Cidade do Rio Grande. Fundac¢4o Universidade do Rio Grande, Museu Oceanografico, Rio Grande. 328 pp., 102 pls. Stanley, S. M. 1970. Relation of the Shell Form to Life Habits of the Bivalvia (Mollusca). Memoir 125 of the Geological Society of America. Colorado. 296 pp. Yonge, C. M. 1949. On the structure and adaptation of the Tellinacea, deposit-feeding Eulamellibranchia. Philosophical Transactions of the Royal Society of London. Series B 234:29-76. Warmke, G. L. and R. T. Abbott. 1962. Caribbean Seashells. Livingston Publishing Co., Pennsylvania. 348 pp. Date of manuscript acceptance: 16 December 1992 Origin and decline of the estuarine clam Rangia cuneata in the Neches River, Texas Richard C. Harrel Department of Biology, P.O. Box 10037, Lamar University, Beaumont, Texas 77710, U.S. A. Abstract: The origin and decline of the brackish water clam, Rangia cuneata (Gray, 1831), in the Neches River were investigated by records of naviga- tion improvements, salt water encroachment, water quality, and demographic data. The origin was probably very recent (since 1900) resulting from construc- tion and improvements to the deep water navigation channel. These modifications formed a suitable salinity environment and allowed the planktonic larvae to be carried upriver from Sabine Lake. By 1951, after industrialization along the navigation channel, all Rangia beds located below river km 40.5 had been eliminated by wastewater effluents and frequent dredging. In 1971, the Rangia population consisted of 45 beds located between river km 40.5 and 57.3. The average density was 238 clams/m2 and the total area of the beds was 113, 115 m2. Since 1971, many Rangia beds have disappeared and all remaining beds examined exhibited decreased density to <1 to 2 clams/m? with an increase in average clam size. These changes were due to alterations in the river discharge pattern that decreased the frequency of salt water intrusion required for spawning and survival of larvae, and natural and cold weather mortalities. By 1985, the permitted BOD waste load in the lower river had been reduced 96%, but Rangia has not yet recolonized this section of the river. Rangia cuneata (Gray, 1831) is an important estuar- ine bivalve (family Mactridae) that is distributed from Maryland to Florida along the Atlantic coast and from northwestern Florida to Campeche, Mexico along the Gulf of Mexico (Hopkins and Andrews, 1970; Andrews, 1971). Rangia is a permanent resident in the oligohaline (0.5-5 ppt) and mesohaline (5-18 ppt) regions of estuaries and is often the dominant species in terms of biomass (Odum and Copeland, 1969; Cain, 1975). Rangia converts organic detritus and algae into organic biomass which can be uti- lized by many secondary consumers, including crustaceans, fishes, water fowl, and humans (LaSalle and de la Cruz, 1985). The shells are used for construction and manufac- ture of many industrial products (Hopkins and Andrews, 1970; Gooch, 1971) and provide a suitable substratum for attachment of epifauna (Hoese, 1973). In addition, Rangia is an excellent biomonitor of some hazardous substances, including the metals cadmium, chromium, copper, and lead, as well as the chlorinated hydrocarbons dioxins and furans (Harrel and McConnell, unpub. data). Cain (1973, 1975) and Hopkins et al. (1973) report- ed that the distribution of Rangia populations is due to fac- tors that control spawning and survival of the larvae, not adult physiology. They reported that gametogenesis was stimulated by temperatures around 15°C or higher, but spawning would not occur unless salinity changed, up from low salinity or down from high salinity. If salinity change does not occur, the gametes will undergo cytolysis when the temperature decreases to about 17°C. Once spawning occurs, early larvae will not survive unless the salinity is between 2 and 10 ppt. However, after the larvae have devel- oped past the planktonic stage and settled to the bottom, salinity is no longer a critical factor. Fairbanks (1963) reported an average life span of eight years and Hopkins et al. (1973) established a maximum life span of 15 years; thus, larval recruitment to populations can occur at long time intervals. Established beds of Rangia are concentrated where salinity seldom exceeds 18 ppt (LaSalle and de la Cruz, 1985) and have never been reported where salinity is continually higher than 15 ppt (Hopkins et al., 1973). Many investigators have reported on the distribution and abundance of Rangia in different estuaries and on vari- ous aspects of its physiology. These studies were reviewed by Hopkins et al. (1973) and LaSalle and de la Cruz (1985). However, no long term studies on a single popula- tion of Rangia have been conducted, but are necessary to understand how anthropogenic changes and natural envi- ronmental variations affect the distribution and abundance of the species. This study traces the environmental history, origin, and decline of R. cuneata in the Neches River estu- ary of Texas. The results substantiate information from other studies and yield new information on the ecology of this important species. METHODS The historical distribution of Rangia cuneata was postulated by correlating ecological requirements of the American Malacological Bulletin, Vol. 10(2) (1993):153-159 153 154 AMER. MALAC. species with the environmental history of the Sabine- Neches estuary. Modern distribution of Rangia was estab- lished by locating and sampling beds along the entire length of the Neches River estuary in 1971. Individual beds were located by sighting dead shells along the shoreline and by probing the bottom with a metal rake from a small boat. When a bed was located, clams were retrieved by hand to determine if they were alive. Only live beds were studied, but the presence of dead beds was noted. The length and width of all live beds were measured and all clams from several quadrats (30.48 cm x 30.48 cm) were removed by hand for determination of density and size dis- tribution. The number of quadrats sampled per bed varied (6 to >30) with the size of the bed and the density of the clams. The quadrats were located at various intervals along the entire length of the beds in water 30 cm to 1.5 m deep. Characteristics that delineated bed boundaries (i.e. depth, change of substratum, or physical barriers) were noted. The greatest length of all clams, or of the first 500 to 600 clams collected from the quadrats, was measured to the nearest mm with vernier calipers or with a measuring board. Density and size distribution of clams at Bed 2, located at river km 40.5, were determined in 1969, 1971, 1977, 1978, 1981, and annually thereafter. One meter square quadrats were used during these analyses and the number of quadrats sampled varied from three during 1969, the year of highest density, to 18 in 1991, the year of lowest density. The number of quadrats sampled during other years varied from five to 15. Additional data on density and size distribution were collected at Bed 10, located at river km 48, and Bed 45, located at river km 56, in 1988 and 1992. During these analyses five 1.0 m2 quadrats were sam- pled. The hand removal of clams from quadrats, as was used throughout this study, probably missed small individu- als (<25 mm), and thus underestimated density. However, all collections throughout the study were conducted by the same technique and therefore reflect real changes in the population. Air temperature data were obtained from the United States Department of Commerce (1969 - 1990). RESULTS ENVIRONMENTAL HISTORY AND DISTRIBUTION The occurrence of Rangia cuneata in the Neches River is probably very recent and can be traced with records of development of the Sabine-Neches deep water navigation channel, salt water encroachment up the river, and industrialization along the river. Kane (1959), in a study of late Pleistocene and Recent sediments, faunas, and geomorphology developed a history of the Sabine Lake- Sabine Pass area, into which the Neches River flows and BULL. 10(2) (1993) where the river's parent population of Rangia had its origin (Fig. 1). Kane (1959) found whole shells and fragments of R. cuneata shells in sediments and cores from the northern and central regions of Sabine Lake and reefs of Crassostrea virginica (Gmelin) in the southern region. However, no salt water encroachment occurred up the Neches River prior to the deepening, widening, and shortening of the natural channel (Lower Neches Valley Authority (LNVA), 1961; U. S. Army Corps of Engineers, 1981). Thus, R. cuneata could not have existed in the river until development of the deep water navigation channel from the Gulf of Mexico to Beaumont, which allowed salt water encroachment and extended the distribution of Rangia from Sabine Lake up the Neches River (Fig. 1). Cain (1975) reported that Rangia larvae could be carried upstream by passive transport or by selectively swimming in the more saline water associated with flood tides. The earliest navigation project in the Sabine-Neches system was completed in 1878 when a 3.05 m deep cut was made through the outer bank in the Gulf of Mexico (U. S. -\ Weiss Bluff ( Salt Water Barrier iLawson' 's Crossing fe Water Intake ’ 4 =< i > o < , “4 ‘ 4 yas as vy! ‘oh? / ‘ rd \ See ! aN os * SABINE LAKE Fig. 1. Sabine-Neches estuary with locations of selected Rangia cuneata beds. Darkened section of Neches River channel is location of live beds. HARREL: NECHES RIVER RANGIA 155 Army Corps of Engineers, 1981). By 1900 a 7.6 m deep channel had been cut along the west bank of Sabine Lake to Port Arthur. At this time forestry, lumber milling, and rice farming were the most important features of the area econ- omy, and navigation up the Neches River by deep-draft ves- sels was seasonal (U. S. Army Corps of Engineers, 1981). In 1901, the Spindletop oil field was discovered and the trend toward oil refining and related industries spurred enlargement of existing waterways. In 1907 a 3.05 m deep channel was completed to the mouth of the Neches River and in 1914 a 7.6 m deep channel was completed to Beau- mont (present day river km 35). At this time, salt water en- croachment first became a problem and several pump sta- tions on the lower river, which pumped water for irrigation of rice, had to be abandoned [Lower Neches Valley Authority (LNVA), 1961]. Also, in 1914 the Beaumont municipal water intake located at river km 36 had to be abandoned and a new intake was constructed at Lawson's Crossing (river km 41.6). In 1915 the water intake was extended upriver to Bunn's Bluff (river km 48.8) and in 1927 the city was forced to extend its canal and water intake upriver to Wiess Bluff (river km 66.7) (Fig. 1) (LNVA, 1961). During 1924-1929 a 9.1 m deep channel was con- structed to Beaumont and in 1926 the LNVA Lakeview Canal was constructed so that water for irrigation, indus- tries, and municipalities along the lower river could be taken at river km 65.6 during periods of salt water en- croachment (Fig. 1). In 1927 salt water intrusion reached the LNVA Canal and since that time, during periods of salt water intrusion, temporary sheet steel barriers have been erected across Pine Island Bayou at km 4.8 and across the river at km 60 to prevent saltwater contamination of this water supply (LNVA, 1961). Since 1932, the first year records are available, the Neches River barrier has been required during 26 years and the Pine Island Bayou barrier during 32 years. The river barrier dam diverted almost all of the flow of the river through the LNVA Canal for distribu- tion to municipalities and industries and for irrigation of rice fields. This allowed salt water from the Gulf of Mexico and the planktonic larvae of R. cuneata to move up river to the Pine Island Bayou and Neches River barrier sites (Fig. 1). Additional navigation improvements that increased the frequency and duration of salt water intrusion in the river include: (1) 1937-1943 - a 10.5 m deep channel, (2) 1950 - an 11 m deep channel, (3) 1962 - a 12.2 m deep channel, and (4) 1984 - a 13.2 m deep channel (U. S. Army Corps of Engineers, 1975, 1981, 1982). After the First World War, and especially after World War II, the petrochemical industry grew at an extremely fast rate and the lower river became grossly pol- luted. Most industries dumped wastes into the river without treatment turning the water black, and oil slicks degraded the shoreline (Patrick et al., 1992). Whenever the salt water barrier dams were in place, salt water from the Gulf and waste effluents from the many industries along the lower river moved up to the barriers, and tidal action flushed the salt-wastewater back and forth, causing it to become more concentrated the longer the barriers were in place (Harrel, 1975; Harrel et al., 1976). During many years the barriers were in place for as long as six months and the resulting pollution killed all but the most tolerant fish and inverte- brates trapped below the barriers. Consequently, by 1951, the Rangia population in the Neches River was restricted to areas between river km 40.5 and river km 60, the saltwater barrier site (S. H. Hopkins, pers. comm., 1968). Although the upriver population had been subjected to low dissolved oxygen concentrations (<2.0 mg/1) for long time periods, it seemed to thrive (Harrel, 1975; Harrel et al., 1976; Harrel and Hall, 1991). Living Rangia beds had also existed in the lower river, as indicated by the presence of Rangia shells in the channels of the old meanders that were cut through when the navigation channel was straightened and in dredge spoils from early navigation projects. These beds were probably eliminated by dredging and toxic pollutants. Intensive surveys of this section of the river conducted by the Texas Department of Water Resources during 1969- 1973 (Warshaw, 1974) and 1980 (Davis, 1984) reported 19 and 22, respectively, priority pollutants in water, sediments, or tissues at significant concentrations. These included heavy metals, phenols and cresols, polycylic aromatic hydrocarbons, phthalate esters, and general inorganics. In 1968, when pollution abatement first began in the Neches River, the permitted biochemical oxygen demand (BOD) waste load for this section of the river was 123, 125 kg/d (220,000 Ib/d), discharged primarily from oil refiner- ies, petrochemical plants, and a large paper mill (Davis, 1984; Twidwell, 1986). The Texas Water Commission ranked the tidal Neches River as the second most polluted waterway in the state (Warshaw, 1974). Studies conducted on the tidal Neches from 1967 through 1972 examined the effects of wastewater and salt water intrusion on water quality and community structure of the macrobenthos (Harrel, 1975; Harrel et al., 1976). The results of these studies confirmed the above distribution of Rangia, and 45 living Rangia beds were located between river km 40.5 and km 57.6 (Fig. 1). An additional 19 Rangia beds were locat- ed in the lower 4.8 km of Pine Island Bayou, but are not included in this analysis. In the early 1970s Federal and State regulatory agencies required all industries along the Neches estuary to upgrade their wastewater treatment systems to at least sec- ondary treatment (U. S. Environmental Protection Agency, 156 AMER. MALAC. BULL. 10(2) (1993) 1980). Since then, all wastewater treatment plants have been improved and two large regional treatment plants have been constructed. The permitted BOD waste load has been reduced 96%, to 8,717 kg/d (19,217 lb/d). Many non-BOD contributing wastes, such as heavy metals and inorganic suspended solids, were also greatly reduced. In 1986 the tidal Neches was ranked 48 of 311 classified segments (Twidwell, 1986). Harrel and Hall (1991) reported on water quality and macrobenthic community structure at the same seven collection sites before (1971-1972) and after (1984- 1985) pollution abatement in the Neches River estuary. Evidence of improved water quality after pollution abate- ment included: (1) an increase in the annual number of taxa collected from 50 to 104, (2) minimum densities in 1984- 1985 exceeding maximum densities for 1971-1972 at most stations, and (3) patterns of species dominance, Sorenson's similarity index, and Shannon's diversity index. Patrick et al. (1992) used the Neches River estuary for their evalua- tion of environmental laws on surface water quality. During their 1984-1985 study Harrel and Hall (1991) found young Rangia, less than one year old, at five collecting stations located between river km 4.8 and river km 40.5. However, no mature Rangia have become estab- lished along the navigation channel in the lower river. This could be due to maintenance dredging of the navigation channel and frequent oil and chemical spills in this section of the river. In 1990 a bed of mature Rangia was located at river km 39.2, above the navigation channel, and the size of the clams indicated that this bed had been established for at least seven or eight years. During 1971 many shells but no living clams were present at this location. DENSITY AND SIZE During 1971, the density of the 45 live Rangia beds located in the Neches River varied from 16 to 665 clams/m?2 and mean density was 238/m2. There was no correlation between density and distance upriver (r = 0.213; p = 0:078) (Fig. 2). However, mean shell length of clams significantly increased with distance upriver (r = 0.827; p = 0.0001) and lower variance occurred above Pine Island Bayou (river km 48) (Fig. 3). Ladd (1951), Gunter (1961), and Pfitzenmeyer and Drobeck (1964) reported that specimens of R. cuneata in fresher waters were larger than those from more saline waters, but gave no explanation. The smallest clam collect- ed from the quadrats was 23 mm long and the largest clam was 74 mm long. The age of these specimens would be a little over one year and about 14 years, using the von Bertalanffy growth curve of Wolf and Petteway (1968). The correlation coefficient between bed density and mean clam length indicated a weak relationship (r = -0.331; p = 0.012). The size of individual beds ranged from 42 square meters for Bed 19 at river km 53.4 to 27,360 square meters 8 4 * Y = -0.006 X + 50.683 r = 0.213; p = 0.078 8 fo.) 8 — BED DENSITY (NO/M?) 8 8 40° 45 50 55 60 RIVER KILOMETER 1971 1988 1992 ad A re) Fig. 2. Rangia cuneata density versus river kilometer in 1971 with best fit line, and at Beds 2, 10, and 45 in 1988 and 1992. for Bed 6 at river km 44. Larger beds and higher variance were found below Pine Island Bayou, and the correlation coefficient between bed size and river km was -0.347 (p = 0.009) (Fig. 4). The boundaries of individual beds were delineated by depths greater than 1.5 m, a change in sub- stratum from mixed substratum (sand, silty clay, and detri- tus) to pure substratum (sand or clay), or sunken logs, barges, or boat docks. The total area of all 45 beds was 113,115 m2 (11.3 hectares). Density and size data are reported for Rangia at Bed 2, located at river km 40.5, for 1951 from Hopkins (1970) and Hopkins and Andrews (1970) and by this investigator for 1969 to 1992 (Table 1). Hopkins (1970) and Hopkins and Andrews (1970), using data collected in 1951 and data collected by this investigator in 1969, reported that the length-frequency and density were remarkably similar and 70 a a Qa oOo Y = 0.806 X + 7.8671 r = 0.827; p = 0.0001 MEAN LENGTH (mm) 40 45 50 55 60 RIVER KILOMETER 1971 1988 1992 —+— A e) Fig. 3. Rangia cuneata mean shell length versus river kilometer in 1971 with best fit line, and at Beds 2, 10, and 45 in 1988 and 1992. HARREL: NECHES RIVER RANGIA Ly Table 1. Mean shell length, length extremes, and density of Rangia cuneata at Bed 2 in the Neches River from 1951 to 1992. Year Mean Length Length Range Density (mm) (mm) (No./m2) 1951° 42 37 - 56 >250 1969 45 35 - 55 496 1971 44 23 - 61 391 1977 55 46 - 67 48 1978 58 34 - 67 43 1981 56 40 - 73 92 1982 53 40 - 65 141 1983 57 36 - 69 142 1984 56 40 - 67 15 1985 59 40 - 70 23 1986 58 40 - 74 22 1987 60 32 - 72 26 1988 58 45 - 69 42 1989 61 49-71 34 1990 59 43 - 70 2 1991 59 54 - 64 250 clams/m? during both years; although the calculated density using the 1969 data was actually 496/m2. Between 1969 and 1992 the mean length of clams at Bed 2 increased (r = 0.745; p = 0.0006) and density de- creased (r = -0.836; p = 0.0001) (Table 1; Fig. 5). The cor- relation coefficient between mean shell length and bed den- sity was -0.87 (p = 0.001). Data collected in 1988 and 1992 at Bed 10, located at river km 48, and Bed 45, at river km 56, indicated that these trends were occurring at all Rangia beds along the river (Fig. 2 and 3). 30 25 Y = -0.0005 X + 50.1613 r = -0.347; p = 0.009 N o BED AREA (NP) (Thousands) ° 40 45 50 55 60 RIVER KILOMETER Fig. 4. Rangia cuneata bed size versus river kilometer in 1971 with best fit line. 180 160 120 F 1900 F FREQUENCY Ra = a | 5 | | 24 27 30 33 36 39 42 45 48 51 54 57 6O 63 66 69 LENGTH (3 mm Intervals) MM 9071 L] 1983 EEE 1084 N= S17 N = 502 N = 67 FREQUENCY 39 42 45 48 51 54 57 60 63 66 69 72 LENGTH (3 mm Intervals) L] 1990 BEE 1992 N = 168 N = 23 N = 15 Fig. 5. Length-frequency histograms of Rangia cuneata at Bed 2. The decrease in density was not constant and con- sisted of three years with large decreases in density (1971, 1983, and 1989) followed by several years with slight increases in density (Table 1). The sharp decreases in densi- ty between 1983-1984 and 1989-1990 resulted from high Rangia mortality following extreme cold fronts that occurred during December, 1983 and December, 1989. During both of these cold fronts ice formed along the shoreline and about 2 m of previously submerged shoreline was exposed due to north winds, resulting in low water lev- els. Two weeks after these cold fronts, the only living Rangia found were in water one meter or more deep, near the outer boundary of the bed where a two meter drop-off occurred. All size classes or ages of clams were affected by the cold, but the most abundant size classes had the highest mortality. The size range remained similar following both die-offs (Fig. 5). During the 1983 and 1989 cold fronts, 113 and 94 continuous hours of below freezing temperatures 158 AMER. MALAC. BULL. 10(2) (1993) occurred, and new low temperature records were set (U. S. Department of Commerce, 1983 and 1989). The Texas Parks and Wildlife Department reported cold water mortali- ties of fish, shrimp, and crabs in many bays, coastal lakes, and estuaries along the entire Texas coast during both cold fronts. Similar cold fronts occurred during January, 1973 and January, 1976 (U. S. Department of Commerce, 1973 and 1976) and could have caused the large decrease in den- sity between 1971 and 1977 (Table 1). However, the de- structiveness of colds waves to organisms is explicable on the basis of acclimatization and depends more upon the rapidity of the temperature drop and the health of the organisms than the low temperature attained (Gunter and Hildebrand, 1951). Hopkins et al. (1973) reported that Rangia had a lower condition index (ratio of meat wet weight/internal shell volume) and meat glycogen content during November and December. This condition and time correlates with the period of major Fall spawning or cytoly- sis of unspent gametes, and could be the most stressful period of the year to Rangia. Gallagher and Wells (1969) reported a winter mortality of R. cuneata in the Elk River, Maryland, which is near the northernmost geographic range of the species. These are the first documented winter mor- talities of Rangia in a Gulf of Mexico estuary. DISCUSSION The increase in size of clams with increasing dis- tance upriver as occurred in 1971 (Fig. 3) and the increase in clam size at Bed 2 from 1969 through 1992 (Table 1) may be explained in terms of energy allocations for growth, reproduction, and survival during stressful times as dis- cussed by Begon et al. (1990). Clams located farther down- river were more frequently subjected to salinity fluctuations that induce spawning, resulting in less energy being avail- able for somatic growth that is beneficial for survival dur- ing stressful times. Thus, clams farther downriver would be smaller and have a shorter longevity. Likewise, clams locat- ed farther upriver could go several years without spawning, and consequently more energy can be used for growth and production of somatic biomass that could be beneficial dur- ing stressful periods and increase longevity. However, this would decrease recruitment into the population and result in a gradual increase in size and decrease in density as aging and mortalities occurred. River discharge pattern, which effects the frequency and duration of salt water intrusion in the river, has had a profound impact on R. cuneata in the Neches River. Between 1951 and 1971 salt water intrusion occurred every year but one and the Pine Island Bayou salt water barrier was erected every year except 1968. The Neches River bar- rier was erected during 17 years of this 20 year time period. Since 1972, the Pine Island Bayou barrier was required dur- ing eight years, but only once since 1981. The Neches River barrier has been required six times since 1972, most recently in 1982. White and Perret (1974) reported that reg- ulated river discharge from Toledo Bend Reservoir on the Sabine River had caused very low summer salinities in Sabine Lake. Freshwater flow into the Neches River estuary is controlled by releases from Sam Rayburn and Steinhagen Reservoirs, located upriver, which allow increased river dis- charge during summer months. Also, demands for diversion of water from the Neches River by the LNVA for irrigation of rice fields have decreased. Since 1972 the area of rice fields irrigated decreased from more than 24,281 hectares (60,000 acres) to less than 12,140 hectares (30,000 acres). Thus, time periods of salt water intrusion necessary to induce spawning of Rangia and suitable salinity for sur- vival of larvae in the river have been very infrequent and of short duration during the last 20 years. The lack of recruit- ment and natural and cold weather mortalities have resulted in the loss of many Rangia beds that existed in 1971 and low densities (<1 to 2 clams/m2) and an increased average size or age of clams at all surviving beds. If the river discharge and cold weather trends of the past 20 years continue, Rangia cuneata could disappear from the Neches River. However, as pointed out by Hop- kins and Andrews (1971), R. cuneata is a very resilient species and suitable conditions for a few years could restore the population to its former level. Also, if the third round of the U. S. Environmental Protections Agency's National Pollution Discharge Elimination System (NPDES) permit system, which began in 1987, is successful and eliminates toxic pollutants from surface waters R. cuneata could again become established in the lower Neches River. ACKNOWLEDGMENTS I thank Tommy Hebert for records on installation of the salt water barriers, Margaret O. Welch and Rolanda Alford for collection of the 1971 field data, and Thomas S. Bianchi, John T. Sullivan, and two anonymous reviewers for critical comments on early drafts of this paper. LITERATURE CITED Andrews, J. 1971. Sea Shells of the Texas Coast. University of Texas Press, Austin. 298 pp. Begon, M., J. L. Harper and C. R. Townsend. 1990. Ecology: Individuals, Populations, and Communities. 2nd ed. Blackwell Scientific Publications, Boston. 945 pp. Cain. T. D. 1973. The combined effects of temperature and salinity on embryo and larvae of the clam Rangia cuneata. Marine Biology 21:1-6. Cain. T. D. 1975. Reproduction and recruitment of the brackish water clam Rangia cuneata in the James River, Virginia. United States HARREL: NECHES RIVER RANGIA 159 National Marine Fishery Bulletin 73:412-430. Davis, J. R. 1984. Intensive survey of the Neches and Sabine Rivers Segments 0601 and 0501. Texas Department of Water Resources Report No. IS-60. TDWR, Austin. 51 pp. Fairbanks, L. D. 1963. Biodemographic studies of the clam Rangia cuneata Gray. Tulane Studies in Zoology 10:3-47. Gallagher, J. L. and H. W. Wells. 1969. Northern range extension and winter mortality of Rangia cuneata. Nautilus 83:22-25. Gooch, D. M. 1971. A study of Rangia cuneata Gray, in Vermillion Bay, Louisiana. Master's Thesis, University Southwestern Louisiana, Lafayette. 48 pp. Gunter, G. 1961. Some relations of estuarine organisms to salinity. Limnology and Oceanography 6:182-190. Gunter, G. and H. H. Hildebrand. 1951. Destruction of fishes and other organisms on the south Texas coast by the cold wave of January 28- February 3, 1951. Ecology 32:731-736. Harrel, R. C. 1975. Water quality and salt water intrusion in the lower Neches River. Texas Journal of Science 26:107-117. Harrel, R. C., J. Ashcraft, R. Howard and L. Patterson. 1976. Stress and community structure of macrobenthos in a Gulf Coast riverine estu- ary. Contributions in Marine Science 20:69-81. Harrel, R. C. and M. A. Hall. 1991. Macrobenthic community structure before and after pollution abatement in the Neches River estuary, Texas. Hydrobiologia 211:241-252. Hoese, H. D. 1973. Abundance of the low salinity clam, Rangia cuneata in southwestern Louisiana. Proceedings of the National Shellfishery Association 63:99-106. Hopkins, S. H. 1970. Studies on the brackish water clams of the genus Rangia in Texas. Proceedings of the National Shellfishery Assoc- iation 60:5-6. Hopkins, S. H. and J. D. Andrews. 1970. Rangia cuneata on the east coast: Thousand mile range extension or resurgence? Science 167:868- 869. Hopkins, S. H., J. W. Anderson and K. Horvath. 1973. The brackish water clam Rangia cuneata as indicator of ecological effects of salinity changes in coastal waters. United States Army of Corps Engineers Waterways Experimental Station, Vicksburg, Mississippi. Report No. DACW39-7 1-C-0007. 250 pp. Kane, H. E. 1959. Late Quaternary geology of Sabine Lake and vicinity, Texas and Louisiana. Transactions Gulf Coast Association Geological Society 9:1-11. Ladd, H. S. 1951. Brackish-water and marine assemblages of the Texas coast with special reference to mollusks. Publications Institute of Marine Science University of Texas 2:12-164. LaSalle, M. W. and A. A. de la Cruz. 1985. Species profiles: Life histories and environmental requirements of coastal fishes and invertebrates (Gulf of Mexico) - common rangia. United States Fish and Wildlife Service Biology Report 82(11.31). United States Army Corps of Engineers, TR EL-82-4. 16 pp. Lower Neches Valley Authority. 1961. Statement on salt water barrier on Neches River below mouth of Pine Island Bayou. Submitted at pub- lic hearing by United States Army Corps of Engineers, Beaumont, Texas - November 14. 59 pp. Odum, H. T. and B. J. Copeland. 1969. A functional classification of the coastal ecological system. /n: Coastal Ecological Systems of the United States, H. T. Odum, B. J. Copeland, and E. A. McMahan, eds. pp. 9-86. Federal Water Pollution Control Administration, Washington, D. C. Patrick, R., F. Douglas, D. M. Palavage and P. M. Stewart. 1992. Surface Water Quality: Have the Laws Been Successful? Princeton University Press, Princeton, New Jersey. 198 pp. Pfitzenmeyer, H. T. and K. G. Drobeck. 1964. The occurrence of the brackish water clam, Rangia cuneata, in the Potomac River, Maryland. Chesapeake Science 5:209-215. Twidwell, S. R. 1986. Intensive survey of the Neches River tidal segment 0601. Texas Water Commission Report No. IS 86-05. TWC, Austin. 92 pp. United States Army Corps of Engineers. 1975. Final environmental state- ment - Maintenance dredging Sabine-Neches waterway, Texas. United States Army Engineer District, Galveston. 82 pp + appen- dices. United States Army Corps of Engineers. 1981. Neches River and tribu- taries, saltwater barrier at Beaumont, Texas: Phase 1 general design memorandum. United States Army Engineer District, Galveston. 139 pp. + appendices. United States Army Corps of Engineers. 1982. Sabine-Neches waterway, Texas feasibility report and environmental impact statement, chan- nel improvements for navigation. United States Army Engineer District, Galveston, 360 pp. + appendices. United States Department of Commerce. 1969 - 1990. Local climatologi- cal data, annual summary and comparative data, Port Arthur, Texas. National Oceanic and Atmospheric Administration Environmental Data Service. Asheville, North Carolina. United States Environmental Protection Agency. 1980. National accom- plishments in pollution control: 1970-1980, some case histories. United States Environmental Protection Agency, Washington, D. C. 153 pp. Warshaw, S. 1974. Water quality segment report for segment 0601 Neches River tidal. Texas Water Quality Board Report No. WQS-2. TWQB, Austin. 44 pp. Wolfe, D. A. and E. N. Petteway. 1968. Growth of Rangia cuneta Gray. Chesapeake Science 9:99-102. White, C. J. and W. S. Perrett. 1974. Short term effects of the Toledo Bend Project on Sabine Lake, Louisiana. In: Proceedings of 27th Annual Conference of Southeastern Association of the Game and Fish Commission, pp. 710-721. Hot Springs, Arkansas. Date of manuscript acceptance: 4 February 1993 Research Note Two common European viviparid species hybridize Andrzej Falniowski’, Andrzej Kozik’, Magdalena Szarowska’ ‘Zoological Museum, Institute of Zoology, Jagiellonian University, ul., Ingardena 6, 30-060 Krakéw, Poland "Department of Biochemistry, Institute of Molecular Biology, Al., Mickiewicza 3, 31-120 Krakéw, Poland Abstract: During a morphological/electrophoretic study on the European Viviparidae, a hybrid but fertile female specimen [Viviparus contectus (Millet, 1813) x V. viviparus (Linnaeus, 1758)] was found in the Niepolomice Forest, South Poland. The specimen showed intermediate characters in its shell, anatomy, and embryonic shell. Polyacrylamide gel electrophoresis of that specimen confirmed its hybrid origin, which was marked in eight enzyme systems: specific homozygotes in both species along with a typical heterozygote in the hybrid. Interspecific differences in allozyme pattern, with different alleles in at least one locus in most enzyme systems studied, seem to indicate that V. contectus and V. viviparus are very old species, and their isolating mechanisms pre- venting hybridization could have become not efficient enough after such a long time since the speciation event. This is all the more probable, now that they occur sympatrically very rarely. Viviparus contectus (Millet, 1813) and V. viviparus (Linnaeus, 1758) are widely distributed and common in Central Europe. However, their diagnostic characters are labile (Falniowski, 1989, 1990). In particular, this concerns shell characters, in which the presence of the umbilicus (which can be partly covered but is always visible) in V. contectus contrary to its absence in V. viviparus seems the most constant difference. Falniowski (1990: Plate 111, phot. 16-19) presented photographs of shell intermediates of the two species. On the other hand, those intermediates were undoubtedly determined anatomically as V. contectus. The only anatomical difference between the two species has been found in the female reproductive organs (Falniowski, 1989, figs. 45-52, also Falniowski, 1990, fig. 362A, B). In V. contectus the duct of the receptaculum seminis is usually narrow while leaving the receptaculum, whereas in V. vivi- parus it is strongly (rather strongly) widened; at the duct section parallel to the receptaculum, in V. contectus there commonly (not always) is a bulbous widening, which is absent in V. viviparus. In the literature, one can find numerous opinions on ecological vicariance of the two species (e.g. Zhadin, 1928). In fact, the distribution pattern of the two species in Poland is slightly different from the one described by Zhadin (1928), but the species rarely occur sympatrically (Falniowski, 1989). In 1976, a giant female of Viviparus (Fig. 1) was found in Drwinka River in Niepolomice Forest (about 30 km E from Krakow, Southern Poland). The shell measured 55 x 42 mm (normally, in V. contectus and V. viviparus it does not exceed 45 x 36 mm and 40 x 28 mm, respective- ly). The specimen was identified (based on shell characters alone; anatomical differences had not been known until then) as V. contectus f, hungaricus Hazay, 1881 (Falniow- ski, 1989). The shell, although more similar to V. contectus than to V. viviparus, was certainly not typical of V. contec- tus, and resembled the shell of the Danube V. acerosus (Bourguignat, 1862). The weakness of diagnostic characters within the family Viviparidae is as already mentioned above, and rela- tionships and taxonomy within the group are poorly under- stood. The species are described solely upon a basis of shell characters and no further study has confirmed their distinct- ness, although viviparids are often used as laboratory ani- mals in various studies. Such are the reasons for starting morphological and electrophoretic studies (Table 1) on the European representatives of the family. Until now, we have already found constant differences in numerous enzyme systems among Viviparus contectus, V. viviparus and V. acerosus (Table 2), coupled with an extremely low propor- tion of polymorphic loci in the three species. The study is in progress. Among viviparids collected from a small pond at Ispina, the Niepolomice Forest, in August 1991, we found one big specimen, the shell of which resembled the one of Viviparus acerosus, as well as the one collected in 1976 (Fig. 1) in the Drwinka River (the same region, one locality a couple of kilometres from the other). The pond is situated between the Vistula River and its right-bank dyke that is the north border of the Niepolomice Forest, Wislisko-Kobyle American Malacological Bulletin, Vol. 10(2) (1993):161-164 161 162 AMER. MALAC. BULL. 10(2) (1993) Table 1. Application of polyacrylamide gel electrophoresis technique. Electrophoresis: in slabs (180 x 130 x 0.7 mm) of 7.5% polyacrylamide gel in a continuous high-pH buffer system of Davis (1964). Reservoir buffer: Tris-glycine (pH 8.3); 3 g Tris and 1.4. glycine per 1 / water. Stacking gel buffer: Tris-HC1 (pH 6.8); 6 g Tris titrated to pH 6.8 with 1M HC! in 100 ml final volume. Resolving buffer: Tris-HCI (pH 8.8); 36.3 g Tris and 48 ml 1M HCl mixed and diluted to 100 ml final volume. Acrylamide-bisacrylamide solution: 30 g acrylamide and 0.8 g bisacry- lamide diluted to 100 ml final volume and filtered. Stacking gel: 2.5 ml acrylamide-bisacrylamide solution, 5 ml stacking gel buffer, 2.5 ml 0.004% riboflavin, 10 ml water and 0.015 TEMED mixed and photopolymerized. Resolving gel: (7.5%): 15 ml acrylamide-bisacrylamide solution, 7.5 ml resolving gel buffer, 39.5 ml water and 0.03 ml TEMED polymer- ized with 3 ml 1.5% ammonium persulfate as the catalyst. Rns: Fourteen samples, 20 ul each, applied for a slab. Typically, a current 20-30 mA maintained for about 4 hrs until a marker dye (bromo- phenol blue) passed all the slab. Reserve. The pond is about 200 m long and 50 m wide; its depth does not exceed about 2 m; there is water in the pond all year round. The shores are sandy, while the deeper parts are muddy. Along the shores, some scarce aquatic plants can be found. Besides the acerosus-like specimen, only V. viviparus were found occurring in mass along the shore. There is another, small pond, situated about 500 m down the river from the one described above. Its distance from the dyke is the same as that of the former pond, but its length and width are smaller by half. Its character is also quite different: the depth is less than | m; the vegetation is rich, and there is no mud. The small pond temporarily des- iccates. It is inhabited by a dense population of Viviparus contectus. Although the shells of V. contectus at this site are small and show some characters, in which they resemble V. viviparus, they are typical V. contectus, both anatomically and electrophoretically. The two ponds are temporarily Table 2. Enzyme systems showing constant differences (no allele in com- mon in at least one locus) between Viviparus contectus and V. viviparus, and typical heterozygotic bands in the hybrid specimen. Enzyme name Enzyme E.C. © staining after number . B-hydroxybutyrate dehydrogenase =E.C. 1.1.1.30 Wurzinger (1979) I 2. Phosphoglucomutase EC. 2.7.5.1 Wurzinger (1979) 3. Leucine aminopeptidase E.C. 3.4.11.1 Rudolph and Burch (1987) 4. Lactate dehydrogenase E.C. 1.1.1.27 Richardson et al. (1986) 5. Malate dehydrogenase E.C. 1.1.1.37 Wurzinger (1979) 6. Phosphohexose isomerase EC. 5.3.1.9 Wurzinger (1979) 7. Hexokinase EC. 2.7.1.1 Richardson et al. (1986) 8. Adenylate kinase E.C. 2.7.4.3 Richardson et al. (1986) linked by floods, which makes gene flow possible. The big, acerosus-like specimen had the shell 44.0 mm high and 33.8 mm broad. The umbilicus was a cross between the ones of Viviparus contectus and V. viviparus, while the shell colouration with distinct bands resembled rather the latter species. Also the operculum showed inter- mediate characters. The specimen was a female. The recep- taculum seminis with its duct (Fig. 2) resembled more the one typical of V. viviparus, but its diagnostic characters were intermediate enough to make determination hardly possible. There were as many as 32 embryos in the brood pouch, which is a number typical of the Polish viviparids, especially of V. contectus (Falniowski, 1989). From among the embryos, 17 were in early stages, with no shell; eight had small shells; seven were ready to be born. The maxi- mum shell dimension of the embryos in the latter group was 4.4 - 5.0 mm; the shell height was up to 4.2 mm. The shells had a covered umbilicus (like in V. viviparus), and a blunt apex (but sharper than in V. viviparus). There were two rows of vestigial bristles along the body whorl. Similar Fig. 1. Viviparus contectus f. hungaricus Hazay, 1881, from the Drwinka River, Niepolomice Forest (after Falniowski, 1989, plate II, phot. 9), spec- imen in the collection of the Zoological Museum of Jagiellonian University, Krakow. MZUJF00217. FALNIOWSKI ET AL.: VIVIPARID HYBRID Fig. 2. Diagnostic characters in viviparid female reproductive system: receptaculum seminis (rs) with its duct (rsd): A, Viviparus contectus; B, the hybrid specimen; C, Viviparus viviparus. viviparus acerosus Nit aes * contectus X viviparus Fig. 3. Zymogram of &-hydroxybutyrate dehydrogenase of Viviparus viviparus, V. contectus, V. acerosus and the hybrid specimen (marked with two “X’’). 163 bristles but longer are characteristic of young V. contectus, whereas young V. viviparus have no bristles at all. Polyacrylamide gel electrophoresis (Table |) of the hepato-pancreas tissue of this curious specimen revealed its hybrid character (Viviparius contectus x V. viviparus), which was detected at eight loci of eight different enzymes (Table 2, Figs. 3, 4). At the same time, no similarity in allozyme pattern was found between the specimen and V. acerosus. %-hydroxybutyrate dehydrogenase (Fig. 3) pro- vides a good example, with specific homozygotes in each species, and a typical heterozygote in the hybrid specimen. The above data look interesting, but their more pro- found interpretation requires a better understanding of viviparid relationships. It seems that Viviparus viviparus and V. contectus can produce fertile hybrids in natural con- ditions. Such hybrids are not common: among about 150 specimens collected at the same locality in May 1992, no such shell was found. However, the hybrid origin of the specimen presented in figure | is very likely. It is also not sure that the hybridization of the two species has always to lead to such heterosis. Hence, only an electrophoretical study on rich material from the Niepolomice Forest can give exact data on hybrid frequencies. Such a study is planned. The observed interspecific differences in allozyme pattern, different alleles being found in at least one locus in almost any enzyme system studied, seem to indicate that Viviparus contectus and V. viviparus are very old species 2 3 4 5 6 7 8 Fig. 4. Schematical diagram of isozyme electrophoretic patterns obseved in Viviparus viviparus, V. contectus, and the hybride individual Only these zones (presumable loci) are presented, which show no allele shared by the two species. For each enzyme system, the heterozygotic hybride specimen is in the mid- dle, V. viviparus in the left, and V. contectus in the right. Numbers of enzyme systems as in Table 2 (2, the slower of two activity zones; 3, the fastest of sev- eral zones detected; 4, the faster of two zones; 8, one of several zones detected). 164 AMER. MALAC. BULL. 10(2) (1993) and the isolating mechanisms that prevent their hybridiza- tion could have broken down after such a long time since the speciation event (the case discussed by Wiley, 1981) which is the more probable, now that they occur sympatri- cally very rarely. LITERATURE CITED Davis, J. B. 1964. Disc electrophoresis - II. Method and application to human serum proteins. Annals of New York Academy of Sciences 121:404-427. Falniowski, A. 1989. Przodoskrzelne (Prosobranchia, Gastropoda, Mollusca) Polski. I. Neritidae, Viviparidae, Valvatidae, Bithyniidae, Rissoidae, Aciculidae - De Prosobranchiis in Polonia obviis. Pars I. Zeszyty Naukowe Uniwersytetu Jagiellonskiego, CMX, Prace Zoologiczne 35:1-148 + i-xx plates (in Polish, with an English sum- mary). Falniowski, A. 1990. Anatomical characters and SEM structure of radula and shell in the species-level taxonomy of freshwater prosobranchs (Mollusca: Gastropoda: Prosobranchia): a comparative usefulness study. Folia Malacologia 4:53-142 + 78 plates. Richardson, B. J., P. R. Baverstock and M. Adams. 1986. Allozyme Electrophoresis. A Handbook for Animal Systematics and Population Studies. Academic Press, Sydney. 410 pp. Rudolph, P. H. and J. B. Burch. 1987. Inheritance of alleles at ten enzy- matic loci of the freshwater snail Stagnicola elodes Lymnaeidae). Genetical Research, Cambridge 49:201-206. Wiley, E. O. 1981. Phylogenetics: The Theory and Practice of Phylo- genetic Systematics. Wiley Interscience, New York. 439 pp. Wurzinger, K.-H. 1979. Allozymes of Ethiopian Bulinus sericinus and Egyptian Bulinus truncatus. Malacological Review 12:51-58. Zhadin, V. I. 1928. Issledovanija po ekologii i izmencivosti Vivipara fas- ciata Muller. Monografii Volzanskoi Biologiceskoi stancii, Saratov 3:1-94. Date of manuscript acceptance: 9 November 1992 Sententia The Archaeogastropoda A clade, a grade or what else?’ Gerhard Haszprunar Institut fiir Zoologie der Leopold-Franzens-Universitat, Technikerstrasse 25, A-6020 Innsbruck, Austria Abstract: The classic concept of Thiele’s Archaeogastropoda includes the Docoglossa (now Patellogastropoda), Neritacea (now Neritimorpha), Cocculinacea (now Cocculiniformia), Zeugobranchia and Trochacea (now Vetigastropoda). The recent discovery of many new archaeogastropod groups mainly from deep waters, and in particular from the hydrothermal vent habitat, necessitates a reevaluation of the taxon. Archaeogastropoda can still be clear- ly defined by protoconch characters and by the diagnostic streptoneurous and hypoathroid nervous system (close association of pleural and pedal ganglia). This definition also includes the Neritimorpha and the architaenioglossate groups. The amount of convergence is very high among archaeogastropods, most characters of advanced gastropods occurred several times in parallel in evolution. As in Thiele’s time, the taxon “Archaeogastropoda” is regarded as the basic gastropod stem group and should be classified as a paraphyletic taxon. Whether this stem group gave rise to a single or two lines of higher gastropods is still a matter of debate. The taxon Archaeogastropoda was introduced by Thiele (1925) and cemented by the author in his famous “Handbuch der Systematischen Weichtierkunde”’ (Thiele, 1929). Five subgroups were included, namely the “Stirps” Zeugobranchia (those with two gills, now included in Vetigastropoda), the Patellacea or Docoglossa (now also called Patellogastropoda), the Trochacea (now includ- ed in Vetigastropoda), the Neritacea (now Neritimorpha), and the Cocculinacea (now Cocculiniformia). Although quite weakly defined - there was not a single diagnostic character given - the use and value of the taxon Archaeogastropoda has not been seriously questioned until very recent times [see Hickman (1988) for detailed historical review]. The first reason to question the use of Archaeo- gastropoda is due to the many recent discoveries of new archaeogastropod species and groups from the deep-sea, in particular but not exclusively from the strange habitat of the hydrothermal vents. Starting with the enigmatic Neomphalus fretterae McLean, 1981, these now include several groups, for which various ranks between genus and order have been proposed (e.g. Hickman, 1984; Marshall, 1988; McLean, 1988, 1989a, b, 1990a, b, 1992; Warén and Bouchet, 1989; Warén, 1989, 1991, 1992; Beck, 1992a, b). Anatomical investigations of these forms (e.g. Fretter, 1988, 1989, 1990; Haszprunar, 1989a, b) as well as of other long- named archaeogastropods such as the Cocculiniformia [see 1This contribution was provided as part of the 1992 AMU Symposium on Advances in Gastropod Phylogeny. Haszprunar (1988a, b) for review, also 1992c] show that at least certain archaeogastropods are by far more advanced and more variable than previously thought. Obviously the assumption is no longer valid that Archaeogastropoda de- scribes a specific level of organization, as it may be holo- phyletic, paraphyletic or polyphyletic. Consequently, the question arises: “What is an archaeogastropod?” or better: “Is it possible to define a taxon Archaeogastropoda and what status has that taxon?” A second reason to question the validity of the taxon Archaeogastropoda is based on the cladistic point of view. Most cladists argue to use in phylogenetic systems only monophyletic (sensu stricto, i.e. holophyletic), i.e. groups with a common ancestor, which include all descen- dents of that ancestor (e.g. Hennig, 1966; Wiley, 1981). Because most authors regard Archaeogastropoda in the sense of Thiele (1929) as a paraphyletic (e.g. Haszprunar, 1988b; see also below) or even as a polyphyletic (e.g. Hickman, 1988), the taxon should no longer be used any more according to that view. In contrast, the author (Haszprunar, 1986, 1988b) has argued to retain paraphylet- ic taxa in classifications, if (and only if) (1) they are quali- fied as such and (2) the relationships between taxa are expressed exactly. In this paper various characters, which have been used to define the taxon Archaeogastropoda, or which are considered primitive for gastropods in general, will be ana- lyzed with respect to homology questions and distribution. American Malacological Bulletin, Vol. 10(2) (1993):165-177 165 166 AMER. MALAC. BULL. 10(2) (1993) After that the status and validity of the taxon Archaeo- gastropoda will be discussed. HOW TO DEFINE ARCHAEOGASTROPODA? — CHARACTER ANALYSIS GENERAL REMARKS In the following, ‘““Archaeogastropoda” is used in the sense of Haszprunar (1988b), including the Patello- gastropoda (i.e, Docoglossa, including the Neolepetopsi- dae, the former “hot-vent group C”’), Cocculiniformia, Neritimorpha, Melanodrymia, Peltospiroidea (formerly “hot-vent group A’), Neomphaloidea, Vetigastropoda (in- cluding also the Seguenzioidea, see below), and the archi- taenioglossate groups (Cyclophoroidea, Ampullarioidea). Fossil groups could be added, if shell characters suggest their archaeogastropod nature (see below). The term “character analysis” is used here in the sense that homology questions as well as distribution pat- terns are discussed. To a certain extent this has been done already earlier (Haszprunar, 1988b), nevertheless I think it is justified to present an updated review: Firstly, there are several new groups to be considered, secondly, certain characters (e.g. heterostrophy) have been differently inter- preted by other authors, and finally, certain new characters (e.g. developmental timing) have been proposed mean- while. TELEOCONCH It is obvious that shell shape cannot be used to define Archaeogastropoda, because all types of shell mor- phology are found. However, so-called “symmetrical” limpets, which do not show any coiling of the juvenile teleoconch, are not found in any other gastropod group. Confusion of symmetrical limpets with the neopilinids (Monoplacophora) can be ruled out if (1) the shell apex is posterior; or (2) a deformed protoconch is present (see below); or (3) muscle scars are visible; or (4) shell structure is considered. The presence of nacre is restricted to the archaeogastropod grade (among Gastropoda), although many groups have lost or replaced it secondarily. Whereas a shell slit/hole(s) is characteristic for zeugobranch groups (Scissurelloidea, Fissurelloidea, Haliotioidea, and Pleurotomarioidea), it is not diagnostic at all, since shell slits are also found in Siliquariidae (Cerithioidea) and many Bellerophontida, the gastropod nature of which is still questionable [see Haszprunar (1988b) for recent review]. PROTOCONCH Scanning electron microscopy has revealed many new characters useful for gastropod systematics. In particu- lar, protoconch features (Fig. 1) are very useful to define distinct groups. According to the information available, four types of protoconch formation are typical for archaeo- gastropods: (1) In the Patellogastropoda, the embryonic shel 1 (i.e. protoconch I, mineralized at once by the shell gland) is usually not really coiled but more or less bent [e.g. Bandel, 1982 (figures labeled as Cocculina reticulata Verrill, 1884 (p. 35, Abb. 26; Tafel 8, Figs. 4, 5, 9) and Cocculina cf. spinigera (p. 36, Abb. 27A; Tafel 8, Figs. 3, 6, 8) very probably show lepetids rather than cocculinids. Compare with SEM-photos of cocculinid limpets in Marshall (1986); Warén, 1988; Fig. 1A]. Depending on the amount of yolk the embryonic shel1 may or may not be symmetrical (see Bandel, 1982). After metamorphosis the embryonic shell is lost together with the early teleo- conch after the formation of a distinct septum leaving a characteristic scar (Smith, 1935; Bandel, 1982; Warén, 1988). Very often the axes of the embryonic and adult patellogastropod shell form a characteristic angle (e.g. Thompson, 1912). This angle has been interpreted as remi- niscent of a coiled ancestor by Lindberg (1988). In this case [but see contrary arguments in Haszprunar (1988b: 370-372)], such an ancestor probably had an hyperstrophic rather than an orthostrophic teleoconch, (see legend of Fig. 1 for definitions) judging from the orientation of the axes. (2) In the lepetelloid limpets (Choristellidae show a vetigastropod-like protoconch, see below) the apex of the embryonic shell is fused in a very characteristic way with the teleoconch, resulting in lateral pouches (cf. Marshall, 1986; Fig. IB). (3) The neritimorph condition (Fig. 1C) is unique among the archaeogastropods in showing a true, multispiral larval shell (i.e. protoconch II, formed and mineralized successively by the mantle margin). As outlined by Bandel (1982, 1992) this type of larval shell is diagnostic for the Neritimorpha and can be used to infer close relationships between this group and the extinct Platyceratoidea. (4) The majority of archaeogastropods show the so- called “trochoid” condition, in which the embryonic shell is immediately followed by the teleoconch. The morpholo- gy of the embryonic shell is highly variable (e.g. Hickman, 1992: fig.5; Fig. 1D,E,F) and could well be used to diagnose minor taxa. In contrast, caenogastropods have orthostrophic larval shells (Fig. 1G-H), and heterobranchs show hyper- strophic larval shells (Fig. 1I1-K), although the condition may be cryptic in the case of lecithotrophic development. Recently, Hadfield and Strathmann (1990) de- scribed “hyperstrophic” protoconchs resp. “heterostrophy” (see legend of Fig. 1 for definitions of terms) of certain trochoid species (see also Hickman, 1992: fig. SL). However, the term “heterostrophy” describes a relationship HASZPRUNAR: ARCHAEOGASTROPODA 167 Fig. 1: Protoconchs of selected gastropods. A. Patellogastropoda-Lepetidae (200-800 m off SE-coast of USA). The embryonic shell (lateral view from the left) is bulbous, the deformation is not planispiral in this case (“Cocculina reticulata” from Bandel, 1982: fig. 26). B. Cocculiniformia-Pseudococculinidae: Mesopelex zelandica Marshall, 1986: Embryonic shell (about 135 um wide and 210 um long) with subreticulate sculpture from (a) lateral right and from (b) anterior showing bilateral symmetry (after SEM-photos of Marshall, 1986: figs. 10C, D). C. Neritimorpha-Neritidae: Smaragdia sp. (Red Sea): Larval shell in (a) transparent to show whorl contours, in (b) external view (from Bandel, 1982: fig 73). D. Vetigastropoda-Scissurellidae: Sinezona sp. (Gmelin, 1791) (Canary Islands): Embryonic shell with strong axial ribs (from Bandel, 1982: fig. 20). E. Vetigastropoda-Fissurellidae: Fissurella angusta Woodring, 1928: Embryonic shell with apertural ridge and sculpture caused by deformation during torsion (from Bandel, 1982: fig. 21). F. Vetigastropoda-Trochidae: Microgaza vetula L., 1767: Embryonic shell with a mixture of spiral ribs and tubercules as sculpture (from Bandel, 1982: fig. 18). G. Coiled Caenogastropoda: (a) Thais haemostoma (Muricoidea): (b) Cerithium litteratum (Born, 1778) (Cerithioidea): Embryonic and larval shel1 show different sculptures because of planktotrophic development (from Bandel, 1982: fig. 87). H. Caenogastropod limpets (Hipponicoidea): (a) Hipponix conicus (Schumacher, 18417): Embryonic and larval shell show different sculpture; (b) Hipponix antiquatus (L., 1767): Embryonic and larval shell cannot be distin- guished by sculpture (from Bandel, 1982: fig. 82). I. Allogastropoda-Architectonicidae: Philippia krebsii (Morch, 1875): The larval shell is hyperstrophic 1), whereas the adult shell starts orthostrophically !). Note anastrophic 2) relationship between larval and adult shell (after Robertson, 1974 from Haszprunar, 1985b: fig. 1c). J. Opisthobranchia-Acteonidae: Acteon tornatilis L., 1758: The larval shel1 is hyperstrophic!), whereas the adult shell starts orthostrophi- cally }). Note heterostrophic (sensu stricto 2) relationship between larval and adult shell [after Thorson, (1946) from Haszprunar (1985b: fig. 1d)]. K. Pulmonata-Siphonariidae: Wil liamia krebsii (M6rch, 1877): The embryonic she11 can be clearly distinguished from the hyperstrophic !) larval shell (from Bandel, (1982: fig. 83). 1) “Orthostrophy” and “hyperstrophy” refer to the relationship between the orientation (left or right-handed) of a helicoid (larval or adult) shell and the orien- tation of the soft body. If shell and soft body are equally handed (regular gastropods, but also triphorids) this is called “orthostrophic’’, if they are differ- ently handed (e.g. larval heterobranchs, neotenic Euthecosomata, also the adult ampullariid Lanistes), it is called “hyperstrophy”’. 2) “Heterostrophy” (sensu Jato) means the different orientation of the axes of the larval and the adult shell. If the angle of axes is about 90° this is called "het- erostrophy sensu stricto", if the angle is about 180° this is called “anastrophy”. 168 AMER. MALAC. BULL. 10(2) (1993) between a larval shell (see Fig. | for definitions) and the teleoconch. In this sense it was taken by Robertson (1985), Haszprunar (1985, 1988b), Bandel (1982, 1988b, 1992), and Ponder (1991). The formation and mode of coiling of an “hyperstrophic” embryonic shell are the results of an entirely different process compared with a hyperstrophic larval shell (Bandel, 1982: 27-36; Hickman, 1992; Warén and Bouchet, 1993: 49). Therefore the “heterostrophy”’ of these trochoids does not corroborate the concept of het- erostrophy respectively of the hyperstrophic larval shell as a basic synapomorphy of heterobranch gastropods (Robert- son, 1985; Haszprunar, 1985, 1988b; Ponder, 1991). How- ever, the presence of hyperstrophic-like embryonic shells among archaeogastropods supports the statement made by Haszprunar (1988b) that torsion and the mode of shell coil- ing are primarily independent features. (5) All the above mentioned characters concern marine species. Freshwater and terrestrial gastropods gener- ally show modifications in shell ontogeny. Whereas nearly nothing is known concerning shell ontogeny of cyclo- phoroids, the shell formation of ampullariids has been studied in some detail (e.g. Demian and Yousif, 1973a, b; Honegger, 1974; Lehmann, 1992). In particular Lehmann (1992) pointed out major differences between ampullariid and archaeogastropod (including neritimorph) shell devel- opment. According to his results it appears more likely that the Ampullariidae have reached the caenogastropod level concerning shell formation. Summing up, protoconch characters enable one to distinguish clearly certain groups of (marine) archaeogas- tropods, caenogastropods and heterobranchs. It also shows that diagnostic features exist to define some fossil archaeogastropod groups, where protoconch data are avail- able. This has been already applied to split up such “lump- ing pots” as the extinct “Euomphalacea” (e.g. Bandel, 1988a) or the recent “‘skeneimorphs” (e.g. Warén, 1992). SHELL MUSCLES As reviewed earlier (Haszprunar, 1985c, 1988b), ontogenetic as well as anatomical investigations show that in many archaeogastropods the adult shell muscles are paired, whereas they are unpaired in caenogastropods and heterobranchs. However, certain archaeogastropods also show the unpaired condition with a left muscle only (e.g. Neomphalus, many trochoids) MANTLE CAVITY The archaeogastropod mantle cavity is usually fully torted, having its opening anteriorly situated. Certain lepe- telloidean limpets show a somewhat detorted orientation of the rectal-nephridial complex, whereas Neomphalus is unique in showing an hypertorted condition. It is generally accepted that a paired set of pallial organs (osphradia, ctenidia, hypobranchial glands) is the primitive condition among the gastropods. Indeed, retention of this character state is found only within the archaeogas- tropod grade, although there are many groups that have lost one or more of the right-side organs. CTENIDIA As outlined elsewhere (Haszprunar, 1988b: 377- 383) gastropod ctenidia are highly variable. The only gas- tropod gill-type, which is exclusively found within the Archaeogastropoda, is bipectinate and supplied by skeletal rods. Hickman (1988) recently proposed that this character should be diagnostic for a restricted use of “Archaeo- gastropoda” equivalent to Vetigastropoda. However, the same character set is found in Neomphalus and certain coiled peltospiroids (but not in Melanodrymia; see Hasz- prunar, 1989b), whereas on the other hand certain vetigas- tropods such as Temnocinclis, Temnozaga, Fissurisepta, many skeneids, or seguenziids) show monopectinate gills (see Cowan, 1969; Haszprunar, 1988b, 1989a, unpubl. data). CIRCULATORY AND EXCRETORY SYSTEM Comparable with the conditions of the pallial organs, the retention of two auricles (diotocard condition) or two kidneys is restricted to archaeogastropods. As demonstrated by the lepetodriloid and trochoid vetigas- tropods the loss of the right gill does not necessarily imply the loss of the right auricle. Again, however, many forms have independently reached the monotocardian or single (left) kidney condition of higher gastropods. Parallel events of loss are probable also with respect to the penetration of the pericardium by the rectum. GENITAL SYSTEM, GAMETES, AND REPRODUCTION Archaeogastropods are usually considered to be “primitive” with respect to reproduction in showing free (ectaquatic) fertilization. This is correlated with the so- called “primitive type” of spermatozoa, which (among gas- tropods) is indeed restricted to archaeogastropods. How- ever, entaquatic (in the female’s mantle cavity) or internal fertilization occurs frequently among archaeogastropods, in particular (1) in very small forms; (2) in deep-water species, including those from the hydrothermal vent habi- tat; (3) in freshwater or terrestrial groups. It should be stressed that in fact all so-called “advanced” conditions concerning molluscan reproduction such as internal fertil- ization, paraspermatozoa, spermatophores, copulatory HASZPRUNAR: ARCHAEOGASTROPODA 169 organs, receptacula, or brooding, are found in various archaeogastropod groups. Even within the Trochoidea, developmental data (e.g. time of hatching, planktonic ver- sus non-planktonic mode of development) vary consider- ably (Hickman, 1992). Concerning reproduction, Archaeo- gastropoda is certainly not a grouping of the same level of organization. Recently, Jamieson (1991) pointed out for teleost fishes that the “primitive type” of spermatozoa (respective- ly ectaquatic fertilization) is probably an advanced feature in that group. The same conclusion has been reached for solitary ascidians (e.g. Franzén, 1992). Considering early gastropods to be very small animals (Haszprunar, 1988b, 1992a), entaquatic fertilization (by sperm of the “primitive type’) could be the more primitive archaeogastropo condi- tion. CLEAVAGE PATTERN Most recently Van den Biggelaar (1993) paid atten- tion to distinct differences between archaeogastropods and higher groups (Caenogastropoda and Heterobranchia) in the spatiotemporal cleavage pattern. (1) Whereas in archaeogastropods (Patella vulgata L., 1758, Haliotis tuberculata L., 1758 and Gibbula magus (L., 1767) have been investigated) the mesentoblast (i.e. the 4d—cell) is formed at the transition of the 63- into the 64-cell stage, this occurs already at the transition of the 24- into the 25-cell stage in caenogastropods (e.g. Littorina, Crepidula, Ilyanassa), Opisthobranchs (e.g. Haminoea, Aplysia, Doris), and pulmonates (e.g. Physa, Lymnaea, Biomphalaria). Because the formation of the mesentoblast occurs even later (72-73 cell stage) in the Polyplacophora (e.g. see Heath, 1912), the archaeogastropod condition is considered as plesiomorphic. (2) In the gastropods there is a distinct trend to retard the formation of the first quartet cells and the forma- tion of the prototroch not only in relation to the develop- ment of the mesentoblast, but also in relation to absolute cell numbers. In Patella, the trochoblasts already divide between the 32-cel1 and the 40-cel1 stage and thus clearly before the formation of the mesentoblast. In Haliotis, this occurs between the 52-cel1 and 60-cel1 stage; in Gibbula, between the 55-cell to 64-cell stage, thus nearly parallel with the formation of the mesoentoblast In caenogas- tropods the first division of the trochoblasts follows the formation of the mesentoblast, whereas in opisthobranchs the formation of the first quartet cells is even more re- tarded. (3) The acceleration of the formation of the pro- totroch in caenogastropods, opisthobranchs and pulmonates is further associated with a considerably smaller number of cells, which build up the prototroch. This is probably corre- lated with the fact that the trochophore stage is free in Patella, partly free in Haliotis and Gibbula, and within the egg-capsule in higher gastropods. In the latter groups the prototroch is further transformed to the velum, which is built up by very many cells. It is unknown at present whether the trochoblasts are dividing again after differenti- ation or whether cells of different source are included in the velum (Van den Biggelaar, pers. comm.). Although the present number of investigated species is still very low (e.g. no data on marine Neritoidea or allogastropods), such data on the spatiotemporal pattern of development might become very useful for phylogenetic purposes. RADULA AND SUPPORTING STRUCTURES The main basis of Thiele’s Archaeogastropoda was the uniting of groups with a rhipidoglossate or docoglossate radula. However, the docoglossate (stereoglossate) condi- tion, i.e. simple rasping without longitudinal bending of the radular membrane and magnetite in the lateral teeth, is likewise found in neopilinids and chitons. Moreover, the recent discovery of valvatoideans (pers. obs. on anatomy; sperm data from J.M. Healy, 1993b) with rhipidoglossate radula (but with entirely different buccal apparatus, pers. obs.), the Hyalogyrinidae (Warén and Bouchet, 1993; Warén et al., 1993) omits also the second type as a simple diagnostic character for Archaeogastropoda. In addition, there are several archaeogastropod groups, in particular the Cocculiniformia (e.g. Hickman, 1983) but also several trochoid groups such as Trochaclidinae or Thysanodontinae (cf. Hickman and McLean, 1990), the radula of which do not fit any of the standard categories. The presence of several pairs of radular cartilages is restricted to the Archaeogastropoda. Certain archaeogas- tropods and caenogastropods in general have a single pair or none; Heterobranchia lack true cartilages, some have secondary ones. The combination “docoglossate or rhipi- doglossate radula with massive cartilages” is restricted to archaeogastropods, however. Most archaeogastropods are provided with a so- called radular diverticulum (cf. Haszprunar, 1988: 392), which is lacking only in the architaenioglossate groups. ALIMENTARY TRACT As outlined elsewhere (Salvini-Plawen and Hasz- prunar, 1987; Haszprunar, 1988b) the presence of oesophageal pouches, which can be simply structured or papillate, is restricted to Archaeogastropoda (among Gastropoda). The same is true for the so-called “anterior loop” of the intestine. In both cases, however, certain 170 AMER. MALAC. BULL. 10(2) (1993) archaeogastropods have reached independently the ad- vanced stage. NERVOUS SYSTEM Pedal cords are common among archaeogastropods, but are also present in certain caenogastropods such as Lavigeria (Cerithioidea) (Moore, 1899 as Nassopsis) or in the Cypraeidae (e.g. see Riese, 1930). Therefore the charac- ter cannot be used to define Archaeogastropoda. The pres- ence of a labial commissure is restricted to archaeogas- tropods, again however, many members have lost it in par- allel to caenogastropods and heterobranchs. As stated repeatedly (Salvini-Plawen and Hasz- prunar, 1987; Haszprunar, 1988a, b) the condition of a hypoathroid cerebropedal ring (i.e. pleural and pedal gan- glia being close together) is the only character that is pre- sent in all archaeogastropods including the architaenioglos- sate groups. In the Viviparidae alone the hypoathroid condi- tion is restricted to the left side, whereas the right side is epiathroid (close up of cerebral and pleural ganglia). This condition is independent of habitat (marine, freshwater, ter- restrial), habit (mode of nourishment), or other modifica- tions of the central nervous system (see above). In contrast, all caenogastropoda and primitive heterobranchs show an epiathroid condition. The hypoathroid condition is also found in certain euthyneuran (i.e. osphradial ganglion at the right side; cf. Haszprunar, 1985b, 1988b) groups such as Aplysiomorpha or Gymnosomata (cf. Hoffmann, 1932-39) or in the Eupulmonata (Trimusculoidea, Ellobioidea, and most Stylommatophora; cf. Haszprunar and Huber (1990) for review). Therefore, I have defined the Archaeogastropoda by a “streptoneurous and hypoathroid nervous system” (Haszprunar, 1988b). Moreover, the hypoathroid-like con- ditions of euthyneurans differ in two fundamental aspects from that of the Archaeogastropoda: (1) Several authors agree that in the euthyneurans the original pleural ganglion is split off into the pleural sensu stricto and so-called parietal ganglion (Regondaud et al., 1974; Brace, 1977; Schmekel, 1985, Haszprunar, 1988b; Haszprunar and Huber, 1990). Usually the hypoathroid condition of euthyneurans concern the pleural ganglia alone, whereas the parietal ganglia are not included, but are fused with the suboesophageal or supraoesophageal ganglion. (2) So far known the hypoathroid condition of euthyneurans is a secondary phenomenon, because during ontogeny an epiathroid condition respectively a common cerebropleural anlage is primarily established as in caenogastropods or in epiathroid heterobranchs. This is well documented in Aplysia (e.g. Kandel et al., 1980; Jacob, 1984; Fig. 2E-F), in the Ellobiidae (Ruthensteiner, 1991, 1992; Fig. 2G-H), and in the Stylommatophora (e.g. Henchmann, 1890). In contrast, the anlage of the pleural ganglion is always close to the pedal ganglion in the Archaeogastropoda. This has been described in Patella (Smith, 1935), Haliotis (Crofts, 1937; Barlow and Truman, 1992; fig 2A-B), Theodoxus (Ruthensteiner, 1991: 72), Ampullarius (Honegger, 1974), and Marisa (Demian and Yousir, 1975; fig. 2C-D). The single exception is again Viviparus (left side hypoathroid, right side epiathroid as adults; see above), for which Andersen (1924) described a common cerebropleural anlage as in caenogastropods or heterobranchs. Most recently, Page (1992a, b) claimed an hypo- athroid condition in a dendronotoid nudibranch Melibe leonina (Gould, 1852). Her results are based on fine-struc- tural studies and include: (1) the pleurals originate from a post-trochal placode as in archaeogastropods; (2) the pleu- rals are situated in front of the pedal ganglia; (3) the pleu- rals are connected to the labial lobe of the cerebral ganglia. Nevertheless, the presented interpretation causes serious problems: (1) the results are in direct contrast to all previ- ous ones on neurogenesis in nudibranchs (e.g, Thompson, 1958, 1962; Tardy, 1970, 1974).; (2) Her “labial lobe” of pyramidellids (“sensory lobe” of Fretter and Graham, 1949) is a rhinophoral ganglion [pers. obs.; see also Haszprunar and Huber (1990) for discussion], and it is like- ly that the same statement can be made about the “labial lobe” of Melibe; (3) Melibe leonina (Gould, 1852) would be the only gastropod (mollusc), in which the visceral loop does not start from the pleural ganglia, but directly from the cerebral ganglia. Moreover, there is no connection between the labial lobe and the pleural ganglia in any other gastropod; (4) the “pleural” ganglia of Melibe strongly resemble the so-called propodal ganglia of Onchidoris bil- ammelata L., 1767, which are likewise connected with the cerebral ganglia (Chia and Koss, 1989). All these argu- ments suggest that the interpretation of Page (1992a, b) is incorrect. Nevertheless, modern neuroanatomical trace methods such as antibody-staining or cobalt-filling [cf. Heimer and Zaborszky (1989) for review] are necessary to finally accept or reject this proposal. However, even if the interpretation of Page (1992 a, b) would be correct, the conditions of Melibe are clearly not directly comparable with those of the Archaeogastropoda and do not influence the validity of the hypoathroid nervous system as a diag- nostic character of the Archaeogastopoda. SENSE ORGANS Eyes lacking a cornea are restricted to the Archaeo- gastropoda, although several groups have developed closed eyes. A subradular organ is found only in certain archaeo- gastropods. Epipodial tentacles also occur in certain HASZPRUNAR: ARCHAEOGASTROPODA 171 0.1 Fig. 2. Comparative view of ontogenesis of the nervous system in selected gastropods to demonstrate primary versus secondary hypoathroid conditions. Scale bars in millimeter. (C - cerebral ganglion; CP1 - cerebropleural ganglion; E - eye; Os - osphradial ganglion; P - pedal ganglion, Pa - parietal ganglion; P1 - pleural ganglion; PSp - fused parietal and supraoesophageal ganglion; Sb - suboesophageal ganglion; Sp - supracesophageal ganglion; St - Statocyst; V - visceral ganglion; VSb - fused visceral and suboesophageal ganglion). A-B. Haliotis tuberculata L., 1758 (Vetigastropoda - Haliotioidea) with hypoathroid nervous system from the beginning. A. 3 days old veliger. B. About 2 months old post-veliger (modified after Crofts, 1937). C-D. Marisa cor- nuarietis (L., 1758) (Architaenioglossa - Ampullarioidea) with hypoathroid nervous system from the beginning. C. Embryos stage VIII (90 hours at 25- 30°C respectively 3 days at 15-20°C). D. Embryo stage X (5 days at 25-30°C respectively 14 days at 15-20°C) (modified after Demian and Yousif, 1975). E-F. Aplysia californica Cooper (Opisthobranchia - Aplysioidea): E. 3 weeks old veliger with epiathroid condition. F. 2 months old juvenile with pleural ganglion in intermediate position (adults are fully hypoathroid) (modified after Kandel et al., 1979) G-H. Ovatella (Myosotella) myosotis (Draparnaud, 1804) (Pulmonata - Ellobioidea). G. Veliger 12 days old with fused cerebropleural ganglion. H. Hatchling about 21 days old with pleural ganglion in inter- mediate position (adults are fully hypoathroid, (modified after Ruthensteiner, 1991). 172 AMER. MALAC. BULL. 10(2) (1993) caenogastropods such as Alaba or Litiopa (Cerithioidea) (Houbrick, 1987), epipodial sense organs appear to be restricted to archaeogastropods. So-called bursicles (Szal, 1971) have been consid- ered to be a synapomorphy of vetigastropods (Salvini- Plawen and Haszprunar, 1987; Haszprunar, 1988b: 398- 399). However, bursicles have since been found also in several lepetelloid families (Haszprunar, 1988a; unpubl.), in the enigmatic Melanodrymia aurantiaca Hickman, 1984 (cf. Haszprunar, 1989b), and in the seguenziids (Hasz- prunar, unpubl.). Whereas the latter can be reasonably included within the vetigastropods, lepetelloids and Melanodrymia are still considered as outgroups. Up to now, bursicles have not been reported from any caenogastropod or heterobranch. SUMMARY OF CHARACTER ANALYSIS Summing up, the diagnosis of “an archaeogastro- pod” among the Gastropoda is: - shell shape is a “symmetrical” limpet (no juvenile coiling); - nacre is present; - the protoconch is patellogastropod-like (lost with a part of the teleoconch by a septurn), lepetelloid (embryonic shell apex fused with teleoconch), neritimorph (several largely overlapping whorls of larval shell), or trochoid-like (marine forms only); the adult shell muscle(s) is(are) paired; both ctenidia or osphradia or hypobranchial glands are retained; - the ctenidia are bipectinate and supported by skeletal rods; - both auricles are retained; - both kidneys are retained; the rectum runs through the pericardium; ectaquatic or entaquatic fertilization by the “‘prim- itive” type of spermatozoa; the paired mesentoblast (4d-cell) is formed between the 63-cell and 64-cell stage, the for- mation of the prototroch occurs before or parallel of to this event and includes many cells; - a docoglossate or rhipidoglossate radula type with massive cartilages is present; - aradular diverticulum is present; - oesophageal pouches are present; the “anterior loop” of intestine is present; a labial commissure is present; a streptoneurous and hypoathroid nervous system is present at least at the left side (THE ONLY DIAGNOSTIC CHARACTER); - the eyes lack a cornea; - a subradular organ is present; - epipodial sense organs are present; - bursicles are present. With the single exception of the hypoathroid ner- vous system all characters listed above are valid but not diagnostic, meaning that there are archaeogastropods, which do not fulfill the specific requirement (Table 1). Because most characters listed are probably plesiomorphic (see Haszprunar, 1988b for reasoning), they also describe the gastropod archaetype (stem species or HAG = Hypo- thetical Ancestral Gastropod). In other words, the first gas- tropod was by definition an archaeogastropod. “ARCHAEOGASTROPODA”’ A PARAPHYLETIC TAXON MONOPHYLY OF ARCHAEOGASTROPODA That archaeogastropod diagnostic characters are plesiomorphic strongly suggests that Archaeogastropoda is a paraphyletic taxon. This view is supported by the fact that the architaenioglossate groups (and certain trochoids?; cf. Healy, 1990) are linked by several characters, in particu- lar by sperm morphology (see Healy (1988) for recent review), with the Caenogastropoda (Cerithioidea). If this is accepted, then the Archaeogastropoda in the given diagno- sis cannot be clade. To decide between polyphyletic versus paraphyletic status the monophyly of Archaeogastropoda respectively of the Gastropoda as a whole has to be demon- strated. Up to now very few people have doubted the holo- phyly (monophyly sensu Hennig, 1966) of the Gastropoda. The only group, for which a separation has recently been proposed, are the Patellogastropoda (cf. Golikov and Starobogatov, 1975; Shilenko, 1977): However, all available evidence suggests that Patellogastropoda and all remaining gastropods have a common origin. The torsion process itself (cf. Crofts, 1955) as well as its various anatomical consequences (see review in Haszprunar, 1988b: 406) are essentially identical in all cases studied. Moreover, many pecularities of the Patellogastropoda such as the symmetrical limpet she11, many shell muscle bundles, or shallow mantle cavity, are also present in the Cocculiniformia. Thus, the latter group links the Patellogastropoda with the remaining archaeo- gastropod groups. Finally, the shared condition of a hypoathroid nervous system, which does not depend on the torsion process, is an independent character supporting the common origin of Patellogastropoda and all remaining Gastropoda. HASZPRUNAR: ARCHAEOGASTROPODA 173 TABLE 1. Characters useful for a definition of Archaeogastropoda. TAXON 12 3 45 67 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Patellogastropoda ++ 4+ 4+ -—- + ¥* + - -—- + + + + + F +E + + + + FF + +E + | = Cocculinoidea + - + + 2? + + + + - - + = =a Lepetelloidea +- + + - - * - - = + + + 2? 2? + + + + + + ~ + Neritimorpha -- ++ -- ++ - + - + - 2? + + + + + + = + = = = Melanodrymia - + + - - + 2 ed Set + + + - = + * = = 4+ Peltospiroidea —- +424 -- - 4 4 =—- + - = 2? 2? + = + + + + = + *¥ + - = Neomphaloidea —- + 4+ ---- + + = = SP ee Be Se a eS OE ee aS Scissurelloidea -- + + + + + + + + + + + $F FF + = + + + + + + - + + Lepetodriloidea —- + + + - + + + + + + =~ 2 2 + = + + + + + =e ee Fissurelloidea —- -—- + + + + + + + + + + + + 72 &£ = + F + +F F F = + + + Haliotoidea — + ++ 4+ + + + + + + F +F + + F = F Ff + HF Ff + +F + F + Pleurotomarioidea — + + - + + + +t Ft Ff F Ff Ff FF F = FH + HF +H FF K + + + Trochoidea —- + +4 +4 - - +++ 4+ + + + + + + = + + + + 4+ + + + + + Seguenzioidea - + 4+4---+--+- + - + = = ? an Sy gh a ee SG et Cyclophoroidea —--—- *---- * * = =- = = = 2? = = = + + A Ampullarioidea - - * - - - + e + + > J Caenogastropoda -- ------ + =) s2: 2.32. 2 =a Heterobranchia - - + + ye i: =) OS oe Characters: (1) "symmetrical" limpet; (2) nacre; (3) archaeogastropod protoconch (see text for various types); (4) paired adult shell muscle(s); (5) paired ctenidia; (6) paired osphradia; (7) paired hypobranchial glands; (8) bipectinate gills; (9) skeletal rods in gill leaflets; (10) paired auricles; (11) two kidneys; (12) rectum runs through the pericardium; (13) fertilization is ect- or entaquatic; (14) "primitive" type of spermatozoa; (15) late formation of mesoentoblast , many trochoblasts; (16) docoglossate or rhipidoglossate-like radula type; (17) three or more pairs of radular cartilages; (18) radular caecum; (19) oesophageal pouches; (20) "anterior loop" of intestine; (21) pedal cords; (22) labial commissure; (23) streptoneurous and hypoathroid nervous system (see text for hypoathroid condition in heterobranchs); (24) open eyes; (25) subradular organ; (26) epipodial sense organs; (27) bursicles. Legend of characters: (+) present; (—) absent; (+) both conditions are present within the taxon; (*) no relevant data; (?) no data available. MONOPHYLY OF HIGHER GASTROPODA During recent years, general agreement has been reached in regarding the Caenogastropoda (the majority of former Mesogastropoda and the Neogastropoda) and the Heterobranchia (allogastropods and euthyneurans) as good clades. Disagreement exists, however, whether both clades have a common or a separate origin. Whereas Salvini-Plawen and Haszprunar (1987) and Haszprunar (1988b) claimed a common origin with the epiathroid nervous system as most important synapomorphy, Ponder (1991) favored an inde- pendent origin of the Heterobranchia out of the Archaeo- gastropoda. Also, a most recent computer-aided reanalysis of the subject (Ponder and Lindberg, 1992) is still equivocal in this respect. Haszprunar (1988b) has proposed that the common ancestor of Caenogastropoda and the Heterobranchia prob- ably was a large animal (centimeter-range) and showed lar- val planktotrophy. If so, the metatrochal ciliary bands of the veliger larva would be directly homologous in both group- ings. Homology was also stated for the orthostrophic larval shell of the Caenogastropoda with the hyperstrophic larval shell of the Heterobranchia. In contrast, Ponder (1991) as- sumed a small (millimeter-range) heterobranch stem species with lecithotrophic development and an independent evolu- tion of larval planktotrophy and of the larval shell in Caenogastropoda and Heterobranchia. Several indices support the monophyletic version: (1) The lecithotrophic heterobranchs show a more or less distinct larval shell, at least their protoconchs always show more than one whorl, calling for a planktotrophic ancestor; (2) Fossil omalogyrids and orbitestellids are lecithotrophic or planktotrophic (Bandel, 1988a, 1991); (3) Both groups, Caenogastropoda and Heterobranchia (opisthobranchs were investigated) have metatrochal ciliary bands with the so- called "discoidal reticulate lamellae", a specific type of gly- cocalix (Bonar and Maugel, 1982). This structure is lacking in the metatrochi of bivalvian planktotrophic larvae and in "trochophore-like" planktotrophic larvae of other spiralian phyla. Unfortunately the presence or absence of discoidal reticulate lamellae could not yet be established in the planktotrophic Neritidae. Nevertheless, this highly specific and complex structure calls for direct homology of meta- trochi in Caenogastropoda and Heterobranchia and thus for common origin of larval planktotrophy in both groups; (4) Also the mentioned spatio-temporal shifts in the formation of the mesentoblast and the prototroch call for a common origin; (5) Finally, "the significant number of spermiogenic features shared by Caenogastropoda and Heterobranchia suggests either that these two groups arose from a common ‘archaeogastropod' source possessing these features or that heterobranchs were derived from early caenogastropds" (Healy, 1993b). 174 AMER. MALAC. BULL. 10(2) (1993) On the other hand, recent investigations on the osphradial fine-structure of Campanile symbolicum Iredale, 1917 (cf. Haszprunar, 1992b) have revealed that this enig- matic relict species should no longer be regarded as a link between primitive caenogastropods and heterobranchs as previously suggested (Haszprunar, 1988b), but should be classified as the earliest offshoot among the Caeno- gastropoda. The same conclusions has been reached inde- pendently by spermatological investigations on a related family, Plesiotrochidae (Healy, 1993a). In addition, the discovery of primitive hetero- branchs with rhipidoglossate radula (Hyalogyrinidae; see above) seem to support Ponder’s (1991) view, although a distinct archaeogastropod sister-group cannot be estab- lished at present (Haszprunar, in prep.). “ARCHAEOGASTROPODA” IN CLASSIFICATION So far the conclusion has been reached that Archaeogastropoda is a paraphyletic taxon, from which one or two lines (Caenogastropoda and Heterobranchia) have evolved (see above). There is a long-lasting debate in systematics, whether or not paraphyletic taxa should be allowed in phylogenetic classifications (1.e. unequivocal retranslation in the basic cladogram is possible; Wiley, 1981). Starting from the cladistic point of view (no para- phyletic groups), what would be the alternatives?: The holophyly of Patellogastropoda and of Neritimorpha is well established by synapomorphies (Haszprunar, 1988b), and certain authors even regard them as distinct orders out of Archaeogastropoda (e.g. Lindberg, 1988; Bandel, 1992). Holophyly of Cocculiniformia is more difficult to establish (Haszprunar, 1988a, b), but also this group might be excluded as an order proper. Although there is not a single autapomorphy known for a taxon Architaenioglossa, holo- phyly cannot be ruled out, and the taxon has a long tradi- tion. The real problem are the remaining groups (Melanodrymia, Neomphaloidea, Peltospiroidea, Veti- gastropoda, Seguenzioidea). To these groups several new ones such as the Pendromidae (cf. Warén, 1991; synonym Trachysmatidae Thiele, 1925) are or will be added in the near future. Such a taxon “Archaeogastropoda” is probably again a paraphyletic assemblage, and (even more serious) no distinct diagnosis can be given for a uniting taxon. Hickman (1988) proposed to replace Vetigastropoda (zeugobranchs, Lepetodriloidea, Trochoidea, probably also Seguenzioidea due to their recently found bursicles and epipodial sense organs) by Archaeogastropoda. But if so, what to do with the remaining groups? In other words: A restricted use of Archaeogastropoda does not solve any of the above mentioned problems, but adds the major one of a lacking diagnosis. Therefore I don’t think that abolishment or a restricted use of Archaeogastropoda is helpful. Hennig (1966, 1974) mentioned two main reasons to abolish paraphyletic groups. (1) His main argument, the equal use of para- and holophyletic taxa leads to serious confusion about the basic cladogram, is still fully valid. However, already Wiley (1981) mentioned in his Rule 1: “Taxa classified without qualification are monophyletic sensu Hennig (1966). Non-monophyletic groups can be added, if they are clearly qualified as such”. As outlined by the author (Haszprunar, 1986, 1990) a specific marking of paraphyletic groups and a general sequential arrangement of subtaxa overcomes Hennig’s (1966) main argument. Hennig’s (1974) second argument, that paraphyletic groups are some kind of polyphyletic group, must be rejected: In contrast to polyphyletic taxa, paraphyletic groups have a common ancestor (are monophyletic sensu lato) and represent like holophyletic (monophyletic sensu stricto) taxa a continuous genealogical line. The case of Archaeogastropoda also clearly demonstrates that a paraphyletic group is not by definition defined by a “lack of x” characters, but by positive charac- ters, which are nevertheless plesiomorphic (see summary of character analysis). The final cladistic argument, that only holophyletic taxa are “natural entities” and that paraphyletic groups are “arbitrary constructions” is rejected on following reasons: (1) Classification does not concern a group or its phylogeny itself, but a reconstruction of phylogeny, the nature of which is always hypothetical and probabilistic. According to Darwin (1872) “natural” means “strictly genealogical” (see also Wiley, 1981: rule 1); this requirement is fulfilled by the marked use of paraphyletic groups. (2) All para- phyletic groups once were holophyletic, meaning that they included all their descendents. Until the descent of Caeno- gastropoda and Heterobranchia “Archaeogastropoda” in the given definition was a holophyletic group. The use of marked paraphyletic taxa has a number of additional advantages (cf. Haszprunar, 1986, 1990): (1) More stability: many traditional or even nomenclatorically conserved (genera, species) paraphyletic taxa such as “Archaeogastropoda” can be still used in a phylogenetic system, if they are clearly marked. (2) So-called “chrono- species” such as (+) “Homo erectus” can be expressed unequivocally and clearly distinguished from offshoots such as (+) Homo sapiens neanderthalensis. (3) Com- bination of Wiley’s (1981) sedis mutabilis and the marking of paraphyletic taxa enables clear expression of so-called “metataxa” (cf. Gauthier, 1986: i.e, holo- or paraphyletic) such as in the case of “Architaenioglossa” with Ampul- larioidea and Cyclophoroidea both with sedis mutabilis (s. Haszprunar, 1988b). In the case of “Archaeogastropoda” an additional HASZPRUNAR: ARCHAEOGASTROPODA 173 argument can be made in favor of its marked retention: there are numerous species and groups, extinct or recent (e.g. Pendromidae or Adeuomphalus and Palazzia, cf. Warén, 1991), where conchological or radular or morpho- logical evidence clearly allows their inclusion among the archaeogastropods in the given definition, but where classi- fication within one of the subgroups cannot yet be estab- lished. Retaining “Archaeogastropoda” as a formal taxon allows a much clearer defined “pot” for such forms. Summing up, I recommend the retention of “Ar- chaeogastropoda” as a formal taxon in gastropod classifica- tion (cf. Haszprunar, 1988b). The given definition, which is mainly based on protoconch and neural characters, is valid for most extinct and recent forms. However, as expressed by its marking: have in mind that “Archaeo- gastropoda” is not a clade but a paraphyletic taxon. In fact, “Archaeogastropoda” is the stem group of the Gastropoda, yet it shows the widest secondary radiation of all gastropod groups. ACKNOWLEDGMENTS I thank the A.M.U. for the invitation to present this topic during the annual meeting in August 1992 in Sarasota, Florida at the International Symposium “Advances in Gastropod Phylogeny”, which was organized by Terrence Gosliner. I am indebted to Prof. Van den Biggelaar (Uni- versity of Utrecht) for making available to me his results on cleavage pat- terns prior to publication. Finally, I am grateful to Winston F. Ponder (Australian Museum, Sydney), James H. McLean (Los Angeles County Museum) and an anonymous referee for critical review of the paper. LITERATURE CITED Andersen, K. 1924. Entwicklungsgeschichtliche Untersuchungen an Paludina vivipara. 2. Teil. Die Entwicklung des Nervensystems bei Paludina vivipara, zugleich eine kritische Studie tiber die Torsion und Chiastoneurie der Gastropoden. Morphologische Jahrbiicher 54:157-204. Bandel, K. 1982. Morphologie und Bildung der friihontogenetischen Gehause bei conchiferen Mol lusken. Facies (Erlangen) 7:1-198, plates 1-22. Bandel, K. 1988a. Reprasentieren die Euomphaloidea eine natiirliche Einheit der Gastropoden? Mitteilungen des Geologisch- Paldontologischen Institutes der Universitit Hamburg 67:1-33. Bandel, K. 1988b. Early ontogenetic shell and shell structure as aids to unravel gastropod phylogeny and evolution In: Prosobranch Phylogeny, Ponder, W. F. ed., Malacological Review Supplement 4:267-272. Bandel, K. 1991 Ontogenetic changes reflected in the morphology of the molluscan shell. In: Constructional Morphology and Evolution, N. Schmidt-Kittler and K.Vogel, eds., pp. 211-230. Springer Verlag, Berlin. Bandel, K. 1992. Larger gastropod units present in the Triassic of St. Cassian formation. Proceedings of the 10th International Mala- cological Congress, Tiibingen 1989 (ISSN 1018-9092):497-502. Barlow, L. A. and J. W. Truman. Patterns of serotonin and SCP immunoreactivity during metamorphosis of the nervous system of the red abalone, Haliotis rufescens. Journal of Neurobiology 23:829-844. Beck, L. A. 1992a. Symmetromphalus hageni sp.n., anew neomphalid gas- tropod (Prosobranchia: Neomphalidae) from hydrothermal vents at the Manus Back-Arc Basin (Bismarck Sea, Papua New Guinea). Annalen des Naturhistorischen Museums Wien 93 B:243-257. Beck, L. A. 1992b. Two new neritacean limpets (Gastropoda: Proso- branchia: Neritacea: Phenacolepadidae) from active hydrothermal vents at Hydrothermal Field 1 ““Wienerwald” in the Manus Back- Arc Basin (Bismarck Sea, Papua-New Guinea). Annalen des Naturhistorischen Museums Wien 93 B:293-275. Bonar, D. B. and T. K. Maugel. 1982. Particle exclusion by discoidal reticulate lamellae in larval gastropod feeding structures. Journal of Ultrastructural Research 81:88-103. Brace, R. C. 1977. Anatomical changes in nervous and vascular system during the transition from prosobranch to opisthobranch organiza- tion. Transactions of the Zoological Society of London 34:1-25. Chia, F.-S. and R. Koss. 1989. The fine structure of the newly discovered propodial ganglia of the veliger larva of the nudibranch Onchidoris bilammelata. Cell and Tissue Research 256:17-26. Cowan, I. M. 1969. A new species of gastropod (Fissurellidae, Fissurisepta) from the Eastern North Pacific Ocean. Veliger 12:24-26. Crofts, D. R. 1937. The development of Haliotis tuberculata, with special reference to organogenesis during torsion. Philosophical Transactions of the Royal Society of London B 228:219-268, plates 21-27. Crofts, D. R. 1955. Muscle organogenesis in primitive gastropods and its relation to torsion. Proceedings of the Zoological Society of London 125:711-750. Demian, E. S. and F. Yousif. 1973a. Embryonic development and organo- genesis in the snail Marisa cornuarietis (Mesogastropoda: Ampullariidae). I. General outlines of development. Malacologia 12:123-150. Demian, E. S. and F. Yousif. 1973b. Embryonic development and organo- genesis in the snail Marisa cornuarietis (Mesogastropoda: Ampul- lariidae). IV. Development of the shell gland, mantle and respira- tory organs. Malacologia 12:195-211. Demian, E. S. and Yousif. 1975. Embryonic development and organo- genesis in the snail Marisa cornuarietis (Mesogastropoda: Ampul- lariidae). V. Development of the nervous system. Malacologia 15:29-42. Franzén, A. 1992. Spermatozoan ultrastructure and spermatogenesis in aplousobranch ascidians, with some phylogenetic considerations. Marine Biology 113:77-87. Fretter, V. 1988. New archaeogastropod limpets from hydrothermal vents, superfamily Lepetodrilacea. Part 2: Anatomy. Philosophical Transactions of the Royal Society of London B 319:33-82. Fretter, V. 1989. The anatomy of some new archaeogastropod limpets (Super-family Peltospiracea) from hydrothermal vents. Journal of Zoology, London 281:123-169. Fretter, V. 1990. The anatomy of some new archaeogastropod limpets (order Patellogastropoda, suborder Lepetopsina) from hydrothermal vents. Journal of Zoology, London 222:529-555. Fretter, V. and A. Graham. 1949. The structure and mode of life of th Pyramidellidae, parasitic opisthobranchs. Journal of the Marine Biological Association of the U.K. 28:493-532. Gauthier, J. 1986. Saurishian monophyly and the origin of birds. In: The Origin of Birds and the Evolution of Flight, Paidan, K. ed., pp. 1- 55. Memoires of the Californian Academy of Sciences Vol. 8. Golikov, A. N. and Y. I. Starobogatov. 1975. Systematics of prosobranch Gastropoda. Malacologia 15:185-232. Gosliner, T. M. 1991. Morphological parallelism in opisthobranch gas- tropods. Malacologia 32:313-327. 176 AMER. MALAC. BULL. 10(2) (1993) Hadfield, M. G. and M. F. Strathmann. 1990. Heterostrophic shells and pelagic development in trochoideans: Implications for classifica- tion, phylogeny and palaeoecology. Journal of Molluscan Studies 56:239-256. Haszprunar, G. 1985a. The fine morphology of the osphradial sense organs of the Mollusca. I. Gastropoda — Prosobranchia. Philo- sophical Transactions of the Royal Society of London B 307:457- 496, plates 1-5. Haszprunar, G. 1985b. The Heterobranchia - a new concept for the phy- logeny of the higher Gastropoda. Zeitschrift fiir zoologische Systematik und Evolutionsforschung 23:15-37. Haszprunar, G. 1985Sc. On the innervation of gastropod shell muscles. Journal of Molluscan Studies 51:309-314. Haszprunar, G. 1986. Die klado-evolutionare Klassifikation - Versuch einer Synthese. Zeitschrift fiir zoologische Systematik und Evolutionsforschung 24:89-109. Haszprunar, G. 1988a. Comparative anatomy of cocculiniform gastropods and its bearing on archaeogastropod systematics. In: Prosobranch Phylogeny, Ponder, W. F. ed., Malacological Review Supplement 4:64-84. Haszprunar, 1988b. On the origins and evolution of major gastropod groups, with special reference to the Streptoneura. Journal of Molluscan Studies 54:367-441. Haszprunar, G. 1989a. New slit limpets (Scissurellacea and Flssurellacea) from hydrothermal vents. Part 2. Anatomy and relationships. LACM Contributions in Science 408:1-17. Haszprunar, 1989b. The anatomy of Melanodrymia aurantiaca Hickman, a coiled archaeogastropod from the East Pacific hydrothermal vents (Mollusca, Gastropoda). Acta Zoologica (Stockholm) 70:175-186. Haszprunar, G. 1990. Systematik und Wahrscheinlichkeit - die klado- evolutionaére Synthese. Verhandlungen der deutschen Zoologischen Gesellschaft 83:506-507. Haszprunar, G. 1992a. The first molluscs - small animals. Bollettino di Zoologia 59:1-16. Haszprunar, G. 1992b. Ultrastructure of the osphradium of the Tertiary relict snail, Campanile symbolicum Iredale (Mollusca, Streptoneura). Philosophical Transactions of the Royal Society of London B 337:457-469. Haszprunar, G. 1992c. On the anatomy and relationships of the Chor- istellidae (Archaeogastropoda: Lepetelloidea). Veliger 35:295-307. Haszprunar, G. and G. Huber. 1990. On the central nervous system of Smeagolidae and Rhdodopidae, two families questionable allied with the Gymnomorpha (Gastropoda: Euthyneura). Journal of Zoology, London 220:185-199. Healy, J. M. 1988. Sperm morphology and its systematic importance in the Gastropoda. In: Prosobranch Phylogeny, Ponder, W. F. ed., Malacological Review Supplement 4:251-266. Healy, J. M. 1990. Euspermatozoa and paraspermatozoa in the trochoid gastropod Zalipais laseroni (Trochoidea: Skeneidae): Marine Biology 105:497-507. Healy, J. M. 1993a. Transfer of the gastropod family Plesiotrochidae to the Campaniloidea based on sperm ultrastructural evidence. Journal of Molluscan Studies 59:135-146. Healy, J. M. 1993b. Comparative sperm ultrastructure and spermiogenesis in basal heterobranch gastropods (Valvatoidea, Architectonicoidea, Rissoelloidea, Omalogyroidea, Pyramidelloidea) (Mollusca). Zoologica Scripta (in press). Heath, H. 1899. The development of Ischnochiton. Zoologische Jahrbiicher, Abteilung Anatomie 12:567-656, plates 31-35. Heimer, L. and Zaborsky, L. 1989. (eds.) Neuroanatomical Tract-Tracing Methods 2 (ISBN 0-306-43165-3). Plenum Publishing Corporation, New York. Henchman, A. 1890. The origin and development of the central nervous system of Limax maximus. Bulletin of the Museum of Comparative Zoology (Harvard) 20:169-208, plates 1-10. Hennig, W. 1966. Phylogenetic Systematics. University of Illinois Press, Urbana. 263 pp. Hickman, C. S. 1983. Radular patterns, systematics, diversity, and ecology of deep-sea limpets. Veliger 23:189-194. Hickman, C. S. 1984. A new archaeogastropod (Rhipidoglossa, Troch- acea) from hydrothermal vents on the East Pacific Rise. Zoologica Scripta 13:19-25. Hickman, C. S. 1988. Archaeogastropod evolution, phylogeny and sys- tematics: a reevaluation. In: Prosobranch Phylogeny, Ponder, W. F. ed. Malacological Review Supplement 4:17-34. Hickman, C. S. 1992. Reproduction and development of trochacean gas- tropods. Veliger 35:245-272. Hickman, C. S. and J. H. McLean. 1990. Systematic Revision and Suprageneric Classification of Trochacean Gastropods. Science Series no. 35, Natural History Museum of Los Angeles County (ISSN 0079-0943), Los Angeles, California. 169 pp. Hoffman, H. 1932-39. Opisthobranchia. In: Bronn's Klassen und Ordnungen des Tierreiches. Vol. 3: Akademische Verlagsgesell- schaft, Leipzig: 1-1247. Honegger, T. 1974. The embryonic development of Ampullarius (Gastropoda, Prosobranchia). Zoologische Jahrbiicher, Abteilung Anatomie 93:1-76. Houbrick, R. S. 1987. Anatomy of Alaba and Litiopa (Prosobranchia: Litiopidae). Systematic implications. Nautilus 101:9-18. Jacob, M. H. 1984. Neurogenesis in Aplysia californica resembles nervous system formation in vertebrates. Journal of Neuroscience 4:1225- 1239. Jamieson, B. G. M. 1991. Fish Evolution and Systematics. Evidence from Spermatozoa. Cambridge University Press, Cambridge. 319 pp. Kandel, E. R., A. Kriegstein and S. Schacher. 1980. Development of the central nervous system in terms of the differentiation of its specific identificable cells. Neuroscience 5:2033-2063. Lehmann, N. 1992. Strategies of shell formation in ampullariid ontogene- sis. Proceedings of the 10th International Malacological Congress. Tiibingen 1989 (ISSN 1018-9092):517-519. Lindberg, D. R. 1988. The Patellogastropoda. In: Prosobranch Phylogeny, Ponder, W. F., ed. Malacological Review Supplement, 4:35-63. Marshall, B. A. 1989. Recent and tertiary Cocculinidae and Pseudo- cocculinidae (Mollusca, Gastropoda) from New Zealand and New South Wales. New Zealand Journal of Zoology 12:505-546. Marshall, B. A. 1988. Skeneidae, Vitrinellidae and Orbitestellidae (Mollusca, Gastropoda) associated with biogenic substrata from bathyal depth off New Zealand and New South Wales. Journal of Natural History 22:949-1004. McLean, J. H. 1981. The Galapagos Rift limpet Neomphalus: Relevance to understanding the evolution of a major Paleozoic-Mesozoic radi- ation. Malacologia 21:291-336. McLean, J. H. 1988. New archaeogastropod limpets from hydrothermal vents, superfamily Lepetodrilacea. Part 1: Systematic descriptions. Philosophical Transactions of the Royal Society of London B 319:1-32. McLean, J. H. 1989a. New archaeogastropod limpets from hydrothermal vents, new family Peltospiridae, new superfamily Peltospiracea. Zoologica Scripta 18:49-66. McLean, J. H. 1989b. New slit limpets from hydrothermal vents: Family Scissurellidae (Scissurellacea) and new family Clypeosectidae (Fissurrellacea). Part 1. Systematic descriptions and comparisons based on shell and radular characters. Los Angeles County Museum Contributions in Science 407:1-29. McJean, J. H. 1990a. A new genus and species of neomphalid limpet from the Mariana Vents, with a review of current understanding of rela- HASZPRUNAR: ARCHAEOGASTROPODA 177 tionships among Neomphalacea and Peltospiracea. Nautilus 104:77- 86. McLean, J. H. 1990b. Neolepetopsidae, a new docoglossate limpte family from hydrothermal vents and its relevance to patellogastropod evo- lution. Journal of Zoology, London 222:485-528. McLean, J. H. 1992a. A new species of Pseudorimula (Fissurellacea, Clypeosectidae) from hydrothermal vents of the Mid-Atlantic Ridge. Nautilus 106:115-118. McLean, J. H. 1992b. Systematic review of the family Choristellidae (Prosobranchia: Lepetellacea), with descriptions of new species. Veliger 35:273-294. Moore, J. E. S. 1899. The molluscs of the Great African Lakes. 4. Nassopsis and Bythoceras. Quarterly Journal of Microscopical Sciences 42:187-202. Page, L. R. 1992a. New intepretation of a nudibranch central nervous sys- tem based on ultrastructural analysis of neurodevelopment in Melibe leonina. 1. Cerebral and visceral loop ganglia. Biological Bulletin 182:348-365. Page, L. R. 1992b. New interpretation of a nudibranch central nervous system based on ultrastructural analysis of neurodevelopment in Melibe leonina. II. Pedal, pleural and labial ganglia. Biological Bulletin 182:366-381. Ponder, W. F. 1991. Marine valvatoideans - implications for early hetero- branch evolution. Journal of Molluscan Studies 57:21-32. Ponder, W. F. and D. L. Lindberg. 1992. Gastropod phylogeny - a cladis- tic evaluation. Abstracts of the 58th Annual Meeting of the American Malacological Union, Sarasota 1992:35. Regondaud, J., P. Brisson and M. de Larambergue. 1974. Considérations sur la morphogénése et l'évolution de la commissure viscérale chez les Gastéropodes Pulmonés. Haliotis 4:49-60. Riese, K. 1930. Phylogenetische Betrachtungen tiber das Nervensystem von Cypraea moneta auf Grund seiner Morhologie und Histologie. Jenaische Zeitschrift fiir Naturwissenschaften 65:36 1-486. Robertson, R. 1985. Four characters and the higher category systematics of gastropods. American Malacological Bulletin, Special Edition 1:1-22. Ruthensteiner, B. 1991. Beitrage zur Entwicklung der Ellobiidae (Pul- monata, Gastropoda). Doctoral Thesis, Institut fiir Zoologie der Universitat Wien, Austria, 218 pp. Ruthensteiner, B. 1992. Development of Ovatella (Myosotella) myosotis (Draparnaud) (Pulmonata, Ellobiidae). Proceedings of the 10th International Malacological Congress, Tiibingen 1989 (ISSN 1018- 9092):45-46. Salvini-Plawen, L. v. and G. Haszprunar. 1987. The Vetigastropoda and the systematics of streptoneurous gastropods (Mollusca). Journal of Zoology, Lodnon A 211:747-770. Schmekel, L. 1985. Aspects of evolution within the Opisthobranchia. Jn: The Mollusca. Vol. 10: Evolution, E. R. Trueman and M. R. Clarke, eds., pp. 221-267. Academic Press, London. Shilenko, A. A. 1977. The symmetry of the Docoglossa and the problem of the origin of the order. Bulletin of the Naturalist's Society of Moskow, Biological Series 81:60-65 (in Russian) Veliger 21:494 (English summary). Smith, F. G. 1935. The development of Patella vulgata. Philosophical Transactions of the Royal Society of London B 225:95-125. Tardy, J. 1970. Contribution a I' étude des métamorphoses chez les Nudi- branches. Annalés des Sciences naturelles, Zoologie 12:299-370. Tardy, J. 1974. Morphogénése du systéme nerveux chez les Mollusques Nudibranches. Haliotis 4:61-75. Thiele, J. 1925. Gastropoda. In: Handbuch der Zoologie, Kiikenthal, W. ed. Vol. 5(1,2):38-155. De Gryter, Berlin. Thiele, J. 1929. Handbuch der Systematischen Weichtierkunde. Teil 1: Loricata/Gastropoda 1 (Prosobranchier). Gustav Fischer Verlag, Jena. Thompson, T. E. 1958. The natural history, embryology, larval biology, and post-larval development of Adalaria proxima (Alder and Hancock) (Gastropoda, Opisthobranchia). Philosophical Trans- actions of the Royal Society of London B 242:1-58. Thompson, T. E. 1962. Studies on the ontogeny of Tritonia hombergi Cuvier (Gastropoda, Opisthobranchia). Philosophical Transactions of the Royal Society of London B 245: 172-218. Thompson, W. F. 1912. The protochonch of Acmaea. Proceedings of the Academy of Natural Sciences of the U.S.A. 64:540-544. Van den Biggelaar, J. A. M. 1993. Cleavage pattern in embryos of Haliotis tuberculata (Gastropoda, Prosobranchia) and gastropod phylogeny. Journal of Morphology 216:121-128. Warén, A. 1988. Neopilina goesi, a new Caribbean monoplacophoran mollusk dredged in 1869. Proceedings of the Biological Society of Washington 101:676-681. Warén, A. 1989. New and little known Mollusca from Iceland. Sarsia 74:1-28. Warén, A. 1991. New and little known Mollusca from Iceland and Scandinavia. Sarsia 76:53-124. Warén, A. 1992. New and little known "skeneimorph" gastropods from the Mediterranean Sea and the adjacent Atlantic Ocean. Bollettino di Malacologico 27:149-248. Warén, A. and P. Bouchet. 1989. New gastropods from East Pacific hydrothermal sites. Zoologica Scripta 18:67-102. Warén, A. and P. Bouchet. 1993. New records, species, genera, and a new family of gastropods from hydrothermal vents and hydrocarbon seeps. Zoologica Scripta 22:1-90. Warén, A., S. Gofas and C. Schander. 1993. Systematic position of three European heterobranch gastropods. Veliger 36:1-15. Wiley, E. O. 1981. Phylogenetics. The Theory and Practice of Phylo- genetic Systematics. John Wiley and Sons, New York. 439 pp. Date of manuscript acceptance: 26 May 1993 SYMPOSIUM ON BIOLOGY OF CARIBBEAN MOLLUSKS Organized by RUDIGER BIELER FIELD MUSEUM OF NATURAL HISTORY AMERICAN MALACOLOGICAL UNION SARASOTA, FLORIDA 3 - 4 AUGUST 1992 179 i Giving and receiving: the tropical Atlantic as donor and recipient region for invading species Geerat J. Vermeij' and Gary Rosenberg’ ‘Department of Geology, University of California, Davis, California 95616, U.S.A. Philadelphia Academy of Natural Sciences, Philadelphia, Pennsylvania 19103, U.S.A. Abstract: After the middle Pliocene uplift of the Central American seaway (3.1 to 3.6 million years ago), the western Atlantic fauna became isolated from that of the eastern Pacific, but connections with the tropical Indo-West-Pacific and eastern Atlantic were maintained. By analyzing the distibution, fos- sil record, and relationships of shallow-water shell-bearing molluscs (those living in less than 100 m depth) in the western Atlantic, we ascertained the extent to which the western Atlantic has served as a recipient and as a donor region for invading taxa. At least 33 species in the western Atlantic are late Pliocene or Pleistocene invaders from the Indo-West-Pacific (17 species) or eastern Atlantic (16 species), whereas at least 39 species dispersed eastward across the Atlantic from the Americas to West Africa. Eleven species derived from the Indo-West- Pacific are included in both tallies, because they probably first dispersed westward from the Indian Ocean around southern Africa to Brazil and the Caribbean region before spreading eastward across the Atlantic to West Africa. Most of this dispersal is probably by means of planktonic larvae, but some species could have been spread as rafting adults. Oceanic currents and prior extinction histories determine the pattern of interchange among tropical marine biotas. Within the tropics, the western Atlantic suffered the greatest molluscan extinctions since the early Pliocene (about 60 to 70%); it is also the region in which the great majority of immigrants have become common and geographically widespread. Extinction in the eastern Atlantic, eastern Pacific, and Indo-West-Pacific has been much less, and immigrants to these regions often have restricted geographical distribution there, and could be represented by populations that are not self-sustaining. Biogeographers working with living species have aries should be drawn, we can ask to what extent regions divided the biosphere into a number of provinces, each of have served as sources for invading species, as refuges for which is characterized by species found only in that region. species that have contracted their ranges, and as places suit- The often unstated assumption is that these provinces have able for the successful establishment of species that origi- a certain continuity and cohesiveness, and that therefore nated elsewhere. they can be treated as biologically meaningful units above Answers to these questions require a fossil record the species level. A consequence of this interpretation is that is biogeographically and statigraphically sufficient to that biogeography has been pervaded by attempts to define trace the histories of many lineages. The Neogene record of boundaries between provinces and to establish criteria for shell-bearing molluscs, though necessarily incomplete and recognizing a region as biogeographically distinct. insufficiently known in some regions, is, nonetheless, ideal As we learn more about the phylogenetic relation- for this purpose. Coupled with cladistic and molecular stud- ships and fossil record of species, the idea of provincial ies, analyses of this record could reveal much about the purity and integrity is giving way to the realization that dynamics of species distributions and about the integrity of many geographical regions support species whose geo- biogeographical provinces. graphical origins and histories vary widely. Some evolved In this paper we devote particular attention to bio- in the region they presently occupy, whereas others arose or geographical connections between the tropical western evolved elsewhere. There is abundant evidence that the dis- Atlantic and other tropical marine regions. Before mid- tributional limits of species change over the course of time. Pliocene time (about 3.1 to 3.6 million years ago), the For example, if a species is initiated as a small geographical Central American seaway connected the western Atlantic isolate, it often expands its range subsequently. Other with the eastern Pacific, so that many species were held in species undergo range contractions as local populations common between these two regions. The connection was become extinct. These considerations lead to a more interrupted by the mid-Pliocene uplift of the Central dynamic view of provinces, one that admits extinctions and American isthmus. Connections between the western invasions of species. Changes in the distributional limits of Atlantic and the two other tropical marine regions, the species affect the distinctiveness as well as the boundaries Indo-West-Pacific and eastern Atlantic, were maintained, of provinces. Instead of ascertaining where these bound- however. Dispersal of planktotrophic larvae and rafted American Malacological Bulletin, Vol. 10(2) (1993):181-194 181 182 AMER. MALAC. BULL. 10(2) (1993) adults westward or eastward across the Atlantic maintains a faunal connection between the eastern and western Atlantic (Scheltema, 1971). The link with the Indo-West-Pacific is by way of dispersal around South Africa from the south- western Indian Ocean to the central and western Atlantic. By analyzing the distribution, history, and relation- ships of western Atlantic shell-bearing molluscs, we ask two related questions in this paper. First, to what extent and by which molluscan lineages has the western Atlantic been invaded from elsewhere in the tropics during the last three million years? Second, to what extent has the western Atlantic contributed invaders to other tropical regions dur- ing the last three million years, and which lineages have taken part in this invasion? METHODS We have compiled data on the geographical distri- bution, fossil record, and relationships of shell-bearing mol- luscs living in the tropical western Atlantic, an area we define as extending from Cape Hatteras, North Carolina, USA, to the vicinity of Rio de Janeiro, Brazil. Our aim was to identify species that invaded either westward or eastward across the Atlantic following the mid-Pliocene uplift of the Central American isthmus. To that end, we compiled a list of all amphi-Atlantic species (those living on both sides of the Atlantic), based largely on the primary taxonomic and systematic literature. We have also relied heavily on the faunistic accounts of von Cosel (1982), for the Cape Verde Islands, and Leal (1991), on the fauna of islands and sea mounts off Brazil. Some of the purported cases of amphi- Atlantic distributions are not based on a critical comparison of eastern and western Atlantic specimens. Some of the cases we report could therefore turn out to represent pairs of closely related eastern and western Atlantic species. In his compilation of amphi-Atlantic species, Talavera (1982) included many species whose distribution on both sides of the Atlantic is clearly in error or not well documented. We have therefore chosen not to include Talavera's amphi- Atlantic records unless other sources corroborate such a distribution. Wood-boring pholadid and teredinid bivalves were also excluded. Species were inferred to have invaded during the last three million years to the western Atlantic from else- where in the tropics if (1) they or their likely ancestors are unknown as fossils in the western Atlantic before mid- Pliocene time, (2) they or their ancestors have a fossil record earlier than Pliocene elsewhere in the tropics, and (3) they have no close relatives (ancestral or sister species) in the tropical eastern Pacific. The rationale for the first and second criteria is obvious, but the third criterion requires brief explanation. If a western Atlantic species has a close fossil or living relative in the eastern Pacific, we can infer that the species or its ancestor was already present in tropi- cal America during or before mid-Pliocene time, when the Central American seaway was still open. The vast majority of western Atlantic species with close eastern Pacific rela- tives indeed have fossil records extending to early Pliocene and often Miocene time. We have excluded from this analysis all species that live only in waters more than 100 m deep. As a rule, deep- water taxa are not well represented in the fossil record, so that direct historical evidence from the fossil record is often lacking. Moreover, an amphi-Atlantic distribution could reflect a continuous benthic population of adults rather than a link between eastern and western populations maintained by overwater dispersal. Potential invaders to the western Atlantic were divided into two groups. The first is composed of species that probably came from the Indo-West-Pacific around southern Africa at a time when the coastal waters of south Africa were warmer than today. Species belonging to groups with a long fossil record in the Indo-West-Pacific but not the eastern Atlantic were assigned to this group of invaders. The second group is composed of amphi-Atlantic species whose fossil record in the eastern Atlantic extends back beyond mid-Pliocene time. Amphi-Atlantic species found only on offshore islands in either the eastern Atlantic (Cape Verde and Canary Islands, for example) or the west- ern Atlantic (Lesser Antilles or Fernando de Noronha, for example) were distinguished from those that occur on the coasts of the continents. The western Atlantic could have contributed species to other tropical regions since the mid-Pliocene. Potential invaders from the western to the eastern Atlantic were iden- tified in cases where the eastern Atlantic fossil record is no older than mid-Pliocene. Many amphi-Atlantic species do not fall into either group of invaders. Some species will have achieved their amphi-Atlantic distribution before the mid-Pliocene. Others are not well understood taxonomically, biogeographically, or paleontologically. We have compiled a list of all shal- low-water shell-bearing amphi-Atlantic species but we have not attempted to trace their history of invasion. INVADERS FROM THE INDO-WEST-PACIFIC At least 20 western Atlantic gastropods are inferred to have arrived from the Indo-West-Pacific region since mid-Pliocene time (Appendix 1). Of these, 11 species (indi- cated by an asterisk) are endemic to the Atlantic but have their closest relatives in the Indo-West-Pacific. For exam- VERMEIJ AND ROSENBERG: DISPERSAL ACROSS THE TROPICAL ATLANTIC 183 ple, the trochid Synaptocochlea picta of the western Atlantic is the only member of its genus in the Atlantic, and is very similar to S. concinna (Gould, 1845), known from the Pleistocene and Recent of the Indo-West-Pacific (see Kay and Johnson, 1987; Leal, 1991). The cassid Casmaria atlantica is the only Atlantic species of its genus. It is extremely similar to the Indo-West-Pacific C. ponderosa (Gmelin, 1791), especially to the Indian Ocean form some- times given the name C. cernica (Sowerby, 1888) (see Ab- bott, 1968). The only eastern Pacific species of Casmaria, C. vibexmexicanum (Stearns, 1894), is in the C. erinacea (Linnaeus, 1758) species complex and is not close to either C. atlantica or C. ponderosa (Abbott, 1968). The genus Casmaria has a Pleistocene fossil record in the western Pacific and Indian Oceans but is not known as a fossil in the Atlantic. Gibson-Smith and Gibson-Smith (1981) tenta- tively suggested that C. atlantica arose via Phalium (Tylocassis) cicatricosum (Gmelin, 1791) from the New World P. granulatum line, which appeared during the early Miocene. They pointed to similarities in protoconch charac- ters and in the lack of sculpture on the later teleoconch whorls. This scenario requires that the outer-lip prickles characteristic of Casmaria evolved independently in the Indo-West-Pacific member of this genus and in C. atlanti- ca. We reject this scenario in favor of the simpler idea that C. atlantica is derived from an Indian Ocean form of C. ponderosa, which entered the Atlantic during relatively recent time. The tonnid Eudolium crosseanum of the Atlantic is close to the Indo-West-Pacific E. pyriforme Sowerby, 1914 (see Marshall, 1992). The genus Eudolium is known from the Oligocene onward in the eastern Atlantic and from the early and late Miocene of the eastern United States, but E. crosseanum and the Atlantic and Indo-West- Pacific E. bairdii may not have descended from these earli- er forms and are themselves not known as Atlantic fossils. Malea (Quimalea) noronhensis, known only from islands off Brazil (Leal, 1991), is very similar to the widespread Indo-West-Pacific Malea pyrum (Linnaeus, 1758). It is the only New World representative of the subgenus Quimalea. All other fossil and living species of Malea in the Americas belong to the nominate subgenus (see Petuch, 1989). The genus Jonna is unknown as a fossil in the Americas. T. galea is known living in both the Atlantic and the Indo- West-Pacific region, but 7. maculosa is endemic to the Atlantic, being closely similar to the Indo-West-Pacific T. perdix (Linnaeus, 1758). The ranellid Cymatium (Septa) occidentale is the only Atlantic representative of the C. rubeculum (Linnaeus, 1758) species complex of the Indo- West-Pacific (Beu, 1986). C. (Ranularia) ridleyi of the western Atlantic is very similar to C. sarcostomum (Reeve, 1844) from the Indo-West-Pacific. Among the Bursidae, the western Atlantic Bursa natalensis is very close to, and sometimes synonymized with, B. latitudo Garrard, 1961 (for a discussion see Leal, 1991). Bursa thomae is extreme- ly similar to the Indo-West-Pacific B. rhodostoma (Sowerby, 1871). A fossil possibly related to B. thomae was described by Jung (1969) from the Pliocene Melajo Clay of Trinidad. Bufonaria (Marsupina) bufo is the only member of its genus in the Atlantic; all other members are Indo- West-Pacific in distribution. Because B. bufo differs at the subgeneric level from other members of Bufonaria, the pos- sibility exists that this species has been in the Atlantic since before mid-Pliocene time. However, it is unknown as a fos- sil and has no close relatives in the eastern Pacific fauna. Finally, the architectonicid Psilaxis krebsii of the Atlantic is extremely close to the Indo-West-Pacific P. oxytropis (A. Adams, 1855). It is known fossil from the late Pliocene or early Pleistocene Bowden Formation of Jamaica (as the subspecies lampra Woodring, 1928) and may therefore have arrived from the Indo-West-Pacific as early as the late Pliocene (see Robertson, 1973). Besides Bursa thomae and Psilaxis krebsii, only one other species inferred to have arrived in the Atlantic from the Indo-West-Pacific after uplift of the Central American isthmus has a fossil record in the New World. Robinson (1990) reported Gyrineum louisae from the early Pleisto- cene Moin Formation of Costa Rica. It is possible that several Atlantic taxa with eastern Pacific counterparts also arrived in the Atlantic directly from the Indo-West-Pacific, and that they are therefore not derived from ancestors living in the New World before the closure of the Central American isthmus. Examples include the ranellids Cymatium aquatile, C. cynocephalum, C. labiosum, C. mundum, C. muricinum, C. nicobaricum, C. parthenopeum, Charonia tritonis variegata, Linaetella cau- data (see e.g. Beu and Cernohorsky, 1986; Beu and Knud- sen, 1987; Beu and Kay, 1988). The eastern Pacific repre- sentatives probably arrived in the New World from the west after uplift of the isthmus, along with numerous other species, and independently of the invasion of the same stocks to the tropical Atlantic (see Emerson, 1991). Which scenario is correct cannot be resolved with the presently available evidence, but for now we have chosen to be con- servative by not including the above species as immigrants to the Atlantic directly from the Indo-West-Pacific. The inference of invasion from the Indo-West- Pacific to the Atlantic is based in part on negative evidence, namely, the absence of a pre-Isthmian New World fossil record. Such evidence is, of course, always suspect. It is always possible that some of these species were already present in the Atlantic at the same time the Central Amer- ican seaway closed. This could apply to Bursa (Lampadop- sis) grayana (Dunker, 1862) (= B. pacamoni Matthews and Coelho, 1971), a species today known only from Brazil. 184 AMER. MALAC. BULL. 10(2) (1993) This bursid is very close to B. davidboschi Beu, 1986, from the Indian Ocean, and could have been descended from an Indo-West-Pacific population. Beu (1986), how- ever, has recorded B. grayana from the early Pliocene or late Miocene Gurabo Formation of the Dominican Republic. If the species did arrive from the Indo-West- Pacific region, it could have done so well before the isth- mus in Central America was uplifted and without leaving eastern Pacific relatives. It is possible that other Atlantic species whose affinities lie with Indo-West-Pacific species will also be recovered as pre-isthmian fossils in the New World. In the absence of such evidence, however, and given that there are no close relatives in the eastern Pacific, we prefer the interpretation that these Atlantic species are post- isthmian invaders from the Indo-West-Pacific. We have not included the ranellid Linatella suc- cincta in the list of invaders to the Atlantic from the Indo- West-Pacific. The species is apparently unknown as a fos- sil, and is closely related or identical to forms from the eastern Pacific and Indo-West-Pacific (Emerson, 1991). Within the Atlantic, the species occurs on the African coast only in Gabon (Beu and Cernohorsky, 1986). It is possible that L. succincta is an invader from the Indo-West-Pacific, but the evidence in favor of this conclusion is not persuasive. Of the 20 invaders from the Indo-West-Pacific to the Atlantic (Appendix 1), 11 (55%) occur on both sides of the Atlantic (Appendix 2). Six of these 11 species, however, have extremely limited distributions in the eastern Atlantic. Bursa ranelloides, B. thomae, Cymatium occidentale, and Tonna maculosa occur in the eastern Atlantic only on off- shore islands (Appendix 2), whereas Eudolium bairdii and E. crosseanum are known from only a few Mediterranean records (Marshall, 1992). This implies that these species or their ancestors came directly from the Indo-West-Pacific to the western Atlantic and subsequently dispersed to the east- ern Atlantic. Further support for this interpretation comes from the fact that six of the Indo-West-Pacific immigrants or their descendants are known in the Atlantic only from the American side, and that no immigrants are limited in the Atlantic to the African side. We believe that that princi- pal dispersal route was from the southwestern Indian Ocean around southern Africa to St. Helena and thence northward and westward to Brazil and the Caribbean Region. Such dispersal is likely only at times when southern Africa was warmer than it is today. St. Helena's marine fauna of mol- luscs and crustaceans is well known for the presence of taxa known otherwise only from southern Africa or the Indo-West-Pacific (E. A. Smith, 1890; Chace, 1966). DISPERSAL ACROSS THE ATLANTIC We recognize 123 shell-bearing molluscs as occur- ring in shallow waters (less than 100 m deep) on both sides of the tropical and subtropical Atlantic (Appendix 2). Of these, 108 are gastropods, 15 are bivalves, and one is a scaphopod. Eleven of the amphi-Atlantic species (all gas- tropods) are inferred to have arrived from the Indo-West- Pacific after the Central American seaway was closed; they were dealt with in the preceding section. Most of the re- maining 113 amphi-Atlantic species or their ancestors already existed in the Atlantic basin when the Central American seaway was still open. Some of these species could already have had an amphi-Atlantic distribution at that time. At least 43 species (37 gastropods, 6 bivalves) are inferred here to have dispersed either eastward or westward across the Atlantic following the mid-Pliocene uplift of the Central American isthmus (Tables 1-3). Of these 44 species, 28 dispersed eastward from the Americas to the coasts of Europe and West Africa, whereas 15 dispersed westward. These numbers exclude the previously discussed immigrants from the Indo-West-Pacific, which are inferred to have dispersed first westward across the south Atlantic and then secondarily eastward again to the eastern Atlantic. A majority (22 species, 79%) of the 28 taxa inferred to have dispersed eastward from the Americas to the African side of the Atlantic have very limited distributions in the eastern Atlantic. For example, the fissurellid keyhole limpet Diodora cayenensis is known in the eastern Atlantic only from the Canary Islands, whereas on the American side it is widespread (Leal, 1991). As a fossil, this species is known not only from the western Atlantic, but also from the eastern Pacific, where Schremp (1981) recorded it from the early Pliocene Imperial Formation of Southern California. The small littorinid Nodilittorina (Fossarilittorina) meleagris is known in the eastern Atlantic only from Ghana (Rosewater and Vermeij, 1972; Rosewater, 1981), but in the western Atlantic it has a wide distribution. The species is unknown as a fossil, but all other species of the subgenus Fossarilittorina occur in the western Atlantic and eastern Pacific (Rosewater, 1981; Reid, 1989). The calyptraeid Crepidula (Bostrycapulus) aculea- tus, which is known in the eastern Atlantic only from the Cape Verde Islands (von Cosel, 1982), lives in both the western Atlantic and eastern Pacific. It has a fossil record in eastern North America extending back to the middle Miocene (Hoagland, 1977). The helmet shell Cassis tuberosa (Cassidae) occurs in the Cape Verde Islands and nowhere else in the eastern Atlantic (von Cosel, 1982). The lineage of C. tuberosa was present during the early Miocene in Florida, and was repre- sented in the eastern Pacific during the early Pliocene (Vokes, 1990). The cassid Phalium (Tylocassis) granulatum is also insular in the eastern Atlantic, but in the Americas VERMEIJ AND ROSENBERG: DISPERSAL ACROSS THE TROPICAL ATLANTIC 185 the subgenus Tylocassis occurs in the late Miocene Gatun Formation of Panama and is represented by living species in both the western Atlantic and eastern Pacific (Abbott, 1968). Exactly the same applies to the bursids Bursa corru- gata and B. granularis, the buccinid Colubraria testacea, and the bivalves Arca imbricata (Arcidae), Arcopsis adamsi (Noetiidae), Dendostrea frons (Ostreidae), and Pseudo- chama radians (Chamidae) (see Woodring, 1964, 1973; Beu, 1980; von Cosel, 1982; Leal, 1991). A number of amphi-Atlantic species confined in the eastern Atlantic to offshore islands perhaps also have dis- persed eastward from the Americas, but the evidence from the fossil record and systematic relationships is weaker. Species in this category include the hipponicids Hipponix antiquatus and H. subrufus, the naticids Natica livida and Polinices lacteus, and the ranellids Cymatium aquatile, C. labiosum, and C. muricinum. Most of these species are closely related or identical to Indo-West-Pacific as well as to eastern Pacific species, so that we cannot be sure when or in what way they achieved an amphi-Atlantic distribu- tion. We have included them among eastward dispersers (Table 1) because there is no fossil record of these species in the eastern Atlantic. In the case of C. labiosum, there are late Pliocene or early Pleistocene records in the western Atlantic from the Bowden Formation of Jamaica and the Moin Formation of Costa Rica (Beu and Knudsen, 1987). C. aquatile is known from the Pliocene of Ecuador in the eastern Pacific (Beu and Kay, 1988). TABLE 1. Eastward-dispersing species in the tropical Atlantic. Diodora cayenensis Lucapinella limatula Cheilea equestris Crepidula aculeata Hipponix antiquatus H. subrufus Nodilittorina meleagris Natica livida Polinices lacteus Bursa corrugata B. granularis Cymatium aquatile C. labiosum C. muricinum Linatella caudata Cassis tuberosa Phalium granulatum Thaisella coronata Trachypollia nodulosa T. turricula Colubraria testacea Coralliophila caribaea Arca imbricata Arcopsis adamsi Nodipecten nodosus Dendostrea frons Pseudochama radians Papyridea soleniformis Only six eastward-dispersing species have broad ranges in the eastern Atlantic. Cheilea equestris (Calyptrae- idae), the muricids Thaisella coronata, Trachypollia nodu- losa, and T. turricula, and the bivalves Nodipecten nodosus (Pectinidae) and Papyridea soleniformis (Cardiidae) are widespread in West Africa as well as in the western Atlantic, and all have counterparts in the eastern Pacific. Fossils referred to Cheilea equestris by Vokes (1975) from the early Miocene Chipola Formation of Florida indicate that this species has been in the western Atlantic for a very long time. It could have been derived from a species of the Eocene of Europe (Vokes, 1975). The genus Cheilea was present in Europe from Eocene to late Miocene (Tortonian) time, but the species C. equestris probably represents a secondary invasion of the eastern Atlantic from the west. Species of Thaisella are known in the Americas since at least middle Miocene time. 7: coronata is very sim- ilar or identical to T. trinitatensis (Guppy, 1869), which is known from the early Pliocene Cercado Formation of the Dominican Republic and still lives in the western Atlantic (Vokes, 1989). Typical T: coronata is also known from the western Atlantic, and has been in the eastern Atlantic since at least the Pleistocene (Rosso, 1974). The scallop Nodipecten nodosus is widely if patchi- ly distributed in the eastern Atlantic. In tropical America, N. nodosus can be traced back to the late Miocene (J. T. Smith,1991), but in the eastern Atlantic the species has been reported as a fossil only from the late Pleistocene of the Canary Islands (Meco Cabrera, 1982). The cardiid bivalve Papyridea soleniformis is wide- spread in the western Atlantic but has a very patchy distrib- ution in the eastern Atlantic, where it is known from the Cape Verde Islands and Angola (Voskuil and Onverwagt, 1989). There is a closely related eastern Pacific species, S. aspersa (Sowerby, 1833). Among westward-dispersing species (Table 2), only four (27% of 15 species) have or had very limited ranges in the western Atlantic. Cymatium kobelti has only been rec- ognized in the western Atlantic as a fossil from the Plio- cene of Colombia (Beu and Knudsen, 1987). The only western Atlantic occurrence of C. trigonum is off the coast of Venezuela (Finlay and Vink, 1982). Two other species, the bursids Bursa marginata and B. scrobiculator, are represented by Miocene to Recent populations in the east- ern Atlantic, but have been reported in the western Atlantic only as late Pliocene or early Pleistocene fossils in the Moin Formation of Costa Rica (D. G. Robinson, pers. comm.). Only one other purportedly amphi-Atlantic species has a markedly limited distribution in the western Atlantic. 186 AMER. MALAC. BULL. 10(2) (1993) TABLE 2. Westward-dispersing species in the tropical Atlantic. Littoraria angulifera Cerithium guinaicum Hinea lineata Charonia lampas Cymatium kobelti C. trigonum Bursa marginata B. scrobiculator Ranella olearia Epitonium albidum Pugilina morio Pseudomalaxis zanclaeus Spirolaxis centrifuga Mathilda barbadensis Siphonaria pectinata Vasum globulum Lamarck, 1816, is reported to occur in the Lesser Antilles as well as in West Africa (see Vokes, 1966, for a review). Faber (1988), however, has pointed out that the West African records of this species are erroneous, and that V. globulum is a western Atlantic species closely relat- ed to the southern Caribbean V. capitellus (Linnaeus, 1758). Two other western Atlantic species with a very lim- ited range are probably derived from eastern Atlantic species in relatively recent times. Thais nodosa meretricula Roding, 1798, known from islands off Brazil and from Ascension Island in the central south Atlantic, is distinct from but very closely related to the West African T. n. nodosa (Linnaeus, 1758) (see Rosewater, 1975; Leal, 1991). There is no similar species in tropical America, either liv- ing or fossil. The littorinids Nodilittorina vermeiji Bandel and Kadolsky, 1982, from islands off Brazil, and N. miliaris (Quoy and Gaimard, 1833) from Ascension Island, are extremely similar to the west African N. granosa (Philippi, 1848), but have no close relatives in the Americas (see Reid, 1989; Leal, 1991). As in the case of Vasum globulus, we have not included these two probable western Atlantic derivatives of eastern Atlantic taxa in our tally because of the absence of a fossil record. The other westward-dispersing species are widely distributed in the western Atlantic, but most are unknown as fossils and therefore appear to be relatively recent immi- grants to the American coasts. The melongenid Pugilina morio, which is found in mangrove environments from the southern Caribbean to Brazil, as well as in West Africa, is unknown as a western Atlantic fossil but is reported from the Pleistocene of West Africa (Rosso, 1974). Melongenids are abundantly represented as fossils in Miocene and Pliocene formations in the western Atlantic and eastern Pacific, but none of the species resembles P. morio and none can be referred to the genus Pugilina. The pulmonate limpet Siphonaria (Mouretus) pecti- nata is not closely related to any other western Atlantic siphonariid. Although the species has no fossil record, we interpret it as a westward disperser because it belongs to a subgenus that is otherwise known only from southern Africa and the northwestern Indian Ocean (see Hubendick, 1945, 1946). Morrison (1963) believed that S. pectinata was introduced by humans from West Africa, where it is abundant on rocky shores, to the western Atlantic. He pointed to the patchy distribution of the species in the Caribbean basin as evidence for this intepretation. Wave- exposed species such as S. pectinata are rarely susceptible to accidental human transport. Moreover, S. pectinata has a planktotrophic larva, which would under favorable circum- stances make overwater dispersal possible. We therefore believe that S. pectinata dispersed from West Africa to the Americas in relatively recent times without the aid of people. We have included the planaxid Hinea lineata among westward dispersers for two reasons. First, there is no fossil record of Hinea in tropical America, and no species of this genus occurs in the eastern Pacific (Houbrick, 1987, 1992). Second, although Hinea itself is not known as a fossil in the eastern Atlantic, there were middle Miocene species such as Planaxis (Dalliella) dautzenbergi Glibert, 1949, from the Helvetian (= Tortonian) of the Loire Basin in France, that have some features in common with Hinea (see Glibert, 1949). H. lineata is abundant and widespread on both sides of the tropical Atlantic. Another likely recent immigrant from the east is the mangrove-associated littorinid Littoraria (Littorinopsis) angulifera. The subgenus Littorinopsis is diverse and wide- spread in the Indo-West-Pacific, and has existed in the east- ern Atlantic since the early Eocene (Reid, 1986, 1989). It is unknown in the eastern Pacific, where the four extant man- grove periwinkles belong to the subgenera Littoraria s.s. and Bulimilittorina (see Reid, 1989). Woodring 1957) reported a small, almost completely decorticated, fossil from the early Miocene Culebra Formation of Panama as “Littorina sp. cf. L. angulifera”. One of us (GJV) has examined this specimen at the U. S. National Museum of Natural History in Washington. The shell has a lower spire than do most species of Littorinopsis, but there are no pre- served features that make distinction between Littorinopsis and Littoraria s.s. possible. The rich fossil record of the Pliocene of Florida contains several species of Littoraria (Petuch, 1991), but none is similar to L. angulifera and none belongs to the subgenus Littorinopsis. We therefore believe that L. angulifera arrived quite recently in the west- ern Atlantic. We cannot exclude the possibility that L. angulifera is a late Pliocene or Pleistocene derivative of the Indo-West-Pacific L. scabra (Linnaeus, 1758), and that the Atlantic L. angulifera followed the dispersal route of other VERMEIJ AND ROSENBERG: DISPERSAL ACROSS THE TROPICAL ATLANTIC 187 immigrants from the Indo-West-Pacific. Under this sce- nario, L. angulifera would have originated in the western Atlantic and subsequently spread to the eastern Atlantic. This would imply that Littorinopsis had become locally extinct in West Africa before the arrival of L. angulifera. We know nothing about the Pliocene and Pleistocene fossil record of Littorinopsis in the Old World, but the mangrove biota of West Africa is quite rich in species and does not appear to have been ravaged by extinction. We therefore favor the hypothesis that L. angulifera dispersed westward to the Americas from Africa, and consider less likely the more complex hypothesis that the species invaded the west- ern Atlantic from the Indo-West-Pacific and subsequently spread to West Africa. Three other westward-dispersing species, Ranella olearia (Ranellidae), Epitonium albidum (Epitoniidae), and Mathilda (Fimbriatella) barbadensis (Mathildidae), all belong to lineages with a history extending to the early Miocene or Oligocene in Europe (see Glibert, 1949; Beu, 1976, 1980; Janssen, 1978a, b). To our knowledge, they have no American fossil record, and are not close to any eastern Pacific species. Six of the 16 species inferred to have dispersed westward across the Atlantic are known as late Pliocene or Pleistocene fossils in the Americas. Three of these (Cy- matium kobelti, Bursa marginata, and B. scrobiculator) are not known living in the western Atlantic today, and were discussed above. Spirolaxis centrifuga (Architectonicidae) has been reported from the late Pliocene or early Pleistocene Bowden Formation of Jamaica (Bieler, 1984), and Cerithium guinaicum appears in the correlative Moin Formation of Costa Rica (Houbrick, 1974). The ranellid Charonia lampas occurs in the early Pleistocene Mare Formation of Venezuela (Beu, 1976). None of these species has close relatives in the eastern Pacific, and none can be traced back beyond the late Pliocene in the western Atlantic. The lineages of Bursa marinata, B. schrobicula- tor, Charonia lampas, and Spirolaxis centrifuga extend back to the Miocene in Europe (see Beu, 1970, 1980; Bieler, 1984). The phylogenetic relationships of Cerithium guinaicum have not been worked out, but this species is the only member of its genus in America without either close eastern Pacific relatives or clear ancestors from the Mio- cene or early Pliocene (see Houbrick, 1974). GENERAL DISCUSSION From the analysis presented in the last two sections, we infer that at least 35 shallow-water shell-bearing mol- luscan species in the western Atlantic are late Pliocene or later invaders from the Indo-West-Pacific or eastern TABLE 3. Summary of invasion of molluscs to and from the tropical Atlantic following uplift of the Central American isthmus. Category Number of Species Invasion to western Atlantic Total 35 Invaders from Indo-West-Pacific 20 Invaders from eastern Atlantic 15 Narrowly distributed invaders 6 Invasion to eastern Atlantic Total 39 Invaders from Indo-West-Pacific 11 From western Atlantic 28 Narrowly distributed invaders 22 Atlantic, and that at least 39 species dispersed eastward across the Atlantic during this same time interval (Table 3). Eleven species derived from the Indo-West-Pacific are included in both tallies, because they probably first dis- persed westward from the Indian Ocean to the Americas and subsequently spread eastward to the tropical eastern Atlantic. The number of westward dispersers would be increased by two if we included Nodilittorina vermeiji and Thais nodosa meretricula, which are probably western Atlantic derivatives of eastern Atlantic species. Invaders comprise a small proportion of the faunas of the eastern and western Atlantic. Von Cosel (1982) esti- mated that the tropical eastern Atlantic molluscan fauna contains about 2220 species. The 39 post-isthmian invaders to that region therefore account for only about 2% of the fauna. No comparable estimates are available for molluscan diversity in the western Atlantic. However, Rosenberg (1993) reports that there are more than 2900 gastropod species between Cape Hatteras and Rio de Janeiro. The 35 shallow-water invaders to the western Atlantic therefore account for only 1.2% of this fauna. Locally, the contribu- tion of invaders is higher, especially on oceanic islands. Thus, the molluscan fauna of the Cape Verde Islands com- prises about 400 species (von Cosel, 1982), of which 27 (6.8%) are invaders from the west by our count. Of the 121 prosobranch gastropod species at Fernando de Noronha off Brazil (Leal, 1991), seven (5.8%) are identified by us as invaders from the Indo-West-Pacific or eastern Atlantic. This tally does not include Nodilittorina vermeiji or Thais nodosa meretricula, both of which appear to be derived from eastern Atlantic taxa. Most of the dispersal across the Atlantic and from the Indo-West-Pacific is probably by planktonic larvae. Scheltema (1971) has documented the existence of plank- tonic larvae of many amphi-Atlantic species in the open ocean far from land, and has provided evidence that larvae remain competent to settle for many weeks or even months. It is possible, however, that several mangrove-associated 188 AMER. MALAC. BULL. 10(2) (1993) species dispersed by rafting as adults on wood. Littoraria angulifera, Thaisella coronata, and Pugilina morio are typ- ical of mangrove environments, and all can be found on logs as well as on living trees. Whether dispersal is by planktonic larvae or by raft- ing adults, the routes of dispersal reflect the pattern of oceanic circulation in the Atlantic. In the southern Atlantic, the strongest trans-oceanic currents flow westward (Leal, 1991), whereas in the North Atlantic the strong Gulf Stream flows eastward (see also Scheltema, 1971). The fact that several invaders to the western Atlantic from the Indo- West-Pacific and eastern Atlantic are known mainly from South and Central America suggests to us that westward dispersal of these species was largely across the south Atlantic. Similarly, many eastward-dispersing species are restricted in the eastern Atlantic to the Cape Verde Islands, Canary Islands, or Mediterranean area. This, we believe, implies that eastward dispersal is chiefly a North Atlantic phenomenon. Ocean currents provide the means for dispersal, but they are not alone in determining the pattern of invasion of species. Propagules can be brought to a potential recipient region, but if they fail to settle, thrive, and reproduce, the invasion is ephemeral. Conditions in the recipient biota must therefore be understood if we are to explain observed patterns of interchange among biotas. The available evidence indicates that the tropical western Atlantic received approximately the same number of shallow-water immigrants (35 species) as this region exported to the eastern Atlantic (28 species). The eastern Atlantic received more species (39) than it exported (16). However, the majority of invaders to the eastern Atlantic (28 of 39 species, 72%) have extremely limited distribu- tions in the eastern Atlantic, where they are confined either to offshore islands or to small sectors of the mainland coast. By contrast, most of the immigrants to the western Atlantic have wide ranges there; only four of the 35 species (11%) can be described as narrowly distributed in the western Atlantic. These data imply that penetration of the eastern Atlantic by immigrants has been marginal, whereas inva- sion of the western Atlantic has been geographically much more extensive. The status of the eastern and western Atlantic as donor and recipient regions for invaders can be compared with that of other tropical marine regions. Since the Central American seaway closed during the middle Pliocene, the eastern Pacific has received a large number of immigrant species from the Indo-West-Pacific, especially in reef habi- tats (Vermeij, 1987, 1991b; Grigg and Hey, 1992). Emerson (1991) recorded 61 shell-bearing gastropods as invaders from the Indo-West-Pacific to the eastern Pacific. To our knowledge, the eastern Pacific has not exported species to other tropical regions since mid-Pliocene time. Although many of the Indo-West-Pacific immigrants are found only on offshore islands, at least 26 (46%) have penetrated to the Pacific mainland coast of the Americas. The Indo-West- Pacific has served as a donor region for many invaders to the eastern Pacific and Atlantic. We know of only one per- suasive case of recent immigration to the Indo-West-Pacific from elsewhere. Lozouet (1986) has marshalled strong arguments in favor of the hypothesis that the lagoonal potamidid gastropod Potamides conicus Blainville, 1828, invaded the Indian Ocean from the Mediterranean during late Pliocene or Pleistocene time. It is, of course, possible that other taxa have a similar history, but detailed documen- tation is lacking. The four tropical marine regions can thus be ranked according to their status as donor and recipient regions for recent molluscan immigrants. The Indo-West-Pacific has been chiefly a donor region, whereas the eastern Pacific has been only a recipient region. The eastern and western Atlantic regions fall in the middle. These patterns of invasion could reflect the extent of prior extinction of species in the various tropical regions (Vermeij, 1991b). Among the four biotas, that of the Indo- West-Pacific probably suffered the smallest loss in overall diversity during the Pliocene and Pleistocene. Only a few taxa at the generic level disappeared from the region, although many genera and species apparently underwent contractions in range during the last three million years (Vermeij, 1986, 1991c). The western Atlantic, on the other hand, has been greatly affected by extinction. Some 60 to 70% of species and 32% of subgenera became extinct in Florida after mid-Pliocene time. Although all of this loss was compensated for, mainly by speciation, the Pliocene represents a time of very high species turnover in Florida and in other parts of the western Atlantic (see Vermeij and Petuch, 1986; Allmon et al., 1993). Extinction in the east- ern Pacific at the generic level was less than half that in the western Atlantic; at the species level, there was also sub- stantial extinction, but most of the losses were compensated for by extensive speciation during the early Pleistocene, so that there may have been no net loss in diversity since the early Pliocene. No published estimates of extinction exist for the tropical eastern Atlantic, because no substantial Pliocene deposits are known from West Africa. Brébion (1979) summarized the Pliocene and Pleistocene molluscs of Morocco (now in subtropical West Africa). From his list we estimate that 21 of 87 species (24%) recorded from the Pliocene and early Pleistocene of Morocco are extinct. There are only two genera that no longer occur in the east- ern Atlantic. This calculation suggests that extinction in this part of the world was much less severe than in the VERMEIJ AND ROSENBERG: DISPERSAL ACROSS THE TROPICAL ATLANTIC 189 western Atlantic. Tropical West Africa and the tropical eastern Pacific have both been recognized as major refuges for taxa that disappeared from southwestern Europe and the western Atlantic during the Pliocene and early Pleistocene (Vermeij, 1986, 199 1c). The western Atlantic thus stands out as the region that has suffered the greatest loss of diversity during the Pliocene and Pleistocene. Although the known number of invaders in this region is small, most of the invading species have achieved wide distributions in the western Atlantic and have become common elements in the habitats they occupy. The eastern Atlantic has received a large number of immigrant species, but few have become geographically widespread and very few have become common in shallow waters. Indeed, it is not at all certain that many of the immi- grant species maintain self-perpetuating populations in the eastern Atlantic. The same situation applies to the eastern Pacific. Many species have immigrated to this region, but very few have become common, and many eastern Pacific populations of Indo-West-Pacific species may be sustained only by the periodic influx of dispersing larvae from the west (see also Richmond, 1990; Emerson, 1991; Grigg and Hey, 1992). Immigration to the Indo-West-Pacific has been negligible and ecologically marginal. Thus, the greatest penetration of immigrants has occurred in the region that suffered the greatest magnitude of extinction and the great- est net loss of diversity since the Pliocene. Biotas in which the loss of diversity has been lower have been less exten- sively infiltrated by species invading during late Pliocene or Pleistocene time. Similar patterns have been documented for other cases of marine as well as terrestrial interchange (Vermeij, 199 1a, b). There is some evidence that patterns of invasion before the middle Pliocene were different from those later. In particular, eastward dispersal in the Atlantic may be a relatively recent phenomenon. Although systematically col- lected data for trans-Atlantic faunal interchange are not available for the Miocene, detailed studies of pectinid bivalves (Waller, 1991) and muricid gastropods (Vokes, 1988) indicate that many Miocene and early Pliocene taxa in tropical America owe their origins to immigration from the east. Examples include the pectinid scallop lineages of Aequipecten mucosus (Wood, 1828) and Argopecten, and the muricid lineages “Hexaplex” brassica (Lamarck, 1822) and Muricanthus. Much stronger ties link the faunas of the eastern Atlantic and tropical America during the Eocene. Dozens of genera and species occurred on both sides of the Atlantic, and may have dispersed from east to west (see e.g. Squires, 1987; Givens, 1989, 1991; Allmon, 1990). We know of no documented cases of eastward dispersal until middle Pliocene time. In the tropical Pacific, westward dis- persal may also have been the rule during the early Cenozoic (Grigg and Hey, 1992). These patterns of disper- sal probably reflect the predominantly westward flow of ocean currents at a time when a more or less continuous belt of tropical seaway existed at low latitudes before sea- ways through the Middle East, Indonesia, and Central America were constricted. Eastward dispersal in the Atlantic may have become possible when circulation in the North Atlantic (especially that in the Gulf Stream) became more vigorous during the middle Pliocene (Kaneps, 1979; Maier-Reimer et al., 1990; Crowley, 1991; Crowley and North, 1991). Whether prior loss of diversity also played a role in the pattern of interchange during Miocene and earli- er Cenozoic time is unknown. Our knowledge of the taxon- omy and distribution of Miocene and older species is still too sketchy to permit firm conclusions to be drawn. We hope that studies of Cenozoic molluscs on a worldwide scale will be done, so that important questions about faunal interchange and extinction can be answered. LITERATURE CITED Abbott, R. T. 1968. The helmet shells of the world (Cassidae). Indo- Pacific Mollusca 2 (9):1-202. Allmon, W. D. 1990. Review of the Bullia group (Gastropoda: Nassariidae) with comments on its evolution, biogeography, and phylogeny. Bulletins of American Paleontology 99: 1-179. Allmon, W. D., G. Rosenberg, R. Portell, and K. Schindler. 1993. Diversity of Pliocene-Recent mollusks in the western Atlantic. Science, 260:1626-1629. Beu, A. G. 1970. The Mollusca of the genus Charonia (family Cymati- idae). Transactions of the Royal Society of New Zealand, Biological Sciences 11:205-223. Beu, A. G. 1976. Arrival of Semicassis pyrum (Lamarck) and other ton- nacean gastropods in the southern ocean during Pleistocene time. Journal of the Royal Society of New Zealand 6:413-432. Beu, A. G. 1980. Australian gastropods of the family Bursidae. 1. The families of Tonnacea, the genera of Bursidae, and revision of species previously assigned to Tutufa Jousseaume, 1881. Records of the Australian Museum 33:248-324. Beu, A. G. 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 13:273-355. Beu, A. G. and W. O. Cernohorsky. 1986. Taxonomy of gastropods of the families Ranellidae (= Cymatiidae) and Bursidae. Part 1. Adoption of Ranellidae, and review of Linatella Gray, 1857. New Zealand Journal of Zoology 13:241-266. Beu, A. G. and E. A. Kay. 1988. Taxonomy of gastropods of the families Ranellidae (= Cymatiidae) and Bursidae. Part 4. The Cymatium pileare complex. Journal of the Royal Society of New Zealand 18:185-223. Beu, A. G. and J. Knudsen. 1987. Taxonomy of gastropods of the families Ranellidae (= Cymatiidae) and Bursidae. Part 3. A review of the trifid-ribbed species of Cymatium (Turritriton). Journal of the Royal Society of New Zealand 17:73-91. Bieler, R. 1984. Die Gattungen der Architectonicidae (Gastropoda: “Heterogastropoda’”). Allgemeines und Teil I: Pseudomalaxis. Archiv fiir Molluskenkunde 115:55-103. 190 AMER. MALAC. BULL. 10(2) (1993) Brébion, P. 1979. Etude biostratigraphique et paléoécologique du Quaternaire marocain. Annales de Paléontologie Invertebrés 65:1-42. Chace, F.A. 1966. Decapod crustaceans from St. Helena Island, South Atlantic. Proceedings of the U. S. National Museum 118:622-662. Cosel, R. von. 1982. Marine Mollusken der Kapverdischen Inseln: Ubersicht mit zoogeographischen Anmerkungen. Courier Forschungs-Institut Senckenberg 52:35-76. Crowley, T. J. 1991. Modeling Pliocene warmth. Quaternary Science Reviews 10:275-282. Crowley, T. J. and J. R. North. 1991. Paleoclimatology. Oxford University Press, New York. 339 pp. Emerson, W. K. 1991. First records of Cymatium mundum (Gould) in the eastern Pacific Ocean, with comments on the zoogeography of the tropical trans-Pacific tonnacean and non-tonnacean prosobranch gastropods with Indo-Pacific faunal affinities in West American waters. Nautilus 105:62-80. Faber, M. J. 1988. Studies in West Indian marine mollusks 13. The mala- cological taxa of Gordon W. Nowell-Usticke. Kreukel 24:67-102. Finlay, C. J. and D. L. N. Vink. 1982. New records of Cymatiidae (Gastropoda) in the western Atlantic. Nautilus 96:132-134. Gibson-Smith, J. and W. Gibson-Smith. 1981. Casmaria atlantica Clench (Mollusca: Gastropoda): thoughts on its evolution. Veliger 23:352- 354. Givens, C. R. 1989. First record of the Tethyan genus Volutilithes (Gastropoda: Volutidae) in the Paleogene of the Gulf Coastal Plain, with a discussion of Tethyan molluscan assemblages in the Gulf Coastal Plain and Florida. Journal of Paleontology 63:852-856. Givens, C. R. 1991. Old World Tethyan occurrences of Lyrischapa (Gastropoda: Volutidae) and biogeographic implications. Journal of Paleontology 65:661-670. Glibert, M. 1949. Gastropodes du Miocéne moyen du basin de la Loire. Institut Royal des Sciences Naturelles de Belgique Mémoires (2) 39:1-240. Grigg, R. W. and R. Hey. 1992. Paleoceanography of the tropical eastern Pacific Ocean. Science 255: 172-178. Hoagland, K. E. 1977. Systematic review of fossil and Recent Crepidula and discussion of evolution of the Calyptraeidae. Malacologia 16:353-420. Houbrick, R. S. 1974. The genus Cerithium in the western Atlantic (Cerithiidae: Prosobranchia). Johnsonia 5:33-84. Houbrick, R. S. 1987. Anatomy, reproductive biology, and phylogeny of the Planaxidae (Cerithiacea: Prosobranchia). Smithsonian Contributions to Zoology 445:1-57. Houbrick, R. S. 1992. Simulathena papuensis, a new planaxid genus and species from the Indo-West Pacific. Veliger 35:64-69. Hubendick, B. 1945. Phylogenie und Tiergeographie der Siphonariidae. Zoologiska Bidrag Uppsala 24:1-216. Hubendick, B. 1946. Systematic monograph of the Patelliformia. Kungliga Svenska Vetenskaps-Akademiens Handlingar (3) 23:1-93. Janssen, R. 1978a. Die Scaphopoden und Gastropoden des Kasseler Meeressande von Glimmerode (Niederhessen). Geologisches Jahrbuch (A) 4:1-195. Janssen, R. 1978b. Die Mollusken des Oberoligozans (Chattium) in Nordseebecken. 1. Scaphopoda, Archaeogastropoda, Mesogastro- poda. Archiv fiir Molluskenkunde 109:137-227. Jung, P. 1969. Miocene and Pliocene mollusks from Trinidad. Bulletins of American Paleontology 55:293-657. Kaneps, A. G. 1979. Gulf Stream: velocity fluctuation during the late Cenozoic. Science 204:297-301. Kay, E. A. and S. Johnson. 1987. Mollusca of Enewetak Atoll. In: The Natural History of Enewetak Atoll, Vol. Il, Biogeography and Systematics, D. M. Devaney, E. S. Reese, B. L. Burch, and P. Helfrich, eds. pp. 105-146. Office of Scientific and Technical Information, U. S. Department of Energy, Washington. Leal, J. H. 1991. Marine Prosobranch Gastropods from Oceanic Islands off Brazil: Species Composition and Biogeography. Universal Book Services, W. Backhuys, Oegstgeest. 418 pp. Lozouet, P. 1986. Rédéfinition des genres Potamides et Pirenella (Gastropoda, Prosobranchia) a partir des espéces actuelles et fos- siles: implications phyletiques et biogéographiques. Annales de Paléontologie (Vertebrés-Invertebrés) 72: 163-210. Maier-Reimer, E., U. Mikolajewicz, and T. Crowley. 1990. Ocean general circulation model sensitivity experiment with an open Central American isthmus. Paleoceanography 5:349-366. Marshall, B. A. 1992. A revision of the Recent species Eudolium, Dall, 1889 (Gastropoda: Tonnidae). Nautilus 106:24-38. Meco, Cabrera, J. 1982. Los bivalvos fossiles de las Canarias orientales. Anuario de Estudios Atlanticos 28:65-125. Morrison, J. P. E. 1963. Notes on American Siphonaria. Annual Report of the American Malacological Union (1963):7-9. Petuch, E. J. 1989. New species of Malea (Gastropoda Tonnidae) from the Pleistocene of southern Florida. Nautilus 103:92-95. Petuch, E. J. 1991. New gastropods from the Plio-Pleistocene of south- western Florida and the Everglades Basin. W. H. Dall Pale- ontological Research Center, Special Publication 1:1-59. Reid, D. G. 1986. The Littorinid Molluscs of Mangrove Forests in the Indo-Pacific Region: the Genus Littoraria. British Museum of Natural History, London. 227 pp. Reid, D. G. 1989. The comparative morphology, phylogeny and evolution of the gastropod family Littorinidae. Philosophical Transactions of the Royal Society of London (B) 324:1-110. Richmond, R. H. 1990. The effects of the El Nifio/Southern Oscillation on the dispersal of corals and other marine organisms: Jn: Global Eco- logical Consequences of the 1982-83 El Nifio-Southern Oscillation, P. W. Glynn, ed. pp. 127-140. Elsevier, Amsterdam. Robertson, R. 1973. On the fossil history and intrageneric relationships of Philippia (Gastropoda: Architectonicidae). Proceedings of the Academy of Natural Sciences of Philadelphia 125:37-46. Robertson, R. 1975. Faunal affinities of the Architectonicidae in the east- ern Pacific. Bulletin of the American Malacological Union (1975):51. Robinson, D. G. 1990. Notes on the fauna of the Moin Formation - I. On the occurrence of Gyrineum in the early Pleistocene Moin For- mation of Costa Rica. Tulane Studies in Geology and Paleontology 23:133-135. Rosenberg, G. 1993. A database approach to studies of molluscan taxono- my, biogeography and diversity, with examples from western Atlantic marine gastropods. American Malacological Bulletin 10(2):257-266. Rosewater, J. 1975. An annotated list of the marine mollusks of Ascension Island, South Atlantic Ocean. Smithsonian Contributions to Zoology 189:1-41. Rosewater, J. 1981. The family Littorinidae in tropical West Africa. Atlantide Reports 13:7-48. Rosewater, J. and G. J. Vermeij. 1972. The amphi-Atlantic distribution of Littorina meleagris. Nautilus 86:67-69. Rosso, J.-C. 1974. Contribution a |’ étude paléontologique du Quaternaire Sénégalo-Mauritanien: Mollusques du Nouakchottien de Saint- Louis (Sénégal) I. Gastropodes et scaphopodes. Bulletin de I’ Institut Fondamental d’Afrique Noire (A) 36:1-50. Scheltema, R. S. 1971. Larval dispersal as a means of genetic exchange between geographically separated populations of shallow-water benthic marine gastropods. Biological Bulletin 140:284-322. Schremp, L. A. 1981. Archaeogastropoda from the Pliocene Imperial Formation of California. Journal of Paleontology 55:1123-1136. VERMEIJ AND ROSENBERG: DISPERSAL ACROSS THE TROPICAL ATLANTIC 191 Smith, E. A. 1890. Report on the marine molluscan fauna of the island of St. Helena. Proceedings of the Zoological Society of London (1890):246-317. Smith, J. T. 1991. Cenozoic giant pectinids from California and the Tertiary Caribbean Province: Lyropecten, “Macrochlamis,” Vertipecten, and Nodipecten species. United States Geological Survey Professional Paper 1391:1-155. Squires, R. L. 1987. Eocene molluscan paleontology of the Whitaker Peak area, Los Angeles, and Ventura Counties, California. Contributions in Science, Natural History Museum of Los Angeles County 388:1-93. Talavera, F. G. 1982. Los Moluscos Gasteropodos Amfiatlanticos (Estudio Paleo y Biogeografico de las Especies Bentonicas Litorales). Coleccidn Monografias 10, Secretariado de Publicaciones de la Universidad de La Laguna, La Laguna. 351 pp. Vermeij, G. J. 1986. Survival during biotic crises: the properties and evo- lutionary significance of refuges. In: Dynamics and Extinction, D. K. Elliott, ed. pp. 231-246. Wiley, New York. Vermeij, G. J. 1987. The dispersal barrier in the tropical Pacific: implica- tions for molluscan speciation and extinction, Evolution 41:1046- 1058. Vermeij, G. J. 1991a. When biotas meet: understanding biotic interchange. Science 253:1099-1104. Vermeij, G. J. 1991b. Anatomy of an invasion: the Trans-Arctic inter- change. Paleobiology 17:281-307. Vermeij, G. J. 1991c. Marine extinctions and their implications for conser- vation and biogeography. In: The Unity of Evolutionary Biology, Proceedings of the Fourth International Congress of Systematic and Evolutionary Biology, Vol. I, E. C. Dudley, ed. pp. 143-148. Dioscorides Press, Portland, Oregon. Vermeij, G. J. and E. J. Petuch. 1986. Differential extinction in tropical American molluscs: endemism, architecture, and the Panama land bridge. Malacologia 27:29-41. Vokes, E. H. 1966. The genus Vasum (Mollusca: Gastropoda) in the New World. Tulane Studies in Geology 5:1-36. Vokes, E. H. 1975. Notes on the Chipola Formation - XVIII. Some new or otherwise interesting members of the Calyptraeidae (Mollusca: Gastropoda). Tulane Studies in Geology and Paleontology 11:163- 172. Vokes, E. H. 1988. Muricidae (Mollusca: Gastropoda) of the Esmeraldas beds, northwestern Ecuador. Tulane Studies in Geology and Pale- ontology 21:1-50. Vokes, E. H. 1989. Neogene paleontology in the northern Dominican Republic 8. The family Muricidae (Mollusca: Gastropoda). Bulletins of American Paleontology 97:5-94. Vokes, E. H. 1990. On the occurrence of the gastropod genus Cassis in the Esmeraldas fauna, northwestern Ecuador. Tulane Studies in Geology and Paleontology 23: 121-126. Voskuil, R. P. A. and W. J. H. Onverwagt. 1989. Inventarisation of the Recent European and West African Cardiidae (Mollusca, Bivalvia). Gloria Maris 28:49-96. Waller, T. R. 1991. Evolutionary relationships among commercial scallops (Mollusca: Bivalvia: Pectinidae). In: Scallops - Biology, Ecology and Aquaculture, S. E. Shumway. ed. p. 1-73. Elsevier, Amsterdam. Woodring, W. P. 1957. Geology and paleontology of Canal Zone and adjacent parts of Panama: geology and description of Tertiary mol- lusks (gastropods: Trochidae to Turritellidae). United States Geological Survey Professional Paper 306-A:1-145. Woodring, W. P. 1964. Geology and paleontology of Canal Zone and adjacent parts of Panama: description of Tertiary mollusks (gas- tropods: Columbellidae to Volutidae). United States Geological Survey Professional Paper 306-C:241-297. Woodring, W. P. 1973. Geology and paleontology of Canal Zone and adjacent parts of Panama: description of Tertiary mollusks (addi- tions to gastropods, scaphopods, pelecypods: Nuculidae to Malleidae). United States Geological Survey Professional Paper 306-E:453-539. Date of manuscript acceptance: 26 April 1993 192 AMER. MALAC. BULL. 10(2) (1993) APPENDIX 1. List of western Atlantic species likely to have arrived from the Indo-West- Pacific region after closure of the Central American isthmus. An asterisk indicates that the species is endemic to the Atlantic but very closely relat- ed to an Indo-West-Pacific taxon; other species occur in both Atlantic and Indo-West-Pacific regions. Gastropoda: * Synaptocochlea picta (Orbigny, 1842) * Casmaria atlantica Clench, 1944 * Malea (Quimalea) noronhensis Kempf and Matthews, 1969 Tonna galea (Linnaeus, 1758) * T. maculosa (Dillwyn, 1817) Eudolium bairdii (Verrill and Smith, 1881) * FE. crosseanum (Monterosato, 1869) * Bufonaria (Marsupina) bufo (Bruguieére, 1792) * Bursa (Bufonariella) natalensis (Coelho and Matthews, 1970) B. (Bufonariella) ranelloides (Reeve, 1844) * B. (Lampadopsis) thomae (Orbigny, 1842) Distorsio perdistorta Fulton, 1938 Cymatium (Turritriton) comptum (A. Adams, 1855) C. (Ranularia) gallinago (Reeve, 1844) * C. (Septa) occidentalis Clench and Turner, 1957 C. (Reticutriton) pfeifferianum (Reeve, 1844) * C. (Ranularia) ridleyi (E. A. Smith, 1890) C. (Turritriton) vespaceum (Lamarck, 1822) Gyrineum louisae Lewis, 1974 * Psilaxis krebsii (Morch, 1874) VERMEIJ AND ROSENBERG: DISPERSAL ACROSS THE TROPICAL ATLANTIC 193 APPENDIX 2. List of shallow-water amphi-Atlantic shell-bearing molluscan species. Species occur at depths of 100 m or shallower. An asterisk indicates that there is no close eastern Pacific relative or that the fossil record of the species or its presumed ancestor does not precede the late Pliocene in the western Atlantic. Species without further annotations are known to be living on the continental coasts of both the eastern and western Atlantic. Those living in the eastern Atlantic only on offshore islands (Canary Islands, Cape Verde Islands, or Madeira) are designated “insular EA”. Species known in the western Atlantic only from offshore islands are designated “insular WA”’. Species known in the western Atlantic only as fossils are designated “fossil WA”. Gastropoda: Scissurella cingulata (O. G. Costa, 1861 ) Diodora cayenensis (Lamarck, 1822) (insular EA) E. tuberculosa (Libassi, 1859) Lucapinella limatula (Reeve, 1850) * Littoraria (Littorinopsis) angulifera (Lamarck, 1822) Nodilittorina (Fossarilittorina) meleagris (Potiez and Michaud, 1838) * Benthonella gaza Dall, 1889 Alaba incerta (Orbigny, 1842) Cerithium (Thericium) atratum (Born, 1778) * C. (T.) guinaicum Philippi, 1849 Fossarus ambiguus (Linnaeus, 1767) * Hinea lineata (da Costa, 1778) Crepidula (Bostrycapulus) aculeata (Gmelin, 1791) (insular EA) Cheilea equestris (Linnaeus, 1758) Hipponix antiquatus (Linnaeus, 1767) (insular EA) H. subrufus (Lamarck, 1819) (insular EA) Pedicularia sicula (Swainson, 1840) Trivia (Dolichupis) candidula (Gaskoin, 1836) Lamellaria perspicua (Linnaeus, 1758) * Gyrodes depressa (Seguenza, 1874) * Natica livida (Pfeiffer, 1840) (insular EA) Notocochlis marochiensis (Gmelin, 1791) Polinices lacteus (Guilding, 1834) (insular EA) Cassis tuberosa (Linnaeus, 1758) (insular EA) Cypraecassis testiculus (Linnaeus, 1758) Phalium (Tylocassis) granulatum (Born, 1778) (insular EA) Bursa (Lampadopsis) corrugata (Perry, 1811) (insular EA) * B. (Apsa) marginata (Gmelin, 1791) (fossil WA) * B. ranelloides (Reeve, 1844) (insular EA) B. (Bufonariella) granularis (R6ding, 1798) (insular EA) * B. (Bufonariella) scrobiculator (Linnaeus, 1758) (fossil WA) * B. (Lampadopsis) thomae (Orbigny, 1842) (insular EA) * Distorsio perdistorta Fulton, 1938 Charonia tritonis variegata (Lamarck, 1816) * C. lampas (Linnaeus, 1758) Cymatium (Monoplex) aquatile (Reeve, 1844) (insular EA) * C. (Turritriton) comptum (A. Adams, 1855) C. (Ranularia) cynocephalum (Lamarck, 1816) C. (Turritriton) kobelti (von Maltzan, 1884) (fossil WA) C. (T.) labiosum (Wood, 1828) (insular EA) C. (Monoplex) martinianum (Orbigny, 1845) C. (Ranularia) muricinum (Réding, 1798) (insular EA) C. (Monoplex) nicobaricum (Réding, 1798) * C. (Septa) occidentale Clench and Turner, 1957 (insular EA) C. (Monoplex) parthenopeum (von Salis, 1793) C. (Turritriton) tenuiliratum (Lischke, 1863) * C. (Monoplex) trigonum (Gmelin, 1791) * Gyrineum louisae Lewis, 1974 Linatella caudata (Gmelin, 1791) (insular EA) L. succincta (Linnaeus, 1771) * Ranella olearia (Linnaeus, 1758) * Eudolium bairdii (Verrill and Smith, 1881) * E. crosseanum (Monterosato, 1869) * Tonna galea (Linnaeus, 1758) * T. maculosa Dillwyn, 1817 (insular EA) * Cosmotriphora melanura (C. B. Adams, 1850) Metaxia abruptu (Watson, 1880) Oceanida graduata (de Folin, 1870) Cylindriscala acus (Watson, 1883) * Epitonium (Hyaloscala) albidum (Orbigny, 1842) E. (Gyroscala) lamellosum (Lamarck, 1822) E. striatissimum (Monterosato, 1878) Opalia (Dentiscala) crenata (Linnaeus, 1758) Scalenostoma subulatum (Broderip, 1832) Cytharmorula grayi (Dall, 1889) Stramonita haemastoma (Linnaeus, 1767) Thaisella coronata (Lamarck, 1822) Trachypollia nodulosa (C. B. Adams, 1845) T. turricula (von Maltzan, 1884) * Typhis (Typhina) belcheri Broderip, 1833 * T. (Typhinellus) sowerbii Broderip, 1833 Coralliophila caribaea Abbott, 1958 (insular EA) C. squamosa (Bivona, 1838) Colubraria testacea (Morch, 1852) (insular EA) * Pugilina morio (Linnaeus, 1758) Mitrella ocellata (Gmelin, 1791) * Vasum globulus Lamarck, 1816 (insular WA) Gibberula lavalleeana (Orbigny, 1842) Persicula interruptolineata (von Mihlfeld, 1816) Prunum marginatum (Born, 1778) Conus (Chelyconus) ermineus (Born, 1778) * Corinnaeturris leucomata (Dall, 1881) Drilliola loprestiana (Calcara, 1841 ) Gymnobela agassizii (Verrill and Smith, 1880) Architectonica nobilis (R6ding, 1798) Discotectonica discus (Philippi, 1844) Heliacus bisulcatus (Orbigny, 1842) H. cylindricus (Gmelin, 1791) H. perrieri (Rochbrune, 1881 ) Pseudotorinia architae (O. G. Costa, 1841 ) Psilaxis krebsii (M6rch, 1874) Spirolaxis centrifuga (Monterosato, 1890) Mathilda (Fimbriatella) barbadensis (Dall, 1889) Eulimella scillae (Scacchi, 1835) Pyramidella dolabrata (Linnaeus, 1758) Hydatina physis (Linnaeus, 1758) Micromelo undatus (Bruguiére, 1792) Philine trachyostraca (Watson, 1897) Cylichna discus Watson, 1883 Bulla striata (Bruguiére, 1792) Atys caribaeus (Orbigny, 1842) A. macandrewi E. A. Smith, 1872 Haminoea antillarum (Orbigny, 1841) H. elegans (Gray, 1825) Umbraculum umbraculum (Lightfoot, 1786) Berthella stellata (Risso, 1826) Pira monilis (Bruguiére, 1792) * Siphonaria (Mouretus) pectinata (Linnaeus, 1758) * * * 194 AMER. MALAC. BULL. 10(2) (1993) Scaphopoda Dentalium striolatum Stimpson, 1851 Bivalvia Arca imbricata Bruguiére, 1789 (insular EA) Arcopsis adamsi (Dall, 1886) (insular EA) Pinna rudis (Linnaeus, 1758) Lima lima (Linnaeus, 1758) Perna perna (Linnaeus, 1758) Lithophaga aristata (Dillwyn, 1817) Nodipecten nodosus (Linnaeus, 1758) Dendostrea frons (Linnaeus, 1758) (insular EA) * Neopycnodonte cochlear (Poli, 1795) * Parahyotissa mcgintyi Harry, 1895 Pseudochama radians (Lamarck, 1819) (insular EA) Papyridea soleniformis (Bruguiére, 1792) Gastrochaena hians (Gmelin, 1791) Barnea truncata (Say, 1822) Pholas (Thovana) campechiensis (Gmelin, 1791 ) The evolution of “Chlamys” (Mollusca: Bivalvia: Pectinidae) in the tropical western Atlantic and eastern Pacific Thomas R. Waller Department of Paleobiology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, U.S.A. Abstract: The phylogenetic relationships and paleontological history of six species of “Chlamys” living in the Caribbean were analyzed for the pur- pose of determining how each originated and is related to geminate species elsewhere in the world ocean. Particular emphasis is on origins and dispersal rel- ative to the closure of seaways connecting the tropical western Atlantic and eastern Pacific. The resulting systematic revision demonstrates that these species are not a monophyletic assemblage, nor are they a single genus, nor are any of them members of the genus Chlamys in a strict sense. The Caribbean species “Chlamys” multisquamata is placed in a new genus Laevichlamys, which originated in the western Indo-Pacific and dis- persed westward to the eastern Atlantic, thence to the western Atlantic. Arrival in the Caribbean region occurred after the mid-Pliocene seaway closure, and L. multisquamata has no geminate species in the eastern Pacific. True Hinnites, represented at present by the relic species H. corallinus from west Africa and by Mio-Pliocene H. crispus of the Mediterranean, is shown to be derived from a stem species in Laevichlamys. Four of the Caribbean species, Chlamys sentis, C. ornata, C. mildredae, and C. imbricata, comprise a new monophyletic genus Caribachlamys characterized by a unique lecithotrophic-type larval shell. This group also originated in post-closure time, probably in the late Pliocene, and is restricted to the tropical western Atlantic. Caribachlamys is placed in a new tribe Crassadomini with the more plesiomorphic genus Crassadoma. The definition of the latter is expanded to include not only its type species, C. gigantea of the eastern Pacific, but also C. multistriata and C. pusio of the eastern Atlantic. The phylogenetic histories of these species are discussed and it is shown that the cemented habit of C. gigantea and C. pusio arose independently and at different times. Lastly, “Chlamys” benedicti of the Caribbean is made the type species of a new genus Spathochlamys, which has a long history in the tropical western Atlantic but originated from ancestors in the eastern Atlantic. Entry of members of the genus into the eastern Pacific occurred via Central American seaways probably in the late Miocene, giving rise to S. vestalis (=S. lowei), which has now dispersed as far as the Gal-pagos Islands with some morphologi- cal divergence. The results suggest that: (1) eurytopy bestows resistance to both extinction and speciation; (2) long-distance dispersal facilitates allopatric specia- tion but so also does lecithotrophy in reef habitats; (3) evolution within some scallop clades has been very rapid; and (4) gene flow between Bermuda and the Antilles was interrupted during the late Pleistocene. Finally, marine evolutionary events in the Neogene of the tropical Americas are related only incidentally to seaway closure; many of the differences that separate modern Caribbean and tropical eastern Pacific faunas arose both well before and well after closure. In addition to the introduction of two new tribes, three new genera, and one new fossil species, lectotypes are designated for certain species named by Linnaeus (1758), Gmelin (1791), Poli (1795), Lamarck (1819), and Reeve (1853). It has long been known that the modern Caribbean mation on both the timing and the effects of seaway clo- marine fauna was formerly part of a larger Tertiary biogeo- sure. Studies of the physical stratigraphy and biostratigra- graphic province that extended into the tropical eastern phy of the Isthmus point to a final closure date of about 3.5 Pacific through connecting seaways in Central America and million years ago, in the middle Pliocene (Coates et al., northern South America (Woodring, 1966). It has been 1992, and references therein). Allmon et al. (1993) com- assumed that the final closure of these seaways set off a piled species-level data on gastropods and found that the divergence between the Atlantic and Pacific parts of a once modern warm western Atlantic gastropod fauna is not less continuous fauna. The Panamic fauna is said to have diverse than the eastern Pacific one but has undergone con- remained relatively unchanged, whereas the Caribbean siderably more taxonomic turnover since closure. Budd et fauna declined in diversity and species richness. These al. (1992), in a study of reef coral species, also found that assumptions have been based mainly on data stemming taxonomic turnover, not diversity decline, was the major from taxic censuses, particularly of gastropods at the level event in the post-closure Caribbean. Caribbean extinctions of genus and subgenus (Woodring, 1966; Petuch, 1982b; increased after closure, but so also did originations. Jones and Hesson, 1985; Vermeij and Petuch, 1986; Furthermore, these and other recent studies (Jackson et al., Vermeij, 1993; Allmon et al., 1993). 1993; Knowlton et al., 1993; Vermeij, 1993) indicate that New approaches, however, are yielding new infor- many of the differences that now exist between the faunas American Malacological Bulletin, Vol. 10(2) (1993):195-249 195 196 AMER. MALAC. BULL. 10(2) (1993) on the two sides of the Isthmus did not originate exactly at the time of final closure. Jackson and Jung (1992), using species-level data, found that strong contrasts between the mollusk faunas of the western Atlantic and eastern Pacific were already present 2 Ma before final closure and that extinction in the Caribbean did not become massive until 1.8 to 1.5 Ma after final closure. Collins (1992) found that the main restructuring of the benthic foraminiferal faunas of the upper continental slope to inner continental shelf of the Caribbean occurred in the latest Miocene well before seaways closed. Cheetham and Jackson (1992) concluded from a phylogenetic study of some cheilostome bryozoans that cladogenesis reached a peak in the Late Miocene “well before final closure of the Panamanian portal, but after for- mation of the sill that disrupted oceanic circulation patterns throughout the region.” Knowlton et al. (1993), in a study of the biochemical and reproductive divergence of trans- isthmian pairs of snapping shrimp, found “that isolation was staggered rather than simultaneous.” It is apparent from these recent studies that species-level analyses provide new insight to the dynamics of evolutionary change in the tropical American marine faunas before and after seaway closure. Aside from a few species-level studies, however, these efforts still depend mainly on taxonomic census data, albeit at a finer level than that in previous studies. The objective of the present paper is to examine the general question of how the Caribbean marine fauna was assembled before and after seaway closure by means of a species-level examination of phylogeny in a subset of Caribbean mollusks. Specifically, for each species consid- ered, it will be asked where, when, and from what it origi- nated. Are the stem groups for Caribbean species in the Caribbean or elsewhere? Is there a center of origin for Caribbean species, and how are origins distributed through time? Lastly, is there evidence for geographic subspeciation within the tropical western Atlantic? If so, what is the polarity of morphological differences among subspecies and how might this polarity be explained? The bivalve family Pectinidae has been selected for three reasons: (1) my research concentrates on this group, and the results of earlier phylogenetic studies (Waller, 1991) can be readily applied to a new problem; (2) pectinid shells, owing to their largely calcitic composition, are commonly preserved where wholly aragonitic mollusk shells are not; (3) although Neogene pectinids appear to have low survivorship and greater extinction rates than other molluscan groups (Stanley, 1986a), they also appear to be capable of more rapid evolution and are thus ideally suited for an examination of events of the past few million years. A subset of Pectinidae consisting of species tradi- tionally placed in the genus Chlamys has been selected because of their generally similar byssate life habits and alleged congeneric status. The species analyzed are the six Caribbean species of “Chlamys” and their close relatives elsewhere in the world ocean (Table 1). It will be shown that these species, far from being congeneric, in fact repre- sent four extant genera distributed among three tribes. This has necessitated a taxonomic revision which has been Table 1. Extant American “Chlamys” and related species of the eastern Atlantic analyzed in the present study. Original name Genus in this study Tribe Pecten multisquamatus Dunker, 1864 Laevichlamys Chlamydini new genus Lima gigantea Gray, 1825 Crassadoma Crassadomini new tribe Ostrea multistriata Poli, 1795 Crassadoma Crassadomini O. pusio Linnaeus, 1758 Crassadoma Crassadomini P. sentis Reeve, 1853 Caribachlamys Crassadomini new genus P. ornatus Lamarck, 1819 Caribachlamys Crassadomini P. (Chlamys) imbricatus mildredae Caribachlamys Crassadomini Bayer, 1941 O. imbricata Gmelin, 1791 Caribachlamys Crassadomini Chlamys benedicti Spathochlamys Mimachlamydini Verrill and Bush, 1897 new genus new tribe P. (Chlamys) lowei Hertlein, 1935 Spathochlamys Mimachlamydini [= Pecten vestalis Reeve, 1853] Provenance W. Atlantic & Caribbean E. Pacific E. Atlantic & SW Indian O. E. Atlantic W. Atlantic & Caribbean W. Atlantic & Caribbean W. Atlantic (SE Florida) W. Atlantic & Caribbean W. Atlantic & Caribbean E. Pacific Stratigraphic range Pleist.-Recent Lower? Mio.-Recent Lower Mio.-Recent Pleist.-Recent Plio.?-Recent Lower? Pleist.-Recent Upper Plio.?-Recent Lower? Pleist.-Recent Upper? Plio.-Recent Upper Mio.-Recent WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 197 extended to two eastern Atlantic species in order to estab- lish the morphological integrity of a new tribe and to assist in the demonstration that cementation, present in one American species, is a highly polyphyletic phenomenon among pectinids. Table 2 is a practical morphological key to the identification of the four genera listed in Table | plus four that are relevant to understanding phylogeny and biogeog- raphy in the tropical American regions. MATERIALS AND METHODS This study is based mainly on the extensive Recent and fossil shell collections of the Departments of Invertebrate Zoology and Paleobiology of the United States National Museum of Natural History, supplemented by specimens from the collection of John Waldrop and Druid Wilson, Lake Wales, Florida, the Florida Museum of Natural History Museum, Gainesville, and the Paleonto- logical Research Institution, Ithaca, New York. Additional information stems from earlier examinations of Recent and fossil specimens housed in major natural history museums in the United States, western Europe, and Great Britain. Some new information on West Coast species was obtained after the initial version of the manuscript was reviewed. These late additions resulted from an opportunity to study collections at the United States Geological Survey office in Menlo Park, California, the Museum of Paleontology of the University of California (Berkeley), and the Los Angeles County Museum of Natural History. Morphological terms are explained in Waller (1991) and measurement definitions are given in Waller and Marincovich (1992). In accord with these earlier studies, the terms “commarginal” and “antimarginal” refer to sculptural features that are approxi- mately parallel or perpendicular to the shell margin, respec- tively. In contrast, the term “radial” refers to features that radiate from the beak, meaning that these features may be nearly parallel to the margin at the disk flanks but nearly perpendicular in the midventral region. All specimens in the Smithsonian collections have been examined with a light microscope (Wild M-5) using reflected or transmitted light, the latter particularly useful in the examination of microsculpture in thin-shelled early growth stages. Prodissoconch morphology was studied at x50 with the light microscope, with verification of inferred prodissoconch configuration provided by examination of selected specimens with a scanning electron microscope (SEM). SEM specimens were sputter-coated with gold in a vacuum chamber and examined at accelerating voltages of 10kv. Abbreviations AHF: Catalogue of the Alan Hancock Foundation Collection, now housed at LACM. AM: Zoological Museum, Amsterdam. ANSP: Academy of Natural Sciences of Philadelphia. AOL: length of anterior outer ligament. BMNH: The Natural History Museum, London. Table 2. Key to some genera of “Chlamys” referred to in the text (asterisk (*) marks those genera that are extinct or absent in the tropical American region). 1. Simple non-branching radial ribs introduced early and persisting throughout ONtOgeNy .............ccceecceceseeeeeeeseseeeeeseeeeeceeeeaeeaessssscesecsenseesessesaeesessessesereeeteteeseees 2 Complex pattern of rib introduction, with introductions occurring through most Of ONtOGENY ............cccccecceseeseesceseeseseesecsesecsecsessecaecseeseesessecsessesseseeseseeeatesseaes 3 2. Rib interspaces with a single medial riblet, at least in early ONtOQENY oo... cece cece ceceseeseseeseeseesecseseseeseeseesenseeeceacuecsecsecsessessessecsesaeseesaeseeseesssssecseesscsassaesasseeseeas 4 Rib interspaces lacking medial riblets and crossed by antimarginal, oblique, or herringbone striae; internal rib carinae limited to juvenile stage OF ADSM... cece eceesceeeseeseeseeeseeseeseeesseeaecseeseesenessecsecsecseeesecaeeeeeeeeeeeseesecessesseesesseesessaesee Mimachlamys* 3. Ribbing coarse and scaly; commarginal lirae commonly present in interspaces at least in early ontogeny; posterior: margins Of POsteriOraUriCles: CONCAVE: svececcecsse; sveeeccase cave desecsacavcssvacestcccsacs seit asevensa weseisnusges dia Geseareuesvan su aedseiptvi dates; 1¥4 009d: 6524 0; 5430 foteasealt sends dacrazeeresedSiad 6 Ribbing fine, with repeated introductions of new ribs by medial intercalation on both valves, filling interspaces; commarginal lirae absent; posterior margins of posterior AUTICIES CONVEX 00.0... .eeeecesceecesesesseesesseseeseeseeseueesecsecseesecseesecsecsessessessssieesesseeseeseeseess Laevichlamys 4. Scales atop ribs concave toward dorsum; commarginals present in rib interspaces at least in early ontogeny; ribs carinate on internal shell surface near Margin 0.0.2... eccececccseeseeseseeseeseeseseesecseceesecsessecsessessesseeaesseseessessessessessessessesateascsecseesscaesesieeseeese 5 Scales atop ribs convex toward dorsum; commarginals in rib interspaces obscure or absent; ribs without internal carinae ................0..c:c000 Dimarzipecten* 5. Rib interspaces in early ontogeny crossed by more or less straight commarginal lirae .........0...cccceeceesescseeesseeeesesseseecseseeseescseesesecacseeasstessseeseseees Spathochlamys Rib interspaces in early ontogeny having wavy or looped commarginal lirae of the Aequipectinine type .........0..c0cccccccccccscscsccescesesesscscseesseevseesestesees Genus A* Gs, Prodissoconch with short Plistage:and long PIl‘sta ge: c.ceiscccscsscacsesalsstssacessuvestusavevevees ove stuiius cscueardncas eeveic duels Sivesdivs fvsvasstessadveviis sstacuddvvarstoevidinibeatavinaseadiplavizeceaye vi Prodissoconch with PI stage large and PII stage limited to a narrow fringe 7. Antimarginal striae but not commarginal lirae dominant at start of ribbed stage; shagreen microsculpture sometimes present on parts Of ValVeS..........:ccccceeeseeesereteees CREE TEETER REE CCE TET EE CEE CTC RE eee Chlamys* Commarginal lirae present in rib interspaces at start of ribbed stage at least on left valve; shagreen microsculpture absent .................c:ccccccecce Crassadoma 198 AMER. MALAC. BULL. 10(2) (1993) BRM: Institut royal des Sciences naturelles de Belgique, Brussels. CAS: California Academy of Sciences, San Francisco. DMNH: Delaware Museum of Natural History, Wilmington, Delaware. GNH: Museum d’Histoire naturelle de Geneve, Switzerland. Ht.: height, the maximum dorsoventral dimension perpen- dicular to the hingeline. HUB: Zoologisches Museum, Museum fir Naturkunde der Humboldt Universitat, Berlin. kv: 1,000 volts. LACM: Los Angeles County Museum of Natural History, Los Angeles, California. LM: Rijksmuseum van Natuurlijke Historie, Leiden. Ma: millions of years, the standard abbreviation in the Systeme International d’Unités of the General Con- ference on Weights and Measures (Harland et al., 1982). MNHN: Muséum national d’ Histoire naturelle, Paris. NMB: Naturhistorisches Museum Basel, Basel, Switzerland. POL: length of posterior outer ligament. PRI: Paleontological Research Institution, Ithaca, New York. R/V: Research vessel. SEM: scanning electron microscope or micrograph. TU: Locality register of H.E. and E.H. Vokes, Tulane University, New Orleans. TUI: Dipartimento di Scienze della Terra, Universita degli Studi di Torino, Torino, Italy. UCBL: Centre des Sciences de la Terre, Université Claude Bernard Lyon I, Villeurbanne, France. UCD: Department of Geology, University of California, Davis. UCMP: Museum of Paleontology, University of California, Berkeley. UF: Florida Museum of Natural History, University of Florida, Gainesville. USFC: United States Fish Commission. USGS: Register of Cenozoic localities of the U.S. Geo- logical Survey. USGS(MP): Branch of Paleontology and Stratigraphy, U.S. Geological Survey, Menlo Park, California. USNM: Department of Invertebrate Zoology (Mollusks), National Museum of Natural History, Smithsonian Institution, Washington, D.C. USNM(P): Department of Paleobiology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. YPM: Peabody Museum, Yale University, New Haven, Connecticut. ZMC: Zoological Museum, University of Copenhagen, Copenhagen. SYSTEMATICS Family Pectinidae Wilkes, 1810 [emend. Waller, 1978] Diagnosis.— Pectinacea having a ctenolium at least in early ontogeny. Discussion.— Authorship of the family name is usually attributed to Rafinesque (1815). The earlier use of the name by Wilkes (1810, as “Pectinoidae”) was brought to my attention by Eugene Coan (pers. comm., 1991). The arrangement of the enormous number of liv- ing and fossil taxa of Pectinidae into a coherent phyloge- netic arrangement above the species level is still far from Pectinidae Chlamydinae [+)) 5 5 € us o/s & abo she o38eR 3 Pea ee = € 2 £ auvtas 2 ee ge 5 o 8s © J=O92o9 § ECSeE |S aa Ae ese25 2 =EL8s |E ee wees BEdOoR F FGIS w a fF Bf Suzan & qooa x= So = ® Oo oO a oO Oo = = oO c [s) 2 > K/T Fig. 1. A phylogeny of major groups of Pectinidae. Lineages ending in arrows are extant; the one ending at a cross-bar is extinct. K/T refers to the Cretaceous-Tertiary boundary. Numbered blocks refer to apomor- phies: (1) resilium with non-fibrous core; (2) ctenolium; (3) radial plicae and straightened anterodorsal margin; (4) modified hinge teeth with dorsal and/or intermediate teeth dominant; (5) two-element hinge dentition with resilial teeth emphasized; (6) withdrawai of crossed-lamellar aragonite to inside pallial line; (7) shagreen microsculpture; (8) modified antimarginal striae in early ontogeny; (9) pitted left beak, simple ribs, and internal rib carinae; (10) resilial teeth extended and tending to parallel hinge (see text for further explanation). WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 199 being completed. Waller (1991, fig. 8) presented an hypothesis of the relationships of major groups and dis- cussed what little can presently be said about the synapo- morphies present at each node in a phylogenetic diagram. The phylogenetic events with which the present study is concerned are at finer levels (Fig. 1). In this figure, pairs of clades that comprise sister groups are shown by a U-shaped branching pattern; paraclades, which cannot yet be charac- terized on the basis of unique apomorphies, are shown as continuations of stem lineages. The tribe Chlamydini, for example, is a clade that is the sister group of a clade con- sisting of the tribes Crassadomini (new tribe), Mimachlamydini (new tribe), and Aequipectinini. The tribes Crassadomini and Mimachlamydini are each para- clades in the light of present knowledge. The tribe Crassadomini is the continuation of the stem group for the tribes Mimachlamydini and Aequipectinini; the tribe Mimachlamydini is a continuation of the stem group for the tribe Aequipectinini, the latter being a true clade in that its members share unique derived features. The family Pectinidae (Node 2 in Fig. 1) is rooted in a single but universal apomorphy, a ctenolium along the ventral edge of the byssal notch of the right valve (Waller, 1984). The subfamily Camptonectinae is viewed as the paraphyletic stem group from which all other Pectinidae were derived and is at present characterized by plesiomor- phic features. Specifically, the subfamily is sculpturally simple, without strong radial ribbing, microsculpture domi- nated by a simple pattern of broadly sweeping antimarginal striae (“Camptonectes striae”), and an anterodorsal disk margin that is concave in lateral view. (See Waller (1991) for a more detailed discussion.) The subfamilies Chlamydinae and Pectininae are joined at Node 3 (Fig. 1) by the advent of coarse ribbing and the straightening of the anterodorsal disk margin. As discussed by Waller (1991), these subfamilies (each referred to in that paper as a series of groups) were already in place well before the end of the Mesozoic era, where they are respectively exemplified by the genera Lyriochlamys Sobetski, 1977, and Microchlamys Sobetski, 1977. Although the subfamily names Chlamydinae and Pectininae used in the present study are based on the gener- ic names Chlamys and Pecten, each associated with a well- known shell shape, these shapes are in no way characteris- tic of an entire subfamily. The chlamydoid form is probably plesiomorphic, but the pectinoid form (as in Pecten, sensu stricto) and other forms such as those associated with the genera Amusium and Aequipecten have evolved repeatedly and independently in both subfamilies. At present these subfamilies are based mainly on extensive linkages between extinct lineages (Waller, unpub. data) and are diffi- cult to define or diagnose on the basis of universal synapo- morphies. In general, however, there is an approximate dif- ference in the style of hinge development, with the Chlamydinae (Node 5) emphasizing resilial teeth and the Pectininae (Node 4) emphasizing dorsal and intermediate teeth (Waller, 1991, fig. 6). Also in an approximate way, the Chlamydinae display greater microsculptural complexity than do the Pectininae. Enumeration of apomorphies within the Chlamy- dinae (i.e. above Node 5, Fig. 1) becomes more straightfor- ward and is taken up in the following discussion of tribes present in the Caribbean and adjacent regions. Subfamily Chlamydinae von Teppner, 1922 Discussion.— Based on comparison with taxa in Mesozoic outgroups such as the Camptonectinae and Lyriochlamys in the Chlamydinae, it seems likely that primitive characters in the Chlamydinae include commarginal lirae in inter- spaces, a small posterior auricle with a concave posterior margin, and microsculpture consisting of antimarginal stri- ae arrayed in a broad sweeping “Camptonectes” pattern that is continuous across radial ribbing. This pattern of antimarginal striae begins early in ontogeny at the edge of the prodissoconch on the left valve and at the edge of the prismatic stage on the right valve. The primitive radial rib- bing pattern is complex in the sense that rib introduction by both branching and intercalation is present in the same indi- vidual and commonly on the same valve. The ribs them- selves, however, are simple corrugations, without carinate edges on the shell interior (Waller, 1991, fig. 5). An important change occurred in shell microstruc- ture between Cretaceous Lyriochlamys and the modern Chlamydinae beginning in the late Cretaceous and early Paleocene (Waller, 1991; Waller and Marincovich, 1992; Fig. 1, Node 6). This involved the withdrawal of the area of deposition of crossed lamellar aragonite on the shell interi- or to a position bounded distally approximately by the pal- lial line, in contrast to Cretaceous and earlier Lyriochlamys, in which aragonite deposition extended nearly to the shell margins and comprised most or all of the hinge structures. The subfamily Chlamydinae can be subdivided into four extant tribes related as shown in figure |. The apo- morphies for each branching point are as follows: Node 7 (Chlamydini). The character that forms the basis for the tribe Chlamydini is shagreen microsculp- ture, a screenlike pattern formed by the offset contacts of frilled commarginal lamellae (Fig. 2; see also Waller, 1972, 1991, and Hayami and Okamoto, 1986). Although this structure is not universally present in all extant members of this tribe, the succession of morphologies among extant and fossil species suggests that it was present initially in the basal lineages contained in the clade. In contrast, no 200 AMER. MALAC. BULL. 10(2) (1993) Fig. 2. Shagreen microsculpture in center of left valve of Chlamys islandi- ca, USNM 764635, Recent, off Godhaven, Greenland (ht. of valve = 20 mm, scale bar = 1.0 mm). unequivocal shagreen microsculpture is known among extant or fossil members of the tribes Crassadomini (new tribe), Mimachlamydini (new tribe), and Aequipectinini, which collectively comprise the sister group of the Chlamydini. Where commarginal lirae are present in the Chlamydini, they begin after the start of radial ribbing. This allows the plesiomorphic sweeping pattern of antimar- ginal striae to be prominent in very early ontogeny, before the beginning of the radial stage on the left valve and con- tinuing at least through the early part of that stage. The more derived lineages in the Chlamydini, which will be dis- cussed in greater detail below (under Laevichlamys, new genus), lose both commarginal lirae and shagreen microsculpture altogether and develop numerous intercalat- ed costae in the interspaces of major ribs, tending to fill the interspaces completely (Fig. 4). As will be shown below, true Hinnites and the genus Pedum Lamarck, 1799, are both within this tribe. The subfamily names Hinnitinae Habe, 1977, and Peduminae Habe, 1977 (sic, emended to Pedinae) are therefore reduced in rank and made junior synonyms of Chlamydini. Node 8 (Crassadomini + Mimachlamydini + Aequipectinini). The only synapomorphy that unites these groups is the loss of the plesiomorphic sweeping pattern of antimarginal striae in the early radial stage. Instead, com- marginal lirae tend to dominate microsculpture at least between the proximal ends of the simple radial ribs (Figs. 5, 7). The microsculpture of the left-valve pre-radial region in the Crassadomini is variable, ranging from interrupted antimarginal striae (Fig. 7g) to a pitted or nearly smooth condition (Fig. 5d). There is a trend toward simplification of ribbing patterns. Medial intercalations, present on both valves in many Chlamydini and in Mesozoic Lyriochlamys, become rare on right valves except for a few examples of secondary origin of precisely medial intercalary costae (e.g. in Spathochlamys, new genus, described below). The plesiomorphic state of ribbing cross-sections at Node 8 (Fig. 1) remains as simple corrugations, without internal carinae on mature shells. The hinge also remains in its plesiomorphic state as in the Chlamydini, consisting of a simple two-element structure, with the dorsal teeth and resilial teeth being of about equal strength in most taxa. Node 9 (Mimachlamydini + Aequipectinini). Three apomorphies unite the tribes Mimachlamydini and Aequipectinini: (1) Left-valve beaks display a densely pit- ted microsculpture (Waller, 1991, figs. 4b, 9c,d), although as previously stated this feature was already making its appearance in some Crassadomini; (2) the ribbing pattern becomes further simplified, with the ribs that are introduced in early ontogeny commonly being the only ribs or at least remaining as the major ribs throughout ontogeny (Figs. 10, 11); (3) the edges of the ribs on inner surface of the shell become carinate (Figs. 10h, m). These carinae are variably expressed in populations of extant Mimachlamys varia but consistently developed in many other extant and extinct members of this clade. The hinge of the Mimachlamydini retains its basic plesiomorphic two-element structure, as in the Chlamydini and Crassadomini, although some extinct Mimachlamys show some thickening and incipient bifidity of the resilial teeth of the right valve. Shell shape remains Chlamys-like in the Mimachlamydini, as does the shape of Chlamydini He Laevichlamys | i= 3 | 5 g ERE 8 § e| 8 3| sas a) se os = 2 g oo SS B58 Ra a = ES Ss 2&5 = g > 532 & SES > > 2 = = L= oO € n & aR ro) Ofc 7e 7d 7c 7b 7a 7, 8 6 Fig. 3. A phylogeny showing derivation of the new genus Laevichlamys within the tribe Chlamydini (numbered blocks refer to apomorphies; those numbered 6, 7, and 8 are as shown in figure 1; see text for explana- tion). WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 201 the posterior auricle, which generally retains a concave pos- terior margin. At Node 10 (Fig. 1), the tribe Aequipectinini is marked by the onset of a hinge structure dominated by resilial teeth, which enlarge and extend anteriorly and pos- teriorly, nearly paralleling the dorsal teeth (Waller, 1991, pl. 5, fig. 14). There are attendant changes in valve symmetry, shape becoming more equilateral and the anterior and pos- terior auricles more equal in size. Lastly, there is a sculp- tural change in the early commarginal lirae, which display a ventrally concave looped trend on the rib flanks, at least in the early ontogeny of many taxa. This basic framework of four extant tribes in the Chlamydinae was already in place by the Eocene epoch, 36 million years ago (Waller, 1991). The main evolutionary radiation of these tribes, however, did not become extensive until the beginning of the Miocene, when the proto- Mediterranean was still open at its eastern end to the Indian Ocean and the Caribbean was still open to the eastern Pacific. The search for ancestors of extant Caribbean species in these tribes therefore should not be limited to the western Atlantic and eastern Pacific. Tribe Chlamydini von Teppner, 1922 Diagnosis.— Chlamydinae retaining in early ontogeny a microsculptural pattern of continous antimarginal striae that sweep across early radial ribbing; shagreen microsculpture present at least in early lineages. Discussion.— Six extant species in the Caribbean and adja- cent western Atlantic waters have traditionally been placed in the genus Chlamys, sensu stricto, because of their asym- metric auricles (the anterior one the longer) and deep byssal notch floored by an ontogenetically persistent ctenolium: Chlamys benedicti Verrill and Bush, 1897, C. imbricata (Gmelin, 1791), C. mildredae (Bayer, 1941), C. multisqua- mata (Dunker, 1886), C. ornata (Lamarck, 1819), and C. sentis (Reeve, 1853). In the present study, these species are found to represent three tribes in the Chlamydinae (Table 1), and none can be assigned to Chlamys unless that name is used so broadly as to encompass almost the entire sub- family. The sole extant member of the tribe Chlamydini in the tropical and subtropical western Atlantic is “Chlamys” multisquamata. This species, with its flattened form, flar- ing disk, and fine sculpture (Figs. 4a-i, 1, m), stands apart from the other Caribbean species and is strikingly similar to “Chlamys” squamosa (Gmelin, 1791) of the western Indo- Pacific (Figs. 4j, k, n). The resemblance extends beyond the flattened streamlined form of these species. There are also strong similarities in ribbing pattern, early microsculp- ture, color pattern, and the strong dorsal projection of the dorsal margin of the right anterior auricle. Resemblance even extends to the common presence of a tiny dense pat- tern of curved intersecting white pigment lines that appears on the beaks before the end of the prismatic stage of the right valve. This same pigmentation pattern occurs in sev- eral Indo-West Pacific chlamydoid species but is especially common in “Chlamys” squamosa and its close relatives. To appreciate whether these similarities are shared derived characters, and, if so, how a western Atlantic species can be related to a western Indo-Pacific species with no apparent intermediate geographic links in either the eastern Pacific or the eastern Atlantic, requires an overview of the tribe Chlamydini as here defined. Figure 3 diagrams a phylogenetic pathway through an array of species in the tribe Chlamydini that show important differences in microsculpture and ribbing pat- terns. It is important at this point to review the plesiomor- phic conditions that were present (Node 6) before the origin of the Chlamydini, because these conditions help to polar- ize the derived states that provide crucial phylogenetic information. The plesiomorphic character states at Node 6 are as follows: 1) Shell shape acline or prosocline, not markedly opisthocline or opisthogyrate, with a deep byssal notch. 2) Dorsal auricular margins not sharply folded and not projecting dorsally very far beyond the outer liga- ments. 3) Ontogenetic transgression of an inner layer of foliated calcite ventrally from the dorsal region across the umbonal region absent; aragonite (most commonly crossed- lamellar except in muscle attachment scars) present throughout the region inside the pallial line. 4) Posterior auricles small but with their posterior margins concave in shape, with the overall trend of the mar- gin forming nearly a right angle with the hinge line. 5) Antimarginal striae in a broad sweeping pattern with trends not significantly interrupted by radial ribbing, at least in the early ontogeny of the radial stage. 6) Left-valve beak sculpture exhibiting the same antimarginal pattern as in the early radial stage; beak secon- darily smooth in some taxa, but seldom coarsely pitted. 7) Commarginal lirae present in interspaces dur- ing at least part of ontogeny. 8) Several modes of rib introduction present on same shell, including branching, rib-flank intercalation, sub-medial intercalation, and medial intercalation on a sin- gle valve. 9) Ribs in the form of simple radial folds, without 202 AMER. MALAC. BULL. 10(2) (1993) thickening of edges on the shell interior, i.e., without inter- nal rib carinae. The base of the Chlamydini (Node 7, Figs. 1, 3) is marked by the advent of shagreen microsculpture (Fig. 2). The primitive expression of this is probably as small patch- es in rib interspaces on the disk and auricles. Extensive sha- green over virtually the entire shell seems to be a derived condition that has been reached independently in different species lineages, most notably in the Semipallium group (see Waller, 1991, fig. 10). The lineage marked “Chlamys, etc.” on Figure 3, Node 7a, marks a polytomous assemblage of groups some of which appeared early in the radiation of the tribe. Among these are genera such as Chlamys, Zygochlamys, Talochlamys, Semipallium, and even the Patinopecten group, each characterized by a set of autapomorphies. Chlamys, sensu stricto, contains species exemplified by (1) the type species of the genus, Chlamys islandica (Miller, 1776), living in the northeastern and northwestern Atlantic, (2) a group of closely related northern Pacific species including C. albida (Arnold, 1906, [ex Dall MS]), C. behringiana (Middendorf, 1849), C. rubida (Hinds, 1845), and C. hastata (Sowerby, 1842), and (3) the eastern Atlan- tic cemented species “Hinnites” ercolanianus (Cocconi, 1873). These taxa are all united by a byssal notch that is shallower than the plesiomorphic state, by a foliated-calcite transgression, and by an early irregularity in the antimar- ginal microsculpture before the advent of the first shagreen microsculpture. The expression of shagreen microsculpture is highly variable among and within species of Chlamys, s.s., and even within population samples. In the eastern Pacific, for example, there is some evidence of a southward disappearance of shagreen microsculpture in the sequence C. albida, C. hastata hastata, C. hastata hericea, a sequence that also displays increasingly derived macro- sculptural patterns. Chlamys, as here restricted, appears to have origi- nated in the Pacific and entered the eastern Pacific and northern Atlantic via northern routes (Waller, 1991). Chlamys islandica itself penetrated as far south as the Mediterranean during cold pulses of the Pleistocene (Roger, 1939: p. 170). “Hinnites” ercolanianus (Cocconi, 1873), of which Hinnites absconditus P. Fischer in Locard, 1883, is a junior synonym (Adam, 1960), presently lives in the eastern South Atlantic off the Cape Verde Islands and along the Atlantic coast of Africa from the Gulf of Guinea (4° 44’N) to northern Angola (8° 30’S) at depths in the 100 to 200 m range (Adam, 1960; Waller, unpub. data). The liv- ing members of this species, however, are but a relict of a much more widespread occurrence in the Pliocene of the Mediterranean (Raffi, 1971) and the Miocene of Belgium (Glibert, 1945, referring to the Anversian or Serravalian stages). The ancestor of this species, which is considered to be “H.” brussoni (de Serres, 1829) from the Burdigalian of southern France (Roger, 1939: p. 175), has shagreen microsculpture. In spite of the long history of Chlamys, s. s., no member of this genus is known to be living in the present- day tropical American regions that are the subject of the present study. As will be shown below, taxa from these regions that have been referred to Chlamys in fact belong to other genera with very different histories. Three derived features are important at Node 7b (Fig. 3): (1) the advent of a somewhat opisthocline and opisthogyrate shell form; (2) a straightening (loss of con- cavity) of the posterior margins of the posterior auricles, these margins becoming straight or even convex; and (3) an increase in the frequency of rib introduction by intercala- tion. Shagreen microsculpture is retained at this node, as is also the plesiomorphic absence of a foliated calcite trans- gression on the shell interior. This stage of phylogenetic development is characterized by the genus Azumapecten Habe, 1977, of which A. farreri (Jones and Preston, 1904) is the type species. The shell shape and deep byssal notch suggest that this level represents an increased adaptation to byssate life, a conclusion that is corroborated by increas- ingly common shell irregularity in response to growth in confined spaces. Geographically, this stage of development among extant species seems to be limited to the western Indo-Pacific. At Node 7c (Fig. 3), the following derived fea- tures appear: (1) increase in rib introduction by intercala- tion to the point that rib interspaces are eliminated in late ontogeny; (2) the ontogenetically late appearance of the first radial ribs, allowing a zone of prominent coarse anti- marginal striae to develop in the pre-radial stage of the left valve; (3) disappearance of shagreen microsculpture from at least the center of the disk and usually from the entire shell; (4) trend of posterior margins of posterior auricles forming an oblique angle with the hinge line, the margins themselves being most commonly nearly straight or only slightly concave or convex. Also at this stage is a common presence of a minute net of white pigment lines in very early ontogeny, still within the prismatic stage of the right valve. This stage and the following one characterized by rib crowding and the ontogenetic disappearance of interspaces encompass species placed in Laevichlamys, new genus. The stage at Node 7c is best represented by species such as L. irregularis (G.B. Sowerby II, 1842) and L. wilhelminae (Bavay, 1904) (see illustrations in Waller, 1972b, where these species were referred to as Chlamys irregularis and C. marshallensis Waller, 1972, the latter name being a junior synonym of Bavay’s name). Node 7d (Fig. 3) is marked by (1) decreased shell WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 203 convexity, (2) dorsal projection of the dorsal margins of the right-valve auricles, particularly that of the right anterior auricle, and (3) marked obliquity and common convexity of the posterior auricular margins. This stage is exemplified by Laevichlamys squamosa (Figs. 4}, k, n). Finally, Node 7e is marked by the advent of sharp recurvature of the right anterior auricle’s dorsal margin and a recurvature of the dorsal margin of the right posterior auricle to produce a slight trough along this margin (Fig. 4c, arrow). The convex curvature of the posterior margin of the posterior auricles and the trend of this margin to pro- duce a very oblique angle with the hinge (Fig. 4g) are both more extreme than in L. squamosa (Fig. 4k). These fea- tures represent the most derived states of characters already present at Node 7b, and this stage is represented by a single extant species, L. multisquamata of the western Atlantic. The morphological trends from Azumapecten to advanced Laevichlamys are not exclusively exemplified by these genera, because there are other generic names that have been introduced on the basis of “degree of difference” considerations. The monotypic Indo-west Pacific genus Pedum Lamarck, 1799, for example, exhibits a bizarre shell form because it lives embedded and entrapped in massive hard corals such as Porites (Yonge, 1967; Waller, 1972; Kleemann, 1990). The early growth stage of this taxon, however, has closely packed intercalated ribs, a projecting right anterior dorsal margin, and oblique posterior auricles that suggest derivation from above Node 7d (Fig. 3), proba- bly from Laevichlamys squamosa or a similar species. Coralichlamys Iredale, 1939, a less bizarre, irregularly shaped, secondarily commarginally lamellate genus that lives embedded in coral in the Indo-Pacific, may also be monotypic. Its morphology suggests derivation from above Node 7b (Fig. 3). The Indo-Pacific genus Scaeochlamys Iredale, 1929, represented by two extant species, S. livida (Lamarck, 1819) and S. tegula (Wood, 1828), is character- ized by a strongly opisthogyrate form, flattened right valve, and hypertrophied scales on major ribs. Its auricular shapes and presence of shagreen both indicate that it is above Node 7b. Based on the geographic distribution of living and fossil species, the early evolutionary radiation of the Chlamydini above Node 7b (Fig. 3) was apparently mainly within the tropical to warm-temperate western Pacific and Indian oceans, because no primitive Laevichlamys are known from the eastern Pacific, the western Atlantic, or the eastern Atlantic. This Indo-Pacific evolutionary radiation of Laevichlamys seems to have been well underway by the Late Miocene, but the extent of the radiation by that time is difficult to judge because of the very meager Miocene fos- sil record on low, reef-dominated tropical Indo-Pacific islands. If the radiation of advanced Laevichlamys (Nodes 7c and 7d, Fig. 3) occurred no earlier than Late Miocene, then its dispersal from the Indian Ocean to the Caribbean by way of the proto-Mediterranean would not have been possible, because the eastern portals of the Mediterranean were effectively closed by the end of the early Miocene (late Burdigalian; Adams et al., 1983, 1990; Piccoli et al., 1986; Por, 1989). Alternatively, entry into the Atlantic may have been by way of chance dispersal from the Indian Ocean around the southern tip of Africa. In either case, one would expect some evidence of the presence of advanced Laevichlamys, either living or fossil, in the eastern Atlantic. The search for evidence of dispersal of Laevichlamys between the Indian Ocean and the Atlantic has led to one of the most surprising findings of the present study. A probable descendant from Laevichlamys turns out to be an extant cemented species of the eastern Atlantic, Hinnites corallinus (G.B. Sowerby I, 1827), which in turn may prove to be the same as the presumably extinct type species of the genus Hinnites, H. crispus (Brocchi, 1814), known mainly from the Mediterranean Pliocene. The evi- dence that has led to this interpretation is as follows: 1) Specimens reported by Kensley (1985) from Namibia were determined by me (pers. comm. to Kensley, 1985) to represent two species of “Hinnites”. One of these, found in Cape Province, South Africa, is the familiar “Hinnites” ercolanianus (Cocconi, 1873), previously known from the Angolan coast (see Adam, 1960). As men- tioned above, “H.” ercolanianus is derived from a sha- green-bearing ancestor and is probably referable to Chlamys, sensu stricto. In the same personal communica- tion quoted by Kensley, I suggested that the second species from Namibia is possibly new. It was distinguished from the former on the basis of the macro- and microsculpture of the early Chlamys stage. As shown in Figures 40 to 4r, this Namibian Hinnites has a pattern of rib introduction like that present in Laevichlamys, where rib interspaces are filled by the repeated medial intercalation of new radial costae. Furthermore, a close relationship to L. squamosa is indicated by auricular shapes, the dorsal margin of the right anterior auricle extending prominently above the hinge line and the posterior margin of the posterior auricle exhibiting a convex outline tending to form an oblique angle with the hinge. Lastly, some specimens of the Namibian species show the minute net of white pigment lines that is common in species above Node 7b in Figure 3. Although Kensley (1985) thought that the specimens of this Namibian species were fossil, the immature shells show little sign of wear and have adhering ligamental material. 2) Cosel and Gofas (1984), in a paper that I received after the aforementioned identification for Kensley (1985) was completed, described a new cemented pectinid species, Hinnites spectabilis Cosel and Gofas, 1984, which 204 AMER. MALAC. BULL. 10(2) (1993) they reported as living along the southern part of the coast of Angola. Their description and illustrations leave no doubt that this is the same as the Namibian species. 3) Cosel (pers. comm., August, 1992) examined the type specimen of Hinnites corallinus G.B. Sowerby I, 1827, originally said to come from East Africa, and deter- mined that it is the same as the species that he and Gofas (1984) had described as new. The East African locality is apparently in error. 4) The macro- and microsculpture of the Chlamys stage of Hinnites corallinus is remarkably similar to that of the type species of Hinnites. The latter is H. crispus (Brocchi, 1814), a fossil originally described from the Pliocene of Italy. The fossil species tends to be larger in size and thicker shelled than the living one, but there are broad similarities in form as well as in sculptural detail (see Roger, 1939). It is possible that these differences are ecophenotypic and that these species may be synonymous, in which case the living east African specimens are but a relic of a previously more broadly distributed species. 5) If it is true that Hinnites crispus is derived from an ancestral species in the new genus Laevichlamys, then the fossil record of H. crispus tends to support the idea that Laevichlamys did not enter the Atlantic by westward dis- persal across the proto-Mediterranean. This is because the oldest specimens of H. crispus are known not from the Mediterranean but from the Atlantic side of France (Helvetian of the Aquitaine Basin and Loire Valley; Roger, 1939). Although the species has been reported from the late Miocene (Tortonian) of Austria, Hungary, and Bulgaria (Cuenca, 1980: 62; Raffi: 1971:124), it did not become widely distributed in the Mediterranean until the Pliocene, at which time it also extended as far as the British Isles (Roger, 1939: 174). Roger (1939: 174) thought that the species persisted into the late Pleistocene (Sicilian) in the western Mediterranean based on specimens from the Alpes- Maritimes region of France. Unfortunately, there is no Miocene or Pliocene fossil record of H. crispus (or H. corallinus) in southern Africa, nor is there any fossil record of a preceding free-living Laevichlamys in the eastern Atlantic. 6) The fossil record of the Caribbean species, Laevichlamys multisquamata, discussed below, is no older than Pleistocene. A reasonable interpretation of this evidence is that a species of advanced Laevichlamys had entered the eastern Atlantic from the Indian Ocean by late Miocene time via a route around southern Africa. A derivative of this species adopted a cemented mode of life, evolved into true Hinnites, dispersed into the Mediterranean through its west- ern portal after the end of the Miocene, and nearly became extinct during the late Pliocene and early Pleistocene cli- matic cooling that profoundly affected the early Pliocene fauna of the Mediterranean (Raffi et al., 1985). If the extant Hinnites corallinus is indeed synonymous with H. crispus, then the species survives today in refugia along the west African coast. It seems likely that another propagule of Laevichlamys dispersed to the tropical Western Atlantic and evolved, during the Pleistocene, into Laevichlamys multi- squamata. This species has no geminate sister species in the eastern Pacific because it originated in the tropical western Atlantic region after the closure of connecting sea- ways. Laevichlamys, new genus Etymology.— The name Laevichlamys is derived from the Latin word levis (or laevis), meaning smooth, with refer- ence to macrosculpture of low relief, and the genus name Chlamys. Diagnosis.— Non-cemented Chlamydini with shagreen microsculpture secondarily absent at least on central sector of disk and commonly on entire shell; radial ribs initially low, tending to originate in ontogeny well after end of pris- matic stage, and preceded by zone of uninterrupted anti- marginal striae; rib introduction mainly by repeated interca- lation medially in interspaces of preceding ribs on both valves, nearly or entirely filling interspaces between ribs; regular commarginal lirae absent. Type species.— Pecten multisquamatus Dunker, 1864, liv- ing, tropical western Atlantic. Other species.— In addition to the type species, extant species included in the new genus are as follows: Pecten irregularis G.B. Sowerby II, 1842; P. lemniscatus Reeve, 1853; P. limatulus Reeve, 1853; P. mollitus Reeve, 1853; P. ruschenbergeri Tryon, 1869; Ostrea squamosa Gmelin, 1791; and Chlamys wilhelminae Bavay, 1904. All of these live in the Indo-Pacific region. There is still much to learn about the fossil record of this genus, but two extinct species are included thus far: Pecten (Chlamys) lauensis Ladd in Ladd and Hoffmeister, 1945, Fiji, late Miocene or Pliocene; P. shirahamaensis Nomura and Niino, 1932, Japan, Early Pliocene (Masuda, 1962, p. 185). Geographic range.— Western Indo-Pacific and western Atlantic. Stratigraphic range.— Upper Miocene to present. Discussion.— As argued in the preceding section dealing WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 205 with the tribe Chlamydini, the new genus Laevichlamys probably has a tropical Indo-Pacific origin, and morpholog- ically advanced species entered the eastern Atlantic in the late Miocene probably via a southern African route of dis- persal. The former presence of the genus in the Mediterranean and eastern Atlantic is indicated by the pres- ence in these areas of Hinnites, which likely evolved from a species of Laevichlamys. Laevichlamys is paraphyletic in that specialized, derived genera such as Hinnites DeFrance, 1821, and Pedum Lamarck, 1799, are excluded (see preceding discus- sion of the tribe Chlamydini). The alternative to achieve a holophyletic status would be to broaden the concept of the genus to include all of these, but the name based on priority would then become Pedum, a taxon well-known for its highly unusual form and specialized living habit embedded in massive corals. This would clearly raise havoc with the morphological concepts that have long been attached to particular names. Laevichlamys multisquamata (Dunker, 1864) (Figures 4a-i, 1, m) Pecten multisquamatus Dunker, 1864: 100, Recent, Havana Bay, Cuba. Pecten multisquamatus Dunker, 1864. Dunker, 1865: 67, pl. 23, figs. 1-3. Pecten effluens Dall, 1886: 219, Recent, off Havana, 127 fathoms [232 m]. Chlamys effluens (Dall, 1886). Verrill, 1897: 59. Chlamys multisquamata (Dunker, 1864). Waller, 1973: 41, figs. 9-11. Types.— The specimen illustrated by Dunker (1865), a pair of matching valves, is of the same dimensions given in his original description in 1864 and would be the logical choice for a lectotype. It is not known, however, whether this specimen still exists. I did not find it among Dunker’s material in the collections of the Humboldt University Museum of Berlin, raising the possibility that it may still be in Cuba, perhaps in the collection of J. Gundlach, from whom Dunker received his specimens. According to Dr. José Espinosa of the Instituto de Oceanologia of Cuba (pers. comm., 1993), however, no specimens of Pecten mul- tisquamatus are present in that collection. A second, small- er pair of matching valves from Cuba that bears one of Dunker’s manuscript names is housed with some of Dunker’s collection in the Humboldt University Museum in Berlin. This specimen, a pair of matching valves collected by Felipe Poey from Cuba, is selected herein as the lecto- type of P. multisquamatus. Its dimensions are height, 43.1 mm, length 39.5 mm, convexity of articulated valves 8.0 mm, length of anterior outer ligament 10.8 mm, and length of posterior outer ligament 5.0 mm. Dall (1886) provided no illustration with his origi- nal description of Pecten effluens but referred to the largest valve as having a height of 26.0 mm and a width of 22.0 mm. In the type collection of the National Museum of Natural History, the lot marked “types” (USNM 62236) contains a left valve and a smaller, non-matching right valve. The left valve, although conforming to Dall’s description, is only half the size that he stated. Three years later Dall (1889, pl. 42, fig. 9) illustrated a Pecten effluens which he said had a height of 26 mm. This specimen, a left valve, corresponds exactly in shape and ribbing pattern to the left valve in USNM 62236 that is only half the size. It is assumed that Dall misstated the measurements from an enlarged drawing provided by an artist, and the left valve (ht. = 13.2 mm) in USNM 62236 is herein selected as the lectotype of P. effluens (Fig. 4h). Type locality Havana Bay, Cuba. Diagnosis.— Laevichlamys of low convexity and stream- lined, flaring form, the umbonal angle exceeding 90° for specimens greater than 20 mm in height; dorsal margin of right anterior auricle sharply folded and dorsal margin of right posterior auricle with slight trough; posterior margins of posterior auricles convex, producing distinctly oblique angle with hinge; ribbing retaining low relief throughout ontogeny, with all ribs and riblets having low, tiny, closely spaced scales. Morphological variation.— Laevichlamys multisquamata attains a moderate size, the largest specimens having a shell height of about 70 mm. The amplitude of the first-order radial plicae can vary considerably (compare Figs. 4b and 4f). The surfaces of some specimens are nearly flat, with only the fine costae present or with the first-order radial pli- cae limited to early ontogeny. The yellowish tinge on the umbones illustrated by Dunker (1865) and referred to by Dall (1886) is common but not present on all specimens. Comparison.— The morphology of Laevichlamys multi- squamata is unique among chlamydoid scallops in the western Atlantic and eastern Pacific. The early, Chlamys stage of Hinnites corallinus of the eastern Atlantic off west Africa (see preceding discussion of the tribe Chlamydini) is similar but differs in having more prominent and persistent first-order radial plicae and in having less numerous sec- ondary intercalated radials. Among Indo-Pacific Laevichlamys, the species that is closest in morphology is L. squamosa (Waller, 1972, pl. 3, figs. 38-41, and Figs. 4), k, n herein). The latter differs from L. multisquamata in 206 AMER. MALAC. BULL. 10(2) (1993) Fig. 4. Laevichlamys (a-m) and Hinnites (n-q). a, b. L. multisquamata, USNM 764750, Guadeloupe, West Indies, matching right and left valves, height 26.1 mm. c,e. L. multisquamata, USNM 846201, off Boynton Beach, Florida, dorsolateral view of right dorsal margin showing trough (arrow) on posterior auricle and interior view of right hinge, hinge length 7.2 mm. d, f. L. multisquamata, USNM 710926, south coast at St. Luce, Martinique, West Indies, matching right and left valves, height 47.9 mm. g, i, 1. L. multisquamata, USNM 764749, La Chorerra, Havana, Cuba, right dorsal exterior, hinge length 23.2 mm., non-matching left dorsal exterior, hinge length 14.2 mm., and details of sculpture on same right and left valves, horizontal field widths both 26 mm. h. L. multisquamata, USNM 62236, probably Dall’s illustrated specimen of Pecten effluens, off Havana, Cuba, left valve, height 13.3 mm. j, k, n. L. squamosa, USNM 846209, Manaua Id., Fiji, dorsal exteriors of matching left and right valves, hinge length 20.3 mm., and detail of sculpture on left valve, horizontal field width 25 mm. o-r. H. corallinus, USNM 782400, Bosluisbaai, Namibia, left valve exterior, height 24.4 mm; another left valve, height 38.0 mm; right valve matching preceding; and detail of exterior of preceding left valve, horizontal field width 12.5 mm. WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 207 having a narrower umbonal angle (less than 90°) and hence a less flaring aspect, a tendency for scales to be present only on the higher order radial ribs, and in having a less sharply refolded dorsal margin on the right anterior auricle. Living habits.— Laevichlamys multisquamata lives on tropical or subtropical coral reef fronts, between coral or dead shells or in crevices, at depths generally greater than 30 m to well over 100 m; dead shells have been dredged from depths greater than 200 m (Waller, 1973: 47; Abbott, 1974: 443; Sutty, 1986: 102). Geographic range.— Throughout the Antilles from Barbados northward to southeastern Florida, the Bahamas, and Bermuda (Waller, 1973: 47) and southward to Brazil. The Brazilian record, not previously reported, is based on a right valve found by me in the Paris Museum Collection, where it had been incorrectly identified as Chlamys ornata. It is from Calypso Station 45, 11° 22.5’S, 37° 10’ W, from a depth of 31 m. A second specimen from Brazil, but with- out locality details, is present in the BMNH collection. A label indicates that it had been identified by Bavay as “squamosus var.” In the southern Caribbean, the species extends westward to Panama (based on a living specimen dredged by the R/V Pilsbry of the University of Miami at a depth of 51 m at 9° 24.8’N, 78° 12.7’W (R/V John Elliott Pillsbury Sta. P-417). The species is thus far known from the Gulf of Mexico only from the deep Flower Garden Reef off the Texas coast (Boone, 1978). Stratigraphic range.— Lower Pleistocene to Recent. Laevichlamys multisquamata has not previously been reported from the fossil record. In the present study, specimens of this species were found in unidentified collec- tions from the Pleistocene reef deposits of Barbados (BMNH), from undated beds in northern Cuba that are probably no older than Pleistocene in age (PRI), and from the Moin Formation of Costa Rica (USNM). Also, D. G. Robinson (pers. comm., August, 1992) reported finding this species in the topmost beds of the Moin Formation of Costa Rica between plates of the coral, Agaricia Lamarck, 1801. The Moin Formation, considered in the past to be early to middle Pleistocene in age (Akers, 1972: 44; Robinson, 1990, 1992), has more recently been considered to include upper Pliocene sediments, although the uppermost part of the formation is still dated as early Pleistocene (Coates et al. 1992). Discussion.— As outlined in the previous discussion of evolution within the tribe Chlamydini, it seems very likely that the ancestor of Laevichlamys multisquamata was a species present in the eastern Atlantic during the Late Miocene and Pliocene. Although L. multisquamata at pre- sent has a broad distribution in the Caribbean region and has been collected from the Caribbean coast of Panama, no geminate or sister species are known from the eastern Pacific. This would suggest that the origin or at least the dispersal of the species to the western Atlantic has occurred since the closure of seaways by the rise of the Isthmus of Panama, i.e. since the middle Pliocene. This is corroborated by the negative evidence of the fossil record; thus far no fossils of this species or probable ancestors have been found in rocks that are known to be older than Pleistocene. Material examined.— Recent material: USNM: 16 lots containing 33 specimens, from southeastern Florida, Bermuda, Cuba, Puerto Rico, the Bahamas, Jamaica, Martinique, Barbados, and Panama. DMNH: 2 lots contain- ing 2 specimens, from the Bahamas and Jamaica. BMNH: 2 lots containing 2 specimens, from Guadaloupe and Brazil. MNHN: 2 lots containing 6 specimens, from Guadeloupe and Brazil. HUB: | lot containing | specimen, from Cuba. Fossil material: BMNH: several specimens, Pleistocene Coral Rock at Bathsheba and Highgate, Barbados. PRI 1257k: a single articulated shell from the D.K. Palmer Collection, Locality 898P: Road cut on Carretera Central, just west of Nena Machado Hospital [“= Bermudez 5”], Matanzas Province, Cuba. [Locality pub- lished in Palmer, 1948.] USNM: USNM(P) 474809, a frag- mented left valve, USGS 18693, “colline en démolition,” Lim6n, Costa Rica. Tribe Crassadomini Waller, new tribe Diagnosis.— Chlamydinae with shell shape and two-ele- ment hinge of Chlamys, but without continuous antimargin- al striae before beginning of radial ribs; prominent com- marginal lirae present in rib interspaces in early ontogeny of left valve or both valves. Left beak microsculpture with discontinuous antimarginal striae, pitted, or secondarily smooth. Rib introductions occurring over much of ontoge- ny, leading to variously clustered and ordered ribbing pat- terns; internal rib carinae absent in mature shells, rarely present but weakly developed in juvenile shells. Discussion.— The pre-radial stage of the left valve in the Crassadomini lacks the plesiomorphic continuously striate condition found in the Chlamydini. In the Crassadomini, commarginal lirae are more prominent than antimarginal striae at the start of the radial stage (Figs. 5, 6), whereas in the Chlamydini, antimarginal striae are more prominent than the commarginals at the same growth stage. No sha- green microsculpture has been detected in any member of the Crassadomini, whereas this feature is present in at least 208 AMER. MALAC. BULL. 10(2) (1993) the early lineages in the Chlamydini. The style of rib introduction in the Crassadomini is plesiomorphic, with introduction on the right valve being primarily by branching and that on the left valve by interca- lation (Figs. 5, 6). This differs from the Azumapecten- Laevichlamys clade in the Chlamydini, where intercalation is important on the right valve as well as on the left (Fig. 4). Both the Mimachlamydini and Aequipectinini have simple Fig. 5. Scanning electron micrographs of prodissoconchs (a-c), pre-radial stage microsculpture (d-f), and early radial stage sculpture (g-i) of left valves of species of Crassodoma. a, d, g. C. multistriata, USNM 764329, San Pedro Bay, Sao Vicente, Cape Verde Ids. b, e, h. C. pusio, USNM 196540, Shetland Ids., Scotland. c, f, i. USNM 764941, off Long Point, Santa Catalina Id., California. Arrows point to approximate position of PI/PII boundary. Projecting object on anterodorsal side of b is not part of prodissoconch. Scale bars: a-c = 25 pm, d-i = 200 um. ribbing patterns in which the primary ribs remain dominant over secondary ribs throughout ontogeny; the edges of these primary ribs have prominent carinae on the inner shell surface (Waller, 1991, fig. 5b; Figs. 10h, m, herein). In contrast, rib introductions in the Crassadomini are spread unevenly during ontogeny to produce variously ordered or clustered rib patterns; these ribs in adults lack internal cari- nae. Both the Mimachlamydini and Aequipectinini have a WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 209 Fig. 6. Crassadoma (a-l) and Caribachlamys (n-r). a, b. Crassadoma gigantea, USNM 764946, channel reef, Egmont, British Columbia, Canada, match- ing left and right valves before cementation, height 16.0 mm. c, d. C. multistriata, USNM 196495, Naples, Italy, matching right and left valves, height 30.0 mm. e, f. Lectotype of Pecten tinctus Reeve (= C. multistriata), BMNH 1981247/1, locality unknown, matching right and left valves, height 32.5 mm. g, h. Lectotype of P. effulgens Reeve (= C. multistriata), BMNH 1993039/1, locality unknown, matching right and left valves, height 18.8 mm. i. Lectotype of P. textilis Reeve (= C. multistriata), BMNH 1993040/1, locality unknown, single left valve, height 24.9 mm. j. paralectotype of P. textilis Reeve (= C. multistriata), BMNH 1993040/2, locality unknown, single right valve, height 24.4 mm. k, 1. C. pusio, USNM 196511, Langland Bay, Wales, matching right and left valves before cementation, height 13.7 mm. m, n. Caribachlamys sentis, BMNH 1993041/1, lectotype, locality unknown, matching left and right valves, height 15.9 mm. 0, p. C. sentis, USNM 599341, Venetian Causeway, Miami, Florida, matching right and left valves, height 32.9 mm. q, r. C. ornata, USNM 764717, Plantation Key, Florida, matching right and left valves, height 32.5 mm. 210 AMER. MALAC. BULL. 10(2) (1993) more distinctly pitted microsculpture on the left beak in the pre-radial stage compared to the pattern in the Crassadomini, where there is a complex interplay of dis- continuous antimarginal striae, pits, and commarginal lines (Figs. 5, 6). The new tribe Crassadomini contains only two genera: Crassadoma Bernard, 1986, and Caribachlamys, new genus. Crassadoma Bernard, 1986 Original diagnosis.— ‘Juvenile a typical Chlamys, equiv- alve, biconvex. Right valve ornamented with bifurcating weakly imbricated riblets. Anterior auricle long, with imbricated radial sculpture. Byssal notch deep, ctenolium with six teeth. Posterior auricle small, wide. Left valve with 10 to 15 spinose ribs, separated by three small weakly imbricate riblets. Free or byssiferous. Adult cemented to substrate by right valve. Shell ponderously thickened, irregular, auricles and byssal notch obsolete. Right valve idiomorphic flat or deeply cupped, ornamented with con- centric lamellae or with radial rows of imbricated scales. Left valve flattened, with radial rows of imbricated ribs. Hinge line distorted, displaced resulting in large cardinal area.” (Bernard, 1986: 72). Emended diagnosis.— Byssate or cemented Crassadomini with normal prodissoconch (small PI stage and large PII); antimarginal striae absent or weakly developed between commarginal lirae in rib interspaces in early ontogeny; medial intercalation of secondary riblets in rib interspaces recurrent throughout ontogeny of left valve. Type species.— Lima gigantea Gray, 1825, by original designation (Bernard, 1986). Other species.— Ostrea pusio Linnaeus, 1758, extant, eastern Atlantic; O. multistriata Poli, 1795, extant, eastern Atlantic; Chlamys harmeri Regteren Altena, 1937, Plio- Pleistocene, Europe. Geographic range.— Eastern Pacific and eastern Atlantic. Stratigraphic range.— Early? Miocene, Middle Miocene to present. (See following section on the type species, Crassadoma gigantea.) Discussion.— Bernard (1986) included only one species in his new genus, the extant “Hinnites” giganteus of the east- ern Pacific, and asserted that shell attachment to the sub- strate in this species evolved independently of the same habit among other extant cementing species (see above dis- cussion of tribe Chlamydini). He also contended that Crassadoma is an obligatory cementer, whereas Hinnites is only a facultative one, its right valve merely appressed against the substrate, not cemented. Unfortunately, Bernard (1986: 71) misstated the type species of Hinnites to be Hinnites distortus (DaCosta, 1778) (= H. pusio) and gave no information on H. crispus, the correct type species. Bernard’s (1986) diagnosis of Crassadoma, quoted above, is merely a description of the type species and does not serve to differentiate this from other genera that may or may not assume a cemented life habit. Harper (1991: 193) found that although some specimens of “Hinnites” pusio may be uncemented and merely wedged into the substrate by their irregular growth, the cemented forms are cemented in the same manner as Crassadoma and other “Hinnites”’. In the present study, the concept of Crassadoma is expanded to encompass both non-cemented and cemented Crassadomini that have a plesiomorphic normal prodisso- conch (small PI and large PII stage, Fig. 5). This prodisso- conch morphology distinguishes Crassadoma from the con- tribal new genus Caribachlamys, all species of which share a derived state of the prodissoconch (large PI and small PII stage, Fig. 7). The two genera also differ in the develop- ment of antimarginal striae between the commarginals. In Crassadoma these striae are obscure or absent (Fig. 5); in Caribachlamys they are strong (Figs. 71, j) and in the more derived species cause the commarginal lirae to assume irregular trends (Fig. 71). Cemented species of Crassadoma are well separated on the basis of the macro- and micro- sculpture of the Chlamys stage from other “Hinnites”, most of which fall within the tribe Chlamydini (see preceding discussion of the tribe Chlamydini). Evidence from the fossil record suggests that cementation in the Crassadomini has evolved independently and at different times in Crassadoma gigantea of the east- ern Pacific and in C. pusio of the eastern Atlantic (see fol- lowing sections on species). Like all cemented Chlamydinae, these species display changes in morphology resulting from cementation and growth in a confined space, including ventral migration of the ligament system and increased distance between the pallial line and the shell margin (Yonge, 1951; Waller, 1972, 1991; Harper, 1991). Like many species adapted to cooler waters, whether cemented or not, these cemented species have a prominent transgression of foliated calcite ventrally across the origi- nally aragonitic region of the umbonal interior. These are features that occur in each independent origin of “Hinnites”, including the two examples of cemented species in the Chlamydini described above. A foliated-cal- cite transgression has also evolved in many species in other clades both within and outside of the subfamily Chlamy- dinae (Waller, 1991; Waller and Marincovich, 1992). WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 21 Crassadoma gigantea (Gray, 1825) (Figs. 5c, f, i, 6a, b) Lima gigantea Gray, 1825: 139. No locality specified. Hinnita giganteus Gray, 1826: 103. Pecten (Chlamys) multirugosus Gale, 1928: 92. New name for Hinnita gigantea Gray. Refer to Grau (1959: 134) and to Roth and Coan (1978) for details of synonymy and nomenclature. Types.— The holotype of Lima gigantea Gray, 1825, which Grau (1959: 136) said is in the British Museum (Natural History), was not examined. Type locality.— “Juan de Fuca Strait (between Vancouver Island, British Columbia, Canada, and the state of Washington, U.S.A.” (Grau, 1959: 136). Diagnosis.— Cemented Crassadoma; left valve with scaly radial ribs of at least three orders, with total number of ribs of all orders exceeding 100 at distal margin in mid-ontoge- ny; rib number commonly decreasing in late ontogeny due to disappearance of lower order riblets; first and second order ribs of left valve remaining distinctly higher in relief than intervening lower order riblets in mid-ontogeny; right valve with fasciculated radial ribs in early ontogeny and at least two orders of ribs in mid-ontogeny; interior of shell vivid purple in color on auricles and hinge plate, foliated calcite transgression extensive, completely covering area inside of pallial line except for muscle scars. Morphological variation.— During the Chlamys stage of ontogeny of “Hinnites”, before cementation to the substra- tum by the right valve, shell attachment is by means of a byssus, with the right valve having a deep byssal notch, well-developed ctenolium (Waller, 1984), and uniform shell curvature. At the end of this stage the mantle of the right valve loses its structural integrity and begins to conform to the substratum surface. Regular sculpture disappears where cementation occurs, but sculptural patterns return whenever the mantle lifts away from the substratum. In Crassadoma gigantea, the shell height of the Chlamys stage (Figs. 6a, b) varies from about 15 to 30 mm even within population sam- ples (e.g. USNM 764937, Newport Beach, California). Although no geographic trends in maximum size were noted among specimens examined, Hertlein (1972: 212) indicated that size increases northward. The largest speci- men that I observed is from Alaska (USNM 678351, Prince of Wales Island, ht. = 24 cm), but Hertlein (1972, p. 212) mentions a “huge specimen” of 23 cm shell height from Santa Cruz Island, California. Shell thickness and the degree of ventral migration of the ligament system are both strongly correlated with size. Extreme ventrad ligament migration produces large ligament areas and a resilium that is much higher than long. Comparison.— Crassadoma gigantea differs from its cemented congener of the eastern Atlantic, C. pusio (see below), in being of much greater maximum size and in hav- ing more prominently fasciculated ribs on its right valve and more distinctly ordered ribs on its left (compare Figs. 6a, b with Figs. 6k, 1). The vivid purple color present on the hinge plate of C. gigantea, particularly on the left valve, is absent or faint in C. pusio. Although it is generally assumed that C. gigantea is thicker shelled than is C. pusio, it is doubtful that there is any significant difference when size is taken into account. Both C. gigantea and C. pusio have extensive development of foliated calcite inside the pallial line, and in both species the crossed lamellar aragonite that is present in this region in early ontogeny is covered over by foliated calcite earlier on the left valve interior than on the right, a common phenomenon among pectinids. C. gigantea 1s more derived than C. pusio in the sense that the foliated calcite transgression begins earlier in ontogeny, appearing on the umbonal interior of right valves as small as 11 mm in shell height and on left valves as smati as 7 mm. In C. pusio it has been observed in right valves no smaller than 17 mm and on left valves no smaller than 13 mm. In C. multistriata the plesiomorphic condition persists; there is no foliated calcite transgression over the umbonal interiors at any stage of ontogeny. Adegoke (1969: 103) described a new species of “Hinnites”, Hinnites benedicti Adegoke, 1969, from the Late Miocene Santa Margarita Formation of California. Although this species was found in the same beds as H. multirugosus crassiplicatus (Gale, 1928) [= Crassadoma gigantea], Adegoke regarded it as distinct because of its larger Chlamys stage and more even ribbing. My own examination of Adegoke’s types at UCMP confirmed this distinctness but did not confirm that “Hinnites” benedicti has a cementing habit. The maximum size of the Chlamys stage (34 mm) given by Adegoke is exceeded by the height of the holotype (46.8 mm) and two of the paratypes, none of which show any evidence of cementation. The holotype, a right valve, would be expected to show an abrupt sculp- tural change and xenomorphic growth if cemented. Instead, the degree of shell irregularity is comparable to that of extant Laevichlamys irregularis of the Indo-Pacific, a species with a byssate, nestling living habit. The style of ribbing of H. benedicti and the apparent lack of commar- ginal lirae suggest that it is not a Crassadoma, but its generic assignment awaits examination of better preserved material. Most likely it is a Chlamys, s.s. on the basis of its style of ribbing introduction. 212 AMER. MALAC. BULL. 10(2) (1993) Living habits.— From just below low tide level to 80m (Hertlein, 1972: 212; Grau, 1959: 137; Bernard, 1983: 25); attached to rocks, coarse gravel, and other hard objects. Specimens in northern waters tend to live closer to shore and at shallower depths than those in southern waters (Hertlein, 1972: 212). Geographic range.— Aleutian Islands, Alaska, to Bahia Magdalena, Baja California, Mexico; offshore in Santa Barbara Islands, California, Guadalupe Island, Mexico (Grau, 1959: 137), and Islas de Revillagigedo, Mexico (reported herein, CAS 60266). Stratigraphic range.— Early Miocene?, Middle Miocene to present. The presence of Crassadoma gigantea in the Middle Miocene through Pleistocene of California is appar- ently well established (Moore, 1984: 66). Lower Miocene records, however, are few in number and could require reexamination. The only Lower Miocene record listed by Moore (1984: 66) is from the Painted Rock Member of the Vaqueros Formation. Arnold (1906: 94) also listed a single occurrence, “associated with Turritella hoffmanni” in rocks that he referred to as Lower Miocene. Smith (1991a: 35), in reviewing the changing concepts of the age of the “Vaqueros Stage’, pointed out that some records in faunal lists need redetermination and further collecting to deter- mine stratigraphic position. The earliest reliable records in California appear to be from the ““Margaritan Stage”, which Smith (199 1a, figs. 12, 14) shows as ranging from planktic foraminiferal zone N12 into lower N16 (Langhian into lower Tortonian stages of Europe) and as correlating with the Shoal River and Yellow River Formations of Florida. Discussion.— Previous workers (e.g. Grant and Gale, 1931: 161; Waller, 1991: 23) have assumed that Crassa- doma gigantea evolved from a member of the Chlamys group, e.g. Chlamys hastata, within the eastern Pacific. For reasons given above, however, this is unlikely. C. hastata and its congeners in the eastern and northern Pacific have features that place them squarely within the tribe Chlamydini as herein defined, whereas C. gigantea has a microsculptural pattern shared with the other species placed here in the tribe Crassadomini. Specifically, in the early ontogeny of C. gigantea the plesiomorphic pattern of con- tinuous sweeping antimarginal striae is absent and commar- ginal lirae are prominent in rib interspaces (Fig. 5f). Crassadoma gigantea has a ribbing pattern that is more derived in comparison to the patterns of its eastern Atlantic congeners, and its large size and purple hinge also are derived in comparison to the eastern Atlantic species and to outgroup Chlamydinae. Given the absence of Crassadoma in the Indo-west Pacific, it is highly likely that C. gigantea evolved from species that were present in the Atlantic. There remains the problem, however, of where the fossils are which should corroborate this ancestry. If it is true that C. gigantea was already present in the eastern Pacific by the middle Miocene, it is likely that an ancestor was present in the western Atlantic and that it dispersed through seaways to the Pacific at this time or earlier. If the ancestral species was eurytopic with a preference for warm temperate but not strictly tropical temperatures, it is possi- ble that it avoided the shallow water habitats that predomi- nate in the early Miocene stratigraphic sections thus far known in the Caribbean. The absence of these fossils thus could be the result of a facies sampling bias in the region in which they are most likely to be found. The most likely interval of the middle Miocene for speciation to have occurred may have been during the time of initial emer- gence of the Isthmus of Panama. This uplift cut off the flow of intermediate water between the Atlantic and Pacific and intensified the southward flow of the cool California Current along the Mexican coast, creating a thermal barrier to the west-to-east dispersal of shallow water taxa (Duque- Caro, 1990). Material examined.— Recent material: USNM: 90 lots containing about 150 specimens, from Baja California, Mexico, to Alaska. Fossil material: USNM: about a dozen lots con- taining fewer than 20 specimens, Miocene (Santa Margarita Formation) to Pleistocene, from Baja California, Mexico, and California. Crassadoma mutltistriata (Poli, 1795) (Figs. 5a,d,g; 6c-j) Ostrea multistriata Poli, 1795: 164, pl. 28, fig. 14, living, Sicily. Pecten tinctus Reeve, 1853, species 106, pl. 26, fig. 106, living, locality unknown. Pecten effulgens Reeve, 1853, species 156, pl. 33, fig. 156, living, locality unknown. Pecten textilis Reeve, 1853, species 174, pl. 35, fig. 174, living, locality unknown. Pecten multistriatus (Poli), Bucquoy et al., 1887: 104. Gives extensive synonymy. Chlamys multistriata (Poli), Roger, 1939: 165. Gives ex- tensive synonymy. Types.— Poli’s (1795) work on the living bivalves of Sicily was based largely on his own extensive shell collection (Dance, 1966: 95; Kohn, 1988: 39), which apparently has not been preserved. Previous authors who have discussed WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 213 the taxonomy of this well-known species in detail, e.g. Bucquoy et al. (1887: 104), Sacco (1897: 6), and Roger (1939: 165), have not mentioned type specimens, possibly because Poli provided an illustration the identity of which has never been questioned. Following the International Code of Zoological Nomenclature [Art. 74(b)] and the example of Kohn (1988: 39), the specimen represented by the figure in Poli (1795, Pl. 28, Fig. 14) is herein selected as the lectotype. The types of the three species of Reeve (1853) list- ed in the synonymy were examined at The Natural History Museum, London. All are represented by syntype series. In the interest of nomenclatural stability, lectotype designa- tions are as follows: Pecten tinctus Reeve, syntypes, two complete shells and a single right valve. The specimen of height 32.5 mm, length 27.5 mm, and convexity of the complete shell 13.0 mm, BMNH 1981247/1 is herein selected as the lecto- type (Figs. 6e,f), because it is apparently the specimen that Reeve figured. The paralectotypes are BMNH 1981247/2-3. Pecten effulgens Reeve, syntypes, two complete shells. The darker of the two, BMNH 1993039/1, height 18.8 mm, length 15.0 mm, is represented in Reeve’s figure 156 and is herein selected as the lectotype (Figs. 6g,h). This taxon was recently incorrectly placed in the synonymy of P. cruentatus Reeve, 1853, by Rombouts (1991: 27). Reeve’s P. cruentatus is a Mimachlamys that is closely related to and possibly a junior synonym of the Indo-Pacific Mimachlamys senatoria (Gmelin, 1791). Pecten textilis Reeve, syntypes, a single right valve (Fig. 6j) and single left valve (Fig. 61). The left valve, BMNH 1993040/1, height 24.9 mm, length 20.2 mm, is represented in Reeve’s fig. 174 and is selected herein as the lectotype. Type locality.— Sicily. Diagnosis.— Byssate, non-cemented Crassadoma of small size (less than 40 mm in height); ribs introduced continu- ously throughout ontogeny without distinct clustering or ordering; 70 to 80 continuously scaly ribs and riblets pre- sent at distal margin of mature shells; inner surface of shell inside pallial line lacking foliated calcite throughout ontogeny. Morphological variation.— The only significant geo- graphic variation in Crassadoma multistriata occurs along the Atlantic and Indian Ocean coasts of southern Africa, where specimens tend to have narrower umbonal angles (as narrow as 80°) and correspondingly higher height to length ratios (as high as 1.34). Some authors have considered these narrow forms to be a distinct species, Chlamys tincta (Reeve, 1853) (Figs. 6e,f). The two forms, however, overlap in shell narrowness and are identical in details of ribbing and microsculpture. Some authors have thought that Crassadoma mul- tistriata becomes cemented in the northern part of its range along the Atlantic coast of France and have therefore treat- ed this species and C. pusio as synonyms (see following discussion). On the basis of the evidence on hand, however, I have not been able to substantiate intergradation of these taxa. Bucquoy et al. (1887: 104) noted the presence of both regular and distorted forms in single samples from Brest, France, but illustrated a regular specimen that was still within the size range of the pre-cementation stage of C. pusio. Shell distortion in the absence of cementation is not a diagnostic feature, because it occurs in many independent lineages of byssate, nestling chlamydoid pectinids. In C. multistriata, distortion is common among many specimens in the Atlantic, including specimens at both the northern and southern limits of its geographic range. Pre-cemented C. pusio, however, can generally be distinguished from C. multistriata in the same size range using criteria summa- rized below. Comparison.— Crassadoma multistriata (Figs. 6a-j) closely resembles its western Atlantic counterpart, Cari- bachlamys sentis (Figs. 6m-p), in shell shape, color, and the lack of a foliated-calcite transgression on the inner shell surface in the umbonal region. The former, however, has a normal prodissoconch (Fig. 5a), whereas that of the latter has a prodissoconch dominated by the PI stage (Figs. 7a,g). The umbonal region of the disk of C. multistriata is more inflated and less flattened than that of C. sentis, and the byssal fasciole of C. multistriata generally has a deeply incised groove whereas that of C. sentis does not. Crassadoma multistriata differs from C. pusio (Figs. 6k,1) in lacking cementation and foliated-calcite umbonal transgressions throughout ontogeny. In C. pusio cementation and attendant changes in morphology begin at shell heights between about 12 and 25 mm. The pre-cemen- tation growth stage of C. pusio closely resembles shells of C. multistriata of similar size. Even at this early stage, however, the two species can be distinguished by two fea- tures. First, in C. multistriata, there is no transgression of foliated calcite from the hinge region ventralward across the region inside the pallial line at any stage of growth. In C. pusio, foliated calcite generally begins to form on the inner shell surface at the dorsal edge of the umbonal region at a shell height of about 17 mm in right valves and about 13 mm in left valves, generally before cementation begins. In mature shells over most of the geographic range of C. pusio, this foliated calcite may extend ventrally at least to the level of the top of the adductor scar. Secondly, the beak 214 AMER. MALAC. BULL. 10(2) (1993) of the left valve of C. multistriata has only weak microsculpture, particularly near the distal margin of the preradial stage (Fig. 5d); microsculpture at this same stage is stronger in C. pusio, consisting of distinct, generally dis- continuous antimarginal striae and pits (Fig. Se). Crassadoma multistriata also resembles C. harmeri (Regteren Altena, 1937) of the Plio-Pleistocene of western Europe and Great Britain. The fossil species differs in reaching a larger size (Commonly exceeding a height of 70 mm), in having a minor foliated-calcite transgression on the inner shell surface in the umbonal region of each valve, and in having a distinct anterior-posterior trending ridge on the inner surface of its right anterior auricle. Living habits.— Crassadoma multistriata lives byssally attached to hard objects on the substrate at shallow shelf depths from below low tide level to at least 100 m in nor- mal marine environments. Dead shells, particularly of juve- niles, have been dredged from depths as great as 700 m. The species is apparently eurythermal and possibly has been ecologically generalized throughout much of its histo- ry. Blondel and Demarcq (1990: 250), in a study of the Early and Middle Miocene (late Burdigalian to early Langhian) fossil record of northern Tunisia, refer to the species as eurytopic. Geographic range.— Crassadoma multistriata has an enormous latitudinal range, occurring from the Brittany coast of France southward into the Mediterranean and thence southward along the African coast to southernmost South Africa (Nicklés, 1955). The species occurs through- out the Mediterranean as well as in the Madeira, Canary, and Cape Verde Islands. It is also present in the central South Atlantic at St. Helena (USNM 124060 and 764331). Along the southern and southeastern coasts of Africa, C. multistriata is abundantly represented in USNM collections from False Bay, South Africa, to southern Mozambique (USNM 764201 and 764219) in the Indian Ocean. Stratigraphic range.— Lower Miocene to present. Sacco (1897: 9) showed the stratigraphic range of “Chlamys” multistriata as extending as far down as the Tortonian (Upper Miocene) and that of its lineage ante- cedent, “Chlamys” tauroperstriata Sacco, 1897, sensu stricto, as extending as far down as the lowermost Miocene (Aquitanian). Cox (1927: 42-43) thought that “Chlamys” pusio (which he used as a senior synonym of Chlamys mul- tistriata) was ““well established in Lower Miocene times” and was present in the Miocene of Persia. Roger (1937: 167) noted that “Chlamys” multistriata is common in the Lower Miocene (Burdigalian) in the Mediterranean region from the Rhone Valley to the Vienna Basin. It is generally thought that the species has a Mediterranean origin (Lauriat-Rage, 1981: 43). The stratigraphic history of the species in the eastern Atlantic outside of the Mediterranean, however, is not precisely known. Discussion.— The discrimination and nomenclature of Ostrea multistriata Poli, 1795, and O. pusio Linnaeus, 1758, have long been contentious issues in pectinid taxono- my. Many authors, e.g. Jeffreys (1863: 52), Bucquoy et al. (1887: 107-108), Dautzenberg and Fischer (1925), and Roger (1939: 166), have considered the two species to be intergradational, with cementation absent throughout the ontogeny of Mediterranean specimens but variably present or absent in the later ontogeny of specimens living in the Atlantic. If intergradation in fact exists, then clearly the two names should be synonymized, and one would normally expect the name with priority, O. pusio, to apply to both. Many authors (Bucquoy et al., 1887: 104; Sacco, 1897: 7; Roger, 1939: 165), however, have pointed out that the Linnaean name is poorly founded, not only because of the generalized brief description given by Linnaeus, but also because the species is not represented by a clearly isolated specimen in the Linnaean Collection. These authors have therefore discarded the name O. pusio and have elected to use Poli’s name multistriata instead, applying the name Chlamys multistriata to both byssate and cemented forms. In so doing, these authors have ignored the priority of Pecten distortus Da Costa, 1778, as well as the possible val- idation of Linnaeus’s name by Pennant (1777), who provid- ed an excellent illustration. Da Costa’s name has been used extensively by British malacologists for the cementing form, which is the only form known to be present in British waters. If the only criterion for distinguishing Crassadoma multistriata from C. pusio is the presence of cementation in the advanced growth stages of the latter, then cementation itself cannot be used to settle the problem of intergradation, because its presence becomes a circular argument for the discrimination of the two species. On the basis of the two characters discussed in the preceding comparison (the extent of the foliated-calcite transgression on the inner shell surface and the degree of coarseness of left beak sculpture), I was unable to find evidence of intergradation in the exten- sive eastern Atlantic Jeffreys Collection (USNM). The position taken in the present study, therefore, is that the cementing and non-cementing forms of Crassadoma in the eastern Atlantic are distinct species which, for reasons of priority, must retain the names C. pusio and C. multistriata, respectively. The issue of the type specimen of C. pusio is treated in the following section. Material examined.— Recent material: USNM: 73 lots WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 215 containing about 300 specimens, mainly Mediterranean (Spain, France, Italy, Yugoslavia, Turkey, Tunisia, Algeria) but also eastern Atlantic off Africa (Cape Verde Islands, Guinea-Bissau, Senegal, Angola, Namibia), Indian Ocean (off southern Mozambique), and central Atlantic (St. Helena). In addition to the type specimens discussed above, numerous non-type lots were examined at museums in Great Britain and Europe, particularly at BMNH, BRM, AM, LM, and ZMC. Fossil material: Numerous specimens at BMNH, BRM, UCBL, and TUI. Crassadoma pusio (Linnaeus, 1758) (Figs. 5b,e,h, 6k,1) Ostrea pusio Linnaeus, 1758: 698, no. 169, living, in “O. australiore.” Pecten pusio (Linnaeus). Pennant, 1777, pl. 61, fig. 65, liv- ing, Great Britain. Pecten distortus Da Costa, 1778: 148, pl. 10, figs. 3, 6, liv- ing, England and Scotland (Orkneys). Ostrea sinuosa Gmelin, 1791: 3319, living, in British seas. Pecten spinosus Brown, 1827, pl. 33, fig. 8, living, Northumberland coast, England [name and figure only]; Brown, 1844: 73 [description, with 1827 figure repeated]. Pecten crotilus Reeve, 1853, species 150, pl. 33, fig. 150, living, locality unknown. Type specimens.—In the Linnaean Collection at the Linnaean Society, London, the pair of matching valves of a distorted and cemented “Chlamys” present in the tray bear- ing Collection Sequence No. 179, with height 37 mm and length 35mm, is herein designated the lectotype of Ostrea pusio Linnaeus, 1758. The background and basis for this designation are given in the following discussion. The type or types of Pecten distortus Da Costa, 1778, have not been examined nor has their depository been determined. Dance (1966: 283) reported that some Da Costa types are in The Natural History Museum, London, and some are in the Glasgow University Museum. Ms. Kathie Way, Collections Manager at BMNH (Invertebrates I), reported that no type material for P. distortus has been found there (pers. comm., 1993). The type series of Pecten crotilus Reeve, 1853, BMNH 1993042, consists of four specimens mounted on a board in the Cuming Collection in The Natural History Museum (London). I examined these and determined that they are young Crassadoma pusio. Type locality.— Herein specified as northeastern Atlantic. Diagnosis.— Cemented Crassadoma of moderate size (uncommonly greater than 60 mm in height); ribs intro- duced continuously throughout ontogeny without distinct clustering or ordering; 60 to 80 ribs and riblets present at distal margin of mature shells; inner surface of valves with small foliated-calcite transgression in umbonal region, sel- dom extending past level of dorsal edge of adductor in mature shells. Morphological variation.— The height of right valves at the time of their cementation to the substratum varies from about 12 to 25 mm. There is no evidence in the collections examined for geographic changes in the height at first cementation or in the frequency of cementation within sam- ples. In general, all individuals within populations of Crassadoma pusio seem to become cemented after a certain stage of ontogeny is reached. Harper (1991: 192), however, found that some individuals remain uncemented and merely become lodged into the substratum by means of their irreg- ular growth form. The presence of small foliated-calcite transgressions in the umbonal region of both valves is fairly consistent, again without observed geographic trends. Some exceptional specimens (USNM 543396) collected from floating mines and buoys off the Atlantic coast of Murocco, however, lack foliated calcite in the umbonal region even at a Shell height of 36 mm, well past the stage of initial cem- entation. A foliated-calcite transgression is present, howev- er, in a specimen from deeper water off Morocco (USNM 196510, 132-234 m) and in a specimen from 90 m off Ivory Coast (USNM 764327). All of these specimens have coarse microsculpture in the pre-radial stage of the left valve like that of more typical C. pusio. Comparison.— The pre-cemented growth stage of the shell of Crassadoma pusio closely resembles the shell of C. multistriata of a similar size. Refer to the preceding section on C. multistriata for comparative details. C.pusio differs from other cementing “Hinnites” of the eastern Atlantic in being of smaller maximum size and having less extensive foliated calcite inside the pallial line. Both Hinnites coralli- nus and Chlamys ercolaniana (see above) lack the promi- nent commarginal lirae that are present between ribs in the early ontogeny of C. pusio. The early Chlamys stage of H. corallinus differs from those of the other two cementing species in having ribs that begin late, following an exten- sive pre-radial area having antimarginal striae in a uniform sweeping pattern. The Chlamys stage of Chlamys ercolani- ana differs from that of C. pusio in having a distinctive microsculpture in rib interspaces. This consists of antimar- ginal striae of highly irregular trends, probably a vestige of the shagreen microsculpture present in the ancestry of this species. Contrary to a statement by Harper (1991: 190), C. 216 AMER. MALAC. BULL. 10(2) (1993) pusio lacks shagreen microsculpture. The microsculptural pattern that she interpreted to be shagreen (her fig. 4.4) is a typical plesiomorphic pattern of antimarginal striae. Living habits.— The early growth stage of Crassadoma pusio is byssate on hard objects, particularly rocks or shells, later becoming cemented by the edges of the right valve to the same substratum. Depth records for specimens collected alive range from just below low tide level to about 100 m. Jeffreys (1863: 52) reported that the species lives in depths from 5 to 85 fm (9 to 155 m), with only young spec- imens found near or in the tidal zone. Geographic range.— Common throughout the British Isles from the Orkney Islands to Cornwall and extending across the North Atlantic as far westward as southwestern Iceland (Madsen, 1949: 30); present along European coasts from northern Norway (69° 22.5’N, Soot-Ryen, 1951) southward to Spain and Portugal; uncommon on African coast from Morocco to Ivory Coast; present in Azores; apparently rare in the westernmost Mediterranean. Stratigraphic range.— Pleistocene? to present. The common tendency for authors to combine cemented and non-cemented Crassadoma in a single species referred to as either Chlamys pusio or C. multistria- ta makes it difficult to tabulate stratigraphic occurrences from the literature when the presence of cementation is not specified (e.g. Sacco, 1897; Roger, 1939; Eames and Cox, 1956). Roger (1939: 167) noted that fossils with the form of C. distorta, presumably meaning cemented forms, are quite rare in the fossil record of the Mediterranean but occur in the Pliocene of France. Apparently the only cemented pectinid in the Pliocene Coralline Crag of Great Britain is the gigantic form that Wood (1861: 19) referred to as Hinnites cortesyi De France, 1821. Roger (1939) con- sidered the latter to be a junior synonym of H. crispus. Indeed, Wood’s superb illustration leaves little doubt that this is a true Hinnites and not a Crassadoma (see above). Fossils that Wood (1861: 33) determined to be “Pecten pusio, Pennant” are also present in the Coralline Crag as well as the younger Red Crag. These shells, as noted by Wood (1861: 34), are not cemented and retain their shell regularity to large size (greater than 50 mm). Specimens of these that I examined can be identified with Crassadoma harmeri (Regteren Altena, 1937), which Glibert and Van de Poel (1965: 34) consider to be a subspecies of “Mima- chlamys”’ multistriata (Poli). If these fossil Crassadoma are ancestors of true C. pusio, it is possible that the modern cemented form that is so common in British waters origi- nated within the Pleistocene and possibly very late in the Pleistocene. Harper (1991: 197), who studied the phenome- non of cementation among many bivalve clades, also con- cluded that cementation in this species has a post-Pliocene origin. Discussion.— Malacologists have long acknowledged that the original description of Ostrea pusio by Linnaeus is too generalized to be useful and that the lot in the Linnaean Collection bearing this name contains shells belonging to more than one species. In the first detailed study of the Linnaean Collection after Linnaeus’s death, Hanley (1855) reported that the marked receptacle bearing the name O. pusio in the collection “has unfortunately been converted into a general depository for all the loose valves of the smaller Pectens....’ He acknowledged, however, that the box contained shells belonging to “the pusio of the British writers and the albolineatus of Sowerby’s [1842] Mono- graph,” as well as a “white valve of the young Jslandicus.” He conjured that Linnaeus’s ill-fitting description may have been of a composite consisting of the valves of different species. He concluded, however, that the species then known to British writers as “P. sinuosus (= pusio) ...(of which there are many loose specimens in the cabinet) accords the best with the definition...” This conclusion did little to resolve the problem of a type for Ostrea pusio, and in subsequent years Linnaeus’s name, as well as the names Pecten distortus Da Costa, 1778, and O. sinuosa Gmelin, 1791, have all been used for the cemented pectinid of the northeastern Atlantic. Cox (1927: 43), adamantly defending the use of Linnaeus’s name but applying it to both the cemented form of the Atlantic and the byssate form of the Mediterranean, said that he would propose “a definite specimen from the Linnaean Collection and figure it as the lectotype of the species.” However, I can find no record that he ever did so. Dodge (1952: 178), in a later study of the Linnaean Col- lection, added nothing new to the findings of Hanley (1855). He urged retention of Linnaeus’s name, however, because of its extensive use, particularly by writers in the last half of the nineteenth century. Glibert and Van de Poel (1965: 33) followed Dodge’s advice, but noted further that even if O. pusio of Linnaeus be regarded as an unaccept- able name, P. pusio Pennant, 1777, which applies to the British cemented species, is available and has priority over O. multistriata Poli, 1795 (and also over P. distortus Da Costa, 1778, and O. sinuosa Gmelin, 1791). Wallin (1991: 154), who reported on the contents of the Queen Ulrica Collection (which contains material that may have been examined by Linnaeus prior to the publication of his tenth- edition names in 1758) found specimens of O. pusio to be present. My own examination of the Linnaean Collection at the Linnaean Society in London in 1977 found that the tray labeled in Linnaeus’s hand as O. pusio contains four speci- WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 217 mens: a pair of matching valves of cemented Crassadoma pusio, height 37 mm, length 35 mm, with approximately 40 radial ribs (secondary ribs not counted); a pair of matching valves of Mimachlamys albolineata (G.B. Sowerby II, 1842); a single left valve of M. albolineata; and a single juvenile right valve of Chlamys islandica (Miiller, 1776). Dance (1967: 8) urged that caution is necessary when designating types from the Linnaean Collection because of the checkered history of the specimens in the collection, which now includes non-Linnaean material. In the case of Ostrea pusio, however, it seems likely that spec- imens of this common species of the northeastern Atlantic were available to Linnaeus, as indicated by their presence in both the Linnaean and Queen Ulrica Collections. In view of the long history of usage of the name, the above designa- tion of a lectotype from the Linnaean Collection seems to be justified. Material examined.— Recent material: USNM: 55 lots containing about 250 specimens, mainly from Great Britain but including material from Sweden, France, the Azores, Morocco, and the Ivory Coast. In addition, numerous speci- mens were examined in the collections abroad, mainly at BMNH, BRM, MNHN, AM, LM, and ZMC. Fossil material: No unequivocal cemented Crassa- doma pusio were found among material examined at muse- ums abroad, underscoring the apparent late origin of this cemented species. At the time of my European studies, however, the ontogenetic criteria for differentiating C. pusio and C. multistriata had not yet been discovered. Caribachlamys Waller, new genus Etymology.— The name Caribachlamys combines a prefix signifying Caribbean with the genus name Chlamys. Diagnosis.— Byssate, non-cemented Crassadomini with lecithotrophic-type prodissoconch (large PI stage, short or absent PII stage); strong antimarginal striae present between commarginal lirae in rib interspaces in early ontogeny. Type species.— Pecten sentis Reeve, 1853. Other species.— Pecten ornatus Lamarck, 1819; Cari- bachlamys paucirama Waller, new sp.; Pecten (Chlamys) imbricatus mildredae Bayer, 1941; Ostrea imbricata Gmelin, 1791. Geographic range.— Caribbean Sea and adjacent waters of the warm-temperate to tropical western Atlantic from North Carolina to Brazil and Bermuda. Stratigraphic range.— Upper Pliocene to Recent. Discussion.— The new genus Caribachlamys is based on the discovery that four extant species of the Caribbean and adjacent waters share a unique prodissoconch morphology (Figs. 7a-f) and a unique pattern of commarginal and anti- marginal microsculpture in rib interspaces (Figs. 7g-l). The phylogenetic relationships of these extant species and one new extinct species are shown in Figure 8. Caribachlamys ornata (Node B, Fig. 8) resembles C. sentis very closely in ribbing pattern but differs in developing smooth-crested scaleless ribs that are I-beam shaped in cross-section. In the Bahamas the two species nearly intergrade in that speci- mens identified herein with C. ornata develop I-beam shaped ribs only in very early ontogeny and otherwise resemble C. sentis (see following section on C. ornata). C. paucirama, C. mildredae, and C. imbricata have ribs that have a plesiomorphic rounded cross-sectional shape (not I- beam shaped) like that in the likely stem species, C. sentis, meaning that it is unlikely that C. ornata can have given rise to any other species in the genus. C. paucirama, C. mil- dredae, and C. imbricata are united (Node C, Fig. 8) by decrease in prominence of the early commarginal tirae in rib interspaces and by the onset of substantial irregularity in the trends of these lirae. At Node D, C. paucirama develops an autapomorphy: rib introductions beyond a shell height of 20 mm are few in number or absent altogether, leading to a pattern of ribs of fairly uniform height and spacing at the distal margin. Caribachlamys mildredae and C. imbricata (Node E) share fasciculation of ribs, increase in scale spac- ing on major ribs, a tendency toward cusping of scales, and marked flattening of the left disk. Finally, C. imbricata (Node F) has evolved a unique ribbing pattern consisting of 9 or 10 major plicae with a tendency for secondary ribs to be eliminated. The scales on the ribs of this species com- monly form closed knobs, the left disk is markedly flat- tened, and the maximum observed irregularity and incon- sistency in direction of the early commarginals is reached. Caribachlamys sentis (Reeve, 1853) (Figs. 6m-p, 7a, d, g, j) Pecten sentis Reeve, 1853, species 129, pl. 29, fig. 129, liv- ing, locality unknown. Chlamys sentis Reeve, Abbott, 1954: 363, pl. 34a. Types.— Among Reeve’s types in The Natural History Museum, London, are two pairs of matching valves in a single box labeled “‘sentis Reeve.’ Upon examining these specimens, I found that the larger specimen, which is a deep orange in color, corresponds in dimensions to Reeve’s 218 AMER. MALAC. BULL. 10(2) (1993) Fig. 7. Scanning electron micrographs of prodissoconchs in planar view (a-c), prodissoconchs in posterior view (d-f), pre-radial microsculpture (g-i), and early radial stage sculpture (j-l) of left valves of Caribachlamys. a, d, g, j. C. sentis, USNM 764706, Molasses Reef, Monroe Co., Florida. b, e, h, k. C. ornata, USNM 766643, Carrie Bow Cay, Belize. c, f, i, 1. C. imbricata, USNM 457668, Bahia Honda, Cuba. Arrows point to approximate position of PI/PII boundary. Scale bars: a-f = 50 um, g-l = 200 um. WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” Ze Caribachlamys bY eas Es ® HB & & x S aS) 3 g 9 © S = = Ae} CS Cc c = Q SE 5 o © E S & & 8 3 & F Fig. 8. A phylogeny of species in the new genus Caribachlamys in the Plio-Pleistocene of the tropical western Atlantic. Lineages ending in arrows are extant; the one ending at a terminal cross-bar is extinct. Lettered blocks refer to apomorphies: (A) larval shell with large PI and small PII stage; (B) I-beam shaped ribs with smooth crests; (C) subdued commarginal lirae of irregular trend in early ontogeny; (D) rib introduc- tions limited to early ontogeny, later ribs evenly spaced without significant clustering; (E) fasciculation of ribs, increased scale spacing with tendency toward cusping, and flattening of left valve; (F) reduction of ribs to 9 or 10 major plicae with tendency to eliminate secondary ribs, scales commonly forming closed knobs, and left valve markedly flattened. fig. 125. The smaller specimen, although nearly of the same color, is not the same species but rather is a specimen of Spathochlamys benedicti (Verrill and Bush, 1897) (see following section). The larger specimen, BMNH (ht., 15.9 mm; L, 14.3 mm; length of anterior outer ligament, 2.9 mm; length of posterior outer ligament, 0.9 mm) is herein selected as the lectotype (Figs. 6m, n). Type locality.— Reeve had no locality data for the species that he described. The type locality is herein specified as the western North Atlantic off southeastern Florida. Diagnosis.— Caribachlamys with valves about equal in convexity; ribs closely spaced, fine, continuously scaly, rounded, and of many different heights, introduced continu- ously throughout ontogeny, with more than 70 ribs and riblets at distal margin of mature individuals; major ribs with straight or slightly undercut sides but not I-beam shaped in cross section; scales fine, closely spaced, erect, distally concave, and open; commarginal lirae well devel- oped in rib interspaces in early ontogeny and fairly uniform in trend across rib interspaces; antimarginal striae promi- nent between commarginals but variably present on rib crests. Morphological variation.— Caribachlamys sentis is of small size, seldom exceeding a height of 40 mm; no geo- graphic trends in maximum size were observed. The ribs tend to be continuously scaly beginning at their origins at the beginning of the radial stage of growth; there is in gen- eral no abrupt change in ontogeny from a non-scaly to a scaly condition. Exceptions occur in the western Caribbean along the coast of Panama (USNM 743733, Payardi Island, and USNM 743360, San Blas Archipelago ). The ribs of the left valves of some of the specimens in these samples lack scales in early ontogeny up to a shell height of from 4.5 to 6.5 mm and then change fairly abruptly to a scaly phase. Unlike the non-scaly phase of ribs in C. ornata, however, these ribs have a rounded crest and are not I-beam shaped in cross section. The shell color and color pattern of Caribachlamys sentis are variable. Most commonly, shells are orange or purple with maroon mottling in early ontoge- ny. In some cases an orange phase may change abruptly along a commarginal growth line to a purple phase, but the reverse has not been observed. Rarely the color may be nearly uniform white or yellow, usually with an early orange-mottled phase. Comparison.— Caribachlamys sentis closely resembles C. ornata in ribbing density and pattern. The ribs differ, how- ever, in cross-sectional shape, at least in early ontogeny. In C. sentis the rib shape is high and rounded, with the rib crests not flattened, whereas in C. ornata the ribs at least in early growth have an I-beam cross-section, with the crests flattened and non-scaly and the rib sides strongly undercut. Specimens from the northern Bahamas identified herein as C. sentis are very close to being transitional between the two species in rib shape in early ontogeny. Unworn shells of C. ornata have dense antimarginal striae present on the flattened rib crests, whereas in C. sentis such rib-crest striae are not so strongly developed. Caribachlamys sentis tends to have more uniform coloration than does C. ornata and generally does not develop the strongly contrasting color pattern that is common in most C.ornata, where dark maroon spots or bars form on a stark white background. Caribachlamys sentis and C. ornata also tend to differ in living habits. Where the two occur in the same area, C. sen- tis is generally found at more protected sites on reefs, whereas C. ornata lives in the more turbulent areas of the reef front. Caribachlamys sentis also closely resembles Crassadoma multistriata (eastern and central Atlantic and southwestern Indian Ocean) in ribbing pattern, color, and color pattern. In C. multistriata, however, the prodisso- 220 AMER. MALAC. BULL. 10(2) (1993) conch is of the normal plesiomorphic type, with a PII stage that is relatively large compared to PI (Fig. 5a); in C. sentis the prodissoconch is almost entirely composed of the PI stage, the PII stage being only a narrow fringe or absent (Figs. 7a, d). C. sentis also differs from C. multistriata in having a somewhat less inflated shell and in generally lack- ing a distinct groove along the center of the outer surface of its byssal fasciole. Aside from the difference in prodisso- conch morphology, young individuals of the two species can be distinguished on the basis of microsculpture. In C. sentis the spaces between commarginal lirae of the early commarginal stage are occupied by well-developed anti- marginal striae; in C. multistriata, antimarginal striae between commarginals are either very poorly developed or absent. Living habits.— Byssally attached beneath coral heads and between rocks and coral rubble on reef crests in shallow backreef areas, on near-shore rock jetties, and in deeper fore-reef rubble zones; depth range from just below low tide level to about 30 m. Deeper offshore specimens are dead shells only, known to occur as deep as 52 m. Diaz et al. (1991), reporting on the molluscan fauna of the Santa Marta area of Colombia, found Caribachlamys sentis liv- ing in three types of habitats: (1) at a depth of 22 to 27 m at base of the reef front on a bottom of “conglomerates of coral rubble partly bound together by sponges; dead coral heads and scattered debris patches may also be present.” (2) at a depth of about 8 to 15m, with the bottom described as “upper reef-slope with large coral heads forming caves and an intricate system of cryptic microhabitats. Species of sea fans, sea whips, and sea plumes are also common.” (3) at depths less than 10 m, with the bottom described as “rocky boulders and pebble partly covered by filamentous algae, crusting zoanthids, and fire corals. This zone is present in calm environments in shallow water.” Geographic range.— From Jupiter Inlet, southeastern Florida, southward through the Florida Keys; uncommon in northern, western, and southern Gulf of Mexico [records from Flower Garden Reef, Texas, and off Tamaulipas, Mexico (USNM 764714)]; western Caribbean (Nicaragua, Panama, Colombia); western Atlantic off Brazil as far south as the state of Santa Catarina (29°S). The presence of the species in southern Brazil is based on Rios (1985: 222); I have not examined these specimens. Stratigraphic range.— Lower? Pliocene to present. Discussion.— The only fossil Caribachlamys sentis found thus far is a single right valve (USNM(P) 474635) collected by the late S. E. Hoerle from the north side of the Caloosahatchee River, 3.4 km (2.1 miles) west of Ortona Locks, Glades County, Florida. Unfortunately the precise stratigraphic position of this specimen is unknown, because the stratigraphic section at this site likely includes the Caloosahatchee, Bermont, and Ft. Thompson Formations. All that can be said is that it is likely that the specimen is Pleistocene in age. I have not yet observed unequivocal pre-Pleisto- cene fossils of Caribachlamys sentis, and my estimate of a Pliocene origin is based upon two indirect lines of evi- dence. First, the oldest fossil member of the tribe Cras- sadomini in the western Atlantic thus far identified is possi- bly the specimen figured by Woodring (1925, pl. 7, fig. 10) as “Chlamys (Chlamys) sp.’ from the Bowden Formation (Bowden shell beds) at Bowden, Jamaica. This specimen, USNM(P) 352779, is an abraded juvenile right valve only 4.4 mm in height. Its pattern of rib introduction, by rib- flank intercalation, is remarkably like that at a similar growth stage of C. sentis (cf. USNM 764710). The micro- sculpture of the specimen does not appear to be preserved except for traces of antimarginal striae on the anterior and posterior edges of the disk and disk flanks and some obscure traces of commarginal lirae. The prodissoconch of the specimen is not preserved, but the impression of it remains on the underlying shell material. This impression has a length of 163 um, exactly the length that one would expect for the prodissoconch of a Caribachlamys. The prodissoconchs in species of Crassadoma are, so far as known, of a larger size (about 200 um). On this basis the specimen is placed tentatively in Caribachlamys, and it may well be an early representative of C. sentis. The Bowden beds are now considered to be early Pliocene in age (see below, under Spathochlamys vaginula). Secondly, the next oldest fossil Caribachlamys thus far discovered occur in the “Pinecrest beds” and Caloosahatchee Formation of Florida of late Pliocene and early Pleistocene age (see below). These specimens, how- ever, are identified with either Caribachlamys paucirama, new species, or C. mildredae, both of which have morpho- logical features that are more derived than are the corre- sponding features of C. sentis (see preceding discussion of the genus Caribachlamys and Fig. 8). This suggests that C. sentis itself must have been present before the time of deposition of these beds. Woodring (1982: 590) identified “Chlamys sentis (Reeve)?” from the Emperador Member of the La Boca Formation, Lower Miocene, of Panama, but neither his illustrated specimen (USNM(P) 647216) nor the other two specimens to which he referred are members of this species. Rather they appear to belong to the genus Dimarz- ipecten Ward, 1992, which is discussed below under Spathochlamys, new genus. WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 221 Recently Garrafielo and Tavora (1989) included Chlamys sentis in a list of species from the Pirabas For- mation, early Miocene, Brazil. This identification is doubt- ful, however, because they include Chlamys japericensis Ferreira, 1960, originally described from the Pirabas Formation, as a junior synonym of C. sentis. A specimen of C. japericensis sent to the USNM by Dr. Ferreira has sha- green microsculpture and is not a Caribachlamys. Material examined.—Recent material: USNM: 61 lots containing about 300 specimens, from southern Florida, the Florida Keys, Mexico, Nicaragua, Panama, and Colombia. MNHN: 4 lots containing 4 specimens, from off Brazil (8° 22.5’ S, 12° 56.0 W, and the Abrolhos Archipelago). CAS: 2 lots containing 5 specimens, from Bahia Limon, Panama and Baia de Todos os Santos, Bahia, Brazil. Fossil material: USNM: USNM(P) 474635, | RV, probably Pleistocene, from the Caloosahatchee River area of south Florida (see preceding discussion). Caribachlamys ornata (Lamarck, 1819) (Figs. 6q,r; 7b,e,h,k; 9a,b) Pecten ornatus Lamarck, 1819: 176, from southern Atlantic Ocean. Chlamys ornata (Lamarck), Verrill and Bush, 1897: 59. Types.— Specimens of Pecten ornatus in the Lamarck Collection in Geneva are in two lots. One of these, GNH 1088/75, which contains a single pair of matching valves of length 28.2 mm, is selected herein as the lectotype, because the specimen conforms to Lamarck’s measurement and is the specimen illustrated by Chenu (1845, pl. 35, figs. 3 and 3a). The second lot, GNH 1088/73, contains five pairs of matching valves some of which clearly belong to species other than P. ornatus. Type locality. Lamarck’s reference to the locality of this species as “l’Océan atlantique austral” is overly general- ized. Caribachlamys ornata extends south of the Equator for only a short distance to islands offshore from Brazil. The type locality is herein emended to the Antillean region of the western Atlantic, Diagnosis.— Caribachlamys with valves about equal in convexity or with left valve slightly less convex than right but not markedly flattened; closely spaced, fine ribs with early non-scaly phase followed by scaly phase; ribs intro- duced continuously throughout ontogeny, with more than 70 ribs at distal margin of disk on mature shells; ribs of right valve commonly in clusters of three, those of non- scaly phase of left valve high and I-beam shaped in cross section, with deeply undercut rib flanks; scales fine, closely spaced, erect, distally concave, and open; commarginal lirae well developed in rib interspaces in early ontogeny and fairly uniform in trend and spacing across rib inter- spaces; antimarginal striae prominent between commargin- al lirae and well developed on rib crests. Morphological variation.— Caribachlamys ornata, like its congeners, is of small size, seldom exceeding a height of 40 mm; no geographic trends in maximum size were observed. The degree to which the ribs of C. ornata are non-scaly, smooth crested, and I-beam shaped is highly variable both within and among population samples. Furthermore, these features are less well developed on the right valve than on the left valve. Among 30 measured valves from Florida and the Bahamas, the non-scaly phase of ribs in the central sector of the disk was found to disap- pear at shell heights ranging from 7 to 10 mm on the right valve and from 9 mm to 25 mm on the left valve. There is much less variation in shell color and color pattern than in ribbing. Most C. ornata have dark red, maroon, or purple maculations on a light-colored, generally white, back- ground. Distinct orange coloration in this background is relatively rare, occurring in the collections examined from eastern Puerto Rico through the Virgin Islands to Dominica. Specimens from the northern Bahamas, represented by a good suite of specimens in the Delaware Museum of Natural History, are lightly pigmented, and the extent of the I-beam phase of their ribbing is highly variable. Comparison.— Caribachlamys ornata resembles C. sentis in both the ribbing-introduction pattern and the microsculp- ture of the early commarginal stage. C. ornata differs, how- ever, from C. sentis and all other Caribachlamys in having a distinct non-scaly I-beam-shaped ribbing phase in early ontogeny, persisting on the left valve at least to a shell height of 9 mm (refer to preceding section on C. sentis). Living habits.— Caribachlamys ornata lives byssally attached beneath rocks and coral heads in shallow water on the more turbulent parts of coral reefs, on hard, rocky bot- toms, or in channels with strong currents. The only docu- mented depth records for live specimens are all shallower than 20m. Dead shells, apparently dislodged from reef fronts, have been dredged from as deep as 165 to 180 m (USNM 501824, off Lazaretto, Barbados). Geographic range.— Caribachlamys ornata occurs throughout the Antilles and in the northern Bahamas, southeastern Florida, and the Florida Keys. I have not been able to substantiate its presence in Bermuda, an occurrence which Abbott (1974, p. 443) also questioned. In the south- 222 AMER. MALAC. BULL. 10(2) (1993) GY | be Fig. 9. Caribachlamys. a,b. C. ornata, USNM 683533, The Baths, Gorda Id., Virgin Ids., West Indies, left valve, planar and oblique views of I-beam shaped ribs, horizontal widths of fields 10 mm and 9 mm. c. C. mildredae, USNM 764745, Long Reef, Florida, juvenile right valve showing irregular com- marginal and antimarginal microsculpture in rib interspaces, transmitted light, horizontal field width 4 mm. d, e. C. mildredae, USNM 598977, holotype, Long Key Reef, Dry Tortugas, Florida, matching right and left valves, height 37.0 mm. f, g. C. imbricata, USNM 764733, Florida morph, La Chorerra, Havana, Cuba, non-matching right and left valves, heights 29.1 and 25.2 mm. h, i. C. imbricata, Bermuda morph, USNM 764735, near Bermuda Biological Station, St. Georges, Bermuda, matching right and left valves, height 39.8 mm. j. C. imbricata, Bermuda morph, USNM 743734, suction dredge spoil from near edge of coral reef, Payardi Id., Minas Bay, Panama, single left valve, height 30.6 mm. k, I. C. imbricata, Bermuda morph, USNM 760081, Pico Feo Id., Gulf of San Blas, Panama, matching right and left valves, height 18.3 mm. m, 0. C. paurirama, UF 24894, Caloosahatchee Fm.?, spoil in Cochran Shell Pit, Hendry Co., Florida, detail of sculpture of left valve, horizontal field width 4.6 mm, and anterior view of articulated shell, shell height 36.5 mm. n. C. pauci- rama, USNM(P) 474637, holotype, Caloosahatchee Fm.?, spoil at rock pit east of Ft. Denaud, Hendry Co., Florida, single left valve, height 29.6 mm. p, q. C. paucirama, UF 24479, Caloosahatchee Fm.?, spoil in Cochran Shell Pit, Hendry Co., Florida, right valve, interior of hinge and exterior, hinge length 15.5 mm, valve height 30.3 mm. r,s. C. paucirama, USNM(P) 474638, “Pinecrest Beds,” northeast of Naples and southwest of Immokalee, Collier Co., Florida, left valve, height 30.7 mm, horizontal field width of detail, 14 mm. WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 223 ern and western Caribbean, the species occurs on or near reefs adjacent to islands off Venezuela, Belize, and Quintana Roo, Mexico. In the Gulf of Mexico (other than along the Florida Keys), the species is apparently either very rare or absent. The USNM collections have one poorly documented lot from Veracruz, Mexico (USNM 57655). Rios (1985: 222) reported the presence of this species in the Brazilian islands of Fernando de Noronha and Atol das Rocas. Stratigraphic range.— Lower? Pleistocene to present. The oldest fossil occurrences of Caribachlamys ornata thus far known are specimens reported by Trechmann (1933: 33) from both high (300 m) and low lev- els of the Coral Rock of Barbados. Two specimens collect- ed by Trechmann and deposited in The Natural History Museum, London (Palaeontology Section) were examined by me in 1979. Although both clearly belong to C. ornata, nothing further can be said about their relationship to mod- ern forms without reexamining the specimens. It was noted at the time that one of the specimens, a left valve from “below the Garrison, west side of Barbados,” still retains a remnant of a typical C. ornata color pattern. Trechmann (1933) argued that the Coral Rock of Barbados, now tilted and variously displaced by tectonic sliding, may be of fairly uniform age, mainly early Pleistocene. More recent studies (Mesolella, 1967; Ladd et al., 1990) have emphasized that Barbados emerged throughout the Pleistocene. Specimens of Caribachlamys ornata of Pleisto- cene age have also been collected on San Salvador Island, Bahamas, from a fossil coral reef northwest of the center of Cockburn Town [USNM(P) 474636]. This reef is Sanga- monian in age, based on 234U-230Th dates that range from 130.75 + 1.5 ky to 119.2 + 1.5 ky (Curran and White, 1989). Dall (1898) reported that this species occurs in the Pleistocene of the Florida Keys and the Antilles. Discussion.—There remains some doubt as to whether Caribachlamys ornata is distinct from C. sentis at the species level or whether the two are merely subspecies. Along the southeastern Florida coast and in the Florida Keys, where both forms occur, there is little doubt that they are well separated morphologically. In the Bahamas, how- ever, particularly at Abaco Island, specimens identified herein as C. ornata have only a very short early growth stage with I-beam shaped, non-scaly ribs, and after this phase even shell color can become uniform and similar to that of C. sentis. The specimen of Lamarck chosen here as the lectotype in fact is close to these specimens in having a short I-beam phase and an abrupt shift to a more uniform color pattern. A possible interpretation is suggested by the fact that C. ornata appears to have a very recent origin in the later Pleistocene. It could well be that the species is still in the process of change and that selection pressures have operated more strongly in reef-front situations in the Lesser Antilles and Florida than on the carbonate banks and patch reefs of the Bahamas. Material examined.— Recent material: USNM: 93 lots containing about 250 specimens, from the Florida Keys, the Bahamas, Cuba, Grand Cayman, Jamaica, Haiti, the Dominican Republic, Puerto Rico, the Virgin Islands, Saint-Martin, Antigua, Dominica, St. Lucia, Barbados, Tobago, Aruba, Mexico (Quintana Roo), and Belize. DMNH: many specimens from the Bahamas. AM: several lots, from Saint-Martin, St. Eustatius, Barbados, Bonaire, and Curacao. Fossil material: USNM: USNM(P) 474636, 7 specimens plus fragments from a fossil coral reef northwest of Cockburn Town, San Salvador Island, Bahamas, Pleistocene (Sangamonian, see preceding section on strati- graphic range). BMNH: 2 specimens from the Pleistocene Coral Rock “below the Garrison”, west side of Barbados, collected by Trechmann (1933). Caribachlamys paucirama Waller, new species (Figs. 9m-s) Etymology.— From the Latin, paucus, meaning few, com- bined with ramus, meaning branch, with reference to the relatively few rib introductions in the late ontogeny of the shell. Types.— Holotype: USNM(P) 474637, 1 left valve (Fig. 9n), from rock pit at east edge of Ft. Denaud, Hendry County, Florida, probably from the Caloosahatchee Formation of early Pleistocene age. Ht. = 29.6 mm, L. = 25.7 mm, umbonal angle = 92°, AOL = 9.4 mm, POL = 4.3 mm. Paratypes: USNM(P) 474638 (Figs. 9r,s), 1 left valve, USGS 26543 (TU 1175) from spoil banks along canals south of Florida Highway 858, 3.2km (2 miles) east of junction with Florida Highway 846 (SE!/q, section 24, T48S, R27E), northeast of Naples and southwest of Immokalee, Collier Co., Florida. Pinecrest Beds. Collected by S.E. and R. E. Hoerle. USNM(P) 474639, | juvenile right valve and 2 fragments, rock pit 5.6 km (3.5 miles) west of LaBelle, off Florida Highway 78 on north side of Caloosahatchee River, Glades County, Florida. Caloosahatchee Formation. USNM(P) 474640, 1 right valve, 3 left valves, and | fragment, Cochran Shell Pit, NE!/4 sec. 23, T43S, R28E, Hendry County, Florida. Collected from spoil, Caloosahatchee Formation. 224 AMER. MALAC. BULL.10(2) (1993) UF 24479, 1 right valve (Figs. 9p,q) and 2 left valves, same data as preceding. UF 24894 (Figs. 9m,o), | pair of valves, 3 right valves, and 5 left valves, same data as preceding. UF 24986, | left valve, same data as preceding. UF 41906, | right valve, same data as preceding. UF 55665, 3 right valves and 2 left valves, same data as preceding. UF 9492, 3 left valves, Caloosahatchee River, Florida, precise locality not specified, presumed to be from the Calocsahatchee Formation. Diagnosis.— Caribachlamys with valves about equally convex; well-spaced, fine, continuously scaly, rounded ribs equal to or narrower than interspaces, with new rib intro- ductions uncommon after mid-ontogeny (shell height of about 15 mm); about 40 to 50 ribs and riblets at margin of mature individuals (shell height of about 30 mm); scales fine, closely spaced, distally concave, and open; commar- ginal lirae absent; microsculpture in rib interspaces consist- ing of antimarginal striae of irregular trend in early ontoge- ny, regularly diverging antimarginal striae in later ontoge- ny; crests of ribs without antimarginal striae. Description of holotype.— Exterior: Left valve with height somewhat greater than length (Ht/L = 1.15) and anterior auricle much larger than posterior one (AOL/POL = 2.19); anterior auricle pointed, with deep byssal sinus; posterior auricle oblique, forming an angle with hingeline of about 120°; posterior margin of posterior auricle nearly straight, slightly convex, or slightly concave. Prodis- soconch not preserved; radial ribs originating at shell height of 1 mm, initially 10 in number, increasing to 27 at shell height of 5 mm, 37 at 10 mm, and 42 at distal margin at shell height of 30 mm. Ribs high and rounded in cross-sec- tion, bearing erect, distally concave, open, closely spaced scales throughout ontogeny; ribs remaining narrower than or equal to interspaces in width. Rib introduction initially mainly by intercalation submedially in interspaces of earli- er ribs; microsculpture of pre-radial stage of beak pitted or with short discontinuous antimarginal striae; microsculp- ture in rib interspaces consisting of striae of irregular, main- ly antimarginal, trends; commarginal lirae obscure or absent in early ontogeny. Interior: Hinge dentition of two-element type as in Chlamys; umbonal region aragonitic, without foliated- calcite transgression; adductor and pedal retractor muscle scars not preserved; ribs expressed internally near margin as simple corrugations without internal carinae. Morphological variation.— The oldest specimen, USNM(P) 474638, from the Pinecrest Beds (Figs. 9r,s), dif- fers from most of the other specimens, all from the Caloosahatchee Formation, in having a slightly greater number of ribs at shell heights of 10 mm and 20 mm, and it exceeds most but not all of the Caloosahatchee specimens in number of ribs present at the distal margin of its disk. Comparison.— Caribachlamys paucirama resembles C. sentis in overall aspect but differs in having most of its rib introductions limited to early ontogeny, thereby producing by late ontogeny a far more regular ribbing pattern and a lower total number of ribs at maturity. By the time a shell height of 20 mm is reached, C. paucirama rarely has as many as 50 ribs, whereas C. sentis has between 50 and 60. Specimens of C. sentis that are 30 mm in height or greater have at least 60 radial ribs present along the margins of their disks; in contrast C. paucirama generally has fewer than 50. In microsculpture the two species differ in their first 10 mm of growth. Distinct commarginal lirae of fairly regular trend are present in C. sentis, whereas C. pauci- rama has irregular antimarginal striae. Caribachlamys paucirama also resembles C. mil- dredae. The microsculpture pattern of the two species in early ontogeny is very similar but ribbing patterns differ. C. mildredae generally has fewer than 30 ribs at a shell height of 10 mm, whereas C. paucirama generally has more than 30. The ribs of C. mildredae are distinctly clus- tered on the right valve and distinctly ordered on the left; those of C. paucirama tend to be more uniform in spacing and height. Lastly, the scales on the ribs of C. mildredae, particularly those on its left valve, are more widely spaced than are the scales of C. paucirama, with the scales of some specimens of C. mildredae approaching the closed knobby condition that is common in C. imbricata. Paleoecology.— Caribachlamys paucirama appears to be most abundant in the Caloosahatchee Formation at sites where there are abundant scleractinian corals. Presumably the living habit and habitat of the new species were similar to, if not the same as, that of extant reef-dwelling members of the genus. Geographic range.— Thus far known only from southern Florida. Stratigraphic range.— Middle and Upper Pliocene (Pinecrest Beds) and Lower Pleistocene (Caloosahatchee Formation). The age of the Pinecrest Beds has been somewhat controversial, depending on how the boundaries for this formation are drawn. The Middle to Late Pliocene age is based on studies by Akers (1974), Waldrop and Wilson (1990: 220), and Jones et al. (1991). WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 225 Discussion.— Caribachlamys paucirama is quite clearly a member of the same clade that contains C. mildredae and C. imbricata, the shared derived characters being the reduc- tion of early commarginal lirae in rib interspaces, the irreg- ularity of early antimarginal microsculpture, and the reduc- tion in rib number (see preceding section on genus Cari- bachlamys and Figure 8). In its lack of distinct rib cluster- ing, the new species is the most primitive known member of this clade. Although there is only a single specimen of the new species available from the Pinecrest beds, the rib- bing pattern of this specimen suggests that between Pinecrest and Caloosahatchee time, rib introductions decreased and rib spacing became more regular. This is just the opposite of a trend that produced C. mildredae and C. imbricata, in which the ribs are strongly clustered and ordered, with a common trend toward distant spacing of scales on major ribs and toward cusping of scales. This trend in addition to the peculiar microsculpture of C. pauci- rama, with highly irregular antimarginal striae and a lack of commarginal lirae, suggests that the new species is a sister group of a clade that contains both C. mildredae and C. imbricata (Fig. 8). Material examined.— USNM: the 10 specimens and frag- ments listed above as types. UF: the 22 specimens listed above as paratypes. Caribachlamys mildredae (Bayer, 1941) (Figs. 9c-e) Pecten (Chlamys) imbricatus mildredae Bayer, 1941: 46, pl. 3, figs. 16,17. Pecten (Chlamys) mildredae Bayer, Bayer, 1943: 110, pl. 12, fig. 7. Chlamys mildredae (F. M. Bayer), Abbott, 1954,: 363, pl. 34, fig. c. Holotype.-—USNM 598977, a pair of matching valves, ht. = 37 mm, from Long Key Reef, Dry Tortugas, depth not specified. Bayer (1941) did not give the locality of the holotype in his original description. In a later note (Bayer, 1942) he specified Biscayne Bay as the type locality on the grounds that the greatest number of shells came from the Miami area. Diagnosis.— Caribachlamys with left valve less convex than right and somewhat flattened; fine, continuously scaly, rounded ribs introduced continuously but sparingly throughout ontogeny, with fewer than 50 ribs and riblets at distal margin of disks of mature individuals; ribs distinctly clustered on right valve and with at least three distinct orders of rib height on left; scales large, widely spaced, commonly strongly inclined distally with tendency to form closed knobs; early commarginal lirae obscure, scarcely higher in relief than intervening antimarginal striae and of irregular trend and spacing; antimarginal striae absent from crests of major ribs. Morphological variation.— The distinctive and fairly con- stant feature of Caribachlamys mildredae is its ribbing pat- tern. On the left valve this is commonly expressed as four orders of rib height (Fig. 9e). Ribs of the first order, the highest in amplitude and generally five in number, tend to be more lightly pigmented than the other ribs, which appear to alternate in height between ribs of the next higher order. On the right valve, the ribs tend to be clustered in groups of from two to four (Fig. 9d). As in other species of Caribachlamys and Crassadoma, the predominant mode of rib introduction is by rib-flank intercalation on the right valve and by either rib-flank or sub-medial intercalation in rib interspaces on the left. Because rib introductions occur throughout ontogeny, the ribs at the margin are of many dif- ferent sizes. The estimate in the diagnosis that fewer than 50 ribs are produced in mature individuals is based on the largest individuals examined, which have heights between 30 and 40 mm. The major ribs of some individuals have undercut flanks, but because the crests are rounded rather than flattened, the ribs can hardly be called I-beam shaped in cross section. The left valves of all of the specimens examined have a zone in early ontogeny in which the scales atop the major ribs are very widely spaced (spacing between 1.5 and 2 mm). Most specimens have distally open scales, the closed knobby condition being rare. Coloration of the shell is highly variable, with orange or purple maculations forming in the spaces between major ribs in early ontogeny, the color pattern becoming more uniform later. Some individuals have valve interiors that are yellow, but this is not a constant feature. Comparison.— Caribachlamys mildredae differs from both C. sentis and C. ornata in having fewer ribs which, on the left valve, are more distinctly ordered. The major ribs of the left valve of some C. mildredae have undercut sides but differ from the I-beam shaped ribs of C. ornata in lacking flattened, non-scaly, antimarginally striate crests. C. mil- dredae more closely resembles C. imbricata in color, in- cluding the yellow internal hue, but differs in having four orders of rib height on its left valve and from 40 to nearly 50 ribs and riblets at the distal margin of mature individu- als. C. imbricata, in contrast, has only three orders or rib height, and its major ribs bear fewer and much larger scales than do those of C. mildredae. For a comparison with C. paucirama, see the preceding section. 226 AMER. MALAC. BULL. 10(2) (1993) Living habits.— Caribachlamys mildredae lives in shallow water (1-20 m) on coral reefs and in crevices on rock jet- ties, its living habit thus overlapping that of C. sentis. Geographic range.— Thus far Caribachlamys mildredae is known with certainty only from southeastern Florida (West Palm Beach southward through Florida Keys to Dry Tortugas). Reports of its presence in Bermuda (Abbott, 1974: 443) are possibly the result of misidentification of what is herein referred to as the Bermuda morph of C. imbricata (see below). Stratigraphic range.— Middle Pliocene? to present. No unequivocally identified specimens of Cari- bachlamys mildredae are known from the fossil record. There is, however, a small (15.5 mm) left valve from the Pinecrest Beds of Florida that could be an early representa- tive of this species (USNM(P) 474641). It is from the same locality (USGS 26543) that yielded the single Pinecrest specimen of C. paucirama in Collier County (see above). This specimen has a greater number of ribs at a shell height of 10 mm than do specimens of extant C. mildredae and also a greater number of ribs at its disk margin than would be expected in C. mildredae of comparable size. The speci- men differs from co-occurring C. paucirama, however, in having its ribs distinctly ordered, the largest ribs lacking scales in early ontogeny and later having scales that are much more widely spaced that are those on intervening riblets. The microsculpture of the specimen is consistent with that of either C. paucirama or C. mildredae. Discussion.— In his original description of Pecten (Chlamys) imbricatus mildredae, Bayer (1941: 46) gave as his basis for linking the new taxon to P. imbricatus “the similar scheme of ribbing; the enlarged, sometimes cupped scales; the yellow and purple interior; and the large size of individuals.” He did not, however, describe any intergrada- tion between the two taxa, nor did he mention any ecologi- cal or geographic separation, both of which would be expected if the two taxa are indeed subspecies. Bayer (1943: 110) later elevated his new “variety” to species rank, while still maintaining that P. imbricatus is its closest rela- tive. Abbott (1974: 443) mentioned that Chlamys mil- dredae “may be a hybrid between sentis and ornata” but gave no reason why hybridization should not be between Caribachlamys sentis and C. imbricata, the latter being the species to which Bayer thought C. mildredae to be most closely related. At any rate, the hybridization hypothesis seems to have little merit. C. mildredae shows no more variation than do associated species; there is no indication of a “hybrid swarm” such as is known to occur commonly in hybridizing situations. It seems more likely, particularly in view of the new species C. paucirama described above, that C. mildredae is a relict species and that its morphology represents an early stage in the evolution of the C. imbrica- ta lineage. This could also explain its present-day rareness and its limited geographic distribution. Material examined.— Recent material: USNM:11 lots containing 11 specimens; ANSP: 3 lots containing 3 speci- mens, from southeastern Florida and Florida Keys. Fossil material: USNM(P) 474641, 1 Right valve, USGS 26543 (TU 1175) from spoil banks along canals south of Florida Highway 858, and 3.2 km [2 miles] east of junction with Florida Highway 846 (SE!/4, section 24, T48S, R27E), northeast of Naples and southwest of Immokalee, Collier Co., Florida. Pinecrest Beds. Collected by S.E. and R. E. Hoerle. . Caribachlamys imbricata (Gmelin, 1791) (Figs. 7c,f,1,1; 9f-j) Ostrea imbricata Gmelin, 1791: 3318, living, “in mari rubro” [erroneous]. Pecten imbricatus (Gmelin), Lamarck, 1819: 171. Chlamys imbricata (Gmelin), Abbott, 1954: 364, pl. 34f. Type specimens.— The specimen represented by the figure that Gmelin cited, which is in Chemnitz (1784), vol. 7, pl. 69, fig. G, is designated as the lectotype of Ostrea imbrica- ta Gmelin. Many of the specimens illustrated by Chemnitz were from the Spengler Collection housed at the Zoological Museum of Copenhagen (Keen, 1966). It is possible that the illustrated specimen of O. imbricata could be in that collection. I examined specimens labeled as Pecten imbri- catus by Lamarck in Paris and Geneva; they all conform to existing concepts of the western Atlantic taxon. Type locality.— Corrected herein to the Antillean region, western Atlantic. Diagnosis.— Caribachlamys with left valve less convex than right and distinctly flattened; major ribs few in num- ber, commonly 10 on right valve and nine on left, with five ribs on left valve tending to be more prominent and with higher scales than others; secondary riblets variably devel- oped, of much lower height than major ribs; major ribs somewhat trigonal in cross section on right valve, rounded on left; scales on major ribs large, widely spaced, distally concave, commonly forming closed knobs; early commar- ginal lirae obscured by stronger antimarginal striae and of highly irregular trend; antimarginal striae absent from crests of major ribs. WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 22] Morphological variation.— Caribachlamys imbricata dis- plays substantial geographic variation, but on the basis of the collections examined this variation appears to be geo- graphically clustered rather than clinal. For purposes of discussion, two intergradational extremes of variation are referred to herein as the Florida morph and the Bermuda morph. Parallel differences between the two are as follows: Florida morph (Figs. 9f, g): (1) In the early ontogeny of the right valve of the Florida morph, secondary costae that are close to or on the flanks of the primary ribs appear by a shell height of 5 mm. By a height of 10 mm, these secondary costae either disappear or are represented by the expanded, rounded rib flanks of the primary ribs; (2) The posterior auricle of the right valve has unevenly distrib- uted costae, the dorsal one or two closer together and much stronger than the others; (3) On the left valve, third-order riblets are usually absent at least in early ontogeny (up to a shell height of 10 mm) and commonly absent throughout ontogeny; (4) On the left valve, there is usually only one riblet present on the anterior side of the anteriormost prima- ry rib in early ontogeny, increasing to two in later growth. These anteriormost riblets lack scales; (5) On the left valve, radial riblets are unevenly distributed on both auri- cles, the dorsalmost riblets commonly much stronger than the others. Bermuda morph (Figs. 9h-l): (1) On the right valve, secondary riblets are distinct and well separated from the primary ones at a shell height of 5mm. Ata shell height of 10mm, these secondary costae are still distinct from the major ribs, the major ribs being without expanded flanks; (2) The posterior auricle of the right valve has fairly evenly distributed riblets of even amplitude; (3) On the left valve, third-order riblets are commonly present by a shell height of 10 mm and may be numerous by this stage of ontogeny; (4) On the left valve, there are usually two distinct and fair- ly strong riblets present on the anterior side of the anterior- most primary rib, commonly increasing to three in later growth. One of these anterior riblets may bear cuspate scales; (5) On the left valve, radial riblets are numerous on both auricles, with only slight or no increase in amplitude toward the dorsal margin. Caribachlamys imbricata is generally lightly pig- mented, the darkest parts being the dark reddish or purple maculations that occur in earlier ontogeny between the pri- mary ribs. Solidly colored shells, such as are common for C. sentis, are relatively rare. They have been observed, however, among specimens of the Bermuda morph from the Caribbean coast of Panama. The valve interiors of both morphs are commonly yellow, sometimes brilliantly so. Comparison.— Caribachlamys imbricata can be distin- guished from its congeners by its broad, high-amplitude major ribs. Unlike congeneric species, the scales on these major ribs remain widely spaced throughout ontogeny and commonly form closed knobs. Living habits.— Caribachlamys imbricata lives byssally attached in crevices, beneath coarse rubble and coral heads, and between coral branches on reefs and in both back-reef and fore-reef areas. Depth records for live-collected speci- mens in museum collections are all shallow, ranging from low subtidal to 20m. Geographic range.— Caribachlamys imbricata appears to have a tropical distribution. It occurs in Bermuda, along the North American mainland only from Miami southward through the Florida Keys, in the Bahamas, and through the Antilles to Curacao and Aruba along the Venezuelan coast. It appears to be absent from the northern Gulf of Mexico, and the only record from the southern Gulf is from Alacran Reef on the north side of the Yucatan Peninsula. Few records are as yet available from the Caribbean coast of Central America with the exception of Belize (offshore cays) and Panama, which are the best sampled areas. The species is not known from the Atlantic coasts of South America. All specimens of Caribachlamys imbricata from Bermuda examined by me are members of the Bermuda morph, but this morph is also present in Jamaica, Panama, Colombia, Venezuela, and possibly within samples that also contain the Florida morph in the Bahamas. Stratigraphic range.— Lower? Pleistocene to present. The oldest fossils of Caribachlamys imbricata thus far found are specimens in the Trechmann collection (BMNH) from the Pleistocene Coral Rock of Barbados at “Cane Garden” and “Highgate”. These did not appear in the lists published by Trechmann; the species determinations are based on my examination of the specimens in 1979. As reviewed above in the section on C. ornata, Trechmann (1933) considered the Coral Rock to be mainly of early Pleistocene age. I also found fossil Caribachlamys imbricata (USNM(P) 474642) in limestone blocks on Windley Key in the Florida Keys. These blocks presumably came from the Windley Key quarry, which exposes the Key Largo Limestone of late Pleistocene age (Multer, 1969; Stanley, 1966). According to Multer (1969: 109), this limestone represents a back-reef environment and is only about 100,000 years old. Although the specimens are fragmen- tary, they have ribbing characteristics that suggest that they belong to the Bermuda morph. The only other known occurrences of fossil Caribachlamys imbricata, from Cuba, Haiti, and the 228 AMER. MALAC. BULL. 10(2) (1993) Dominican Republic, also appear to be Pleistocene in age (see below under material examined). Discussion.— The two extremes of morphology described above as the Florida morph and the Bermuda morph are of considerable interest in that the former is clearly more derived than the latter when they are compared to outgroup taxa within Caribachlamys, particularly C. mildredae. Among the fossil specimens examined, those from the Windley Key Limestone of Florida (USNM(P) 474642) have ribbing characteristics that suggest that they belong to the Bermuda morph, and this also applies to the specimens from Haiti (USNM(P) 481997 and 481998). The sole Cuban specimen on hand, however, belongs to the Florida morph. A likely explanation for the extant Bermuda popu- lations, which belong exclusively to the Bermuda morph so far as known, is that they represent dispersal to Bermuda early in the evolution of the species at some time during the Pleistocene. Apparently modern oceanographic conditions do not permit continued dispersal of this species to Bermuda, and most members of populations elsewhere have assumed a more derived condition than those in Bermuda. The presence of Bermuda morphs among some extant populations outside of Bermuda suggests that the Bermuda populations should not be referred to as a geo- graphic subspecies. Material examined.— Recent material: USNM: 60 lots containing about 150 specimens, from southeastern Florida, Florida Keys, Bermuda, Bahamas, Mexico (Alacran Reef), Jamaica, Cuba, Puerto Rico, Virgin Islands, Anguilla, Barbuda, Curacao, Aruba, Belize, Panama, Colombia, and Venezuela; MNHN: several specimens from Guadeloupe and 2 specimens in the Lamarck Collection without locali- ty; GNH: 3 specimens in the Lamarck Collection, without locality; AM: specimens from Bonaire, Curacao, and Aruba; UCD, collection of Geerat J. Vermeij: a complete shell of the Bermuda morph from a depth of 5 m in a cave near Runaway Bay, north coast of Jamaica. Fossil material: USNM: USNM(P) 474642, 7 valves or fragments, Windley Key, Florida, Key Largo Limestone, Upper Pleistocene; USNM(P) 481977 and 481978, 2 LV and 1 RV fragment, USGS 9764, on coast 100 m west of old fort on west side of entrance to Port-de Paix harbor, Departement du Nort Ouest, Haiti, “Bed 2 of section, Pleistocene, collected by W. P. Woodring, 1921; USNM(P) 481998, 1 LV, same locality as preceding; USNM(P) 474810, 1 RV, USGS 7943, Naval Station, Guantanamo Bay, Cuba, 6 to 18 m (20 to 60 ft) above sea level, Pleistocene, collected by P. Bartsch, 1917. BMNH: Barbados, Coral Rock, Pleistocene (see “Stratigraphic Range” for details). FIELD SPECIMENS: Fossils in place in the lowest (and youngest) Pleistocene reef terrace along the south coast of the Dominican Republic east of Boca Chica. (These were too brittle and fragile to remove from the rock.) Tribe Mimachlamydini, New Tribe Diagnosis.— Chlamydinae with pitted microsculpture in pre-radial stage of left valve, the pits commonly extending into early radial stage; ribbing pattern simple, with the first ribs introduced in ontogeny tending to remain as major ribs and commonly the only ribs throughout ontogeny; ribs on shell interior usually with carinate edges; commarginal lirae in rib interspaces absent, obscure, or limited to early ontogeny; microsculpture between ribs commonly in a divaricating or herringbone pattern. Discussion.— The new tribe Mimachlamydini is represent- ed by the extant genera Mimachlamys Iredale, 1929, of the eastern Atlantic and Indo-Pacific and Spathochlamys, new genus, of the western Atlantic and eastern Pacific. It also includes the extinct genus Dimarzipecten Ward, 1992, of the tropical western Atlantic. The tribe forms a bridge between the tribes Crassadomini and Aequipectinini and is a sister group of the Aequipectinini in that both share the common presence of a coarsely pitted left beak before the start of the radial stage and a simple ribbing pattern in which the first ribs introduced tend to remain the major ribs and commonly the only ribs throughout the remainder of ontogeny. In both the Mimachlamydini and Aequipectinini these ribs develop carinate edges on the shell interior. In the Mimachlamydini, however, internal rib carinae are absent in the late ontogeny of the more primitive members of the tribe, whereas in the Aequipectinini these carinae are uni- versally present throughout ontogeny. Mimachlamydini differ from Aequipectinini in having a plesiomorphic chlamydinid two-element hinge with the resilial teeth not elongated toward the free edges of the auricles. In the Aequipectinini the resilial teeth become larger in size than the dorsal teeth and tend to dominate the hinge structure. Lastly, the Aequipectinini develop charac- teristic wavy and looped commarginal lirae not present in the Mimachlamydini. The greatest number of extant species in the Mimachlamydini occurs in the western Indo-Pacific region. The genus Mimachlamys in particular contains at least a dozen species but with scores of species-group names avail- able due to the prolific overnaming that characterized typo- logical systematics of earlier days. Among these species are the well known M. senatoria (Gmelin, 1791) complex, the colorful M. nobilis (Reeve, 1852) of Japanese waters, and WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 229 the giant M. townsendi (G. B. Sowerby III, 1895) of the Red Sea, the last being one of the largest of extant pec- tinids. In the eastern Atlantic and Mediterranean, the tribe is represented at present only by M. varia (Linnaeus, 1758). Phylogenetic relationships of some of these species were recently reviewed by Waller (1991: 31). The tribe is absent in the eastern central Pacific including the Hawaiian Islands, but a pair of extant geminate species assigned here- in to Spathochlamys, new genus, is present in the tropical American region, one member of the pair in the eastern Pacific, the other in the western Atlantic. The tribe Mimachlamydini has an extensive fossil record in the Tertiary of Europe beginning in the Paleocene (Waller, 1991: 32) but is poorly represented in the Amer- ican Tertiary. The new genus Spathochlamys is apparently the only member of the group that dispersed through sea- way connections from the tropical Atlantic into the eastern Pacific. Spathochlamys, new genus Etymology.— Based on the Greek “spathe,” meaning spade, in reference to concave-up scales, combined with the genus name Chlamys, which in turn is based on the Greek word “chlamys,” meaning “mantle”. Diagnosis.— Mimachlamydini with rounded or broadly trigonal ribs separated by interspaces each containing a sin- gle very narrow riblet; crests of major ribs bearing narrow erect scales that are concave on their upper (dorsal) sides; edges of ribs on inner surfaces of valves strongly carinate; microsculpture in early rib interspaces obscure or dominat- ed by commarginal lirae which cross interspaces without strong curvature. Type species.— Pecten benedicti Verrill and Bush in Verrill, 1897, extant, western Atlantic. Other Species.— Pecten vestalis Reeve, 1853 [= Pecten (Chlamys) lowei Hertlein, 1935], extant, eastern Pacific, and P. vaginulus Dall, 1898, fossil, Miocene-Pliocene, Caribbean region and southeastern United States. Geographic range.— Western Atlantic from at least Massachusetts to Brazil and throughout the Gulf of Mexico and Caribbean; eastern Pacific and Gulf of California from southern California to Ecuador and westward to the Galapagos Islands. Stratigraphic range.— Upper Miocene to present. Discussion.— The new genus differs in a number of quali- tative features from Mimachlamys. Both the extant and fos- sil species of Spathochlamys have much more strongly developed carinate edges on the ribs on the inner shell sur- face than do species of Mimachlamys, where these carinae are sometimes minimally developed or restricted to early growth stages (as in M. varia, an extant species of the east- ern Atlantic). Concave-up scales, which form atop at least some of the ribs in Spathochlamys, are unknown among extant species of Mimachlamys, where scales are concave- down (i.e. concave side facing toward the ventral margin). Some extinct European middle and upper Tertiary Mimachlamydini, e.g., M. angelonii (De Stefani and Pan- tanelli, 1878), have concave-up scales but lack medial riblets, and other aspects of their morphology suggest that they have a phyletic history that is independent of that of Spathochlamys. There is a strong resemblance in ribbing pattern between Spathochlamys and Dimarzipecten Ward, 1992, a genus recently introduced to accommodate what its author believed to be a single species, Pecten crocus Cooke, 1919, that was originally described from the early Miocene (or late Oligocene) of Anguilla. Ward (1992) stressed that Dimarzipecten has a narrow stratigraphic range (restricted to lower Miocene and possibly upper Oligocene) and a broad geographic range extending from the Antilles to the Belgrade, Edisto, and Tampa Formations and correlative stratigraphic units of the Carolinas, Georgia, and Florida. However, I have identified a second apparently undescribed species of Dimarzipecten in the Chipola Formation of Florida (USNM(P) 474643 and 474644, from Farley Creek, NE!/4 sec. 21, TIN, R9W, Calhoun Co.). This extends the stratigraphic range of the genus into the upper lower Miocene (Burdigalian in age; see Akers, 1972: 10). Like Spathochlamys, Dimarzipecten has a Chlamys-like shape, ribs that become trigonal in cross-sec- tion at least in late ontogeny, narrow scales atop the ribs, and a fairly regular ribbing pattern with a single medial riblet in each interspace. Dimarzipecten, however, differs from Spathochlamys in lacking internal rib carinae and in having convex-up rather than concave-up scales on the rib crests. Furthermore, Dimarzipecten has a plesiomorphic microsculptural pattern of coarse antimarginal striae in the rib interspaces. The configuration of these striae is like that present in many modern Mimachlamys, including a herring- bone-like configuration in rib interspaces in the center of the disk in early ontogeny. In Spathochlamys, antimarginal striae are absent or obsolete in early ontogeny except on and near the disk flanks, and commarginal lirae are present in the rib interspaces. A third genus, possibly new and referred to herein as Genus A pending further research, is characterized by Pecten sansebastianus Maury, 1920, of which P. (Chlamys) 230 AMER. MALAC. BULL. 10(2) (1993) portoricoensis Hubbard, 1920, is a junior synonym. This species occurs in the San Sebastian Shale and lower Lares Limestone of Puerto Rico (Maury, 1920; Hubbard, 1920), formations that are considered middle to late Oligocene in age (Maurrasse, 1990, fold-out correlation chart, Column 20). Genus A has a high, narrow shape like that of Dimarzipecten, and as in members of that genus, a small medial riblet appears in most rib interspaces in late ontoge- ny. Like Spathochlamys, Genus A has minute concave-up scales on its rib crests and the ribs have strongly carinate edges on the inner surfaces of the valves. Genus A differs from both Dimarzipecten and Spathochlamys in having a distinctive Aequipecten-like microsculpture in rib inter- spaces in early ontogeny. This pattern consists of wavy commarginal lirae that are strongly looped in a dorsal direc- tion on the rib flanks and in a ventral direction in the rib interspaces; the looped lirae on the rib flanks may become cuspate, as in some Aequipecten and in most Cryptopecten. Like Dimarzipecten and Spathochlamys, Genus A lacks a foliated-calcite transgression on the inner surface of its umbonal region, but this is a plesiomorphic state that can- not serve as an indicator of close relationship. Tentatively, Genus A is considered a member of the tribe Aequi- pectinini based on its microsculptural pattern. If Spathochlamys originated from a species of Dimarzipecten, then it seems likely that this origin was within the middle Miocene. The latest Dimarzipecten thus far uncovered is the Chipola (Burdigalian) species referred to above, and the earliest undoubted Spathochlamys are the specimens of S. vaginula, late Miocene and early Pliocene, referred to below. Although fossil specimens of Spatho- chlamys are rare, those that have been brought together for this study suggest that S. vaginula of the Miocene and early Pliocene gave rise to both of the extant species but at differ- ent times. S. vestalis of the eastern Pacific probably origi- nated in the late Miocene; S. benedicti of the western Atlantic probably originated in the late Pliocene. Spathochlamys benedicti (Verrill and Bush, 1897) (Figs. 10a-k) Pecten mundus Reeve, 1853, species 151, pl. 33, fig. 151, locality unknown, non Pecten mundus M’Coy, 1844, p. 97, Carboniferous of Ireland. Chlamys benedicti Verrill and Bush, In: Verrill, 1897: 74- 75, “off Marthas Vineyard, in 1356 fath., dead; West Indies, in 25 to 72 fath., living.” Chlamys benedicti Verrill and Bush, Verrill and Bush, 1898, 20(1139): 834-835, pl. 84, figs. 1-2. Chlamys verrilli Dollfus, 1898: 180, new name for Chlamys benedicti Verrill and Bush, 1897, non Pecten benedictus Lamarck, 1819. [invalid new name. ] Pecten (Chlamys) mundus Reeve, Bavay, 1902: 404-406, pl. 8, figs. 8, 9. “European” [erroneous]. Pecten (Chlamys) nympha Bavay, 1906:. 246-247, pl. 7, figs. 3,4. “Caribeeum mare?” Pecten mundus Reeve, Bavay, 1913: 26, “Bahia” [Brazil]. Chlamys (Chlamys) benedicti Verrill and Bush, Weisbord, 1964: 139-142, pl. 14, figs. 8-11, Playa Grande Formation (Catia Member), “Pliocene” [now dated as Pleistocene], Distrito Federal, Venezuela. Chlamys (Chlamys) munda Reeve, Fischer-Piette and Testud, 1967: 184-185, Brazil. Types.— Johnson (1989: 23) designated as lectotype of Chlamys benedicti Verrill and Bush a specimen from the Yale Peabody collection (YPM 8833) on the grounds that the “figured holotype” that was supposed to be in the USNM could not be located. The lectotype, a single right valve illustrated by Johnson (1989, pl. 8, fig. 4), is from R/V Albatross Stations 2369-2374, 29° 11’ 15”N, 85° 29’ 32” W, south of Panama City, Florida, 25-27 fm (= 46- 49m). In fact, Verrill and Bush (1897) did not specify a holotype in their original description, which referred to specimens from two localities. When they republished the same description a year later, they presented a drawing of one of their specimens, an articulated shell, USNM 202999, from United States Fish Commission [R/V Albatross] Stations 2369 to 2374, 25 to 27 fm, Gulf of Mexico between the Mississippi Delta and Cedar Keys, Florida. This is clearly the same set of stations from which the lec- totype designated by Johnson (1989) came. Although Johnson’s lectotype stands, the specimen illustrated in 1898 (USNM 202999) has now been located in the USNM col- lections. The holotype of Pecten mundus Reeve, 1853, is a pair of matching valves, the left broken, in The Natural History Museum, London (BMNH 1950.11.14.48). The right valve is refigured herein (Figs. 10a,b). The holotype of Pecten (Chlamys) nympha Bavay, 1906, is the specimen illustrated herein (Figures 10c and 10d). The specimen, which is in Paris (MNHN), is the one that Bavay illustrated and apparently was the only specimen that was available to him at the time that he wrote his description. Type locality.— Gulf of Mexico, 29° 11’ 15”N, 85° 29’ 32” W, south of Panama City, Florida, 25-27 fm (= 46- 49m). There is a potential source of confusion regarding the type locality of Spathochlamys benedicti. In their origi- nal description Verrill and Bush (1897: 75) referred to two WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 231 Fig 10. Spathochlamys. a, b. Holotype of Pecten mundus Reeve {= S. benedicti), BMNH 1950.11.14.48, locality unknown, right valve uncoated, height 13.3 mm, and detail of surface, horizontal field width 4 mm. c, d. Holotype of P. nympha Bavay (= S. benedicti), Caribbean?, Paris Museum, matching right and left valves, height 15.6mm. e-g. S. benedicti, USNM 523035, Gulf of Mexico off Carabelle, Florida, matching right and left valves, height 17.0 mm., and detail of left exterior, horizontal field width 5.7 mm. h. S. benedicti, USNM 764660, Gulf of Mexico off Campeche, Mexico, internal rib carinae of left valve, horizontal field width 9 mm. i. S. benedicti, USNM(P) 474645, James City Fm., right bank of Neuse River, 3.2 km below James City, Craven Co., North Carolina, right valve, height 21.9 mm. j. S. benedicti, USNM(P) 474656, Pleistocene?, Minnitimmi Creek, Bocas Island, Panama right valve, height 17.9 mm. k. S. benedicti, USNM(P) 474655, Pleistocene?, Bocas Island, Panama, left valve, height 19.2 mm. 1, m, q, r. S. vestalis, mainland morph, USNM 774188, Gulf of California off Angel de la Guardia Id., Mexico, matching right and left valves: right hinge of length 10.0 mm; internal rib carinae, horizontal field width 9.3 mm; right and left exteriors, valve height 17.8 mm. n-p. S. vestalis lectotype, BMNH 1992115, “West Indies” [erroneous], matching right and left valves, valve height 22.0 mm, and right hinge of length 9.3 mm. s, t. S. vestalis, mainland morph, USNM 774191, off North Isle, Pinas Bay, Panama, matching left and right valves, height 14.9 mm. 252 AMER. MALAC. BULL. 10(2) (1993) localities, one “off Martha’s Vineyard, in 1356 fath., dead,” the other “West Indies, in 25-72 fath., living” [italics added]. Specimens from the first locality are present in the Smithsonian collection, catalogued as USNM 52542, USFC Station 2571, with the locality data from that station corre- sponding to that given in the original description. The lot consists of two right valves, the larger of which is exactly 6mm in height and lemon yellow in color; it is clearly the “largest specimen” referred to by Verrill and Bush in their original description. In the following year, Verrill and Bush (1898: 834- 835, pl. 84, figs. 1, 2) repeated their original description, but this time they provided a drawing of a pair of matching juvenile valves (USNM 202999). Because the description is identical to the first one, it can be inferred that the two localities to which they refer in the second description are also the same as in the first. The deeper station, USFC Sta. 2571, is certainly the same, but the second, from which the illustrated specimen came, presents a problem. In the 1898 description the second locality is given as “stations 2369 to 2374, in 25 to 27 fm [italics added],” and this time no men- tion was made of the West Indies. The station-book entries for USFC Stations 2369 through 2374 verify the depth range of 25 to 27 fm and give the locality as the Gulf of Mexico between the Mississippi Delta and Cedar Keys, Florida, 1885. Because there is no material in the Smithsonian collections of Chlamys benedicti from the West Indies bear- ing either the depth range of 25-72 fm as originally pub- lished or 25 to 27 fm as published a year later, it is inferred that the references to “West Indies” and “72 fathoms” in the original description were errors and that the specimen later illustrated by Verrill and Bush (1898) from the Gulf of Mexico is one that was before them when they wrote the original description. Diagnosis.— Spathochlamys with length of anterior outer ligament less than twice length of posterior outer ligament; major ribs commonly scaly throughout ontogeny; rib-flank costae poorly developed, not becoming scaly until late ontogeny; posterior auricles erect, with posterior auricular margins forming an angle with hinge line of 90° or less. Morphological variation.— No apparent geographic trends in size, shell shape, or sculpture were detected across the broad geographic range of this species. The largest specimen examined has a shell height of only 23 mm, and most specimens are in the range of 10 to 15 mm. Comparison.— Spathochlamys benedicti closely resem- bles its eastern Pacific geminate species, S. vestalis. Mature shells of the former have relatively larger posterior auricles, the AOL/POL ratio being less than two, whereas in S. vestalis this ratio is greater than two. The posterior auricles of S. benedicti have an erect appearance, tending to form a right or acute angle with the hinge, whereas those of S. vestalis are uniformly obtuse. Lastly, the rib-flank costae of S. benedicti are generally not well developed and seldom bear prominent scales; in S. vestalis these same costae tend to develop early in ontogeny and form scales that rival those of the major ribs in height. Spathochlamys benedicti resembles the extinct western Atlantic S. vaginula (Dall, 1898) in shape of disk and auricles but differs in having consistently scaly and trigonal ribs throughout ontogeny, even in the central sector of the disk. In S. vaginula the ribs lack scales in early ontogeny in the central part of the disk and are commonly rounded rather than trigonal in cross section. Lastly, S. vaginula from the Neogene of the Florida Peninsula attains a much larger size (commonly 30 to 35 mm in shell height). Living Habits.— The species lives in tropical to temperate waters across a broad range of depths, from about 2 to 800 m, but generally seems to be most common on the middle shelf at depths ranging from about 40 to 90 m (Waller, 1973: 47). The deep byssal notch, prominent active cteno- lium, adhering byssal threads on mature dried shells, and data taken at the time of collection all indicate that S. bene- dicti is probably byssally attached throughout life. In- dividuals attach to a great variety of substrates, including coral debris, sponges, and algal mats (Waller, 1973, and unpub. data). Reed and Mikkelsen (1987) referred to Chlamys benedicti as being a common member of the com- munity associated with the scleractinian coral Oculina vari- cosa Lesueur, 1830, off eastern Florida. Geographic Range.— Although dead specimens have been collected as far north as Latitude 40° 9’ N, southeast of George’s Bank (the locality referred to by Verrill and Bush in their original description), the distribution of this species is primarily from off Cape Hatteras southward throughout the Gulf of Mexico and Caribbean and as far south as Brazil. Rios (1985: 222) reported the species (as Chlamys munda) in Brazil from Amapa to Espirito Santo and in the Abrolhos and Trindade Islands. The species is also present in Bermuda (Waller, 1973). Stratigraphic Range.— Upper? Pliocene to present. Fossils of Spathochlamys benedicti are reported herein for the first time from the James City Formation of North Carolina (Fig. 101i). There is now substantial agree- ment that the age of this formation is early Pleistocene (Cronin et al., 1984: 40; Ward and Blackwelder, 1987: 114; Miller III, 1989; Cronin, 1990). Weisbord (1964) reported WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 233 the species from the Playa Grande Formation (Catia Member) of Venezuela. The age of the Playa Grande Formation is generally considered to be early Pleistocene (Bermiidez, 1980: 303), not early Pliocene as originally stated by Weisbord (1964), but the stratigraphic relation- ships of formations in the Cabo Blanco area of Venezuela are still controversial (e.g. see, for example, Gibson-Smith, 1976: 4). The fossil occurrences in Costa Rica are from beds now placed in the Moin Formation, which most work- ers have considered to be Pleistocene in age (e.g. Akers, 1972: 42; Cassell and Sen Gupta, 1989: 147; Robinson, 1990; Lyons, 1991: 159). Coates et al. (1992: 821), howev- er, date what are apparently the same beds as late Pliocene- early Pleistocene. The Panamanian fossils (Figs. 10j,k) are all from Bocas del Toro area from beds that contain a fauna that is clearly younger than that of the Miocene Gatun Formation and is assumed herein to be of late Pliocene or early Pleistocene age. Coates et al. (1992: 819) have re- cently described a stratigraphic sequence in this general area that extends in age from the late Miocene through the Pliocene, but they do not specifically mention what is pre- sent on Bocas Island and Nancy Cay, the sites that yielded S. benedicti. The oldest known representative of S. benedic- ti could be a specimen [USNM(P) 474649] from the Quebradillas Limestone of Puerto Rico. This limestone apparently forms the upper part of the Camuy Formation and may be early Pliocene in age but younger than the Gurabo Formation of the Dominican Republic and the Bowden shell beds of Jamaica (Bermudez and Sieglie, 1970). Discussion.— The holotype of Pecten mundus Reeve, BMNH 1950.11.14.48, was examined. It is a pair of matching valves, both broken, the right valve (Fig. 10a) having a restored height of 12.7 mm. Reeve’s species is clearly a senior synonym of Chlamys benedicti but is a junior primary homonym of P. mundus M’Coy, 1844, a fos- sil from the Carboniferous of Ireland. Dollfus (1898: 180) introduced the new name Chlamys verrilli to replace Chlamys benedicti Verrill and Bush on the erroneous assumption that Verrill and Bush’s name is a junior homonym of Pecten benedictus Lamarck, 1819. The endings of the two names, however, differ in case, not in gender, and this is sufficient to prevent homonymy. Bavay (1902) applied the name Pecten (Chlamys) mundus Reeve to two small shells in a collection said to come from Corsica and concluded that the species is “European and even French” [translation]. Although the specimen that he illustrated appears to be what is here called Spathochlamys benedicti, there is no reason to believe that the locality data are correct. My own studies of European pectinid collections have not turned up any authentic records of the species in the eastern Atlantic. In a later publication, Bavay (1913: 26) identified shells from Salvador (Bahia), Brazil, as P. mundus Reeve, concluding that the species is present on both sides of the Atlantic. In support of this contention, he reported that P. commutatus Monterosato, a well-known Mediterranean species, also occurs in Bahia, Brazil. It seems more likely, however, that his “P. commutatus” was in fact Argopecten noronhensis (Smith, 1885), a species which is known only from the western Atlantic and which may co-occur with S. benedicti (Waller, 1973). Bavay (1906) described Pecten (Chlamys) nympha on the basis of a shell in the Paris museum collection. Because the shell was found glued to a board that also held a specimen of P. antillarum Récluz, 1853, a well-known Caribbean species, Bavay assumed that his new species also came from that region. The holotype, a pair of match- ing valves marked by Bavay as the figured type, is refigured herein (Figs. 10c, d). It is clearly a junior synonym of Spathochlamys benedicti. Weisbord (1964: 139), in his report of fossil Chlamys benedicti from the Pleistocene Playa Grande Formation of north central Venezuela, referred (p. 141) to “the type” of C. benedicti and to “illustrations of the type” but gave no references. He was apparently assuming that the specimen illustrated by Verrill and Bush (1898) is the holotype, although it was never specified as such. Fischer-Piette and Testud (1967) applied the name Chlamys munda Reeve to specimens from Brazil, apparent- ly drawing support from Bavay’s (1913) previous report of its occurrence there. Apparently neither Fischer-Piette and Testud (1967) nor Bavay (1902, 1913) were aware that the name C. benedicti was already in use for this same species in the western Atlantic. Material examined.— Recent material: USNM: about 500 lots, from throughout the geographic range of the species; BMNH: syntypes of Reeve (1853) listed in above syn- onymy; BRM: 7 lots, from Antilles and South America; LM: 18 lots, mainly from off Guyana; MNHN: 7 lots, from South America. Fossil material: USNM: North Carolina USNM(P) 474645, 1 RV, USGS D273, “Croatan Sand”, right bank of Neuse River, 3.2 km [2 miles] below James City, Craven County. Collected by MacNeil and Malde, 1954. USNM(P) 474646 and 474647, 2 RV, James City Formation, about 0.4 km [1/4 mile] upstream from Johnson Point and at Johnson Point, right bank of Neuse River, 234 AMER. MALAC. BULL. 10(2) (1993) Craven County. Collected by T .R. Waller, 1963. Mississippi USNM(P) 474648, 10 RV, 12 LV, USGS 24729, shell bed between Clay Units I and II (Morgan et al., 1968, p. 151,152), on Mudlump 90, a mudlump island to the west of the entrance to the South Pass of the Mississippi River delta. Collected by T. R. Waller, 1969. Puerto Rico USNM(P) 474649, 1 RV, USGS 19814, Que- bradillas Limestone, from sink on Ramey Air Force Base, | km west and 3.3 km south of edges of quadrangle, Aquadilla Quadrangle. Collected by C. W. Cooke and A. D. Watt, 1955. Costa Rica USNM(P) 474650, 1 LV, USGS 5884b, section exposed in railroad cut, from third fossiliferous zone above level of rails, Moin Hill. Collected by D. MacDonald, 1911. USNM(P) 474651, 1 LV, USGS 18693, “colline en démolition,’ Limon. Collected by Pittier, 1898 or 1899. USNM(P) 474652, 1 RV, 1 LV, USGS 21035, hill- side cut and spoils dump at new site of buildings of Colegio de Lim6n in northeastern outskirts of Puerto Limon. Collected by E. Malavassi, A. A. Olsson, and W. P. Woodring, 1958. USNM(P) 474653, 2 RV, USGS 21036, scraped hillside slope in Barrio Cementario distinct in southern out- skirts of Puerto Limon, east of cemetery. Collected by E. Malavassi, A. A. Olsson, and W. P. Woodring, 1958. Panama USNM(P) 474654, 1 LV, USGS 8306, Bocas Island. Collected by A. A. Olsson, 1917. USNM(P) 474655 (Fig. 10k), USGS 8307, 1 LV, Bocas group, Bocas Island. Collected by A. A. Olsson. USNM(P) 474656 (Fig. 10j), USGS 8309, 3 RY, 1 LV, Minnitimmi Creek, Bocas Island. Collected by A. A. Olsson. USNM(P) 474657, 11 RV, 7 LV, USGS 8349, Bocas group, Conch Point member, Bocas Island. Collected by A. A. Olsson and Sears, 1919. USNM(P) 474658, 2 RV, USGS 8494, “Spondylus group” [apparently with reference to a group of strata], Nancy Cay. Collected by A. A. Olsson, 1919. Spathochlamys vestalis (Reeve, 1853) (Figs. 10 1-r; 11 a,b) Pecten vestalis Reeve, 1853, species 154, pl. 33, fig. 154, “West Indies” [erroneous]. Pecten (Chlamys) lowei Hertlein, 1935: 308-311, pl. 19, figs. 1, 2, 7, 8. “Gulf of California; Galapagos Islands. ?Catalina Island, California.” Types.— The syntypes of Pecten vestalis Reeve comprise three specimens: a pair of matching valves and two smaller specimens, one a right valve and the other a non-matching left valve. The first, BMNH 1992115 (Figs. 10n,p), ht. = 22.0 mm, L = 19.3 mm, is the specimen figured by Reeve (1853, pl. 33, species 154) and is selected herein as the lec- totype. The holotype of Pecten (Chlamys) lowei Hertlein is a pair of matching valves, California Academy of Sciences, no. 6878, from Carmen Island, Gulf of California, depth 37 m (20 fm) (Hertlein, 1935: 308, pl. 19, figs. 1,2,7,8). Type locality.x— Eastern Pacific off Baja California, Mexico. Diagnosis.— Spathochlamys with length of anterior outer ligament greater than twice length of posterior outer liga- ment; rib-flank costae beginning on right valve at shell height of less than 12 mm; rib crests with scales ranging from very high in relief to low and sometimes absent on ribs in center of disk; free margin of posterior auricle slightly concave or straight, forming an angle with dorsal margin exceeding 110°. Morphological variation.— Spathochlamys vestalis, like S. benedicti, is small in size, seldom attaining a shell height greater than 23 mm. Unlike its western Atlantic counter- part, S. vestalis displays some geographic variation in mor- phology, particularly between mainland populations and those of the Galapagos Islands. These end-points of varia- tion are referred to below as the “mainland morph” and the “Galapagos morph.” The mainland morph (Figs. 10 1-t), which includes the lectotype (Figs. 10n-p), is typified by populations from Santa Catalina Island, California, southward along Baja California, Mexico, throughout the Gulf of California, and along the Mexican coast. These morphs are characterized by the abundant well-developed scales both on the crests of the major ribs and on the secondary rib-flank and medial costae. They also have a conspicuous commarginally lirate stage in early ontogeny. In this early stage, the commargin- al lirae are prominent in the rib interspaces and are of high- er relief than the proximal ends of the medial costae, which are interrupted by the lirae. The extent of the lirate stage is variable, extending to shell heights from 6 to 10 mm on the right valve and slightly less on the left valve. The number of major ribs is generally 20 or 21, but there may be as few as 17 or as many as 23. WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 235 Fig. 11. Spathochlamys. a, b. S. vestalis, Galapagos morph, USNM 774206, Tagus Cove, Albemarle Id., Galapagos Ids., Ecuador, matching right and left valves, height 12.2 mm. c-f. S. vaginula, lectotype (right valve) and paralectotype (non-matching left valve), USNM(P) 135786, USGS Loc. 2580, Lower Pliocene, Bowden Beds, Bowden, Jamaica: right valve, height 14.4 mm; ribbing of right valve, horizontal field width 5.5 mm; left valve, height 14.5 mm; ribbing of left valve, horizontal field width 7 mm. g-i. S. vaginula, USNM(P) 474659, Upper Miocene or Lower Pliocene, Gurabo Fm., Rio Yaque del Norte, Dominican Republic: left valve, height 13.8 mm; non-matching right valve missing anterior ear, height 12.8 mm; detail of ribbing on left valve at start of scaly phase, horizontal field width 6 mm. j. S. vaginula, USNM(P) 474660, Upper Miocene or Lower Pliocene, Gurabo Fm., Rio Gurabo, Dominican Republic, left valve, height 14.8 mm. k-o. S. vaginula, Tamiami Fm. sensu Waldrop and Wilson (1991) [= “Bed 11”], APAC Pit, near Sarasota, Sarasota Co., Florida: k) USNM(P) 474661, anterior view, height 30.2 mm; 1) USNM(P) 474662, right valve, height 33.1 mm; m) USNM(P) 474663, right valve, height 22.8 mm; n) USNM(P) 474664, left valve, height 33.9 mm; 0. USNM(P) 474665, left valve, height 29.1 mm. p. S. vaginula, USNM(P) 474666, Upper Miocene or Lower Pliocene, near base of Jackson Bluff Fm., Jackson Bluff, Leon Co., Florida, right valve, height 22.2 mm. 236 AMER. MALAC. BULL. 10(2) (1993) On the basis of the few specimens available for study, mainland morphs from off Panama (Fig. 10s, t) and Colombia and some of those from off Isla Clarion, Islas de Revillagigedo, Mexico, differ from the typical mainland morphs in having smaller scales and smoother rib inter- spaces in early ontogeny. They tend to lack conspicuous commarginal lirae in the interspaces of the right valve before the start of medial costae, while their left valves retain a commarginally lirate early growth stage. The absence of commarginal lirae is not due to wear, because fine antimarginal microsculpture is still present in the inter- spaces. The Galapagos morph (Figs. 1 1a,b) differs in hav- ing a broader umbonal angle (95 to 100°, in contrast to that of the mainland morph, which has an umbonal angle rang- ing from 85 to 96°). The Galapagos morph is also some- what more inflated even though it is smaller in size. The most distinctive feature of the Galapagos morph, however, is the diminutive scales present on the rib crests. The rib crests are somewhat flattened while still retaining a narrow central keel bearing the scales. With few exceptions, speci- mens from the Galapagos Islands are of this morphology. This morph also occurs, however, off Isla Gorgona, Colombia (LACM: AHF 221-34), and off La Plata Island, Ecuador (LACM: AHF 213-34). Both of these islands are near the shelf edge. These morphs are not separated by a distinct mor- phological gap. Transitional specimens occur, although rarely, in the Galapagos Islands and also along the Ecuador coast. Comparison.— Spathochlamys vestalis differs from both S. benedicti and S. vaginula in having a more asymmetric hinge line, with the anterior outer ligament much longer than the posterior one. The AOL/POL ratio in specimens greater than 12 mm in shell height is generally above 2.05 in S. vestalis and is commonly in the range of 2.2 to 2.7. In S. benedicti the AOL/POL ratio is generally less than 2.0 and commonly in the range of 1.8 to 1.9. The posterior auricles of S. vestalis compared to S. benedicti are not only smaller relative to the anterior auricles, but they also are more oblique with a straighter, more inclined posterior mar- gin. This difference in posterior auricular obliquity increas- es with ontogeny. The secondary rib-flank costae, one on each side of each major rib in the central region of the disk, generally appear earlier in S. vestalis than in the other species although there is considerable overlap in time of appearance with that in S. benedicti. The medial costae on the right valve of S. vestalis generally begin later in ontoge- ny than in S. benedicti. In S. vestalis these costae seldom originate before a shell height of 3 mm is reached; in S. benedicti the origination point is generally at shell heights ranging from 1.5 to 2.5 mm. Hertlein (1935: 309) compared his new species, Pecten (Chlamys) lowei (= P. vestalis), to several fossil species. These fossil taxa, however, are not closely related, and most are now known to belong to the tribe Aequipec- tinini. This is also true for Chlamys corteziana Durham, 1950 (see also Moore, 1984:. B22), which Durham (1950, p. 64) thought resembled C. lowei. C. onzola Olsson, 1964, thought by Olsson (1964: 35) to be “probably closest to C. lowei (Hertlein)’, is also not closely related. I examined the holotype of Olsson’s species and found that it is more closely related to the extant eastern Pacific C. incantata Hertlein, 1972, of the Galapagos Islands. Hertlein (1972) thought that this Galapagos species “bears a general simi- larity to that of illustrations of P. (Chlamys) nympha Bavay”, a species that is now known to be a junior syn- onym of Spathochlamys benedicti (see above). Sculptural details of both C. onzola and C. incantata, however, indi- cate that these species both belong in Veprichlamys Iredale, 1929, an Indo-Pacific genus that can be placed in the tribe Chlamydini on the basis of its pre-radial microsculpture and lack of internal rib carinae. Living habits.— The total depth range of live Spatho- chlamys vestalis, based on specimen data in the USNM col- lection, is from 9 to 183 m, with most specimens having been dredged between 40 and 90 m. Grau (1959) and Bernard (1983) listed depths as shallow as one and two meters, but such shallow occurrences do not appear to be common for living specimens. There do not appear to be any significant differences in the depth preferences of the two morphs described above nor among different parts of the geographic range of the species. Bernard (1983) gave the sea-bottom temperature range of the species as 10 to 29°C. Data on substrate preference and living position are few. Grau (1959: 93) concluded that the species occurs on “rocky, sandy or muddy bottoms, associated with algae, kelp, bryozoa, coral, coralline and sponge.” The only spec- imens in the USNM collections with the bottom type indi- cated as mud were collected as dead valves. Some speci- mens show nearly total cover by thin encrusting sponge, and the persistent byssal notch suggests persistent byssal attachment. Geographic range.— From Santa Catalina Island, Cali- fornia (USNM 774180) southward to Isla La Plata, Ecuador (1° 16’ S. Lat.; USNM 774194), in Gulf of California as far north as Bahia San Felipe (USNM 774189), and in the Galapagos Islands (Grau, 1959: 92). Stratigraphic range.— Upper? Miocene to Recent. The Upper Miocene limit is based on the occur- WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” Za rence of this species in the Imperial Formation of southern California (Powell, 1986, 1988), discussed below. Discussion.— Reeve gave the locality for this species as “West Indies”, but the morphology of the syntypes (Figs. 10n-p) clearly indicates that they belong to the eastern Pacific species Pecten (Chlamys) lowei Hertlein, 1935, which thus becomes a junior synonym of P. vestalis Reeve, 1853. The features of Reeve’s syntypes that are characteris- tic of the eastern Pacific species are as follows: (1) The ratio of the length of anterior outer liga- ment to length of posterior outer ligament (AOL/POL) is 2.56 in the lectotype and 2.41 and 2.46 in the paralecto- types. These ratios are higher than any found in the western Atlantic species (AOL/POL = 1.76 to 2.03 where height is greater than 12 mm, n=6) but are within the normal range of variation of the eastern Pacific species (2.04-2.88 where height is greater than 12 mm, n=9). (2) The rib-flank costae of the right valves of the syntypes of Pecten vestalis begin at heights of 7.5 and 10 mm. This is earlier than the beginning of comparable costae in the western Atlantic species but is consistent with the ontogeny of these costae in the eastern Pacific species. (3) The numbers of major ribs on the right valves of the syntypes of Pecten vestalis are 21 (lectotype) and 19 (both paralectotypes), lower than the number of major ribs present on right valves of the western Atlantic species (22- 24). Provenance data in nineteenth century works such as Reeve’s Conchologia Iconica are notoriously inaccurate, and I have uncovered other locality inconsistencies in Reeve’s work. That shells from the Pacific side of North America were available to Reeve is indicated by the fact that he described two other pectinids from this region, Pecten arthriticus [now Nodipecten arthriticus; see Smith, 1991a:. 88) and P. leucophaeus [=Argopecten circularis (Sowerby); Waller, unpub. data]. Both of these species were listed by Reeve as from unknown localities. Vokes (1991) has commented on similar errors. The only known reports of Spathochlamys vestalis from the fossil record are those of Powell (1986, 1988 reported as Chlamys lowei) from the Imperial Formation of southern California. As detailed above, fossil taxa consid- ered closely related to “Chlamys lowei” by Hertlein (1935, 1972) and Olsson (1964) are neither ancestral nor con- generic. Moore (1984) listed no records of S. lowei in her compendium of Tertiary Pectinidae of California and Baja California, and my own search of the extensive Pacific Coast collections of the USNM did not turn up any speci- mens. The possibility that this species will eventually be found in strata as old as Pliocene is based on its inferred derivation from an Atlantic ancestor while seaways were still open. I examined the collection from the Imperial Formation on loan from UCMP to C.L. Powell II at USGS(MP). Two specimens of Spathochlamys vestalis were found, both from UCMP Locality A-1415, a part of the section that Powell places in the lower part of the upper member of the formation and interprets as possibly repre- senting a storm deposit formed at inner-shelf depths (Powell, 1993, pers. comm; also see Powell, 1985). One of these specimens is a small incomplete left valve (ht., 9.4 mm) on which rib-flank costae have not yet developed. Medial costae appear to begin late (at about 4.5 mm). These ontogenetic features are consistent with those in S. vestalis. The second specimen is a badly crushed and incomplete valve missing its dorsal region. Its original height was probably about 12 mm. Scaly rib-flank costae are present and appear to have formed at a reconstructed shell height of about 7 mm, also consistent with this feature in S. vestalis. The age of the Imperial Formation has been a mat- ter of some debate, but it apparently includes sediments of both Miocene and Pliocene age (Powell, 1985, 1986, 1988; Smith, 1991b). The deposits in the northern Salton Trough area of Riverside County, California, that yielded the speci- mens of S. vestalis appear to be no younger than late Miocene. This age constraint is based on a radiometric (K/Ar) date of about six milltion years obtained from a basalt flow in overlying beds (Powell, 1986, 1988; Smith; 1991b; based on data in Matti et al., 1985). Material examined.— Recent material: USNM: Pacific Ocean off southern California and Mexico, 7 lots, 22 speci- mens; Gulf of California, 17 lots, 33 specimens; Panama, 4 lots, 9 specimens; Colombia, | specimen; Ecuador, | speci- men; Galapagos Islands, 16 lots, about 40 specimens. LACM: Gulf of California, 11 lots, about 50 specimens; Mexico: 7 lots, 10 specimens; Panama: 2 lots, 3 specimens; Colombia: 2 lots, 2 specimens; Ecuador mainland: 3 lots, 4 specimens; Galapagos Islands: 22 lots, 72 specimens. Fossil material: UCMP Loc. A-1415(RAB 168A): “About the center of the north boundary of NE!/4 of NW!/4 of section 36, T2S, R3E, [Desert Hot Springs 7.5’ Quadrangle, Riverside County, California]. About Im (50 ft.) south of the divide, in the stream bed of the head of the north directed branch [which separates from the main canyon about 1370 m (4500 ft) from its mouth] of the sec- ond canyon east of the Whitewater River. Pebble-bearing siltstone. Collected by R. Bramkamp. (=Burrobend Mem- ber).” (Powell, 1986, Appendix 2.) Two valves, both incomplete. 238 AMER. MALAC. BULL. 10(2) (1993) Spathochlamys vaginula (Dall, 1898) (Figs. 11c-p) Pecten (Chlamys) ornatus Lamarck? var. vaginulus Dall, 1898: 715-716, “Oligocene”, Bowden, Jamaica. Pecten vaginulus Dall, Maury, 1917: 186, pl. 34, fig. 7. Tertiary, Samba Hills, Dominican Republic Chlamys (Chlamys) vaginulus (Dall), Woodring, 1925:65- 66, pl. 8, figs. 1, 2, “Miocene”, Bowden, Jamaica. Types.— Dall (1898: 715-716) mentioned having “seven small valves” of his new variety from the Bowden Beds of Jamaica but provided no illustration, no measurements, and no type designation. This type series was subsequently referred to by Schuchert et al. (1905: 490) as “cotypes” and was catalogued as USNM(P) 135786. The entry for this number in the catalogue, however, mentions only six valves. The discrepancy is apparently due to the recogni- tion, after Dall’s 1898 publication, that one of the seven valves does not belong to the same species. (In fact it is an Argopecten, not a Spathochlamys.) Woodring (1925, pl. 8, figs. 1, 2) provided the first illustrations of members of the type series, a right valve and a non-matching left valve, but continued to refer to these as cotypes. The right valve illus- trated by Woodring (1925, pl. 8, fig. 1) is herein selected as the lectotype and refigured (Figs. 11c-f). The type locality is USGS 2580 (see following section on materials exam- ined) Diagnosis.— Spathochlamys with length of anterior outer ligament commonly greater than twice length of posterior outer ligament, especially where shell height exceeds 20 mm; early ribs commonly rounded rather than trigonal in cross section, with scales variably developed or absent; rib- flank costae not appearing until late in ontogeny, generally at shell heights exceeding 12 mm; medial costae variably developed, commonly weak in early ontogeny of left valve; ontogenetic persistence of commarginal lirae in rib inter- spaces variable, commonly persisting throughout ontogeny, with height of lirae greater than that of medial costae in early ontogeny; posterior margins of posterior auricles slightly concave to nearly straight, forming oblique angle with dorsal margin. Morphological variation.— In addition to the six valves of the type series from the Bowden Shell Beds of Jamaica, there are about 30 additional valves and fragments from the type locality (USGS 2580) in the USNM collections. The range of variation of these specimens seems to encompass the morphology of specimens from the Gurabo Formation of the Dominican Republic. In general these Bowden and Gurabo specimens have robust simple ribs that tend to be rounded in cross section in early ontogeny, becoming trigo- nal later. Rib-flank costae are present only on the largest specimens and commonly do not begin to form in ontogeny until a shell height of about 15 mm is reached. The Bowden and Gurabo specimens share rib counts within the range of 21 to 25. Trigonal ribs are common among the specimens from the Dominican Republic but rare among the speci- mens from Bowden, Jamaica. The Bowden specimens in general have more prominent commarginal lirae in rib interspaces and medial costae that begin later in ontogeny than in the Dominican Republic. Specimens of Spathochlamys from the Tamiami Formation in a pit near Sarasota, Florida (see below), also appear to be within the range of variation of S. vaginula. The Tamiami specimens (Figs. 11k-o) differ from those of the Bowden Shell Beds of Jamaica and the Gurabo Formation of the Dominican Republic mainly in being of much greater size, with shell heights ranging from 18 to 34 mm (mean = 27 mm) compared to the maximum shell height of 17 mm for the other samples. Like the Bowden specimens, however, those from the Tamiami Formation have ontogenetically persistent commarginal lirae in rib interspaces, robust rounded major ribs, and late-appearing medial costae. A single right valve 22 mm in height [USNM(P) 474666, Fig. 11p] was collected by the author from near the base of the Jackson Bluff Formation of Florida (see follow- ing discussion of stratigraphy). It resembles the specimens from the Bowden, Gurabo, and Tamiami Formations in having poorly developed scales on the crests of ribs in the center of its disk and in lacking clearly delimited rib-flank costae. The Jackson Bluff specimen differs, however, in having fewer ribs (only 18 at the valve margin compared to 21 to 25 in the other samples). Furthermore, the ribs of the Jackson Bluff specimen are distinctly trigonal in cross-sec- tion, and a few of the central ribs bifurcate. The latter fea- ture has been observed to occur in extant species in response to injury or simply as extremes of variation in oth- erwise simple-ribbed populations. Comparison.— Spathochlamys vaginula differs from both of the extant species, S. benedicti and S. vestalis, primarily in having less scaly rib crests on the central part of the disk at least in early ontogeny and in having less trigonal ribs at an early growth stage. Many but not all specimens of S. vaginula have ontogenetically persistent commarginal lirae in rib interspaces, whereas in the extant species these are limited to early ontogeny. Weisbord (1964: 142) included both “Chlamys vaginula Dall” and “Pecten (Chlamys) portoricoensis" in his cSmparisons for Spathochlamys benedicti. As discussed above in the description of Spathochlamys, P. portoricoen- WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 239 sis differs from all Spathochlamys species in having an Aequipecten-like microsculpture in early ontogeny and is not considered to be congeneric. Paleoecology.— The molluscan fauna of the Bowden For- mation of Jamaica has been interpreted as representing a paleoecological mixture that includes deep-water elements (Woodring, 1928: 37-38). Specimens of Spathochlamys vaginula from the Dominican Republic appear to be restricted to facies representing deposition in deeper waters. Saunders et al. (1986: 16) noted that depth of deposition increased rapidly in the part of the Gurabo Formation that yielded NMB 15941 and 15835 (see below), with water depths eventually exceeding 200 m. A specimen from about this same level (TU Loc. 1210) is covered by a cemented Dimya sp., a genus that is most commonly found at present on the outer continental shelf and slope (H. E. Vokes, 1979: 37). One of the Rio Yaque samples, TU 1227A, occurred in a lens interpreted to be a turbidity-flow lens, again suggest- ing deep water. Scleractinian corals collected at this same locality suggested to Cairns and Wells (1987: 25) a depth of deposition greater than 200 m. The single valve of S. vagin- ula from the Jackson Bluff Formation of Florida (see below) is from the part of this formation determined by DuBar and Taylor (1962: 356) to represent maximum marine transgression and maximum water depth, which they inferred may have been up to 20 fm (37 m) deep. The S. vaginula-bearing bed in the pit near Sarasota (Unit 11 of Petuch, 1982a) was interpreted by Petuch (1982a: 19) as representing ‘“‘a quiet, deeper water lagoonal habitat with depths of around 10 m,’” but the remains of whales, sharks, and brachiopods suggest to this author that a deeper water environment may have been present. That was also the opinion of E. H. Vokes (1988: 2, footnote), who suggested that the Tamiami deposits in the Sarasota area represent “a more offshore facies” than the overlying Pinecrest Beds. Geographic range.— North Carolina, Florida, Jamaica, and the Dominican Republic. Stratigraphic range.— uppermost Miocene (planktic foraminiferal zone N17) through lower Pliocene (upper planktic zone N19/N20). The definition and age of the Tamiami Formation of south Florida have been matters of considerable confu- sion, the history of which was most recently summarized by Lyons (1991: 137). In a paper that was written while Lyons’s study was in press, Waldrop and Wilson (1990) have further clarified part of this history and have provided some biostratigraphic constraints on the age of the forma- tion in the Sarasota area. They restrict the term Tamiami Formation to beds unconformably underlying the “Pinecrest Beds” in the APAC (or MacAsphalt) Pit at Sarasota. (The latter formational name is preoccupied, and Waldrop and Wilson renamed these beds the Fruitville Formation.) They also report (p. 202) that at some locali- ties near Sarasota the Tamiami Formation interfingers with beds of the upper Bone Valley Formation and that the verte- brate fossils overlying the Tamiami Formation at these sites are of late Hemphillian age. Tedford et al. (1987: 183), on the basis of K/Ar and fission-track dates, gave 4.5 to 6 Ma as the age of the late Hemphillian interval (cited incorrectly as 4.5 to 5.0 Ma in Waldrop and Wilson, 1990: 202). This interval thus spans the Miocene-Pliocene boundary, which most workers now agree is at approximately 5 ma (5.2 ma in Haq et al., 1987: 1158). This would suggest that the specimens of Spathochlamys vaginula from the Tamiami Formation in the Sarasota area are of either latest Miocene or earliest Pliocene age. The overlying Pinecrest Beds (Fruitville Formation of Waldrop and Wilson, 1990) are apparently middle to late Pliocene in age (Akers, 1972; Waldrop and Wilson, 1990; Jones et al., 1991). The specimens of Spathochlamys vaginula from the Gurabo Formation of the Dominican Republic are prob- ably of approximately the same age as those from the Tamiami Formation at Sarasota, Florida. In the extensive collections from along the Rio Gurabo in the Dominican Republic made by Peter Jung and colleagues of the Natural History Museum of Basel, Switzerland, this species occurs in only two samples, NMB 15941 and 15835. These are closely spaced stratigraphically, NMB 15835 occurring only about 2 m above NMB 15941. As stated by Saunders et al. (1986: 17), NMB 15941 lies about 5 m below the Miocene-Pliocene boundary if that boundary is demarcated by the first occurrence of the planktic foraminifer Globorotalia margaritae Bolli and Bermtdez, 1965. They also note that NMB 15941 represents the approximate stratigraphic position of the nannofossil NN11-NN12 zonal boundary, generally taken to be just below the Miocene- Pliocene boundary. This NNI1-NN12 zonal boundary occurs within the upper part of planktic foraminiferal zone N17 (Bybell and Poore, 1991, fig. 2). The only other Rio Gurabo specimens known to me are from two localities, USGS 8548 and TU 1210, both of which are apparently close to but somewhat higher than the site of the NMB samples (Saunders et al., 1986, text-figs. 4, 5). Other speci- mens from the Dominican Republic are from sections along the Rio Yaque del Norte, collected by both the U. S. Geological Survey (USGS 8726) and by H. E and E. H. Vokes of Tulane University (TU 1227A). The Bowden Formation of Jamaica is now gener- ally considered to be within the Globorotalia margaritae zone and hence of early Pliocene age (Bolli and Bermudez, 1965, pp 125, 146; Bolli and Premoli Silva (1973: 479, fig. 240 AMER. MALAC. BULL. 10(2) (1993) 2; Jung, 1989: 11). The presence of G. margaritae in the Bowden suggests that this formation may be younger than that part of the Gurabo Formation that yielded Spatho- chlamys, which is below the G. margaritae zone. In the Dominican Republic the G. margaritae zone begins just above the nannofossil NN11/NN12 boundary (upper zone N17) and extends to the NN15/NN16 boundary (Saunders et al., 1986, table 3), this upper limit being near the top of planktic foraminiferal zone N19 (Bybell and Poore, 1991, fig. 2). Both Woodring (1928: 37-38), on the basis of mol- luscan faunas, and Bold (1971: 327), on the basis of ostra- code faunas, considered the Bowden shell beds to be strati- graphically correlative with the Gurabo Formation. The single, possibly aberrant, valve of Spatho- chlamys vaginula collected from near the base of the Jackson Bluff Formation in western Florida is possibly about the same age as the Gurabo specimens. The lower Jackson Bluff Formation (the Ecphora faunizone) was dated as early Pliocene (middle of Zone N18 to within Zone N19) at its type area at Alum Bluff on the basis of planktic foraminifera, among which is Globorotalia mar- garitae (Akers, 1972: 15 and 20). At Jackson Bluff, from a locality near where the Spathochlamys reported here was obtained, Akers (1972: 20) obtained planktic foraminifera that indicated an age ranging from the upper part of Zone N18 to the upper part of Zone N19, but his samples did not include G. margaritae. Because the precise location of the foraminiferal samples relative to the base of the Jackson Bluff formation was not specified, it is possible that the Spathochlamys collected by me from less than a meter above the base of the formation is near the maximum allowable age (N18) determined by Akers (1972: 15). Zone N18 is just above the Miocene-Pliocene boundary, Discussion.— Spathochlamys vaginula is the closest of the three species of Spathochlamys to the probable outgroup genus, Mimachlamys, in having the greatest extent of left- valve rib development without scales, the least trigonal rib cross-sections in early ontogeny, the ontogenetically latest appearing medial costae in the rib interspaces, and the most ontogenetically persistent commarginal lirae. The derived features of the eastern Pacific species, S. vestalis, relative to the states of characters present in S. vaginula include the following: (1) increase in obliquity of the posterior auricle, (2) reduction in the relative size of the posterior ear, and (3) earlier ontogenetic origin of costae on the flanks of the pri- mary ribs, these rib-flank costae becoming more scaly. In S. benedicti, two of these characters show opposite trends rel- ative to their condition in S. vaginula: (1) The posterior auricle becomes less oblique, its posterior margin forming a 90° or acute angle with the hinge; (2) the posterior auricle increases in relative size. The Gal-pagos specimens of S. vestalis appear to be more derived than their mainland counterparts in having a more inflated form, with the rib crests secondarily flattened except for a very low, narrow crest from which very narrow scales develop. These morphological differences and stratigraphic occurrences imply that Spathochlamys vaginula gave rise to both of the extant species but at different times. Speci- ation seems to have occurred before final seaway closure in the eastern Pacific (based on the occurrence of S. vestalis in the Imperial Formation) but after seaway closure in the western Atlantic (based on the earliest known S. benedicti). Material examined.— Jamaica USGS 2580. Bowden, Jamaica. Fine sandy stra- tum in wagon road cut at foot of a hill that is 300 ft high. Collected by J. B. Henderson and C. T. Simpson, 1894. Bowden Shell Beds. Pliocene. 6 valves including the lecto- type, catalogued as USNM(P) 135786, and about 30 uncat- alogued valves and fragments. Dominican Republic NMB 15835. Rio Gurabo near highway bridge, Dominican Republic. Gurabo Formation. From macrofossil sample collected by P. Jung, 1978. See Saunders et al. (1986, text-fig. 4) for plot of location and stratigraphic position. 4 valves. NMB 15941. Rio Gurabo about 0.6km down- stream from highway bridge. Hard massive silt with diverse fauna scattered and in pockets. Gurabo Formation. From microfossil sample collected by J. B. Saunders, 1978. See Saunders et al. (1986, text-fig. 4) for plot of location and stratigraphic position. 3 valves. USNM(P) 474811. USGS 8548. Right bank of Rio Gurabo, 500 ft [152 m] below lower ford at Gurabo Adentro Bluff A, Distrito de Monte Cristi, Dominican Republic. Collected by T. W. Vaughan and C. E. Cooke, 1919. 1 valve in several pieces. USNM(P) 47812. USGS 8726. La Canela, south side of Rio Yaque del Norte, 15 km west of Santiago, Dominican Republic. Formation not specified but probably the Gurabo Formation. Collected by T. W. Vaughan, C. W. Cooke, and others, 1919. 1 valve. USNM(P) 474659 and 474667. TU-1227A. Turbidity-flow lens (about 30 inches [76 cm] long and 6 inches [15 cm] thick) about 2 feet [61 cm] above base of outcrop at point approximately 75 feet [23 m] downstream from highway bridge, Rio Yaque del Norte, Dominican Republic. Gurabo Formation. Collected by E. H. and H -E. Vokes. 10 valves (Figs. 11g-i). USNM(P) 474660. TU-1210. Rio Gurabo, east WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 241 bank, first bluff downstream from bridge on Los Quemados-Sabaneta road (equivalent to USGS 8544, Maury’s Zone B), Dominican Republic. Gurabo Formation. [See Saunders et al. (1986, text-fig. 5) for plot of locality.] 1 valve (Fig. 11). Florida USNM(P) 474661-474665 and 474813, Mac- asphalt Co. shell borrow pit complex (currently the APAC Pit) on the eastern outskirts of Sarasota, 0.3 km west of Interstate Highway I-75, sec. 12, T36S, R18E, Bee Ridge Quadrangle, Sarasota Country, Florida. Collection from a drainage ditch and pit floor of the pit that was first opened in December, 1988. Tamiami Formation (sensu Waldrop and Wilson, 1990). Collected in July, 1989, and February, 1991 by John Waldrop, Druid Wilson, and Earlene Mitchell. 5 figured specimens (Figs. 11k-o) and 41 not fig- ured. USNM(P) 474666. Marl pit atop Jackson Bluff on left back of Ochlockonee River a short distance down- stream from Lake Talquin Dam, Leon County, Florida. Upper bed of Ecphora faunizone 3.5 to 5 feet [0.9-1.5 m] above base of Jackson Bluff Formation. | valve (Fig. 1 1p). SUMMARY AND CONCLUSIONS The six species of “Chlamys” presently living in the tropical and subtropical western Atlantic (Table 1) are not a monophyletic assemblage, nor are they a single genus, nor are any of them members of the genus Chlamys in a strict sense. The only features held in common by these species are their byssate living habit and its attendant form, where the byssal notch is deep, the active ctenolium persis- tent, and the auricles asymmetric, with the anterior one the larger. The fact that each species is a member of the Caribbean fauna is a coincidence stemming from a variety of circumstances involving very different phylogenetic his- tories and different times of arrival of ancestral lineages in the western Atlantic. The different phylogenetic and biogeographic his- tories of these species are reflected in the fact that their closest extant relatives may be in the western Indo-Pacific, the eastern Atlantic, the western Atlantic, or the eastern Pacific. The morphological study of closely related living and fossil species, coupled with outgroup comparisons based on a continuing study of the Pectinidae on a world- wide basis, has made it possible to construct phylogenies. These phylogenies, coupled with data on present and past distributions, lead to the following conclusions: 1) All six of the extant tropical western Atlantic “Chlamys” species have ancestors to the east, in the eastern Atlantic, the Mediterranean, or the western Indo-Pacific. There is no morphological or paleontological basis for sup- posing that any originated in the eastern Pacific and then dispersed eastward to the western Atlantic. 2) All of the tropical western Atlantic “Chlamys” species are constrained by middle latitudes, meaning that it is highly unlikely that they could have dispersed from the western Atlantic to the eastern Pacific by polar routes. 3) The extant tropical American “Chlamys” species, as well as the extant eastern Pacific “Hinnites”, are distributed among four genera, Laevichlamys, Crassadoma, Caribachlamys, and Spathochlamys, each of which has a different history of entry into or origin within the Caribbean region. The timing of these origins or arrivals relative to the final closure of the seaways connecting the Caribbean and the eastern Pacific is the primary factor that determines whether geminate species are present in the eastern Pacific. These genus-level histories, which are illustrated diagram- matically in Figures 12-14, are as follows: Laevichlamys originated in the western Indo- Pacific in the Miocene and dispersed either through the Mediterranean or, more likely, around southern Africa into the eastern Atlantic and thence to the tropical western Atlantic (Fig. 12). The fossil record suggests that its arrival in the tropical western Atlantic and the origin of the Caribbean species Laevichlamys multisquamata occurred in post-closure time, i.e. in the late Pliocene or Pleistocene. This is corroborated by the absence of both extant and fos- sil Laevichlamys in the eastern Pacific. Crassadoma, the type species of which is a cemented form living in the eastern Pacific, has a long and continuous history in the eastern Atlantic, particularly in the Mediterranean (Fig. 13). This history stems from at least the early Miocene, when the Mediterranean was still open at its eastern end to the Indian Ocean. Dispersal to the western Atlantic by members of Crassadoma apparently occurred twice (Fig. 13). The first was in the early or mid- dle Miocene, based on evidence that extant Crassadoma gigantea of the eastern Pacific is a sister species of C. mul- tistriata of the eastern Atlantic. The second entry into the western Atlantic occurred much later, probably in post-clo- sure time, giving rise within the Caribbean to the new genus Caribachlamys. The key morphological feature of Caribachlamys is its lecithotrophic-type larval shell, possi- bly an adaptation to sequester larvae in reef habitats that were under stress during the rapid sea-level changes of the Pleistocene. Speciation within Caribachlamys has occurred rapidly within the Pleistocene to produce four extant species (C. sentis, C. ornata, C. mildredae, and C. imbrica- ta) and one extinct Plio-Pleistocene species (C. paucirama, new species). The new genus Spathochlamys is represented in 242 AMER. MALAC. BULL. 10(2) (1993) a re = 3 7 L5 oS = EE aes € AUS) S g =P Eo = og: oe 2 S >= Se = oS ® oe = & Ss wo i = = % ~J PLEIST. 7 i] M. MIO. seeepeeseseeeeese L. MIO. eer Peer eee W. ATL. E.ATL. © I.-PAC. Fig. 12. The evolution of Laevichlamys among world oceans. Solid verti- cal lines = physical barriers to dispersal; dashed vertical lines = seaways (Tethyan, south African, or central American) or distance barriers; dashed lineage lines = missing data from the fossil record. Arrows indicate disper- sal directions based on the polarities of morphological changes and on stratigraphic records. the present-day Caribbean by S. benedicti, which is the most populous and widespread of all of the extant tropical western Atlantic “Chlamys” species. It is also the most eurytopic, living on a variety of bottoms across a span of depths centered on the middle shelf. The genus originated in the tropical western Atlantic in pre-closure time from ancestors in the new tribe Mimachlamydini, which has its most extensive history in the eastern Atlantic and western Indo-Pacific (Fig. 14). The known fossil record of Spatho- chlamys in the western Atlantic is sparse but continuous from the late Miocene, and the succession of morphologies suggests that its entry into the eastern Pacific was in the late Miocene, possibly during a general transgression that per- mitted deeper water taxa to disperse across seaways that were at other times too shallow. The geminate sister species of S. benedicti in the eastern Pacific, S. vestalis [=S. lowei], at present also has a broad distribution in similar habitats and has dispersed as far westward as the Galapagos Islands with minor morphological divergence from mainland popu- lations. It is logical to assume that eurytopy bestows resis- tance to extinction and that stenotopy, coupled with small population size, invites extinction (Stanley, 1986b). The eurytopic species Spathochlamys benedicti seems to epito- Caribachlamys Crassadoma Crassadoma i | | | | s gl w & 8 | & & ® Ge ON — + 5 SSELvs o B g S esse 8S 3 > EEsss act = is] Q PLEIST. W. ATL. E. ATL. Fig. 13. The evolution of Crassadoma and Caribachlamys among world oceans (symbols are as in figure 5.) Crassadoma multistriata is presently distributed around southern Africa into the southwestern Indian Ocean. Spathochlamys ake Mimachlamys spp. a > oO a & SG £ Ce @ OE a 2 2 = = £ & 9 > S benedicti p. Dimarzipecten sp Seeeeeeeeeeeseee eeesiae ee ee sieseeese Fig. 14. The evolution of Spathochlamys from Dimarzipecten and Mimachlamys among world oceans (symbols are as in figure 5). WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 243 mize this kind of resistance. It is of interest to note, howev- er, that the lineage of S. benedicti has also been resistant to speciation within the western Atlantic in post-closure time. Extant S. benedicti is probably a lineage descendant of S. vaginula, a species that lived in the tropical western Atlantic during the late Miocene and early Pliocene. The only demonstrable speciation of this lineage is that associ- ated with its dispersal through seaway bottlenecks into the eastern Pacific. Once these seaways were closed, the pre- sumably small founder populations that were isolated in the eastern Pacific were subject either to greater selection pres- sure or to genetic drift, causing them to diverge from the western Atlantic stem group. The same process is possibly in progress today, forcing divergence of the Galapagos pop- ulations of S. vestalis that are separated by a deep-water barrier from their mainland counterparts. Most of the clade originations outlined in the above systematic revision suggest that long-distance disper- sal facilitates allopatric speciation. The exception is the endemic Caribbean genus Caribachlamys, the most spe- ciose of the three western Atlantic genera. The five species (one extinct) in this genus have apparently all originated since the beginning of the Pliocene within a time span of less than five million years, and one of the species has ap- parently originated during the Pleistocene. Although the founding populations of this clade may have arrived in the western Atlantic by long distance dispersal from the east, the synapomorphy that unites the species of the genus, and which presumably contributed to its rapid speciation, is a derived prodissoconch morphology (large PI stage and short PII stage) that suggests lecithotrophy and limited dis- persal ability. The key to understanding why a decrease in dis- persal ability is associated with rapid speciation in Cari- bachlamys could lie in the common habitat preferences of these species. All are associated with reef-delimited habi- tats, either reef fronts, reef platforms, or back-reef areas. At present such habitats are not contiguous and are spatial- ly limited; presumably they were even more fragmented during the extensive sea-level fluctuations of the Pleistocene. Reef-adapted byssate pectinid bivalves may have evolved diminished planktic dispersal ability under selective pressure to keep their young in place and to maxi- mize the reproductive potential of local populations. By the same token, if only a very short dispersal phase is present, the possibilities for allopatric speciation are increased by chance transport of these larvae over even relatively short distances. Because all of these species are ctenidial feeders on suspended food particles and do not appear to be gregar- ious, there is little possibility for direct competition for food. Once reproductive isolation has occurred, daughter species can be reintroduced by chance dispersal into the habitats of their founders. Indeed, the present overlapping geographic ranges of the four extant and to some degree successive species is testimony that this has occurred. The only strong indication of geographic infraspe- cific variation within the Caribbean species analyzed here occurs in Caribachlamys imbricata from Bermuda. Both outgroup comparison and the fossil record suggest, howev- er, that the Bermuda populations are primitive rather than derived compared to populations from throughout the Antilles and in southeastern Florida. The Bermuda popula- tions are therefore relictual and evidence from the fossil record suggests that gene flow to Bermuda has become restricted beginning in the late Pleistocene. There has long been a tendency for researchers to view the final closure of the Isthmus of Panama as an instantaneous event that triggered divergence, speciation, and extinction of once continuous populations of species remaining on the two sides. It is becoming increasingly apparent, however, that the shoaling of seaways was grad- ual and that divergences occurred both well before and well after closure. ACKNOWLEDGMENTS Specimens referred to in this study have been provided by many dedicated collectors and colleagues: Dr. R. Bailey and his students of Northeastern University, Boston; Mr. W. Blow of the National Museum of Natural History, Washington, D.C.; Mrs. E. Bradley, Bradenton, Florida; Mr. J. Coltro and Mr. M. Coltro, Brazil; the late Mrs. S. Hoerle, Florida; Dr. P. Jung, the Natural History Museum of Basel, Switzerland; Dr. Brian Kensley, National Museum of Natural History, Smithsonian Institution, Washington, D. C., Mr. R. Portell, Florida Natural History Museum, Gainesville; Dr. G. Vermeij, University of California, Davis; Drs. E. and H. Vokes, Tulane University, New Orleans; the late Dr. G. Voss, University of Miami; and Mr. J. Waldrop and Mr. D. Wilson, Lake Wales, Florida. For technical assistance and advice I am grateful to the follow- ing colleagues at the National Museum of Natural History, Washington: J. Beck for photographic printing, W. Blow for research assistance and plate construction, M. Parrish for drawings, P. Viola for SEM operation, R. Lindsey, D. Steere, Jr., and C. Hahn for library assistance, and B. Bedette for technical assistance. Access to collections and helpful information were provided by the following: D. Wilson and J. Waldrop, Lake Wales, Florida; E. Lazo-Wasem, Peabody Museum, Yale University; R. Portell, Florida Natural History Museum; D. Robinson, Academy of Natural Sciences of Philadelphia; P. Hoover and W. Allmon, Paleontological Research Institution; C. Powell II and L. Marincovich, Jr., USGS, Menlo Park, California; C. Hickman, Museum of Paleontology, Berkeley; and J. McLean and the late C. Coney, Los Angeles County Natural History Museum. Kathie Way provided information on types at The Natural History Museum, London. Dr. R. Bieler of the Field Museum of Natural History served as organizer of the Caribbean Biogeography Symposium held at the Annual Meeting of the American Malacological Union at Sarasota, Florida, August, 1992, and provided encouragement for this study. Lastly, I am grateful to E. Coan and J. T. Smith, who reviewed 244 AMER. MALAC. BULL. 10(2) (1993) the manuscript and made suggestions for its improvement. Partial funding for this study was made available by the Research Opportunities program of the Smithsonian Institution and the Department of Paleobiology, National Museum of Natural History. LITERATURE CITED Abbott, R. T. 1954. American Seashells. D. Van Nostrand Co., Inc., New York. xiv + 541 pp. Abbott, R. T. 1974. American Seashells. Second Edition. Van Nostrand Reinhold Co., New York. 663 pp. Adam, W. 1960. A propos de Chlamys (F. Acc. Hinnites) abscondita (P. Fischer, 1898) de la Cote de l’Afrique. Bulletin de l|’Insitut royal des Sciences naturelles de Belgique 36(20):1-10. Adams, C. G., A. W. Genry, and P. J. Whybrow. 1983. Dating the termi- nal Tethyan event. Utrecht Micropaleontological Bulletins 30:273-298. Adams, C. G., D. E. Lee, and B. R. Rosen. 1990. Conflicting isotopic and biotic evidence for tropical sea-surface temperatures during the Tertiary. Palaeogeography, Palaeoclimatology, Palaeoecology 77:289-313. Adegoke, O. S. 1969. Stratigraphy and paleontology of the marine Neogene formations of the Coalinga Region, California. University of California Publications in Geological Sciences 80:1-241. Akers, W. H. 1972. Planktonic foraminifera and biostratigraphy of some Neogene formations, northern Florida and Atlantic Coastal Plain. Tulane Studies in Geology and Paleontology 9(1-4):1-139. Akers, W. H. 1974. Age of Pinecrest Beds, south Florida. Tulane Studies in Geology and Paleontology 11(2):119-120. Allmon, W. D., G. Rosenberg, R. W. Portell, and K. S. Schindler. 1993. Diversity of Atlantic Coastal Plain mollusks since the Pliocene. Science 260:1626-1629. Arnold, R. 1906. The Tertiary and Quaternary pectens of California. United States Geological Survey Professional Paper 47:1-264. Bavay, A. 1902 [1903]. Note sur quelques espéces du genre Pecten, nou- velles ou mal connues. Journal de Conchyliologie 50(4):399-406. Bavay, A. 1904. Descriptions de quelques nouvelles espéces du genre Pecten et rectifications. Journal de Conchyliologie 52(3):197-206. Bavay, A. 1906. Sur quelques espéces ou variétés nouvelles du genre Pecten. Journal de Conchyliologie 53(3):243-247. Bavay, A. 1913. Note au sujet des Pecten de la plage de Bahia récoltés par M. Serre,Concul de France. Bulletin du Muséum national d’ Histoire naturelle, Paris 19:25-16. Bayer, F. M. 1941. Notes on Florida Mollusca, with descriptions of two new varieties. Nautilus 55(2):43-46. Bayer, F. M. 1942. The type locality and dates of Pecten imbricatus mil- dredae Bayer. Nautilus 55(3):106. Bayer, F. M. 1943. Observations on marine Mollusca, with descriptions of new species. Nautilus 56(4):109-115. Bermidez, P. 1980. Geologia y bioestratigrafia de los Cerros de Caigiiire, Estado Sucre, Venezuela. Sociedad de Ciencias Naturales La Salle, Memoria 108:295-312. Berm'dez, P. J., and G. A. Sieglie. 1970. Age, paleoecology, correlation and foraminifers of the uppermost Tertiary formation of northern Puerto Rico. Caribbean Journal of Science 10(1-2):17-24. Bernard, F. R. 1983. Catalogue of the living Bivalvia of the eastern Pacific Ocean: Bering Strait to Cape Horn. Canadian Special Publication of Fisheries and Aquatic Sciences 61:vii + 102 pp. Bernard, F. R. 1986. Crassadoma gen. nov. for “Hinnites” giganteus (Gray, 1825) from the northeastern Pacific Ocean (Bivalvia: Pectinidae). Venus, The Japanese Journal of Malacology 45(1):70- 74. Blondel, T. and G. Demarcq. 1990. Les Pectinidae du Burdigalien termi- nal-Langhien de Tunisie: données paléobiologiques et paléogéo- graphiques. Revue de Paléobiologie 9(2):243-255. Bold, W. A. van den. 1971. Ostracoda of the Coastal Group of formations of Jamaica. Transactions Gulf Coast Association of Geological Societies 21:325-348. Bolli, H. M. and P. J. Bermudez. 1965. Zonation based on planktonic foraminifera of Middle Miocene to Pliocene warm-water sediments. Boletin Informativo, Asociacién Venezolana de Geologia, Mineria y Petréleo 8(5):119-149. Bolli, H. M. and I. Premoli Silva. 1973. Oligocene to recent planktonic foraminifera and stratigraphy of the Leg 15 sites in the Caribbean Sea. Initial Reports of the Deep Sea Drilling Project 15:475-497. Boone, Connie. 1978. New Range:Chlamys multisquamata. Shell Collector Magazine [Shell Collector Publications, Ft. Lauderdale, Florida] 1:52. Brocchi, G. B. 1814. Conchiologia fossile subapennina, con osservazioni geologiche sugli Apennini e sul suolo adiacente, vol. 2. Stamperia Reale, Milan. pp. 241-712. Brown, T. 1827. Illustrations of the conchology of Great Britain and Ireland, vol. 1. W. H. Lizars and D. Lizars, Edinburgh. 52 plates and interleaved pages. Brown, T. 1844. Illustrations of the conchology of Great Britain and Ireland, Second Edition. Smith, Elder, and Co., London. xi + 142 pp. Bucquoy, E., Ph. Dautzenberg, and G. Dollfus. 1887. Les mollusques marins du Roussillon, 2(1), Pélécypoda. J.-B. Bailliére and Fils, Paris. 884 pp. Budd, A. F., T .A. Stemann, and K. G. Johnson. 1992. Late Cenozoic turnover in the Caribbean reef coral fauna [abstract]. In: Fifth North American Paleontological Convention, Abstracts and Programs. Field Museum of natural History, June 28-July 1, 1992. The Paleontological Society Special Publication No. 6, S. Lidgard and P. R. Crane, eds. p. 43. Bybell, L. M. and R. Z. Poore. 1991. Calcareous nannofossils and planktic foraminifers from Enewetak Atoll, western Pacific Ocean. United Sates Geological Survey Professional Paper 1513-C:1-21. Cairns, S. D., and J. W. Wells. 1987. Neogene paleontology in the north- ern Dominican Republic. 5. The suborders Caryophylliina and Dendrophyliina (Anthozoa: Scleractinia). Bulletins of American Paleontology 93(328):23-66. Cassell, D. T. and B. K. Sen Gupta. 1989. Pliocene foraminifera and envi- ronments, Limon Basin of Costa Rica. Journal of Paleontology 63(2): 146-157. Cheetham, A. H. and J. B. C. Jackson. 1992. Speciation and diversity of Caribbean Neogene to Holocene cheilostome bryozoans [abstract]. In: Fifth North American Paleontological Convention, Abstracts and Programs. Field Museum of natural History, June 28-July 1, 1992. The Paleontological Society Special Publication No. 6, S. Lidgard and P. R. Crane, eds. p. 55. Chemnitz, J.H. 1784. Neues systematisches Conchylien-Cabinet, 7. Raspischen Buchhandlung, Nuremburg. 356 pp. [See Richardson et al. (1979) for collation and xerographic reproduction of plates.] Chenu, J. C. 1845. Illustrations Conchyliologiques ou Description et Figures de toutes les Coquilles connues Vivants et Fossiles, Classées Suivant le Systeme de Lamark. Part 41, Pecten. Franck, Paris. [Publication date and part number based on the collation of Sherborn and Smith (1911).] Coates, A .G., J. B. C. Jackson, L. S. Collins, T. M. Cronin, H. J. Dowsett, L.M. Bybell, P. Jung, and J.A. Obando. 1992. Closure of the WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 245 Isthmus of Panama: the near-shore marine record of Costa Rica and western Panama. Geological Society of America Bulletin 104:814- 828. Cocconi, G. 1873. Enumerazione sistematica dei Molluschi miocenici e pliocenici delle provincie di Parma e di Piacenza. Memorie. Reale Accademia della Scienze. Istituto di Bologna, serie III, 3:1-372. Collins, L. S. 1992. Timing of environmental change in Caribbean shallow waters relative to the closure of the tropical American seaway: evi- dence from benthic foraminfera [abstract]. Jn: Fifth North American Paleontological Convention, Abstracts and Programs. Field Museum of natural History, June 28-July 1, 1992. The Paleontological Society Special Publication No. 6, S. Lidgard and P. R. Crane, eds. p. 67. Cooke, C. W. 1919. Tertiary Mollusks from the Leeward Islands and Cuba. Carnegie Institution of Washington Publication 291:105-156. Cosel, R. von and S. Gofas. 1984. A new species of Hinnites (Bivalvia: Pectinidae) from Angola. Bollettino Malacologico 20: 297-306. Cox, L. R. 1927. Neogene and Quaternary Mollusca from the Zanzibar Protectorate. In: Report on the Palaeontology of the Zanzibar Protectorate. Government of Zanzibar, distributed through Crown Agents for the Colonies, London. 180 pp. Cronin, T. M. 1990. Evolution of Neogene and Quaternary marine Ostracoda, United States Atlantic Coastal Plain: Evolution and spe- ciation in Ostracoda, IV. United States Geological Survey Professional Paper 1367C:1-43. Cronin, T. W., L. M. Bybell, R. Z. Poore, B. W. Blackwelder, J. C. Liddicoat, and J. E. Hazel. 1984. Age and correlation of emerged Pliocene and Pleistocene deposits, U.S. Atlantic Coastal Plain. Palaeogeography, Palaeoclimatology, Palaeoecology 47:21-51. Cuenca Anaya, A. M. 1980. Actualizacién de la sistematica de Almera y Bofil para los moluscos Pliocénicos del bajo Liobregat y Ilano de Barcelona. 2.: Orden Pteroida. Boletin de la Real Sociedad Espafiola de Historia Natural (Geologia) 78: 49-74. Curran, H. A. and B. White. 1989. The Cockburn Town fossil coral reef of San Salvador Island, Bahamas. In: Pleistocene and Holocene Carbonate Environments on San Salvador Island, Bahamas. 28th International Geological Congress Field Trip Guidebook T175. H. A. Curran, ed. pp. 27-34. American Geophysical Union, Washington, D. C. Da Costa, E. M. 1778. Historia naturalis testaceorum Britanniae, or, The British Conchology. Printed for the author, and sold by Millan, B. White, Elmsley, and Robson, London. xii + 25 pp. Dall, W. H. 1886. Reports on the results of dredging, under the supervi- sion of Alexander Agassiz, in the Gulf of Mexico (1877-78) and in the Caribbean Sea (1879-80), by the U.S. Coast Survey Steamer “Blake,” Lieut.-Commander C.D. Sigsbee, U.S.N., and Commander J.R. Bartlett, U.S.N., commanding. Bulletin of the Museum of Comparative Zoology at Harvard College 12:171-318. Dall, W. H. 1898. Contributions to the Tertiary fauna of Florida with espe- cial reference to the Silex beds of Tampa and the Pliocene beds of the Caloosahatchie River, including in many cases a complete revi- sion of the generic groups treated of and their American Tertiary species. Part IV. I. Prionodesmacea: Nucula to Julia. II. Teleodesmacea:Teredo to Ervilia. Transactions of the Wagner Free Institute of Science of Philadelphia 3(4):571-947. Dance, S. P. 1966. Shell Collecting, an Illustrated History. Univ. of Calif. Press, Berkeley. 344 pp. Dance, S. P. 1967. Report on the Linnaean shell collection. Proceedings Linnaean Society London 178:1-24. Defrance, M. J. L. 1821. Hinnite. In: Dictionnaire des Sciences Nasturelles par plusieurs professeurs du Jardin du Roi et des prin- cipales Ecoles de Paris, vol. 21. Strasbourg, Paris. pp. 169-170. De Stefani, C. and D. Pantanelli. 1878. Molluschi pliocenici dei dintorni di Siena. Bollettino della Societa Malacologica Italiana 4:1-215. Diaz, J. M., L. A. Escobar, and L. E. Velasquez. 1991. Reef associated molluscan fauna of the Santa Marta area, Caribbean coast of Colombia. Anales del Instituto de Investigaciones Marinas de Punta de Betin [Colombia] 19-20:173-196. Dodge, H. 1952. A historical review of the mollusks of Linnaeus. Part 1.The classes Loricata and Pelecypoda. Bulletin of the American Museum of Natural History, New York 100(1):1-263. Dollfus, G. F. 1898. Rectifications de nomenclature. Revue critique de Paléozoologie 2(4):180. Du Bar, J. R. and D. S. Taylor. 1962. Paleoecology of the Choctawhatchee deposits, Jackson Bluff, Florida. Transactions Gulf Coast Association of Geological Societies 12:349-376. Dunker, W. 1864. Fiinf neue Mollusken. Malakozoologische Blatter (con- tinuation of Zeitschrift fiir Malakozoologie) 11:99-102. Dunker, W. 1858-1870. Novitates Conchologicae. Mollusca Marina. Theodor Fischer, Cassel, Germany. 144 pp. [Publication dates of particular sections given in Johnson (1969).] Duque-Caro, H. 1990. Neogene stratigraphy, paleoceanography and paleo- biogeography in northwest South America and the evolution of the Panama Seaway. Palaeogeography, Palaeoclimatology, Palaeoecology 77:203-234. Durham, J. W. 1950. 1940 “E.W. Scripps” Cruise to the Gulf of California. Part II, Megascopic paleontology and marine stratigra- phy. Geological Society of America Memoir 43:1-216. Eames, F. E. and L. R. Cox. 1956. Some Tertiary Pectinacea from East Africa, Persia, and the Mediterranean region. Proceedings of the Malacological Society of London 32(1-2):1-64. Ferreira, C. S. 1960. Contribuigdo 4 paleontologia do estado do Para: Revisao da familia Pectinidae da Formagao Pirabas (Mioceno infe- rior), com a descrigao de novas espécies. VI - Mollusca-Pelecypoda. Arquivos do Museu Nacional 50:135-166. Fischer-Piette, E. and A.-M. Testud. 1967. Campagne de la Calypso au large des cotes Atlantiques de l’Amerique du Sud (1961-1962). I. Résultats Scientifiques des Campagnes de la “Calypso”, Fascicule 8(9). Mollusques lamellibranches: Pectinidae. Masson et Cie, Paris. pp. 183-188. Gale, H. R. 1928. West coast species of Hinnites. Transactions San Diego Society of Natural History 5(9):91-94. Garrafielo Fernandes, J. M. and V. Araujo Tavora. 1989. Estudo sistemati- co da subclasse Pteriomorphia Beurlen, 1944 (Mollusca - Bivalvia) na Formacéo Pirabas (Mioceno Inferior) do acervo do Museu de Geosciéncias -UFPa. Boletim do Museu Paraense Emilio Goeldi (Univ. Federal do Para, Brazil] 1(2):91-118. Gibson-Smith, J. 1976. A new taxon in the genus Charonia (Mollusca: Gastropoda) from the Pliocene of Venezuela, and its implications. Boletin Asociacion Venezolana de Geologia, Mineria y Petrdleo 19(1):1-15. Glibert, M. 1945. Faune malacologique du Miocéne de la Belgique. I. Pélécypodes. Mémoires du Musée royal d’Histoire naturelle de Belgique 103:1-266. Glibert, M. and L. Van de Poel. 1965. Les Bivalvia fossiles du Cénozoique étranger des Collections de I’ Institut royal des Sciences naturelles de Belgique. II. Pteroconchida, Colloconchida et Isofilibranchida. Mémoires Institut royal des Sciences naturelles de Belgique, Second Series, 78:1-105. Gmelin, J. F. 1791. Caroli Linnaei systema naturae per regna tria natu- rae. Ed. 13, aucta, reformata, Vermes Testacea. Lipsiae 1(6), pp. 3021-3910 [for dating see Hopkinson, 1907, and Kabat and Petit, 1988}. Grant IV, U. S. and H. R. Gale. 1931. Catalogue of the marine Pliocene 246 AMER. MALAC. BULL. 10(2) (1993) and Pleistocene Mollusca of California and adjacent regions. Memoirs of the San Diego Society of Natural History 1:1-1036. Grau, G. 1959. Pectinidae of the Eastern Pacific. Allan Hancock Pacific Expeditions 23:1-308. Gray, J. E. 1825. A list and description of some species of shells not taken notice of by Lamarck. Annals of Philosophy (new series) 9(2):134- 140. Gray, J. E. 1826. On a Recent species of the genus Hinnita of DeFrance, and some observations on the shells of the Monomyaires of Lamarck. Annals of Philosophy (new series) 12(2):103-106. Habe, T. 1977. Systematics of Mollusca in Japan. Bivalvia and Scaphopoda. Hokuryukan, Tokyo. 372 pp. Hanley, S. 1855. Ipsa Linnaei Conchylia. The Shells of Linnaeus, deter- mined from his manuscripts and collections. Williams and Norgate, London. 556 pp. Haq, B. U., J. Hardenbol, and P. R. Vail. 1987. Chronology of fluctuating sea levels since the Triassic. Science 235:1156-1166. Harland, W. B., A. V. Cox, P. G. Llewellyn, C. A. G. Pickton, A. G. Smith, and R. Walters. 1982. A Geologic Time Scale. Cambridge University Press, New York. xi+ 131 pp. Harper, E. M. 1991. The evolution of the cemented habit in the bivalved molluscs. Unpublished Ph.D. dissertation, Department of Earth Sciences, The Open University, Milton Keynes, United Kingdom. 323 pp. Hayami, I. and T. Okamoto. 1986. Geometric regularity of some oblique sculpture in pectinid and other bivalves; recognition by computer simulations. Paleobiology 12(4):433-449. Hertlein, L. G. 1935. The Templeton Crocker Expedition of the California Academy of Sciences, 1932. No. 25. The Recent Pectinidae. Proceedings of the California Academy of Sciences, Fourth Series, 21(25):301-328. Hertlein, L. G. 1972. Description of a new species of Chlamys (Mollusca: Pelecypoda) from the Gadpagos Islands. Proceedings of the California Academy of Sciences, Fourth Series, 39(1):1-6. Hertlein, L. G. and U. S. Grant IV. 1972. The geology and paleontology of the marine Pliocene of San Diego, California (Paleontology: Pelecypoda). San Diego Society of Natural History Memoir 2 (Part 2B): 143-409. Hinds, R. B. 1844-1845. The zoology of the voyage of H. M. S. Sulphur, under the command of Capt. Sir Edward Belcher ... during the years 1836-42, Part 2 [Mollusca]. Smith and Elder, London. 72 pp. Hopkinson, J. 1907. Dates of publication of the separate parts of Gmelin’s edition (13th) of the ‘Systema Naturae’ of Linnaeus. Proceedings of the Zoological Society of London 1907(49):1035- 1037. Hubbard, B. 1920. Tertiary Mollusca from the Lares district, Porto Rico. New York Academy of Sciences, Scientific Survey of Porto Rico and the Virgin Islands 3(2):79-164. Iredale, T. 1929. Mollusca from the continental shelf of eastern Australia. Records of the Australian Museum Sydney 17(4):157-189. Iredale, T. 1939. Mollusca, Part I. British Museum (Natural History) Great Barrier Reef Expedition 1928-29 Scientific Reports 5(6):209- 425. Jackson, J. B. C. and P. Jung. 1992. Molluscan diversification and extinc- tion on opposite sides of the Isthmus of Panama [abstract]. Jn: Fifth North American Paleontological Convention, Abstracts and Programs. Field Museum of natural History, June 28-July 1, 1992. The Paleontological Society Special Publication No. 6, S. Lidgard and P. R. Crane, eds. p. 144. Jackson, J. B. C., P. Jung, A. G. Coates, and L. S. Collins. 1993. Diversity and extinction of tropical American mollusks and emergence of the Isthmus of Panama. Science 260:1624-1626. Jeffreys, J. G. 1863. British Conchology, vol. 2. London: John Van Voorst. xiv + 466 pp. Johnson, R. I. 1969. Pfeiffer’s Novitates Conchologicae, Series 1, Land Mollusca, 1854-1879, and Dunker’s Novitates Conchologicae, Series II, Marine Mollusca, 1862-1882. A complete collation. Journal Society for the Biblphyl. of Natural History. 5(3):236-239. Johnson, R. I. 1989. Molluscan taxa of Addison Emery Verrill and Katharine Jeannette Bush, including those introduced by Sanderson Smith and Alpheus Hyatt Verrill. Occasional Papers on Mollusks [Department of Mollusks, Museum of Comparative Zoology, Harvard University] 5(67):1-143. Jones, D. S., and P. F. Hasson. 1985. History and development of the marine invertebrate faunas separated by the Central American isth- mus. In: The Great American Biotic Interchange, F. G. Stehli and S. D. Webb, eds. pp. 325-355. Plenum Press, New York. Jones, D. S., B. J. Macfadden, S. D. Webb, and P. A. Mueller. 1991. Integrated geochronology of a classic Pliocene fossil site in Florida: linking marine and terrestrial biochronologies. The Journal of Geology 99(5): 637-648. Jones, K. H. and H. B. Preston. 1904. List of Mollusca collected during the expedition of H.M.S. “Waterwitch” in the China Seas, 1900- 1903, with descriptions of new species. Proceedings of the Malacological Society of London 6:138-151. Jung, P. 1989. Revision of the Strombina-group (Gastropoda: Columbellidae), fossil and living; distribution, biostratigraphy, and systematics. Schweizerische Paldontologische Abhandlungen 111:1-298. Kabat, A. R., and R. E. Petit. 1988. The two printings of J. F. Gmelin’s Systema Naturae, 13th Edition (1788-96). Nautilus 102(4):164- 166. Keen, A. M. 1966. Some notes on the molluscan collections at the University of Copenhagen. American Malacological Union, Inc. Bulletin 33 (Annual Reports for 1966): 73. Kensley, B. 1985. Cementing bivalve molluscs of the genus Chlamys (‘Hinnites’) in southern Africa. South African Journal of Science 81: 629-631. Kleemann, K. 1990. Coral associations, biocorrosion, and space competi- tion in Pedum spondyloideum (Gmelin) (Pectinacea, Bivalvia). Marine Ecology 11(1):77-94. Kohn, A. J. 1988. Type specimens and identity of the described species of Conus. VIII. The species described 1821-1830. Zoological Journal of the Linnaean Society 93:19-70. Knowlton, N., L. A. Weight, L. A. Solérzano, D. K. Mills, and E. Bermingham. 1993. Divergence in proteins, mitochondrial DNA, and reproductive compatibility across the Isthmus of Panama. Science 260:1629-1632. Ladd, H. S. and J. E. Hoffmeister.1945. Geology of Lau, Fiji. Bernice P. Bishop Museum Bulletin 181:vi + 399 pp. Ladd, J. W., T. L. Holcombe, G. K. Westbrook, and N. T. Edgar. 1990. Caribbean marine geology; active margins of the plate boundary. In: The Geology of North America. Vol. H. The Caribbean Region. Decade of North American Geology Series. G. Dengo and J. E. Case, eds. pp. 261-290. The Geological Society of America, Boulder, Colorado. Lamarck, J- B. P. 1799. Prodrome d’une nouvelle classification des coquilles. Mémoire de la Société Histoire naturelle, Paris 1:63-91. Lamarck, J- B. P. 1819. Histoire naturelle des animaux sans vertébres [Paris] 6(1):vi + 252 pp. Lauriat-Rage, A. 1981. Les bivalves du Redonien (Pliocéne Atlantique de France), signification stratitgraphique et paléobiogéographique. Mémoires du Muséum National d’Histoire Naturelle, N.S., 45:1- 173. WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 247 Linnaeus, C. 1758. Systema naturae per regna tria naturae ... editio deci- ma, reformata 1 [Regnum animale]. Salvii, Stockholm. 824 + ili pp. Locard, A. 1898. Expéditions scientifiques du Travailleur et du Talisman pendant les Années 1880, 1881, 1882, 1883. Mollusques Testacés, Vol. 2. Masson, Paris. 515 pp. Lyons, W. G. 1991. Post-Miocene species of Latirus Montfort, 1810 (Mollusca: Fasciolariidae) of southern Florida, with a review of regional marine biostratigraphy. Bulletin of the Florida Museum of Natural History, Biological Sciences 35(3):131-208. Madsen, F. J. 1949. The Zoology of Iceland, vol. 4, part 63, Marine Bivalvia. Ejnar Munksgaard, Copenhagen and Reykjavik. 116 pp. Masuda, K. 1962. Tertiary Pectinidae of Japan. Science Reports of the Tohoku University, Sendai, Second Series (Geology) 33(2):117-238. Matti, J. C., D. M. Morton, and B. F. Cox. 1985. Distribution and geologic relations of fault systems in the vicinity of the central Transverse Ranges, southern California. United States Geological Survey Open-file Report 85-365:23 pp. Maurrasse, F. J- M. R. 1990. Stratitgraphic correlation for the circum- Caribbean region. In: The Geology of North America. Vol. H. The Caribbean Region, G. Dengo and J. E. Case, eds. The Geological Society of America, Boulder, Colorado. Folded chart. Maury, C. J. 1917. Santo Domingo type sections and fossils: pt. 1. Bulletins of American Paleontology 5(29):1-240. Maury, C. J. 1920. Tertiary Mollusca from Porto Rico. New York Academy of Sciences, Scientific Survey of Porto rico and Virgin Islands, 3(1):1-77. M’Coy, F. 1844. A synopsis of the characters of the Carboniferous Limestone fossils of Ireland. The University Press (M.H. Gill), Dublin. viii + 207 pp. Mesolella, K. J. 1967. Zonation of uplifted Pleistocene coral reefs on Barbados, West Indies. Science 156:638-640. Middendorff, A. T. von. 1849. Beitrage zu einer Malacozoologia Rossica, III. Akademia Nauk SSSR. Memoir. Series 6 (Natural Sciences), vol. 6:517-610. Miller III, W. 1989. Checklist of megafossils from the James City Formation (Lower Pleistocene) at Johnson Point, Craven County, North Carolina. Tulane Studies in Geology and Paleontology 22(3):93-100. Moore, E. J. 1984. Tertiary marine pelecypods of California and Baja California: Propeamussidae and Pectinidae. United States Geological Survey Professional Paper 1228-B:1-112. Morgan, J. P., J. M. Coleman, and S. M. Gagliano. 1968. Mudlumps: diapiric structures in Mississippi Delta sediments. In: American Association of Petroleum Geologists, Memoir 8, J. Braunstein and G.D. O’Brien, eds. pp. 145-161. American Association of Petroleum Geologists, Tulsa, Oklahoma. Miiller, O. F. 1776. Zoologiae Danicae Prodromus, seu animalium Daniae et Norvegiae indigenarum characteres, nomina, et synonyma imprimis popularium. Copenhagen. xxxii + 281 pp. Multer, H. G. 1969. Pleistocene back reef environments. In: Field guide to some carbonate rock environments, Florida Keys and western Bahamas, H. G. Multer. ed. Miami Geological Society, Miami. pp. 109-113. Nicklés, M. 1955. Scaphopodes et lamellibranches récoltes dans l’ouest Africain. Atlantide Report No. 3 (Scientific results of the Danish expedition to the coasts of tropical west Africa, 1945-1946), No. 3. Danish Science Press, Copenhagen. 237 pp. Nomura, S. and H. Niino. 1932. Fossil Moliusca from Izu and Hakone. Tohoku Imperial University, Science Reports, Second Series (Geology) 15(3):169-192. Olsson, A. A. 1964. Neogene mollusks from northwestern Ecuador. Paleontological Research Institution, Ithaca, New York. 256 pp. Palmer, D. K. 1948. List of Palmer Cuban fossil localities. Bulletins of American Paleontology 31(128):1-178. Pennant, T. 1777. British Zoology, vol. 4: Crustacea. Mollusca. Testacea. Printed by W. Eyres, for B. White, London. 278 pp. Petuch, E. J. 1982a. Notes on the molluscan paleoecology of the Pinecrest Beds at Sarasota, Florida with the description of Pyruella, a strati- graphically important new genus (Gastropoda: Melongenidae). Proceedings of the Academy of Natural Sciences of Philadelphia 134:12-30. Petuch, E. J. 1982b. Geographical heterochrony: Contemporaneous coex- istence of Neogene and Recent molluscan faunas in the Americas. Palaeogeography, Palaeoclimatology, Palaeoecology 37:277-312. Piccoli, G., S. Sartori, and A. Franchino. 1986. Mathematical model of the migration of Cenozoic benthic mollusks in the Tethyan belt. Memorie di Scienze Geologiche (Universita di Padova] 38:207-244. Poli, G. S. 1795. Testacea utriusque Siciliae eorumque historia et anatome tabulis aeneis illustrata 2. Parma. |xxvi + 264 pp. Por, F. D. 1989. The Legacy of Tethys: An aquatic biogeography of the Levant. Kluwer Academic Publishers, Boston, xii + 214 pp. Powell, C. L., IJ. 1985. Bivalve molluscan paleoecology of northern expo- sures of the marine Neogene Imperial Formation in Riverside County, California. Western Society of Malacologists, Annual Report 17:29-32. Powell, C. L., II. 1986. Stratigraphy and bivalve molluscan paleontology of the Neogene Imperial Formation in Riverside County, California. Master's Thesis, San Jose State University, San Jose, California, 325 pp. Powell, C. L., II. 1988. The Miocene and Pliocene Imperial Formation of southern California and its molluscan fauna: an overview. Western Society of Malacologists, Annual Report 20:11-18. Raffi, S. 1971. I Pettinidi del Pliocene e Calabriano dell’ Emilia occiden- tale (Piacentino e Parmense). Bolletino della Societa Paleontologica Italiana 9(2):97-135. Raffi, S., S. M. Stanley, and R. Marasti. 1985. Biogeographic patterns and Plio-Pleistocene extinction of Bivalvia in the Mediterranean and southern North Sea. Paleobiology 11(4):368-388. Rafinesque, C. S. 1815. Analyse de la Nature. Palermo. 225 pp. Récluz, M. C. 1853. Description de coquilles nouvelles (G. Pecten, Tellina et Natica). Journal de Conchyliologie 4:152-156. Reed, J. K. and P. M. Mikkelsen. 1987. The molluscan community associ- ated with the scleractinian coral Oculina varicosa. Bulletin of Marine Science 40(1):99-131. Reeve, L. A. 1852-53. Conchologia Iconica: or, Illustrations of the Shells of Molluscous Animals, vol. VIII, Monograph of the genus Pecten. Lovell Reeve, London. Unnumbered pages on which 157 consecu- tively numbered species are described, interleaved with 35 pls. Regteren Altena, C. O. van. 1937. Bijdrage tot de kennis der fossiele, sub- fossiele en recente mollusken, die op de Nederlandsche stranden aanspoelen, en hunner verspreiding. D. van Sijn and Sons, Rotterdam. xii + 184 pp. Richardson, L., R. T. Abbott, and G. M. Davis. 1979. Early references to the figures in the Conchylien Cabinet of Martini and Chemnitz Volumes I-XII. Tryonia (Miscellaneous Publications of the Department of Malacology of the Academy of Natural Sciences of Philadelphia) 2(1-2):iv + 427 pp. Rios, E. C. 1985. Seashells of Brazil. Fundagao Cidade do Rio Grande, Fundacdo Universidade do Rio Grande, Rio Grande, RS, Brazil. 328 pp. Robinson, D. G. 1990. Notes on the fauna of the Moin Formation - I. On the occurrence of Gyrineum in the Early Pleistocene Moin Formation of Costa Rica.Tulane Studies in Geology and 248 AMER. MALAC. BULL. 10(2) (1993) Paleontology 23(4):133-135. Robinson, D. G. 1992. The zoogeographic implications of the prosobranch gastropods of the Molin Formation of Costa Rica (abstract). American Malacological Union, Program and Abstracts, 58th Annual Meeting, 2-7 August 1992, Sarasota, Florida: p. 36. Roger, J. 1939. Le genre Chlamys dans les formations Néogénes de l'Europe. Conclusions générales sur la répartition géographique et stratigraphique des Pectinidés du Tertiaire récent. Mémoires de la Société Géologique de France, N.S. 17(2-4): 294 pp. Rombouts, A. 1991. Guidebook to Pecten shells. Recent Pectinidae and Propeamussiidae of the World. Universal Book Services, Dr. W. Backhuys, Oegstgeest, The Netherlands. xiii + 157 pp. Roth, B. and E. V. Coan. 1978. Nomenclatural notes on Hinnites gigan- teus (Gray). The Veliger 20(3):297-298. Sacco, F. 1897. I Molluschi dei terreni Terziarii del Piemonte e della Ligura. Part 24 (Pectinidae). Carlo Clausen, Torino. 74 pp. Saunders, J. B., P. Jung, and B. Biju-Duval. 1986. Neogene paleontology in the northern Dominican Republic. 1. Field surveys, lithology, environment, and age. Bulletins of American Paleontology 89(323):1-79. Serres, M. de. 1829. Géognosie des Terrains tertiares; ou Tableau des principaux animaux invertébres des terrains marins tertiares, du Midi de la France. Pomathio-Durville, Montpellier and Paris. xcii + 276 pp. Sherborn, C. D., and E. A. Smith. 1911. A collation of J. C. Chenu’s Illustrations Concyliologiques, and a note on P. L. Duclos’ Hist. Nat. gén. et Part. Coquilles. Proceedings of the Malacological Society of London 9(4):264-267. Smith, J. T. 1991a. Cenozoic giant pectinids from California and the Tertiary Caribbean Province. United States Geological Survey Professional Paper 1391:v + 155 pp. Smith, J. T. 1991b. Cenozoic marine mollusks and paleogeography of the Gulf of California. In: The Gulf and Peninsular Province of the Californias. J. P Dauphin and B. R. T. Simoneit, eds. pp. 637-666. American Association of Petroleum Geologists, Memoir 47. American Association of Petroleum Geologists, Tulsa, Oklahoma. Sobetski, V. A. 1977. Bivalve molluscs of the Upper Cretaceous platform seas of SW USSR. [In Russian] Trudy Paleontological Institute, Akademia Nauk SSSR 159:1-256. Soot-Ryen, T. 1951. New records on the distribution of marine Mollusca in northern Norway. Astarte 1:1-11. Sowerby I, G. B. 1827. Observations on the shells of an acephalous mol- luscum of the family of Pectinidae, for which the generic name of Hinnites has been proposed by M. Defrance; together with the char- acters of several species. The Zoological Journal 3(9):67-72. Sowerby II, G. B. 1842. Thesaurus Conchyliorum, Vol. 1. Monography of the genus Pecten. Sowerby, London. pp. 45-82. Sowerby III, G. B. 1895. Descriptions of nine new species of shells. Proceedings of the Malacological Society of London 1:214-217. Stanley, S. M. 1966. Paleoecology and diagenesis of Key Largo Limestone, Florida. Bulletin of the American Association of Petroleum Geologists 50:1927-1947. Stanley, S. M. 1986a. Population size, extinction, and speciation: the fis- sion effect in Neogene Bivalvia. Paleobiology 12(1):89-110. Stanley, S. M. 1986b. Anatomy of a regional mass extinction: Plio- Pleistocene decimation of the Western Atlantic bivalve fauna. Palaios 1:17-36. Sutty, L. 1986. Seashell treasures of the Caribbean. E. P. Dutton, New York. 128 pp. Tedford, R. H., M. F. Skinner, R. W. Fields, J. M. Rensberger, D. P. Whistler, T. Galusha, B. E. Taylor, J. R. Macdonald, and S. D. Webb. 1987. Faunal succession and biochronology of the Arikareean through Hemphillian internal (Late Oligocene through earliest Pliocene epochs) in North America. In: Cenozoic mammals of North America, geochronology and biostratigraphy, M. O. Woodburme, ed. University of California Press, Berkeley. pp. 174- 210. Teppner, W. von. 1922. Pars 15. Lamellibranchiata tertiaria. “Anisomyaria’. In: Fossilium Catalogus, I: Animalia, C. Diener, ed. W. Junk, Berlin. pp. 67-296. Trechmann, C. T. 1933. The uplift of Barbados. The Geological Magazine 70(823):19-47. Tryon, G. W. 1869. Descriptions of new species of marine bivalve Mollusca in the collection of the Academy of Natural Sciences. American Journal of Conchology 5:170-172. Vermeij, G. J. 1993. The biological history of a seaway. Science 260: 1603-1604. Vermeij, G. J. and E. J. Petuch. 1986. Differential extinction in tropical American molluscs: Endemism, architecture, and the Panama Land Bridge. Malacologia 27(1):29-41. Verrill, A. E. 1897. A study of the family Pectinidae, with a revision of the genera and subgenera. Transactions of the Connecticut Academy of Arts and Sciences 10:41-95. Vernill, A. E. and K. J. Bush. 1898. Revision of the deep-water Mollusca of the Atlantic Coast of North America, with descriptions of new genera and species. Part I - Bivalvia. Proceedings of the United States National Museum 20(1139):775-901. Vokes, E. H. 1988. Muricidae (Mollusca: Gastropoda) of the Esmeraldas Beds, northwestern Ecuador. Tulane Studies in Geology and Paleontology 21(1): 1-58. Vokes, E. H. 1991. The genus Truncaria (Gastropoda: Buccinidae) in the Esmeraldas Beds of northwestern Ecuador. Tulane Studies in Geology and Paleontology 24(4):97-100. Vokes, H. E. 1979. Observations on the genus Dimya (Mollusca: Bivalvia) in the Cenozoic faunas of the Western Atlantic region. Tulane Studies in Geology and Paleontology 15(2):33-53. Waldrop, J. S. and D. Wilson. 1990. Late Cenozoic stratigraphy of the Sarasota area. In: Plio-Pleistocene stratigraphy and paleontology of south Florida, December 7-8, 1990. Southeastern Geological Society Annual Field Excursion, Guidebook 3. W. D. Allmon and T. M. Scott, eds. Southeastern Geological Society, Post Office Box 1634, Tallahassee, Florida. pp. 195-227. Waller, T. R. 1972. The Pectinidae (Mollusca: Bivalvia) of Eniwetok Atoll, Marshall Islands. Veliger 14(3):221-264. Waller, T. R. 1973. The habits and habitats of some Bermudian marine mollusks. Nautilus 87(2):31-52. Waller, T. R. 1978. Morphology, morphoclines and a new classification of the Pteriomorphia (Mollusca: Bivalvia). Philosophical Transactions of the Royal Society, London B,284:345-365. Waller, T. R. 1984. The ctenolium of scallop shells: Functional morpholo- gy and evolution of a key family-level character in the Pectinacea (Mollusca: Bivalvia). Malacologia 25(1): 203-219. Waller, T. R. 1991. Evolutionary relationships among commercial scallops (Mollusca: Bivalvia: Pectinidae). In: Scallops: Biology, Ecology and Aquaculture, S. E. Shumway, ed. Elsevier, New York. pp. 1-73. Waller, T. R. and L. Marincovich, Jr. 1992. New species of Camptochlamys and Chlamys (Mollusca: Bivalvia: Pectinidae) from near the Cretaceous/Tertiary boundary at Ocean Point, North Slope, Alaska. Journal of Paleontology 66(2):215-227. Wallin, L. 1991. Catalogue of type specimens. 4. Linnaean specimens. Zoological Museum, Uppsala University, Uppsala, Sweden. 233 pp. Ward, L. W. 1992. Molluscan biostratigraphy of the Miocene, middle Atlantic Coastal Plain of North America. Virginia Museum of Natural History, Memoir 2:1-159. WALLER: EVOLUTION OF TROPICAL AMERICAN “CHLAMYS” 249 Ward, L. W. and B. W. Blackwelder. 1987. Late Pliocene and Early Pleistocene Mollusca from the James City and Chowan River Formations at the Lee Creek Mine. Jn: Clayton E. Ray (ed.), Geology and paleontology of the Lee Creek Mine, North Carolina, II. Smithsonian Contributions to Paleobiology 61:113-283. Weisbord, N. E. 1964. Late Cenozoic pelecypods from northern Venezuela. Bulletins of American Paleontology 45(204):1-564. Wilkes, J. 1810. Conchology. In: Encyclopaedia Londinensis; or, Universal Dictionary of Arts, Sciences, and LIterature. J. Adlard, London. pp. 14-41. Wood, S. V. 1861. A monograph of the Crag Mollusca, with description of shells from the upper Tertiaries of the British Isles. Vol. 2. Bivalvies. London: The Palaeontographical Society, 342 pp., 31 pls. Wood, W. 1828. Supplement to the Index Testaceologicus; or Catalogue of Shells, British and Foreign. Printed for W. Wood, London. iv + 59 pp. Woodring, W. P. 1925. Contributions to the Geology and Palaeontology of the West Indies. Miocene Mollusks from Bowden, Jamaica. Pelecypods and Scaphopods. Carnegie Institution of Washington, Washington, D.C. v + 222 pp. Woodring, W. P. 1928. Miocene mollusks from Bowden, Jamaica. Part Il. Gastropods and discussion of results. Washington, D.C.: Carnegie Institution of Washington. vii + 564 pp. Woodring, W. P. 1966. The Panama land bridge as a sea barrier. Proceedings of the American Philosophical Society 110(6):425-433. Woodring, W. P. 1982. Geology and paleontology of Canal Zone and adjoining parts of Panama. Description of Tertiary mollusks (Pelecypods: Propeamussiidae to Cuspidariidae; additions to fami- lies covered in P 306-E; additions to gastropods; cephalopods). United States Geological Survey Professional Paper 306-F:541- 759. Yonge, C. M. 1951. Observations on Hinnites multirugosus (Gale). University of California Publications in Zoology 55(8):409-420. Yonge, C. M. 1967. Observations on Pedum spondyloideum (Chemnitz) Gmelin, a scallop associated with reef-building corals. Proceedings of the Malacological Society of London 37:311-323. Date of manuscript acceptance: 7 June 1993 The zoogeographic implications of the prosobranch gastropods of the Moin Formation of Costa Rica David G. Robinson Malacology/Invertebrate Paleontology, Academy of Natural Sciences of Philadelphia, 1900 Benjamin Franklin Parkway, Philadelphia, Pennsylvania 19103, U.S.A. Abstract: The deposits of the Moin Formation are located in and around the town of Puerto Limon, on the Caribbean coast of Costa Rica. The fossils date the formation as being Late Pliocene to early Pleistocene in age. The systematics of its constituent prosobranch gastropods were studied, and compar- isons made with Recent and fossil gastropod faunas in the Caribbean area. Seventy-one percent of the taxa are living species and for most, this is their first occurrence in the fossil record. These are primarily Caribbean species, but West African/Mediterranean, Panamic and Indo-Pacific elements are also present. Twenty-one percent of the taxa are endemic to these beds, and 8% are typically Mio-Pliocene species occurring for the last time in the fossil record. A number of species characteristic of the Moin formation have a Holocene distribution in the western Caribbean, extending approximately from the Gulf of Honduras to the northern coast of Colombia, while others have a southern Caribbean distribution, from Costa Rica to across northern South America. These zoogeographic patterns are not well-defined, however, and based on the prosobranch species occurring in the Moin Formation, the existence of specific faunules or other zoogeographic entities in the southwestern Caribbean area cannot be supported. The deposits of the Moin Formation crop out pri- marily west and northwest of the town of Puerto Limén on the Caribbean coast of Costa Rica. Historically, the forma- tion has been considered to consist of some 70 m of dark blue claystone to loosely consolidated blue sandy clay with indistinct bedding due to extensive bioturbation, with inter- bedded lenses of coral reef material. Where bedding is visi- ble, it dips at an angle of 1-2° to the north (Cassell, 1986). With the exposure by bulldozing associated with the urban- ization of the peninsula, these beds weather to a reddish- orange clay, with well-preserved fossils eroding out so that they can be easily collected. Workers in the past have dated the Moin Formation as Miocene or Pliocene in age (Gabb, 1881; Olsson, 1922; Woodring, 1928) using traditional dating methods such as percentages of living species involving Lyellian curves. More recently, using micropaleontological evidence, the Moin Formation has been considered to be early Pleisto- cene in age (Akers, 1972; Taylor, 1975; Robinson, 1990, 1991) and it has been suggested that the topmost beds extend into the middle Pleistocene (Akers, pers. comm.) or even late Pleistocene (Cassell, 1986). However, Coates et al. (1992), have expanded the definition of the Moin Formation, adding another 130 m of underlying sediments, placing most of the formation in the Late Pliocene, with only the topmost strata falling in the Pleistocene. This expanded definition shall remain open to question for the present. METHODS Extensive collections of fossil material from the area were made over several years, and the collections made by earlier Tulane University researchers, primarily Drs. Harold and Emily Vokes, were incorporated. The full geographic extent of the Moin Formation is not clearly established, but is believed to underlie most of the tropical forest on the higher elevations of the Limon Peninsula. The collection of fossil material has tended to be opportunistic in nature, being possible only in “windows” of short dura- tion, occurring when briefly exposed by bulldozing of the vegetation preparatory to urban expansion of Puerto Limon. If not immediately built upon, the land is rapidly reclaimed by secondary forest growth within a few months. The mol- luscs collected from the Moin Formation were compared with taxa from other Caribbean faunas, in particular Mio- Pliocene Gurabo and Cercado Formations of the Domin- ican Republic, the Pinecrest and Caloosahatchee For- mations of Florida, the Agueguexquite Formation of the Isthmus of Tehuantepec, the Limén and Gatun Formations of the Panamic Isthmus, the Cantaure Formation of Vene- zuela, and the Esmeraldas beds of Ecuador. Of particular importance for comparison were the faunas of the Bowden Formation of Jamaica, the Cabo Blanco Group of Vene- zuela and the Bermont of Florida, as they are closer in age to the Moin and provide a chronological context for the study. The systematics of all constituent species had to be American Malacological Bulletin, Vol. 10(2) (1993):251-255 251 252 AMER. MALAC. BULL. 10(2) (1993) examined with care because of the perspective adopted by the first molluscan systematists working with these various formations. In many cases, the assumption was made that these beds were by and large contemporaneous,and the molluscan taxa contained in one particular formation were therefore common to others. With the increasing accuracy of dating using foraminifera and calcareous nannoplankton indicators, these circum-Caribbean beds are now known to range in age from Miocene to Pleistocene. Many of the molluscan species can now be seen as distinct, as opposed to being regional variants of more widespread taxa, particu- larly when they can be studied together in a more clearly- defined geographical and chronological context. During this study, collections of comparative Recent shallow-water material were also made from the western Caribbean region, from the Yucatan Peninsula, and the Caribbean coasts of Honduras, Costa Rica (Robinson and Montoya, 1988), and Panama and Colombia. Deeper water material was collected from the Honduran continental shelf from on board Honduran shrimp boats. Ecological data from living specimens were used to interpret the paleoecological condi- tions under which the Moin Formation was deposited. RESULTS Three hundred and six prosobranch gastropods were identified in this study, and the systematics as well as the paleoecology of these species are currently in prepara- tion. It should be understood that this paper is a preliminary report, with further analysis of the systematics and phylo- genetic relationships of the species currently in progress, but some conclusions of interest can already be made. Effects of the Plio-Pleistocene glaciations A substantial proportion of the rich Moin fauna (65 species), approximately 21% of the total, appears to be restricted to these beds. The deposition of the portion of the Moin Formation focused on in this study is interpreted here as occurring during a warm interglacial period, which re- sulted in a short-lived speciation event. While some of these species have already been described (Table 1), a large pro- portion have not. More collecting needs to be done in the southwestern Caribbean to confirm that all of these species are indeed extinct. Another 8% of the Moin taxa (24 species) represent the last occurrence in the fossil record of a number Mio- Pliocene taxa typical of older fossil beds, including the Bowden, Pinecrest, Cantaure, Gatin and Rio Banano Formations (Table 2). Of interest are a number of species that are illustrative of the effects of the initial Plio- Pleistocene glacial pulses. By the time the Moin Formation Table 1. Previously described taxa endemic to the Moin Formation. Latirus irazu Olsson, 1922 Agaronia mancinella (Olsson, 1922) Calliostoma castilla Olsson, 1922 _—Olivella limonensis Olsson, 1922 Turbo pittieri Dall, 1912 Mitra poas Olsson, 1922 Cochliolepis simplex (Gabb, 1881) Nebularia coralliophila Olsson, 1922 Atlanta ammonitiformis Prunum limonensis (Dall, 1896) Gabb, 1881 Conus trisculptus Cerithiopsis limonensis Pilsbry and Johnson, 1917 (Gabb, 1881) Conus ultimus Aclis microsculpta Gabb, 1881 Pilsbry and Johnson, 1917 Chicoreus prolixus E. Vokes, 1974 Conus limonensis Olsson, 1922 Calotrophon ascensus Compsodrillia moenensis E. Vokes, 1976 (Gabb, 1881) Antillophos limonensis Cerodrillia limonetta (Olsson, 1922) (Olsson, 1922) Knefastia limonensis Cantharus tortuguera Olsson, 1922 (Olsson, 1922) Metula limonensis Olsson, 1922 plus 40 undescribed taxa Nassarius caribbeus (Gabb, 1881) Calliostoma limonensis Olsson, 1922 was deposited, the Caribbean Basin had been subjected to the effects of several glacial episodes, and the composition of the Moin fauna already indicating a significant loss of Mio-Pliocene species, with many groups now considered to be Paciphilic and a number of extinct species characteristic of somewhat older formations, especially the Bowden Formation of Jamaica. It should be noted that although the Bowden is not very much older (considered here as early Late Pliocene) and is geographically close, a substantial portion of its fauna had disappeared by Moin time. Clearly, the differences between the faunas of the Bowden and Moin Formations highlight the effects of one or more of the earlier glacial pulses of the Late Pliocene. A small but notable constitutent of the Moin fauna is defined by those species extending their geographic ranges or migrating southwards due to climatic deteriora- tion during the Tertiary. Among these are Trivia cf. T. crovae (Olsson, 1967), Jenneria cf. J. richardsi Olsson, 1967, and Niso willcoxiana Dall, 1889. These species, or Table 2. Mio-Pliocene species whose last occurrence is in the Moin Formation. Calliostoma guppyana Gabb, 1881 Conus gracillisimus Guppy, 1866 Turbo saltus Olsson, 1922 Conus recognitus Guppy, 1867 Arene lepidota Woodring, 1928 Scobinella morieri Laville, 1913 Microstelma cepula (Guppy, 1866) Polystira barretti (Guppy, 1896) Modulus basileus Guppy, 1878 Miraclathurella vittata Woodring, 1928 Atlanta diamesa Woodring, 1928 Glyphostoma moinica Olsson, 1922 Niso willcoxiana Dall, 1889 Nannodiella amicta (Guppy, 1896) Cancilla limonensis Saccharoturris consentanea (Olsson, 1922) (Guppy, 1986) Conus wiedenmayeri Jung, 1965 Bactrocythara obtusata (Guppy, 1896) plus 6 undescribed taxa ROBINSON: MOIN PROSOBRANCHS ZOOGEOGRAPHY 209 their immediate ancestors, were characteristic members of the faunas of the Pliocene Pinecrest Formation and were absent from contemporaneous fauas in the Caribbean. Depsite their subsequent occurrence in the Moin Forma- tion, they are absent from contemporaneous Late Pliocene and Pleistocene formations in Florida, indicating a com- plete displacement of these species by Plio-Pleistocene glacial episodes to the southern Caribbean. Their appear- ance in the Moin is also their last before complete extinc- tion, resulting from succeeding glacial pulses. Of particular interest is the Conus cedonulli com- plex, represented in the Moin by C. ultimus Pilsbry and Johnson, 1917. The Mio-Pliocene ancestor of this group was C. consobrinus G. B. Sowerby I, 1850, that was dis- tributed throughout the Caribbean basin. The species was subsequently restricted southwards to isolated refugia in the southern Caribbean, where it has speciated into a number of sibling species, each with a very limited distribution (Vink and Cosel, 1985). These species live today in a wide arc from the southern Lesser Antilles, along the coast of north- ern South America and then north to Honduras, although the complex is not as widely distributed as its parent species. Similarily, C. venezuelanus Petuch, 1987, a descendent of the widely-distributed Mio-Pliocene C. planiliratus G. B. Sowerby I, 1850, survives in a small area in the Golfo de Triste, off northern Venezuela. C. venezuelanus evidently had a greater range during Moin time, as it is quite common in the Costa Rica beds. West African/Mediterranean Influences Of interest in the Moin Formation also are those faunal elements that are identified with provinces outside of the typical Caribbean. The occurrence of some West African and Mediterranean elements is not entirely unex- pected. Ranellids have been shown to have the ability to cross the Atlantic (Scheltema, 1971; Laursen, 1981) and the closely-related bursids appear also to have achieved some level of teleplanic efficiency. Bufonaria marginata (Gmelin, 1791) is restricted to the eastern Atlantic and Mediterranean today and throughout its fossil history, except for the occur- rence of four individuals (so far) in the Moin Formation. Teleplanic efficiency may be increased by glacially-induced lowered sea temperatures, as suggested by Beu (1970), by delaying larval metamorphosis just long enough to enable an occasional individual to survive the trans-Atlantic cross- ing. These particular individuals were able to develop into adults here in the southwestern Caribbean, but apparently were never numerous enough to establish a permanent pop- ulation. Beu (pers. comm.) identified several individuals of Bursa scrobilator (Linné, 1758) from the Moin as well. This West African taxon does have a somewhat longer his- tory in the western Atlantic, having been described as B. mexicana Perrilliat, 1963, from the Agueguexquite Formation. B. scrobilator is now restricted to the eastern Atlantic. Other amphi-Atlantic species occurring in the Moin include Cerithium guinaicum Philippi, 1849, Phalium granulatum (Born, 1778), Polinices lacteus (Guilding, 1843), Pterotyphis pinnatus (Broderip, 1833), and Typhis sowerbii Broderip, 1833, although there is not enough evi- dence to determine unequivocally whether these taxa origi- nated in the western or eastern Atlantic. Connections with the Indo-Pacific The deep-water barrier between the Indo-West Pacific and the eastern Pacific has only been breached since Late Pliocene (Vermeij, 1978), or perhaps even later, with the migration of some elements of reef-associated faunas. The occasional incursion of teleplanic Indo-Pacific species today into the Panamic province is well-documented (Emerson, 1967 and 1978; Houbrick, 1968; Montoya, 1983), although the establishment of permanent popula- tions of such species is for the most part restricted to off- shore island groups, including Cocos, Clipperton and the Galapagos Islands (Hertlein, 1937, 1963; Hertlein and Emerson, 1953; Emerson, 1991). Unfortunately, it is not clear whether species of Indo-Pacific origin crossed into the Caribbean prior to the closure of the last trans-Isthmian seaways. Vermeij (1978) suggested that a warming of the waters around southern Africa during the Pliocene permit- ted the invasion of marine angiosperm grasses from the Indo-West Pacific. The occurrence of Smaragdia viridis (Linné, 1758), Turbo pittieri (Dall, 1912) (a turbinid close- ly related to T. petholatus Linné, 1758), Charonia tritonis variegata (Lamarck, 1816), Bursa rhodostoma thomae (d'Orbigny, 1842) and Gyrineum louisae Lewis, 1974 (Rob- inson, 1990), in the Late Pliocene and early Pleistocene Caribbean Basin supports this possible invasion route; belonging to typically Indo-Pacific supra-specific taxa, they are absent from the eastern Pacific (Panamic) waters and there is no fossil record of their ever having been estab- lished in that region. The first “modern” Caribbean fauna The Moin fauna represents the first fossil Caribbean fauna that is essentially modern in aspect. Two hundred and seventeen prosobranch gastropod species identified (approxi- mately 71% of the total) are still extant and for the most part, their Moin occurrence represents their first in the fos- sil record. The environments represented by these species range from the intertidal zones (including mangrove, rock and sand environments), through the various reef zones (lagoonal, back-reef, reef crest and fore-reef environments), with deposition occurring at the base of the fore-reef sys- tem at depths of 100 to 200 m. The systematic composition 254 AMER. MALAC. BULL. 10(2) (1993) of the fauna has also provided interesting insights into the zoogeography of the region. It should be made clear that the majority of the liv- ing species represented in the Moin fauna are pan- Caribbean in distribution. Smaller numbers of species have a more restricted distribution in the western and/or southern Caribbean, but excessive focus on these particular taxa without keeping in perspective the overall Caribbean nature of the fauna could lead to premature conclusions regarding the subdivision of the Caribbean Province as a whole. The general area is poorly known and for many species, their geographic range is not yet defined. Those Moin species that are living today (or that have a recognizable direct liv- ing descendent) and appear to be characteristic of an area extending from the Gulf of Honduras south to Panama and the adjacent northern Colombian coastline, are listed in Table 3. Another group of species (Table 4), appear to have a southern Caribbean bias, extending from Costa Rica south to Panama and east along the northern coast of South Ameica, with some ranging farther south to Surinam and Brazil. It should be noted, however, that some taxa occur- ring as fossils in Costa Rica, have yet to be found along that coast and are currently known from Colombia or Venezuela and farther east and south, e.g. Siratus springeri (Bullis, 1964). Three Moin species that have a southern distribution but occur in two disjunct populations in the southwestern and southeastern Caribbean (not having been reported from the central area), are Cyclostremiscus schrammi (P. Fischer, 1857), Metula lintea Guppy, 1882 and Siponochelus tityrus F. Bayer, 1971. CONCLUSIONS The gradual accumulation of both paleontological and neontological systematic and zoogeographic data, of which the current study represents only a fragment, is slow- ly adding to our understanding of Caribbean zoogeography. The emerging picture is one that is far more complex than Table 3. Moin species that have a western Caribbean distribution. Stephopoma myrakeenae Voluta virescens Lightfoot, 1786 Olsson and McGinty, 1958 (including the various named Cerithioclava garciai forms) Houbrick, 1985 Agaronia mancinella (Olsson, 1922) Cerithiopsis caribbaeus (Gabb, 1881) — (represented by today by Haustellum olssoni Agaronia hilli Petuch, 1987 and (E. Vokes, 1967) A. leonardhilli Petuch, 1987) Muricopsis deformis (Reeve, 1846) | Conus spurius quadratus Siphonochelus bullisi Gertman, 1969 Roding, 1798 Olivella myrmecoon Dall, 1912 Fusinus gabbi Grabau, 1904 Hindsiclava tippetti Petuch, 1987 Table 4. Moin species with a southern Caribbean distribution. Astralium brevispinum (Lamarck, 1822) Sconsia lindae Petuch, 1987 Siratus springeri (Bullis, 1964) Murexsul harasewychi Petuch, 1987 Fusinus caboblanquensis Weisbord, 1962 Scaphella evelina F. Bayer, 1971 (represented in the Moin by an undescribed but related species) Conus baylei Jousseaume, 1872 Conus venezuelensis Petuch, 1987 Hindsiclava cf. H. chazeliei (Dautzenberg, 1900) Persicula interruptolineata (Megerle von Miihlfeld, 1816) that of adjacent provinces. Although the Panamic and Carolinian provinces seem to be relatively homogeneous, with a large proportion of the molluscan species occurring throughout the extent of their respective provinces, the Caribbean appears to be fragmented, with numerous species having quite restricted geographic ranges. The nature of this subdivision, however, remains poorly under- stood. The implications of this study indicate the possibility of the existence of a western Caribbean and a southern Caribbean fauna. It should be stressed again, however, that the evidence so far does not yet justify the designation of zoogeographic entities on the Subprovince level, particular- ly in view of the limited numbers of species involved. Just as there are species that apparently define a faunule, there are others whose distribution extend outside the implied boundaries of such a faunule and may even span two such entities, or may change through time. Earlier attempts to subdivide the western Caribbean into subprovinces or faunules (Petuch, 1981, 1990) have been based on small numbers of species. The taxonomic status of a number of newly described taxa used to define these faunas has yet to be fully evaluated and their relationship with closely related taxa is not clearly understood. The Moin taxa and their liv- ing descendants do not fit into these subprovincial units and they therefore cannot be used to support these hypothetical constructs. A great deal more work needs to be done in terms of systematics and biogeography of the species (non- molluscan, as well as molluscan) assumed to be endemic to the area before subprovinical boundaries can be defined. ACKNOWLEDGMENTS The author gratefully acknowledges the help of many individuals who helped in numerous ways during the course of this project. In particu- lar, Emily H. Vokes of Tulane University, New Orleans, without whose help and support it would never have been completed. For their help in the field, J. Michel Montoya of San Jése, Costa Rica; Goldie Cooper, Ronald Ebanks, and Richard McNab of Roatan of the Bay Islands, Honduras; James Ernest of Panama City, Panam4; Matthew C. Redmond V of Golden, Colorado; and Richard J. Kirshner of New Orleans, Louisiana. ROBINSON: MOIN PROSOBRANCHS ZOOGEOGRAPHY 255 For their input and discussion, I would like to thank Gary Rosenberg and Elana Benamy of the Academy of Natural Sciences of Philadelphia. I am also grateful for the helpful comments and suggestions of two anonymous reviewers of an earlier draft of this paper. LITERATURE CITED Akers, W. H. 1972. Planktonic foraminifera and biostratigraphy of some Neogene formations, nothern Florida and the Atlantic coastal plain. Tulane Studies in Geology and Paleontology 9:1-139. Beu. A. G. 1970. The mollusca of the genus Charonia (Family: Cymatiidae). Transactions of the Royal Society of New Zealand, Biological Sciences 11:205-233. Cassell, D. T. 1986. Neogene foraminifera of the Limon Basin of Costa Rica. Doctoral dissertation, Louisiana State University, Baton Rouge, Louisiana. 323 pp. Coates, A. G., J. B. C. Jackson, L. S. Collins, T. M. Cronin, H. J. Dowsett, L. M. Byell, P. Jung and J. A. Obando. 1992. Closure of the Isthmus of Panama: The near-shore marine record of Costa Rica and western Panama. Geological Society of America Bulletin 104:814-828. Emerson, K. 1967. Indo-Pacific faunal elements in the tropical eastern Pacific, with special reference to the mollusks. Venus 25(3-4): 85-93. Emerson, K. 1978. Mollusks with Indo-Pacific affinities in the eastern Pacific Ocean. Nautilus 92:91-96. Emerson, K. 1991. First records for Cymatium mundum (Gould) in the eastern Pacific Ocean, with comments on the zoogeography of the tropical trans-Pacific Tonnacean and non-Tonnacean prosobranch gastropods with Indo-Pacific faunal affinities in western American waters. Nautilus 105:62-50. Gabb, W. M. 1881. Description of new species of fossils from the Pliocene Clay Beds between Limon and Moen, Costa Rica, together with notes on previously known species from there and elsewhere in the Caribbean area. Journal of the Academy of Natural Sciences of Philadelphia (Series 2) 8:349-380. Gertman, R. L. 1969. Cenozoic Typhinae (Mollusca: Gastropoda) of the western Atlantic region. Tulane Studies in Geology and Pale- ontology 7:143-191. Hertlein, L.G. 1937. A note on some species of marine mollusks occur- ring in both Polynesia and the western Americas. Proceedings of the American Philosophical Society 78:303-312. Hertlein, L. G. 1963. Contribution to the biogeography of Cocos Island, including a bibliography. Proceedings of the California Academy of Sciences (Series 4) 32:219-289. Hertlein, L. G. and W. K. Emerson. 1953. Mollusks from Clipperton Island (eastern Pacific) with the description of a new species of gas- tropod. Transactions of the San Diego Society of Natural History 11:345-364. Houbrick, R. S. 1968. New record of Conus cbraeus in Costa Rica. Veliger 1:292. Knight, J. B., L. R. Cox, A. M. Keen, A. G. Smith, R. L. Batten, E. L. Yochelson, N. H. Ludbrook, R. Robertson, C. M. Yonge and R. C. Moore. 1960. Treatise on Invertebrate Paleontology (ed. R. C. Moore). Part |. Mollusca, 351 pp. Laursen, D. 1981. Taxonomy and distribution of teleplanic prosobranch larvae in the North Atlantic. Dana Report No. 89, 43 pp. Montoya, J. M. 1983. Los moluscos marinos de la Isla del Coco, Costa Rica. 1. Lista anotada de especies. Brenesia 21:325-353. Olsson, A. A. 1922. The Miocene of northern Costa Rica, with notes on its general stratigraphic relations. Bulletins of American Paleontology 9:179-460. Petuch, E. J. 1981. A volutid species radiation from northern Honduras, with notes on the Honduras Caloosahatchian secondary relict pocket. Proceedings of the Biological Society of Washington 94:1110-1130. Petuch, E. J. 1990. A new molluscan faunule from the Caribbean coast of Panama. Nautilus 104(2):57-71. Robinson, D. G. 1990. On the occurrence of Gyrineum in the early Pleistocene Moin Formation of Costa Rica. Tulane Studies in Geology and Paleontology 23:133-135. Robinson, D. G. 1991. The systematics and paleoecology of the proso- branch gastropods of the Pleistocene Moin Formation of Costa Rica. Doctoral dissertation, Tulane University, New Orleans, Louisiana. 749 pp. Robinson, D. G. and J. M. Montoya. 1988. Los moluscos marinos de la costa Atlantica de Costa Rica. Revista de Biologia Tropical 35:375-400. Scheltema, R. S. 1971. Larval dispersal as a means of genetic exchange between geographically separated populations of shallow-water benthic marine gastropods. Biological Bulletin 140:284-322. Taylor, G. D. 1975. The Geology of the Limon area of Costa Rica. Doctoral dissertation, Louisiana State University, Baton Rouge, Louisiana. 116 pp. Vermeij, G. J. 1978. Biogeography and Adaptation: Patterns of Marine Life. Harvard University Press, Cambridge, Massachusetts. 332 pp. Vink, D. L. N. and R. von Cosel. 1985. The Conus cedonulli complex: his- torical review, taxonomy and biological observations. Revue suisse de Zoologie 92:525-603. Woodring, W. P. 1928. Miocene mollusks from Bowden, Jamaica. Part 2, Gastropods and discussion of results. Carnegie Institute of Washington, Publication 385, 564 pp. Woodring, W. P. 1957-1973. Geology and paleontology of Canal Zone and adjoining parts of Panama. Description of Tertiary mollusks. United States Geological Survey, Professional Paper 306 A-E. 539 pp., 82 pl. Date of manuscript acceptance: 28 April 1993 o A database approach to studies of molluscan taxonomy, biogeography and diversity, with examples from western Atlantic marine gastropods Gary Rosenberg Malacology Department, Academy of Natural Sciences, Philadelphia, Pennsylvania 19103, U.S.A. Abstract: A system of data fields and conventions is introduced that will allow workers on any group of mollusks to build interactive databases docu- menting classifications, synonymies, geographic and bathymetric ranges, and other summary information at the species level. This system is used to build a database which is the first comprehensive catalogue of Recent Western Atlantic gastropods ever assembled with geographic coverage extending from Greenland to Antarctica. As of January 1993, the database contained 8370 records, of which 3988 represent currently recognized species, 3491 are syn- onyms, 157 are nomina dubia and the remainder are misidentifications, misspellings, invalidly published or extralimital. There are 3103 currently recognized species of tropical Western Atlantic gastropods (35°N to 24°S); 2641 of these had been named by 1971, when Keen documented 2438 gastropod species in the tropical Eastern Pacific. The common perception that the tropical Western Atlantic fauna is depauperate compared to the Eastern Pacific cannot be supported. Faunal lists corrected for synonymies, variant generic combinations and misidentifications were extracted from the database for eight areas in the tropical Western Atlantic. These are eastern and western Florida, Yucatan, Panama, Jamaica, Puerto Rico, the Netherlands Antilles and northern Brazil. To correct for regional collecting biases, species smaller than 5 mm, those occurring only deeper than 50 meters, and those lacking external shells were exclud- ed from the lists. In 28 pairwise comparisons among the standardized lists, 27 showed faunal similarities greater than 50%. Western Florida, which lacks shallow reefal habitats, had faunal similarities lower than did eastern Florida, which has these habitats. Habitat availability seems as important as geographic distance in determining faunal similarity within the tropical Western Atlantic. None of the eight regions had more than 4% endemic species. Although species tend to be widespread within the tropical Western Atlantic, only 20% are known from other biogeographic provinces. Studies of mollusks in the Western Atlantic and “which would have been inconceivable without a reliable throughout the world are hindered by the lack of up-to-date reference list’; c) other researchers were stimulated to pro- systematic and faunal lists. Researchers attempting to iden- duce similar catalogues of Mediterranean opisthobranchs, tify or describe species have no comprehensive list of can- aplacophorans and cephalopods. Sabelli et al. (1990, 1992) didate species for comparison. Different names are used for themselves recently have finished a 781 page annotated cat- the same species in different regions, making faunal com- alogue of Mediterranean mollusks. parisons difficult. Researchers trying to document the The electronic database represents the next step in effects of extinction, immigration, and speciation on the the production of such catalogues. The potential value of diversity of various faunas find that reliable estimates of databases in biological research is well-known (Allkin and levels of diversity are impossible to obtain even for most Bisby, 1984), but this potential has not been realized in shallow-water faunas. Several published catalogues have malacology. The structure of data fields and conventions covered parts of the Western Atlantic fauna in the Northern introduced here will allow workers on any group of mol- Hemisphere (Dall, 1889b; Maury, 1922; Johnson, 1934; lusks to build interactive databases documenting classifica- Abbott, 1974; Turgeon et al., 1988). No catalogue has ever tions, synonymies, geographic and depth ranges, etc. been assembled for the entire Western Atlantic, nor for the Printed catalogues are static, whereas databases are dynam- tropical Western Atlantic biogeographic province. ic. Information in a database can be reorganized (alphabeti- Even in such intensively studied areas as the Medi- cally, systematically, geographically, chronologically, etc.) terranean, malacological research has suffered from the to suit the needs of the user. It can be queried in numerous lack of comprehensive lists of taxa. Sabelli et al. (1990) ways, limited only by the ingenuity of the researcher and note the salutary effects of the first modern catalogue of the types of raw data utilized by the database. Mediterranean shell-bearing mollusks, that of Piani (1980): A database must be designed to maximize its ability a) many researchers adopted the proposed classification to answer the questions most likely to be asked of it. A sin- thus stabilizing nomenclature; b) the Italian Malacological gle database that could address the needs of all fields of Society began censusing the Italian marine mollusks, molluscan research would be extremely complex and cum- American Malacological Bulletin, Vol. 10(2) (1993):257-266 251 258 AMER. MALAC. BULL. 10(2) (1993) bersome. The species-level database discussed here has a coarser level of focus than a collection database, summariz- ing information about the species overall, rather than about particular samples (lots) of a species. Coverage includes all Recent marine gastropod species reported in the Western Atlantic from Greenland to Antarctica. MATERIALS AND METHODS More than 1,000 publications on Western Atlantic gastropods were scanned for taxonomic, and bathymetric information. These publications are listed in a bibliographic database linked to the species database. Data come primari- ly from the published literature, but have also been taken from the malacological collections at the Academy of Natural Sciences of Philadelphia (ANSP) and solicited from researchers expert on the systematics of particular groups. The database contains references to sources of TABLE 1. Data fields used in the database. Numbers give character length of field; N = numeric, A = alphanumeric. Systematic number N Family A25 Genus A20 Subgenus A20 Species A20 Subspecies A20 Associated name A20 Status 1 Al Status 2 A3 Author A40 Date N Abc Al Attributed author A40 Original genus A20 Parentheses Al Combining genera A60 Combination A60 Citation A20 Figure A50 Other figure ASO Type locality A120 All localities A150 Ocean Al5 Shell Al North N South N East N West N Shallow N Deep N Live shallow N Live deep N Size N Comments A150 References A150 information about each species, to allow verification of the data. For each species-level name applied to a Western Atlantic gastropod, the database attempts to document its status as available, valid, synonymous, dubious, misidenti- fied or misspelled. Note that a species can be valid (i.e. cur- rently recognized), while its currently used name is invalid (e.g. preoccupied). Table 1 lists the fields of information tracked, and the space allocated to them; these fields are described below. Information about geographic, depth and size ranges is entered under the specific name used in a given publication. Geographic, depth, and size data are combined over synonyms to generate summary data for each currently recognized species. If the status of a name is changed from synonymous to valid, or moved from the syn- onymy of one species to another, its data move with it. The database is thus self-correcting to the degree that the litera- ture provides corrections. Figure 1 shows a subset of data- base fields and their contents. The database is currently implemented in Paradox 4.0, but can be easily transported in database or delimited ASCII format to other database programs. It can also be linked to collection databases for comparison of range data and synonymies. The database currently occupies 17 megabytes as a database file, or about 2 megabytes as a delimited ASCII file. A 386 computer or the equivalent is needed for satisfactory performance. [Subsets of the data- base are available electronically, via Internet from Rosenberg @ say.acnatsci.edu.] DATABASE STRUCTURE AND FORMAT Data fields, data types and character limits are listed in Table 1; conventions for data entry are described here. Examples of the contents of some fields are shown in Figure 1. SYSTEMATIC NUMBER: Each family is assigned a number, to allow sorting in systematic order. Alternate clas- sifications can be accommodated by adding fields for num- bering schemes that would allow sorting in different orders. FAMILY: Family names and classification follow Ponder and Warén (1988), Rosenberg (1992) and other recent works. GENUS: Generic classification follows Turgeon et al. (1988), Vaught (1989), and other recent works. Authorities for generic placements are cited in references. SUBGENUS: Conventions are the same as for genus. Few data pertaining to subgenera have been entered because ROSENBERG: DATABASE OF WESTERN ATLANTIC GASTROPODS 29 there is little agreement on subgeneric classifications and many authors of systematic works do not provide subgener- ic placements for the species they treat. SPECIES: Only the valid or currently recognized specific name of a species is entered here; synonymous names are entered under associated name. In cases where species previously synonymized have been recognized as valid by recent workers, citations are given in references. SUBSPECIES: Only the valid or currently recognized sub- specific name of a subspecies is entered here; synonymous names are entered under associated name. Most subspecies reported in the malacological literature occur sympatrically with the nominate subspecies, because malacological work- ers until recently treated forms and varieties as subspecies. Subspecies are therefore considered to be synonyms of the nominate form, except in cases where authors have present- ed evidence for geographical separation of ranges or have raised subspecies to full species. ASSOCIATED NAME: This field contains all species- group names that have been applied to Western Atlantic mollusks. In the case of currently used specific or subspe- cific names, the name is repeated here. Synonymous names, misspellings, and misidentifications are entered only here. For nomina dubia, the specific name is entered both here and under species. If a Western Atlantic species is known from other ocean basins, all known synonyms worldwide are listed. STATUS 1: Describes the status of a given associated name: V = valid (i.e. currently recognized); + = synonym (including misspellings and misidentifications); X = extra- limital (for names formerly attributed to the Western Atlantic fauna); D = nomen dubium (used to mean “‘not yet identified” rather than “‘unidentifiable”). Only one code can be entered for a given name; for synonyms of nomina dubia, ““D” takes priority over “+.” STATUS 2: Gives additional status information about the associated name: ? = questionable synonymy; N = name not available for nomenclatural purposes; S = currently rec- ognized subspecific name; O = objective synonym of the valid name; + = synonym of nomen dubium. More than one code can be entered, possible combinations being ?N, ?+, N+, and ?N+. AUTHOR: The author of a name is here strictly interpreted as the person responsible for a name’s being available. Authors of non-binomial, manuscript or nude names later made available are cited under attributed author. For misidentifications, ‘“auct. non x” is entered; for mis- spellings, just “auct.” The author responsible for the mis- identification or misspelling is entered under attributed author. Variant spellings are treated as misspellings unless there is strong evidence that they are intentional, in which case they are treated as emendations. The author of a justi- fied emendation is the original author, of an unjustified emendation the emending author. In the case of variant spellings in the original publication, the author’s intended spelling is used when this is obvious, with other spellings treated as misspellings. Constructions such as “x in y” are avoided by citing in the bibliographic database the block of text written by x within y’s work. Only in a few cases has this proved impos- sible, for example, Verrill and Smith in Verrill, where scat- tered descriptions in Verrill’s work are attributed to Verrill and Smith. DATE: The date of publication is the year in which a name was made available. The exact day and month of publica- tion, if known, and references thereto, are given in the bib- liographic database, where the stated date of publication is distinguished from the true date of publication. Mis- spellings and misidentifications do not have dates of publi- cation in this sense. ABC: This field distinguishes multiple publications by an author in a single year (e.g. Dall 1889a, 1889b). The species database links to the bibliographic database through the combination of the author, date, and abc fields. ATTRIBUTED AUTHOR: Names are often attributed to authors other than those responsible for their publication and availability. Authors of non-binomial, manuscript and nude names later made available are listed here. Other cases include ones such as names being attributed to Lamarck, 1822 by workers unaware that the names were first intro- duced in Lamarck, 1816. Authors responsible for mis- spellings and misidentifications are also entered here, not under author. ORIGINAL GENUS: The genus in which the name in associated name was placed when first published. Some authors, such as Dall (1889a), often used subgeneric names as if they were generic names, with the generic name placed in a heading. The original genus is taken to be the name that the author stated was the generic name, in accor- dance with ICZN rules. PARENTHESES (): If genus matches original genus, “‘n” (no) is placed in this field, indicating that no parentheses are needed around author and date when citing the species 260 AMER. MALAC. BULL. 10(2) (1993) Figure 1. Selected fields from the database, showing species of Neritopsidae, Phenacolepadidae, and Pleurotomariidae. Genus Species Subspecies Status Associated name Author Date 2: Neritopsis atlantica Vv atlantica Sarastia 1973 Neritopsis atlantica + finlayi Hoerle 1974 Phenacolepas hamillei Vv hamillei Fischer 1857 Phenacolepas hamillei + antillarum Dall 1889 Phenacolepas rushii Vv rushil Dall 1889 Entemnotrochus | adansonianus adansonianus Vv adansonianus Crosse & Fischer 1861 Entemnotrochus | adansonianus bermudensis Vv Ss bermudensis Okutani & Goto 1983 Perotrochus amabilis Vv amabilis Bayer 1963 Perotrochus atlanticus Vv atlanticus Rios & Matthews 1968 Perotrochus charlestonensis V charlestonensis Askew 1988 Perotrochus gemma Vv gemma Bayer 1965 Perotrochus lucaya Vv lucaya Bayer 1965 Perotrochus maureri Vv maureri Harasewych & Askew 1993 Perotrochus maureri + N amabilis auct. non Bayer 1963 Perotrochus midas V midas Bayer 1965 Perotrochus notialis Vv notialis Leme & Penna 1969 Perotrochus pyramus Vv pyramus Bayer 1967 Perotrochus quoyanus quoyanus Vv quoyanus Fischer & Bernardi 1856 Perotrochus quoyanus insularis VS insularis Okutani & Goto 1985 name. If they do not match, “‘y” (yes) indicates that paren- theses are needed. This comparison can be automated, except in cases where the original genus was misspelled. For example, if original genus is “Litorina” and genus is “Littorina,” “n’” is entered. Other common examples in- clude “Actaeon” vs “Acteon,” “Mangilia” vs “Mangelia,” and “Homalogyra” vs “Omalogyra.” COMBINING GENERA: This field lists genera (and sub- genera) with which associated name has been combined in the literature. COMBINATION: Citation of the original combination of names is in the format “Genus (Subgenus) {Section} species status infraspecific-names” along with any nota- tions of uncertainty, such as a genus with a query (?). For “status” the type of name (variety, form, etc.) for infrasub- specific names is indicated. Some authors, such as Dall (1889a) have used sectional in addition to generic and sub- generic names; these are placed in curly braces { }. CITATION: The pages on which a name is introduced. The entire page range is cited, not just the starting page, to assist interlibrary loan requests. Occurrences of names in indices or tables of contents are not cited unless alternate spellings occur there. FIGURE: The plate and figure numbers in the original publication. These are separated from citation to allow them to be sorted independently, to check, for example, whether all of the figures in a given publication have been cited. OTHER FIGURE: Illustrations of type specimens not fig- ured in the original publication are cited here. Such illustra- tions may have been referred to in the original publication (indications), or have been published subsequent to it. Photographs or otherwise improved illustrations of previ- ously figured types are also cited here. TYPE LOCALITY: It is difficult to eliminate a subjective element in the citation of type localities. Authors frequently give only part of the locality information in the original description, with the rest being contained in the introduc- tion of an article, or an appendix giving station numbers. Country is often omitted entirely by authors who consider it to be obvious. Citations of type localities are therefore paraphrased based on all information available in the origi- nal publication, with interpretative comments added in brackets as necessary. In many cases an author lists more than one locality, without explicitly stating a type locality, or mentioning which locality the holotype came from. If an author neglected to state a locality, “not stated’’ is entered; if an author did not know the locality, “unknown” is entered. Restrictions of type localities by subsequent authors are given in brackets at the end of the field. ALL LOCALITIES: Abbreviations were assigned to numerous localities at the level of state, province, and island throughout the Western Atlantic (e.g., Lab = ROSENBERG: DATABASE OF WESTERN ATLANTIC GASTROPODS 261 Fig. 1. (continued) () Original North South West East Shallow Deep Shallow Deep Size genus live live n Neritopsis 23 -20.5 82.5 29.3 0 20 16 n Neritopsis 23 23 82.8 81.3 0 0 17 y Acmaea 28 -28 87 0 3 8 y Scutellina 25 25: 82 0 0 8 y Umbraculum 26 9 80 55 55 10 y Pleurotomaria 26.48 13.06 78.67 59.62 107 482 107 366 146 n Entemnotrochus 32.3 32.3 64.8 64.8 366 366 56 y Mikadotrochus 27.73 23 93 80.86 128 411 219 219 80 n Perotrochus -24 -31 51 133 200 72 n Perotrochus 32.73 32.73 78.09 78.09 213 213 87 n Perotrochus 13.2 13.2 59.6 59.6 183 183 47 n Perotrochus 26.48 26.48 78.67 78.67 320 320 32 n Perotrochus 32.73 30.3 80 78.09 193 366 195 213 60 n 32.73 32.73 78.1 78.1 210 198 210 198 45 n Perotrochus 25.93 25.93 78.12 78.12 650 650 118 y Mikadotrochus -32 -32 51 150 150 74 n Perotrochus 16.29 16.29 61.16 61.16 600 600 48 y Pleurotomaria 21.79 12.55 86.4 59.65 128 549 134 350 57 n Perotrochus 32.3 32.3 64.8 64.8 366 366 54 Labrador; Ber = Bermuda; FVen = Falcoén, Venezuela). These allow species distributions to be documented more precisely than by north, south, east, and west described below. Abbreviations are summarized in a separate data- base table. OCEAN: Distributions of Western Atlantic species in other oceanic regions are tracked with the following abbrevia- tions: WA = Western Atlantic, EA = Eastern Atlantic, IO = Indian Ocean, WP = Western Pacific, EP = Eastern Pacific, AO = Arctic Ocean, SO = Southern Ocean. Data in north, south, and the four depth fields are taken only from Western Atlantic records. SHELL: Four abbreviations (s = shell, i = internal shell, v = vestigial shell, n = no shell) can be used, for example to look at the systematic distribution of shell reduction and loss, or to include only readily fossilizable taxa in compar- isons to the fossil record. NORTH: The farthest north in decimal degrees that the associated name has been reported in the Western Atlantic, from the mid-Atlantic ridge westward, including East Greenland, but excluding Iceland. Latitudes and longitudes are converted to decimal degrees, allowing this and the next three fields to be numer- ic rather than alphanumeric. Database programs allow mathematical operations on numeric fields but not on alphanumeric ones. Latitudes below the equator are entered as negative numbers, e.g. 23°25'S is entered as -23.42. SOUTH: The farthest south in decimal degrees that the associated name has been reported in the Western Atlantic, including the Antarctic Peninsula, Tierra del Fuego, and Ascension Island, but not St. Helena. EAST: The farthest east in decimal degrees that the associ- ated name has been reported in the Atlantic, limited to a minimum of zero. Longitudes east of Greenwich are not reported. WEST: The farthest west in decimal degrees that the asso- ciated name has been reported, to a maximum of 180° for circumtropical or circumpolar species. Longitudes west of 180° are not reported. SHALLOW: The shallowest depth in meters, with “0” indicating intertidal or beach-collected specimens. Note: in this and the other three depth fields, only “proven” depths are reported. If a species has been reported only in one dredge haul from 100-130 meters, “130” is entered in shal- low and “100” in deep, corresponding to the shallowest and deepest it has been proven to occur. Similarly, if it is known from two dredge hauls, one from 100-130 meters, the other from 170-190, “130” is entered in shallow and “170” in deep. DEEP: The deepest recorded depth in meters. Exact con- versions are given from feet and fathoms, with the caveat that this can imply accuracy not inherent in the original number (e.g. 100 fathoms = 183 meters). 262 AMER. MALAC. BULL. 10(2) (1993) LIVE SHALLOW: The shallowest depth in meters report- ed for live-collected specimens, with negative numbers indicating supratidal occurrence. LIVE DEEP: The deepest depth in meters reported for live-collected specimens. SIZE: The maximum size, in millimeters, for any dimen- sion. COMMENTS: Any comments necessary to explain or modify entries in other fields. Preoccupied, replacement, nude, and non-binomial names are noted here. If an author declared a name to be a nomen oblitum during the period when that provision of the ICZN was in force, that is noted here. REFERENCES: Sources of information in the other fields are documented here by citation of author and date fol- lowed by a series of codes: D = maximum depth; d = mini- mum depth; L = maximum live depth; | = minimum live depth; N = north; S = south; E = east; W = west; M = maxi- mum size; C = current classification; + = synonymy; V = valid; T = lectotype or neotype designation or restriction of type locality. Multiple references are separated by semi- colons. A typical entry in references might look like this: Dall (1889a) DLIW; Abbott (1974) dNM; Leal (1991) SE. Other fields beyond those noted here are possible, and can be added depending on the needs of a particular researcher. One could record protoconch size and whorl number; references to protoconch, radular and anatomical illustrations; substrate preference; feeding type; reproduc- tive mode; depositories and catalogue numbers of type specimens; and so on, ad infinitum. RESULTS AND DISCUSSION WESTERN ATLANTIC GASTROPOD DIVERSITY The database currently contains 8370 records for Western Atlantic marine gastropods from Greenland and Northern Canada through Antarctica. Statistics concerning these records are summarized in Table 2. Of these records, 3988 are for species currently recognized as valid. This means that the species has not been synonymized since it was named, or if it has been synonymized, some author has presented strong arguments that it should be taken out of synonymy. Currently recognized species somewhat outnumber the 3491 validly published synonyms. The synonymy ratio is 0.88:1, about half the 1.64:1 estimated by Clench (1959) Table 2. Composition of records in the Western Atlantic marine gastropod database. CURRENTLY RECOGNIZED SPECIES 3988 Tropical 3103 Caribbean 2164 Northern 413 Southern 472 NOMINA DUBIA 157 SYNONYMS 4189 validly published 3491 invalidly published 78 misidentifications 435 misspellings 185 EXTRALIMITAL 36* TOTAL RECORDS 8370 *includes one land snail erroneously described as marine. for the Western Atlantic. Boss (1971) has estimated syn- onymy ratios for mollusks overall to be in the range of 4:1. This ratio would predict 1496 valid species and 5983 syn- onyms among the 7479 names for Western Atlantic gas- tropods, and is clearly far too high for Western Atlantic gastropods. This may reflect that most Western Atlantic workers have introduced names for full species; the tradi- tion of naming varieties and forms is not as strong as it has been historically among European workers. Undoubtedly many species listed as valid will be synonymized once monographic work on particular families is done, but other species will be taken out of synonymy, and new species will continue to be discovered. Synonymy ratios for Western Atlantic gastropods are unlikely to significantly exceed 1:1. Of the 3988 currently recognized species, 3103 (78%) occur in tropical or semitropical areas, here defined as extending from Cape Hatteras to Rio de Janeiro, Brazil (south of 35°N to north of 24°S). Of these, 2164 occur in, but are not necessarily restricted to, the Caribbean region (south of 24°N, north of 8°N; west of 59°W, east of 88°W). About 413 species are restricted primarily to northern areas (north of 35°N) and 472 to southern areas (south of 24°S). The only estimate of the size of the entire Western Atlantic molluscan fauna appears to be that of Clench (1959), who predicted (without documentation) that it “would exceed 6,000 species and subspecies, but would not reach 8,000 species and subspecies.” Abbott (1974) lists 4491 species of marine gastropods in the Americas (Western Atlantic and Eastern Pacific) and 1918 species in other classes, giving a ratio of 2.34 gastropods species per non-gastropod. The total of 3988 marine gastropods in the ROSENBERG: DATABASE OF WESTERN ATLANTIC GASTROPODS 263 database therefore implies that there should be about 1700 non-gastropod mollusks in the Western Atlantic. Thus, known diversity of Western Atlantic mollusks is about 5700 species, substantiating Clench’s prediction that total diver- sity will exceed 6,000 species as knowledge of the fauna increases. The estimate of 3100 species of gastropods in the tropical Western Atlantic is considerably higher than previ- ously thought. The tropical Western Atlantic molluscan fauna is usually considered less diverse than that of the tropical Eastern Pacific. Keen (1971:2) stated “at the moment, preliminary lists suggest that the Pacific side, in spite of its narrow continental shelf, has more species.” However, Keen lists only 2438 species of gastropods named by 1971 in the tropical Eastern Pacific. Of the 3103 Western Atlantic species, 2641 had been named by 1971. Given the uncertainty in these numbers, the diversity levels of these faunas must be considered indistinguishable. Although there has never been a reliable estimate of the diversity of Western Atlantic gastropods, a number of authors have discussed the impoverishment of the fauna as compared to the Eastern Pacific (Olsson, 1961; Woodring, 1966; Vermeij, 1978, 1991; Stanley and Campbell, 1981; Stanley, 1986). Olsson (1961:2) stated, “As compared to its richness in the Miocene, the present-day Caribbean mol- lusks appear strangely modified and greatly impoverished; on the other hand, the Panamic-Pacific molluscan fauna has remained fundamentally unchanged.” Stanley and Camp- bell (1981) demonstrated that Pliocene faunas in the West- ern Atlantic were 70-80% extinct, whereas Pliocene faunas of California were only 30% extinct. They invoked the higher extinction rates in the Western Atlantic to explain the depauperate Recent fauna in the region. As demonstrated here, the gastropod fauna of the tropical Western Atlantic is not depauperate compared to that of the tropical Eastern Pacific. Previous workers have been attempting to explain a myth. It is true that the West- ern Atlantic fauna has undergone substantial extinction, but it is not possible to demonstrate that this has led to an over- all decline in diversity. Other processes, such as speciation and immigration must be balancing extinction (Allmon et al., 1993). It is possible that the Western Atlantic fauna is depauperate in shallow water as compared to that of the Eastern Pacific, with greater diversity in deeper water, where there is considerably more continental shelf area than in the Eastern Pacific. This possibility would be easily testable if data on Eastern Pacific mollusks were available in database form. WESTERN ATLANTIC GASTROPOD BIOGEOGRAPHY Comparisons of the faunas in different parts of the Western Atlantic have been complicated by the different names and combinations in use for a given species. Different emphases in regional sampling cause further problems. In one area workers have concentrated on micro- mollusks, in another the deep water fauna is virtually unknown. The database approach allows standardization of faunal lists by cross-referencing synonyms and generic placements. Collecting biases can be accounted for by excluding particular size, depth, or taxonomic ranges. Eight selected regions of the tropical Western Atlantic serve to demonstrate the value of databases for making faunal comparisons. These are eastern and western Florida, Yucatan, Panama, Jamaica, Puerto Rico, the Netherlands Antilles and northern Brazil. Table 3 summa- rizes the total number of shelled gastropod species in these Table 3. Number of marine gastropod species in selected tropical Western Atlantic faunas, with restrictions by size and bathymetry to standarize faunal comparisons. Region Total depth <50m size>5mm <50m+>5mm shelled # % # % # % Western Florida 709 590.83 577s Bl 468 .66 Eastern Florida 778 663.85 623.80 529 .68 Yucatan 594 528 89 490 82 431 73 Panama 444 423 95 353 —-.80 332 75 Jamaica 443 432 98 381 86 370 84 Puerto Rico 674 546 81 542 80 450 67 Neth. Antilles 678 662 98 491.72 480 7 Northern Brazil 709 593.84 589.83 492 .69 Table 4. Percentage of species in common among the tropical Western Atlantic faunas in Table 3. The percentage reflects the proportion of species in the smaller fauna not reported in the larger fauna. Numbers are calculated based on the species greater than 5S mm occurring in less than 50 meters (column 7 of Table 3). Numbers below and above the diagonal are identical; both sets are included for ease of use. REGION WFL EFL Yuc Pan Jam PR NA _ NBr Western Florida - 75 66 50 51 54 #47 ~~ 156 Eastern Florida 75 - 71 64 66 67 58 62 Yucatan 66 «71 - 67 69 67 4.63 ~~ .56 Panama 50 64 .67 - 65 72 72 ~~ ~«.60 Jamaica my | 66 69 65 - 79 77 61 Puerto Rico 54 67 67 72° 79 - 14 67 Neth. Antilles 47 58 63 72 77) 74 - 59 Northern Brazil 56 62 56 60 61 £67 ~~ 59 - faunas. Shell-less species have been excluded from the comparisons because of great variability in regional report- ing: more than 10% of the Brazilian fauna falls in the shell- less category, but only 3% of the reported Panamanian fauna. Table 3 also gives totals adjusted for depth (species occurring in less than 5Q meters) and size (species with maximum size greater than 5 mm). Table 4 shows for each 264 AMER. MALAC. BULL. 10(2) (1993) pair of faunas the percentage of species that the smaller fauna has in common with the larger one. This is the Simpson Index (Simpson, 1943), which is appropriate for large faunal lists corrected for sampling biases (Alroy 1992). Important sources of information and results con- cerning faunal composition for each of these areas are dis- cussed below. Western and Eastern Florida: Faunal lists for western and eastern Florida north of 25°N (i.e. excluding the Florida Keys) were compiled from Lyons (1989) and all papers on mollusks of Florida cited in the bibliography therein, Springer and Bullis (1956), Perry and Schwengel (1955), Maury (1922), and all papers on mollusks of Florida pub- lished in Nautilus in the last 70 years. Records from Dall (1927) are excluded because the Fernandina, Florida station is actually off Georgia. Uncertain records were checked by reference to the collections at ANSP. Three-quarters of all species from western Florida are also known from eastern Florida. Diversity is higher in eastern Florida than western Florida, probably because of a greater diversity of habitats in the former. Eastern Florida has more reefal, hard substrate areas than western Florida, which has mainly soft bottom. Reefal habitats in western Florida are restricted to patchy areas off-shore, such as the Florida Middle Grounds (Lyons, 1976; Turgeon and Lyons, 1977; Hopkins et al., 1977). Five of the seven faunal regions have lowest similarity with western Florida, reflect- ing the absence of shallow reefal habitat there that is avail- able in the other areas. Western Florida is most similar to eastern Florida and Yucatan, the two areas closest to it geo- graphically. Yucatan: The faunal list for Yucatan (the entire Yucatan Peninsula) is derived mainly from Vokes and Vokes (1983). Additional records, mostly for deep water mollusks, come from Dall (1881, 1889a). Yucatan shows greater faunal similarity to eastern Florida (71%) than to western Florida (66%), although it is geographically closer to western Florida. Yucatan shares with eastern Florida a strong shal- low-water reefal faunal component that is absent in western Florida. Panama: The faunal list for Panama comes mainly from Olsson and McGinty (1958) and Radwin (1969). The few available records deeper than 50 meters come from Petuch (1990). Jamaica: Records from Jamaica come primarily from Humfrey (1975), but species that he listed under “Other Jamaican Gastropods” on pages 198-205 were excluded unless confirmed by other sources. All of C.B. Adams’ Jamaican marine gastropods as documented by Clench and Turner (1950) are included, synonymized as appropriate. Puerto Rico: Puerto Rican records are taken mainly from Warmke and Abbott (1961), Ortiz-Corps (1983) and Dall and Simpson (1901). Many deep water records come from Watson (1886). Puerto Rico shows greatest faunal similari- ty to Jamaica (79%) and the Netherlands Antilles (74%), the two closest points among the regions compared here. Puerto Rico and the Netherlands Antilles are about equidistant from northern Brazil but the Brazilian fauna is more similar to that of Puerto Rico (67%) than to that of the Netherlands Antilles (59%). Puerto Rico and the Netherlands Antilles share 74% of their species. Puerto Rico shares 62 species with northern Brazil that it does not share with the Netherlands Antilles. There is no obvious pattern in taxonomy or habitat preference among these species, but it is possible that they tend to be deeper water species. The average minimum depth for species occurring in less than 5O meters depth is between 3 and 4 meters in both Puerto Rico and the Netherlands Antilles. The species shared by Puerto Rico and Brazil but not the Netherlands Antilles have an average minimum depth of 9.5 meters. Collecting in moderate depths (10 to 50 meters) in the Netherlands Antilles should increase the apparent similarity with the Brazilian fauna. Netherlands Antilles: The primary source of information is Jong and Coomans (1988). Virtually no information is available about deep-water mollusks of this region; only 3% of the gastropods known from the area are restricted to depths below 50 meters. On the other hand, the micromol- luscan fauna is extremely well known because of the work of Jong and Coomans and their colleagues. About one quar- ter of the gastropod species reported from the Netherlands Antilles do not exceed 5 mm at maturity, a higher percent- age than reported anywhere else in the Western Atlantic. Northern Brazil: Because of the enormous extent of the Brazilian coastline, comparisons were restricted to northern Brazil. Only those species whose geographic ranges lie within or cross the zone from 4°N to 6°S (Amapa to Rio Grande do Norte) were included. The two primary sources are Rios (1985) and Leal (1991). Rios (1975) gave more precise station data than in 1985; Rios (1970) provided maps showing the locations of many stations. Leal (1991) documented the faunas of Atol das Rocas and Fernando de Noronha, which are included in this zone. The average similarity of the Brazilian fauna to other tropical Western Atlantic faunas is 0.60, somewhat lower than the average similarities for the other areas, which range from 0.64 to 0.69 (except western Florida at 0.57). ROSENBERG: DATABASE OF WESTERN ATLANTIC GASTROPODS 265 This is consistent with its geographical remoteness from the Caribbean area. Western Atlantic: Of 28 pairwise comparisons between eight Western Atlantic faunas, only a single one yields a similarity lower than 50%: that between western Florida and the Netherlands Antilles. All other values are above 50%, as expected for regions within a single biogeographic province. Of the 3103 gastropod species occurring in the tropical Western Atlantic province, 2497 (80%) are restrict- ed to that region. Coomans (1962) defines a biogeographic province (which he calls an “autonomous zoogeographical province’’) as having at least 50% endemic species. There seems to be little basis for recognizing biogeo- graphic subprovinces within the tropical Western Atlantic, because similarities between regional faunas are deter- mined as much by habitat availability as by geographic proximity. As shown in Table 5, no local region of the Western Atlantic has more than about 4% endemics, excluding species named since 1980 as these are likely to be discovered in other areas once attention has been called to their existence. Such low levels of endemicity are insuffi- cient to make faunal province subdivisions. Coomans (1962) lumped the Virginian area (Cape Hatteras to Cape Cod) into the Boreal Province because it had only 10.5% endemic species. Tropical Western Atlantic endemics are not concen- trated in one area, although they appear more common on continental margins than on islands. Petuch (1990) named the Blasian faunal subregion for the Caribbean coast of Panama and Costa Rica, but Panama (including Costa Rica) has only about 4% endemic species (Table 5). Jong and Coomans (1988) named a number of species from the Netherlands Antilles increasing apparent endemicity, but many of these have been identified in samples from the Bahamas at ANSP (J. Worsfold, pers. comm.). Thus, even as our knowledge of regional Western Atlantic faunas increases, it is unlikely that the percentage of narrow rang- Table 5. Number and percentage of endemic species in the faunas in Table 3. The second column repeats the totals from the seventh column of that table. Region <50m, Endemic >5mm # % Western Florida 468 17 3.6 Eastern Florida $29 9 1.7 Yucatan 431 10 2.3 Panama 332 14 4.2 Jamaica 370 2 0.5 Puerto Rico 450 4 0.9 Neth. Antilles 480 14 2.9 Northern Brazil 492 14 2.8 ing endemics will increase. Diversity in the tropical Western Atlantic is, however, higher than has been com- monly perceived, and many local faunas have been badly under-sampled. In 1901, Dall and Simpson estimated “the average American marine tropical shell-fauna” to contain about 600 species. Five of the eight faunas discussed here exceed 600 species of gastropods alone. A hint of the diversity yet to be discovered comes from the Worsfold collection at ANSP, which documents about 1550 species of Bahamian mol- lusks. Because most species in the tropical Western Atlantic are widespread, as indicated by the high similarities of the eight widely separated faunas studied here, diversity in the Bahamas should be considered typical of the faunal province as a whole. Most local faunas in the tropical Western Atlantic will eventually be demonstrated to have in excess of 1500 species of Recent marine mollusks. ACKNOWLEDGMENTS I thank Robert Robertson (ANSP) for valuable discussions con- cerning Western Atlantic mollusks and for allowing me to use his collec- tion of malacological citations and literature. William G. Lyons and James F. Quinn, Jr. (both Florida Department of Natural Resources) helped me to cross-check records in the database against a manuscript revision of marine gastropods for a second edition of Turgeon et al. (1988), and Jim gave me a manuscript list of Western Atlantic trochids. John K. Tucker (Illinois Natural History Survey) made available his manuscript catalogue of American turrids. Philippe Bouchet (Museum national d’ Histoire naturelle, Paris) brought to my attention recent changes in the classifica- tion of a number of species that I had overlooked. Harry G. Lee (Jacksonville, Florida) made available his manuscript list of mollusks of northeastern Florida. Emily E. Vokes (Tulane University), Riidiger Bieler (Field Museum of Natural History), Kenneth J. Boss (Museum of Comparative Zoology) and David G. Reid (British Museum of Natural History) advised me on the identities of obscure taxa. LITERATURE CITED Abbott, R. T. 1974. American Seashells, 2nd ed. Van Nostrand Reinhold: New York. [viii] + 663 pp., 24 pls. Allkin, R. and F. A. Bisby, eds. 1984. Databases in systematics. The Systematics Association Special Volume no. 26. Academic Press: London. xiii + 329 pp. Allmon, W. D., G. Rosenberg, R. Portell and K. Schindler. 1993. Diversity of Atlantic Coastal Plain mollusks since the Pliocene. Science 260:1626-1629. Alroy, J. 1992. Conjunction among taxonomic distributions and the Miocene mammalian biochronology of the Great Plains. Paleobiology 18:326-343. Boss, K. J. 1971. Critical estimate of the number of Recent Mollusca. Occasional Papers on Mollusks 3(40):81-135. Clench, W. J. 1959. A partial analysis of the molluscan fauna of the Western Atlantic. Johnsonia 3:viii. Clench, W. J. and R. D. Turner. 1950. The Western Atlantic marine Mollusca described by C. B. Adams, Occasional Papers on Mollusks 1(15):233-403. 266 AMER. MALAC. BULL. 10(2) (1993) Coomans, H. E. 1962. The marine mollusk fauna of the Virginian area as a basis for defining zoogeographical provinces. Beaufortia 9(98):83- 104. Dall, W. H. 1881. Reports on the results of dredging, under the supervi- sion of Alexander Agassiz, in the Gulf of Mexico, and in the Caribbean Sea, 1877-79, by the United States Coast Survey Steamer “Blake”. Bulletin of the Museum of Comparative Zoology 9:33-144. Dall, W. H. 1889a. Reports on the results of dredgings, under the supervi- sion of Alexander Agassiz, in the Gulf of Mexico (1877-78) and in the Caribbean Sea (1879-80), by the United States Coast Survey Steamer “Blake”. Bulletin of the Museum of Comparative Zoology 18:1-492, pls. 10-40. Dall, W. H. 1889b. A preliminary catalogue of the shell-bearing marine mollusks and brachiopods of the south-eastern coast of the United States, with illustrations of many of the species. Bulletin of the United States National Museum 37:1-221, 74 pls. Dall, W. H. 1927. Small shells from dredgings off the southeast coast of the United States by the United States Fisheries Steamer “Albatross” in 1885 and 1886. Proceedings of the United States National Museum 70(18):1-134. Dall, W. H. and C. T. Simpson. 1901. The Mollusca of Porto Rico. United States Fisheries Commission Bulletin for 1900:351-524, pls. 53-58. Hopkins, T. S., D. R. Blizzard, and D. K. Gilbert. 1977. The molluscan fauna of the Florida Middle Grounds with comments on it’s [sic] zoogeographical affinities. Northeast Gulf Science 1:39-47. Humfrey, M. 1975. Sea Shells of the West Indies. Collins: London. 351 pp., 32 pls. Johnson, C. W. 1934. List of marine Mollusca of the Atlantic coast from Labrador to Texas. Proceedings of the Boston Society of Natural History 40:1-204. Jong, K. M. de, and H. E. Coomans. 1988. Marine gastropods from Curagao, Aruba and Bonaire. Studies on the Fauna of Curacao and other Caribbean islands 69:1-261. [Also published as a book by Brill: Leiden.] Keen, A. M. 1971. Seashells of tropical West America, 2nd ed. Stanford University Press: Stanford. xv + 1064 pp., 22 pls. Leal, J. H. 1991. Marine Prosobranch Gastropods from Oceanic Islands off Brazil. Backhuys/U.B.S.: Oegstgeest, The Netherlands. x + 419 pp. Lyons, W. G. 1976. Distribution of Cerithium litteratum (Born) (Gastropoda: Cerithiidae) off western Florida. Veliger 18:375-377, 1 pl. Lyons, W. G. 1989. Nearshore marine ecology at Hutchinson Island, Florida: 1971-1974 XI. Mollusks. Florida Marine Research Publications 47:1-131. Maury, C. J. 1922. Recent Mollusca of the Gulf of Mexico and Pleistocene and Pliocene species from the Gulf States. Part 2. Scaphopoda, Gastropoda, Amphineura, Cephalopoda. Bulletins of American Paleontology 9(38):1-142. [Reprinted in 1971 by Paleontological Research Institution, Ithaca, New York.]} Olsson, A. A. 1961. Mollusks of the Tropical Eastern Pacific, Particularly from the Southern Half of the Panamic-Pacific Province (Panama to Peru): Panamic-Pacific Pelecypoda. Paleontological Research Institution, Ithaca, New York. 574 pp., 86 pls. Olsson, A. A. and T. L. McGinty. 1958. Recent marine mollusks from the Caribbean coast of Panama with the description of some new gen- era and species. Bulletins of American Paleontology 39(177):1-58, pls. 1-5. Ortiz-Corps, E. [1985]. An Annotated Checklist of the Recent Marine Gastropoda (Mollusca) from Puerto Rico. Memorias del Quinto Simposio de la Fauna de Puerto Rico y el Caribe. 11 + 220 pp. Perry, L. M. and J. S. Schwengel. 1955. Marine Shells of the Western coast of Florida. Paleontological Research Institution, Ithaca, New York. 318 pp., frontispiece, 55 pls. Petuch, E. J. 1990. A new molluscan faunule from the Caribbean coast of Panama. Nautilus 104:57-70. Piani, P. 1980. Catalogo dei molluschi conchiferi viventi nel Mediterraneo. Bollettino Malacologico 16:113-224. Ponder, W. F. and A. Warén. 1988. Classification of the Caenogastropoda and Heterostropha—a list of the family-group names and higher taxa. Malacological Review, Supplement 4:288-328. Radwin, G. E. 1969. A Recent Molluscan fauna from the Caribbean coast of Panama. Transactions of the San Diego Society of Natural History 15:229-236. Rios, E. C. 1970. Coastal Brazilian Seashells. Museu Oceanografico, Rio Grande. 255 + [1] pp., 4 maps, 60 pls. Rios, E. C. 1975. Brazilian Marine Mollusks Iconography. Museu OceanogrAfico: Rio Grande. 331 pp., 91 pls. Rios, E. C. 1985. Seashells of Brazil. Museu Oceanogréfico: Rio Grande. [ii] + 329 pp., 102 pls. Rosenberg, G. 1992. The Encyclopedia of Seashells. Dorset Press: New York. 224 pp. Sabelli, B., R. Giannuzzi-Savelli and D. Bedulli. 1990-1992. Annotated check-list of Mediterranean marine mollusks. Societa Italiana di Malacologia: Bologna. Vol. 1 (1990), pp. i-xiv, 1-348; vol. 2 (1992), pp. 349-498 [dual pagination in Italian and English]; vol. 3 (1992), pp. 501-781. Simpson, G. G. 1943. Mammals and the nature of continents. American Journal of Science 241:1-31. Springer, S. and H. R. Bullis. 1956. Collections by the Oregon in the Gulf of Mexico. Special Scientific Report—Fisheries 196, ii + 134 pp. Stanley, S. M. 1986. Anatomy of a regional mass extinction: Plio- Pleistocene decimation of the western Atlantic bivalve fauna. Palaios 1:17-36. Stanley, S. M. and L. D. Campbell. 1981. Neogene mass extinction of western Atlantic molluscs. Nature 293:457-459. Turgeon, D. D. and W. G. Lyons. 1977. A tropical marine molluscan assemblage in the northeastern Gulf of Mexico. Bulletin of American Malacological Union for 1977:88-89. 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 and J. D. Williams. 1988. Common and scientific names of aquatic inver- tebrates from the United States and Canada: Mollusks. American Fisheries Society Special Publication 16, vii + 277 pp., [12] pls. Vaught, K. C. 1989. A Classification of the Living Mollusca. American Malacologists: Melbourne, Florida. xii + 189 pp. Vermeij, G. J. 1978. Biogeography and Adaptation. Harvard University, Cambridge, Massachusetts. 332 pp. Vermeij, G. J. 1991. When biotas meet: understanding biotic interchange. Science 253:1099-1104. Vokes, H. E. and E. H. Vokes. 1983. Distribution of shallow-water marine Mollusca, Yucatan Peninsula, Mexico. Middle American Research Institute Publication 54, viii + 183. [Also Mesoamerican Ecology Institute Monograph 1.] Warmke, G. L. and R. T. Abbott. 1961. Caribbean Seashells. Livingston: Narberth, Pennsylvania. x + 346 pp., 44 pls. Watson, R. B. 1886. Report on the Scaphopoda and Gasteropoda collected by H. M. S. Challenger during the years 1873-1876. Report on the Scientific Results of the Voyage of H. M. S. Challenger 15(2), v + 680 pp., 50 pls. Woodring, W. P. 1966. The Panama land bridge as a sea barrier. Proceedings of the American Philosophical Society 110:425-433. Date of manuscript acceptance: 28 April 1993 A bibliography of Caribbean malacology 1826 - 1993 Paula M. Mikkelsen!, Riidiger Bieler2, and Richard E. Petit 1Harbor Branch Oceanographic Institution, 5600 U.S. 1 North, Ft. Pierce, Florida 34946, and Dept. of Biological Sciences, Florida Institute of Technology, Melbourne, Florida 32901, U.S.A. 2Center for Evolutionary and Environmental Biology, Field Museum of Natural History, Roosevelt Road at Lake Shore Drive, Chicago, Illinois 60605, U.S.A. 3North Myrtle Beach, South Carolina, U.S.A. Abstract. A bibliography of over 800 publications lists major works on marine, land and freshwater mollusks, both Recent and fossil, occurring in the Caribbean Sea, its islands, and the bordering regions of Central and South America. Emphasis is placed on papers dealing with zoogeography and/or taxonomy, intended as an introduction to the molluscan fauna of this geographic area. A geographic index is provided. This bibliography covers over 800 malacological pub- lications for the Caribbean. It is a selected listing, admitted- ly incomplete, but intended to provide access to the main body of literature in this field. Geographic coverage includes the Caribbean Sea and its islands, as well as the bordering regions of Central and South America. Western and eastern borders are defined by the eastern edge of the Yucatan Peninsula and the island of Barbados. The northern coasts of Cuba and Hispaniola are also included. Florida and the Bahamas are not included. The coverage extends to marine, land and freshwater mol- lusks, both Recent and fossil. The inclusion of some parts of Central and South American land masses, and the treat- ment of fossils from several periods, make this a very gen- eral concept of “Caribbean.” Emphasis is placed on malacological papers dealing with zoogeography and/or taxonomy relevant to this geo- graphic area. Taxonomic treatments were generally includ- ed only if they covered taxa at the generic level or above, i.e. papers dealing with only one or a few species were not included. Exceptions to this rule were small parts of a large series of papers by a single author or authors. Large monographs including the Caribbean but covering a much wider area (e.g. world-wide treatises) normally were not included. Abstracts and theses were listed only if they pro- vide useful information which has not been formally pub- lished. In all cases, we were more lenient with older litera- ture, papers dealing with areas otherwise not covered, and “obscure” works. Entries are organized alphabetically by authors. A geographic index is also provided. Geographic areas are listed in brackets following the reference only if not apparent by the title of the work. In the interest of space, we have not included the con- tents of the journal Johnsonia (Monographs of the Marine Mollusks of the Western Atlantic), published in 50 parts between 1941 and 1974 by the Department of Mollusks, Museum of Comparative Zoology, Harvard University. Likewise, only selected works have been included from the numerous malacological publications produced in Panama, Venezuela, and Cuba (see Bieler and Kabat, 1991), e.g. the Revista de la Sociedad Malacélogia “Carlos de la Torre” (see published index to the last, Jacobson, 1971). These essential references should be consulted for additional information on the mollusks of this region. The abstract volume for the Primer Congreso Latinoamericano de Malacologia, 15-19 July 1991, Universidad Sim6n Bolivar, Caracas, Venezuela, also provides many useful references. THE BIBLIOGRAPHY Abbott, R. T. 1957. The tropical western Atlantic Province. Proceedings of the Philadelphia Shell Club 1(2):7-11. Abbott, R. T. 1958. The marine mollusks of Grand Cayman Island, British West Indies. Monographs of the Academy of Natural Sciences of Philadelphia 11:1-138, pls.1-5. Abbott, R. T. 1984. Collectible Shells of Southeastern U.S., Bahamas & Caribbean. American Malacologists, Melbourne, Florida, 64 pp. [Also issued under the title Collectible Florida Shells, 1984] Adam, W. 1937. Zoologische Ergebnisse einer Reise nach Bonaire, Curagao und Aruba im Jahr 1930. No. 24. Céphalopodes de iles Bonaire et Curacao (avec une révision du genre Sepioteuthis de la céte Américaine). Capita Zoologica 8(3):1-29. [Bonaire, Curagao, general] Adam, W. 1957. Notes sur les cephalopodes. XXIII. Quelques espéces des Antilles. Bulletin de l'Institut Royal des Sciences Naturelles de Belgique 33(7):1-10, 1 pl. Adams, C. B. 1845. Specierum novarum conchyliorum, in Jamaica reper- torum, synopsis. Proceedings of the Boston Society of Natural History 2:1-17. Adams, C. B. 1846a. [Descriptions of undescribed species of shells from American Malacological Bulletin, Vol. 10(2) (1993):267-290 267 268 AMER. MALAC. BULL. 10(2) (1993) the island of Jamaica.] Proceedings of the Boston Society of Natural History 2:102-103. Adams, C. B. 1846b. [On the Mollusca of the island of Jamaica with remarks on their geographical distribution and habits...] Pro- ceedings of the Boston Society of Natural History 2:132-135. Adams, C. B. 1849a. Catalogue of Land Shells which Inhabit Jamaica; Catalogue of Fresh Water Shells, which Inhabit Jamaica. Amherst, Massachusetts, 4 pp. Adams, C. B. 1849b. Descriptions of forty-four supposed new species and varieties of operculated land shells from Jamaica. Contributions to Conchology 1(1):1-14. Adams, C. B. 1849c. Catalogue of operculated land shells which inhabit Jamaica. Contributions to Conchology 1(1):15-16. Adams, C. B. 1849d. Descriptions of supposed new species and varieties of Helicidae from Jamaica. Contributions to Conchology 1(2):17- 32; 1(3):33-38. Adams, C. B. 1849e. Catalogue of species and varieties of Helicidae which inhabit Jamaica. Contributions to Conchology 1(3):39-41. Adams, C. B. 1849f. Catalogue of Auriculidae which inhabit Jamaica. Contributions to Conchology 1(3):42. Adams, C. B. 1849g. Catalogue freshwater shells which inhabit Jamaica. Contributions to Conchology 1(3):45. Adams, C. B. 1849h. Remarks on the distribution of the terrestrial and fresh-water Mollusca which inhabit Jamaica. Contributions to Conchology 1(3):45-48; 1(4):49-50. Adams, C. B. 1850a. Descriptions of supposed new species of marine shells which inhabit Jamaica. Contributions to Conchology 1(4):56-68; 1(5):69-75. Adams, C. B. 1850b. Descriptions of supposed new species and varieties of terrestrial shells, which inhabit Jamaica. Contributions to Conchology 1(5):76-84; 1(6):90-98; 1(7):101-108. Adams, C. B. 1850c. Remarks on the origin of terrestrial mollusks of Jamaica. Contributions to Conchology 6:85-87. Adams, C. B. 1850d. Descriptions of supposed new species of marine shells, which inhabit Jamaica. Contributions to Conchology 1(7):109-123. Adams, C. B. 1850e. Descriptions of new species and varieties of shells, which inhabit Jamaica. Contributions to Conchology 1(8):129-140. {also Annals of the Lyceum of Natural History of New York 5(1852):45-67, 1851] Adams, C. B. 1851a. Description of new species and varieties of the land shells of Jamaica, with notes on some previously described species. Contributions to Conchology 1(9):153-174. [also Annals of the Lyceum of Natural History of New York 5(1852):77-102, 1851] Adams, C. B. 1851b. Catalogue of the land shells which inhabit Jamaica. Contributions to Conchology 1(9):179-186. [also Annals of the Lyceum of Natural History of New York 5(1852):103-110, 1851] Adams, C. B. 1851c. Catalogue of the fresh water shells which inhabit Jamaica. Contributions to Conchology 1(9):187. [also Annals of the Lyceum of Natural History of New York 5(1852):111, 1851] Adams, C. B. 1851d. On the nature and origin of the species of the terres- trial Mollusca in the island of Jamaica. Contributions to Conch- ology 1(10):189-194. Adams, C. B. 1852a. Catalogue of shells collected at Panama, with notes on synonymy, station, and habitat. Annals of the Lyceum of Natural History of New York 5:229-344 (June); 345-548 (July). [Reprinted as a single volume in the same year by R. Craighead Printers, New York, viii + 334 pp., with slight change in title and the addition of a title page, preface, and list of references.] Adams, C. B. 1852b. Catalogue of species of Lucina, which inhabit the West Indian seas. Contributions to Conchology 1(12):242-247. [Jamaica] Aguayo, C. G. 1932. Notes and descriptions of Cuban mollusks. Occasional Papers of the Boston Society of Natural History 8:31- 36, pl. 3. Aguayo, C. G. 1934. Mollusca Cubana. Addenda et corrigenda. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 8(2): 87-96. Aguayo, C. G. 1935. Espicilegio de moluscos Cubanos. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 9(2):107-128, pl. 9. Aguayo, C. G. 1938a. Los moluscos fluviatiles Cubanos. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 12(3):203- 242 (Parte I, Generalidades); 12(4):253-276, pl.18 (Parte II, Sistematica). Aguayo, C. G. 1938b. Un molusco terrestre africano de reciente introduc- cion en Cuba. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 12(5):367-373. [Includes review of non- marine mollusks introduced to Cuba.] Aguayo, C. G. 1938c. Moluscos Pleistocénicos de Guanténamo, Cuba. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 12(2):97-118. Aguayo, C. G. 1943a. Centenario de los “Moluscos” de d’Orbigny en la obra de la Sagra. Revista de la Sociedad Malacélogica “Carlos de la Torre” 1(1):37-40, 1 pl. [Cuba] Aguayo, C. G. 1943b. Nuevos operculados de Cuba oriental. Revista de la Sociedad Malacologica “Carlos de la Torre” 1(2):69-80, pls. 10-11. Aguayo, C. G. 1944a. Leptinaria lamellata y otros moluscos introducidos en Cuba. Revista de la Sociedad Malacélogica “Carlos de la Torre” 2(2):51-58. Aguayo, C. G. 1944b. Los moluscos comestibles de Cuba. Revista de la Sociedad Malacologica “Carlos de la Torre” 2(1):17-20. Aguayo, C. G. 1944c. Nuevos operculados de la regién oriental de Cuba. Revista de la Sociedad Malacélogica “Carlos de la Torre” 2(1): 1-6, pl. 1. Aguayo, C. G. 1944d. Posibilidades de investigaciédn malacolégica en Cuba. Revista de la Sociedad Malacoélogica “Carlos de la Torre” 2(1):31-33. Aguayo, C. G. 1948. Moluscos fésiles de la Provincia de Oriente, Cuba. Revista de la Sociedad Malacélogica “Carlos de la Torre” 6(2):55- 63. Aguayo, C. G. 1949a. Nuevos moluscos fdésiles de Cuba y Panama. Revista de la Sociedad Malacélogica “Carlos de la Torre” 7(1): 11-14. Aguayo, C. G. 1949b. Nuevos moluscos fésiles de la Reptiblica Dominicana. Revista de la Sociedad Malacélogica “Carlos de la Torre” 6:91-92. Aguayo, C. G. 1953. Algunos nuevos moluscos terrestres de Cuba orien- tal. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 21(3):299-310, pls. 33-35. Aguayo, C. G. 1959. Los origenes de la fauna Cubana. Anais da Academia Ciencias Méd. Fisicas y Naturales, Habana 88:1-23. [not seen] Aguayo, C. G. 196la. Aspecto general de la fauna malacologica Puert- oriquefia. Caribbean Journal of Science 1(3):89-105. [land snails} Aguayo, C. G. 1961b-1963. Notas sobre moluscos terrestres Antillanos. Caribbean Journal of Science 1(4):143 (1961, [part I]); 2(1):9-12 (1962, part II); 2(3):108-112 (1963, part III); 3(1):69-71 [1963, part IV, La familia Sphaeridae (Mollusca: Pelecypoda)]. [Cuba, Puerto Rico] Aguayo, C. G. 1966. Una lista de los moluscos terrestres y fluviales de Puerto Rico. Stahlia, Miscellaneous Papers of the Museum of Biology, University of Puerto Rico, Rio Piedras 5:1-17. Aguayo, C. G. and P. Borro. 1946a. Algunos moluscos Terciarios de MIKKELSEN ET AL.: BIBLIOGRAPHY OF CARIBBEAN MALACOLOGY 269 Cuba. Revista de la Sociedad Malacologica “Carlos de la Torre” 4(2):43-49, pl. 3. Aguayo, C. G. and P. Borro. 1946b. Nuevos moluscos del Terciario supe- rior de Cuba. Revista de la Sociedad Malacélogica “Carlos de la Torre” 4(1):9-12, pl. 1. Aguayo, C. G. and A. de la Torre. 1951. Nuevos ceriénidos de la costa norte de Matanzas. Revista de la Sociedad Malacélogica “Carlos de la Torre” 8:19-22. Aguayo, C. G. and A. de la Torre. 1954. Nuevo helicinido de la Provincia de Matanzas. Revista de la Sociedad Malacélogica “Carlos de la Torre” 9:67-68. Aguayo, C. G. and M. L. Jaume. 1934. Notas y adiciones a la fauna mala- colégica habanera. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 8(1):9-14. Aguayo, C. G. and M. L. Jaume. 1936. Sobre algunos moluscos marinos de Cuba. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 10(2):115-122. Aguayo, C. G. and M. L. Jaume. 1939. Moluscos semifésiles del Bosque de la Habana. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 13(4):229-245. Aguayo, C. G. and M. L. Jaume. 1945. Novedades malacologicas cubanas. Revista de la Sociedad Malacélogica “Carlos de la Torre” 3:95-98, pl. 9. Aguayo, C. G. and M. L. Jaume. 1947a. Nuevos gasterépodos de Cuba. Revista de la Sociedad Malacoélogica “Carlos de la Torre” 5(2): 53-58. Aguayo, C. G. and M. L. Jaume. 1947b-1951a. Catdlogo de los Moluscos de Cuba. Havana. Nos. 1-725 (mimeographed loose leaves, num- bered in accordance with order of publication). Aguayo, C. G. and M. L. Jaume. 1951b. Nuevos ceriénidos de Cuba. Revista de la Sociedad Malacoélogica “Carlos de la Torre” 8(1):1- 18, pls. 1-2. Aguayo, C. G. and M. L. Jaume. 1952. Nuevos moluscos operculados de Matanzas, Cuba. Revista de la Sociedad Malacoélogica “Carlos de la Torre” 8(3):127-128. Aguayo, C. G. and M. L. Jaume. 1953. Moluscos terrestres de la region de Baracoa, Habana. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 21(3):267-280, pl. 31. Aguayo, C. G. and M. L. Jaume. 1954. Descripcién de nuevos [sic] especies de moluscos terrestres cubanos. Revista de la Sociedad Malacélogica “Carlos de la Torre” 9(2):47-66, pls. 6-7. Aguayo, C. G. and M. L. Jaume. 1957-1958. Adiciones a la fauna mala- coldgica cubana. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 23(2):117-142, pls.1-6 (1957, part I); 24(1):91-104, 1 pl. (1958, part ID). Alayo Dalmau, P. 1960. Lista de los Moluscos Litorales de Cuba. Pt. 1. Prosobranchia. Universidad de Oriente, Cuba, 33 pp., 9 pls. [not seen] Alcalde Ledén, O. 1945a-1948. Estudio y revision de los moluscos cubanos del genero Farcimen. Revista de la Sociedad Malacélogica “Carlos de la Torre” 3(1):5-18, pl.1 (1945, [part I]); 3(2):39-50 (1945, part II); 3(3):85-93 (1945, part III); 4(2):37-40 (1946, part IV); 5(1):1-5 (1947, continuation [part V]); 5(3):85-90 (1947, con- tinuation [part VI]); 6(1):1-3 (1948, conclusion [part VII]). Alcalde Led6én, O. 1945b. El genero Farcimoides en Cuba. Revista de la Sociedad Malacélogica “Carlos de la Torre” 3(2):37-38, pls.5-6. Almeida Pérez, P. 1974. Distribucién de los moluscos en la costa centro- occidental (Patanemo-Punta Tucacas) de Venezuela. Comparacién de los habitats litorales. Memorias de la Sociedad de Ciencias Naturales La Salle 34(97):25-52. [not seen] Ancey, C. F. 1886. Une excursion malacologique sur le versant atlantique du Honduras. Annales de Malacologie 2:237-260. Anderson, F. M. 1927. The marine Miocene deposits of north Colombia. Proceedings of the California Academy of Sciences 4th series, 16(3):87-95, pls. 2-3. Anderson, F. M. 1928. Notes on lower Tertiary deposits of Colombia and their molluscan and foraminiferal fauna. Proceedings of the California Academy of Sciences 4th series, 17(1):1-29, pl. 1. Anderson, F. M. 1929. Marine Miocene and related deposits of north Colombia. Proceedings of the California Academy of Sciences 4th series, 18(4):73-213, pls. 8-23. Arango y Molina, R. 1878-1880. Contribucion a la Fauna Malacolégica Cubana. G. Montiel y Comp., Havana, 280 text pp. + 35 index + errata pp. Armow, L., F. St. Clair, and T. Arnow. 1963. The Mollusca of a lagoonal area at Playa de Vega Baja, Puerto Rico. Caribbean Journal of Science 3(2-3):163-172. Arocha, F., L. Marcano, and R. Cipriani. 1991. Cephalopods trawled from Venezuelan waters by the R/V Dr. Fridtjof Nansen in 1988. Bulletin of Marine Science 49(1-2):231-234. Arocha, F. and L. J. Urosa. 1982. Cefalépodos del género Octopus en el area nororiental de Venezuela. Boletin del Instituto de Oceanografico, Universidad de Oriente 21(1-2):167-189. Baboolal, S., S. Johnatty, and Z. Ali. 1981. Studies on the Trinidad chi- tons. Living World, Journal of the Trinidad Field Naturalists Club for 1981:39-45. [not seen] Baker, F. C. 1891. Notes on a collection of shells from southern Mexico. Proceedings of the Academy of Natural Sciences of Philadelphia 43:45-55. [Yucatan] Baker, H.B. 1923-1930. The Mollusca collected by the University of Michigan-Williamson Expedition to Venezuela. Occasional Papers of the Museum of Zoology, University of Michigan 137:59 pp., pls. 1-5 (1923, parts I-II); 156:57 pp., pls. 6-11 (1925, part II); 167:49 pp., pls. 12-19 (1926, part IV); 182:36 pp., pls. 20-26 (1927, part V); 210:96 pp., pls. 27-33 (1930, part VI). Baker, H. B. 1924. Land and freshwater molluscs of the Dutch Leeward Islands. Occasional Papers of the Museum of Zoology, University of Michigan 152:1-158, pls.1-21. [Netherlands Antilles] Baker, H. B. 1934-1935. Jamaican land snails. Nautilus 48(1):6-14, pl. 2 (1934, part 1); 48(2):60-67, pl. 2 (1934, part 2); 48(3):83-88, pl. 3 (1935, part 3); 48(4):135-139, pls. 8-9 (1935, part 4); 49(1):21-27, pl. 2 (1935, part 5); 49(2):52-58 (1935, part 6). Baker, H. B. 1940. New subgenera of Antillean Helicinidae. Nautilus 54(2):70-71. [Puerto Rico] Baker, H. B. 1941la. New Puerto Rican land snails. Notulae Naturae 88:1-6. Baker, H. B. 1941b. Puerto Rican Oleacininae. Nautilus 55(1):24-30. Baker, H. B. 1943. Some Antillean helicids. Nautilus 56(3):81-91, pls. 9-11. Baker, H. B. 1961a. Puerto Rican Xanthonychidae. Nautilus 74(4): 142-149. Baker, H. B. 1961b. Puerto Rican pupillids and clausilioids. Nautilus 75(1):33-36. Baker, H. B. 1962a. Puerto Rican Camaenidae. Nautilus 75(2):64-67. Baker, H. B. 1962b. Puerto Rican holopodopes. Nautilus 75(3):116-121. Baker, H. B. 1962c. Puerto Rican land operculates. Nautilus 76(1):16-22. Baker, H. B. 1962d. Puerto Rican oleacinoids. Nautilus 75(4):142-145. Bakus, G. J. 1968. Zonation in marine gastropods of Costa Rica and species diversity. Veliger 10(3):207-211. Bandel, K. 1974a. Faecal pellets of Amphineura and Prosobranchia (Mollusca) from the Caribbean coast of Colombia, South America. Senckenbergiana Maritima 6(1):1-31. Bandel, K. 1974b. Spawning and development of some Columbellidae from the Caribbean Sea of Colombia (South America). Veliger 270 AMER. MALAC. BULL. 10(2) (1993) 16(3):271-282. Bandel, K. 1975a. Embryonalgehduse karibischer Meso- und Neo- gastropoden (Mollusca). Abhandlungen der Mathematisch- Naturwissenschaftlichen Klasse, Akademie der Wissenschaften und der Literatur, Mainz 1975(1):1-175, 21 pls. [Colombia, general] Bandel, K. 1976a. Egg masses of 27 Caribbean opisthobranchs from Santa Marta, Columbia. Studies on Neotropical Fauna and Environment 11:87-118. [Colombia] Bandel, K. 1976b. Die Gelege karibischer Vertreter aus den Uberfamilien Strombacea, Naticacea und Tonnacea (Mesogastropoda) sowie Beobachtungen im Meer und Aquarium. Mitteilungen aus dem Institut Colombo-Alemdn Invest. Cient., Santa Marta 8:105-139. Bandel, K. 1976c. Morphologie der Gelege und 6kologische Beobach- tungen an Buccinaceen (Gastropoda) aus der siidlichen Karibischen See. Bonner Zoologische Beitrdge 27(1-2):98-133. [Colombia] Bandel, K. 1976d. Morphologie der Gelege und 6kologische Beobach- tungen an Muriciden (Gastropoda) aus der siidlichen Karibischen See. Verhhandlungen der Naturforschenden Gesellschaft in Basel 85(1-2):1-32. [Colombia] Bandel, K. 1976e. Observations on spawn, embryonic development and ecology of some Caribbean lower Mesogastropoda (Mollusca). Veliger 18(3):249-271. [Colombia, Curagao] Bandel, K. 1976f. Spawning, development and ecology of some higher Neogastropoda from the Caribbean Sea of Colombia (South America). Veliger 19(2):176-193. Bandel, K. 1984. The radulae of Caribbean and other Mesogastropoda and Neogastropoda. Zoologische Verhandelingen 214:188 pp., 22 pls. [Colombia] Bandel, K. and E. Wedler. 1987. Hydroid, amphineuran and gastropod zonation in the littoral of the Caribbean Sea, Colombia. Senckenbergiana Maritima 19(1/2):1-129. Bartsch, P. 1930. Explorations for mollusks in the West Indies. Jn: Explorations & Field-work of the Smithsonian Institution in 1929, Smithsonian Institution Publication 3060:99-112. Bartsch, P. 1931. Descriptions of new marine mollusks from Panama, with a figure of the genotype of Engina. Proceedings of the United States National Museum 79(2881, art. 15):1-10, pl. 1. Bartsch, P. 1932. A newly discovered West Indian mollusk faunula. Proceedings of the United States National Museum 81(6):1-12, pls. 1-3. [Haiti] Bartsch, P. 1934a. The first Johnson-Smithsonian Deep-Sea Expedition to the Puerto Rican Deep. /n: Explorations and Field-work of the Smithsonian Institution in 1933, Smithsonian Institution Publication 3235:7-14. Bartsch, P. 1934b. Reports on the collections obtained by the first Johnson-Smithsonian Deep-Sea Expedition to the Puerto Rican Deep; New mollusks of the family Turritidae. Smithsonian Mis- cellaneous Collections 91(2):29 pp., 8 pls. Bartsch, P. 1942. The cyclophorid mollusks of the West Indies, exclusive of Cuba. Jn: The Cyclophorid Operculate Land Mollusks of America, by C. de la Torre, P. Bartsch, and J. P. E. Morrison, pp. 43-141, pls. 9-18. Bulletin of the United States National Museum 181(2):1-306, pls. 1-42. Bartsch, P. 1946. The operculate land mollusks of the family Annulariidae of the island of Hispaniola and the Bahama Archipelago. Bulletin of the United States National Museum 192:iv + 264 pp., pls. 1-38. Bavay, A. 1922. Sables littoraux de la mer des Antilles provenant des abords de Colon et de Cuba. Bulletin du Muséum National d’Histoire Naturelle 28:423-428. Bayer, F. M. 1971. New and unusual mollusks collected by R/V John Elliott Pillsbury and R/V Gerda in the tropical western Atlantic. Jn: Studies in Tropical American Mollusks, F. M. Bayer and G. L. Voss, eds., pp. 111-236. University of Miami Press, Coral Gables, viii + 236 pp. Beau, M. 1858a. Catalogue des coquilles recueillies 4 la Guadeloupe et ses dépendances, précédé d’une introduction, par M. P. Fischer. Revue Coloniale (Paris), December 1857:479-505. Beau, M. 1858b. De I’ utilité de certains mollusques marins vivants sur les cétes de la Guadeloupe et de la Martinique. Journal de Conch- yliologie 7:25-40. Beauperthuy, I. 1967. Los mitilidos de Venezuela (Mollusca: Bivalvia). Boletin del Instituto de Oceanogrdfico, Universidad de Oriente 6(1):7-115, pls. 1-21, 18 figs., 5 tabs., 1 map. Bequaert, J. C. and W. J. Clench. 1933. The non-marine mollusks of Yucatan. Carnegie Institute of Washington Publication 431:525- 545, pl. 68. Bequaert, J. [C.] and W. J. Clench. 1936. A second contribution to the molluscan fauna of Yucatan. Carnegie Institute of Washington Publication 457:61-75, pls. 1-2. [not seen] Bequaert, J. C. and W. J. Clench. 1938. A third contribution to the mollus- can fauna of Yucatan. Carnegie Institute of Washington Publi- cation 491:257-260. Bergh, R. S. 1890. Report 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., and Com- mander J. R. Bartlett, U.S.N., commanding. XXXII. Report on the nudibranchs. Bulletin of the Museum of Comparative Zoology 20(3):155-181, pls. 1-3. [Dominica, Yucatan] Bertsch, H. 1979. Tropical faunal affinities of opisthobranchs from the Panamic Province (eastern Pacific). Nautilus 93(2-3):57-61. Bidard, L. and J. Espinosa. 1989. Moluscos terrestres de Yara, Baracoa, Provincia Guantanamo. Garciana (17):1-2. [Cuba] Bland, T. 1852. Catalogue of the terrestrial shells of St. Thomas, West Indies. Contributions to Conchology 1(11):215-224. Bland, T. 1854. Note on the geographical distribution of the terrestrial mollusks which inhabit the island of St. Thomas, W. I. Annals of the Lyceum of Natural History of New York 6(1858):74-75. Bland, T. 1855. Notes on certain terrestrial mollusks which inhabit the West Indies. Annals of the Lyceum of Natural History of New York 6(1858):147-155. [general] Bland, T. 1861. On the geographical distribution of the genera and species of land shells of the West India islands; with a catalogue of the species of each island. Annals of the Lyceum of Natural History of New York 7:335-361, 2 tables. Bland, T. 1881. On the relations of the flora and fauna of Santa Cruz, West Indies. Annals of the New York Academy of Sciences 2:117- 126. [St. Croix] Bock, W. D. and D. R. Moore. 1971. The Foraminifera and micromollusks of Hogsty Reef and Serrana Bank and their paleoecological signifi- cance. Proceedings of the Sth Caribbean Geological Conference, Geological Bulletin 5:143-146. [Nicaragua] Bolivar de Carranza, A.M. and E. Hidalgo-Escalante. 1990. Lista de moluscos gastrépodos y pelecipodos del Golfo de México y el Caribe. Anales de la Escuela Nacional de Ciencias Biologicas 33(1-4):53-72. [not seen] Boone, L. 1925. Mollusca from tropical east American seas. Scientific results of the first oceanographic expedition of the “Pawnee” 1925. Bulletin of the Bingham Oceanographic Collection \(art. 3):1-20. [Cuba, Honduras, Belize, general] Bordaz, G. 1899. Liste des coquilles recueillies 4 la Martinique. Soc. Hist. Nat. Autun 12:165-184. [not seen] Borkowski, T. V. 1975. Variability among Caribbean Littorinidae. Veliger 17(4):369-377. MIKKELSEN ET AL.: BIBLIOGRAPHY OF CARIBBEAN MALACOLOGY 271 Boss, K. J. 1972. Minute Jamaican prosobranch gastropods: Stoastoma and its congeners. Breviora 393:1-13, pls. 1-2. Boss, K. J. and M. K. Jacobson. 1973a. Monograph of Ceratodiscus. Occasional Papers on Mollusks 3(45):253-279, pls. 45-48. [Cuba, Hispaniola, Puerto Rico] Boss, K. J. and M. K. Jacobson. 1973b. Monograph of the genus Alcadia in Cuba (Mollusca: Prosobranchia: Helicinidae). Bulletin of the Museum of Comparative Zoology 145(7):311-358, 6 pls. Boss, K. J. and M. K. Jacobson. 1974. Monograph of the genus Lucidella in Cuba (Prosobranchia: Helicinidae). Occasional Papers on Mol- lusks 4(48):1-28. Boss, K. J. and M. K. Jacobson. 1975a. Felipe Poey with a catalogue of the Mollusca described by him. Occasional Papers on Mollusks 4(53):105-132. [Cuba] Boss, K. J. and M. K. Jacobson. 1975b. Proserpine snails of the Greater Antilles (Prosobranchia; Helicinidae). Occasional Papers on Mollusks 4(51):53-90. [Cuba, Jamaica] Branson, B. A. and C. J. McCoy Jr. 1963. Gastropoda of the 1961 Uni- versity of Colorado Museum Expedition in Mexico. Nautilus 76(3):101-108. [Yucatan] Brattstr6m, H. 1980. Rocky-shore zonation in the Santa Marta area, Colombia. Sarsia 65:163-226, pls. 1-4. Brattstr6m, H. 1985. Rocky-shore zonation on the Atlantic coast of Panama. Sarsia 70:179-216. Breure, A. S. H. 1974. Caribbean land mollusks: Bulimulidae. I. Bulimu- lus. Studies on the Fauna of Curacao and other Caribbean Islands 45(145):1-80, pls. 1-7. [general] Breure, A. S. H. 1975. Caribbean land mollusks: Bulimulidae. II. Plekocheilus and Naesiotus. Studies on the Fauna of Curacao and other Caribbean Islands 46(152):71-93, pls.6-8. [general] Britton, J. C. 1976. The shallow water marine mollusks of the Swan Islands, Honduras. Bulletin of the American Malacological Union 1975:33-40. Brown, A. P. 1913. Notes on the geology of the island of Antigua. Pro- ceedings of the Academy of Natural Sciences of Philadelphia 1913:584-616, pls. 18-20. Brown, A. P. and H. A. Pilsbry. 1911-1913a. Fauna of the Gatun Forma- tion, Isthmus of Panama. Proceedings of the Academy of Natural Sciences of Philadelphia 63:336-373, pls. 22-29 (pt. 1, 1911); 64(1912):500-519, pls. 22-26 (pt. 2, 1913). Brown, A. P. and H. A. Pilsbry. 1913b. Two collections of Pleistocene fossils from the Isthmus of Panama. Proceedings of the Academy of Natural Sciences of Philadelphia 65:493-500. Brown, A. P. and H. A. Pilsbry. 1914. Fresh-water mollusks of the Oligo- cene of Antigua. Proceedings of the Academy of Natural Sciences of Philadelphia 66:209-213, pl. 9. Buehler, P. J. 1987. Caribbean conch culture. Oceanus 30(4):61. [Part of an entire issue devoted to marine resources of the Caribbean. ] Bullis, H. R. Jr. 1964. Muricidae (Gastropoda) from the northeast coast of South America, with descriptions of four new species. Tulane Studies in Zoology 11(4):99-107, 1 pl. [Venezuela] Bullock, R. C. 1974. A contribution to the systematics of some West Indian Latirus (Gastropoda: Fasciolariidae). Nautilus 88(3):69-79. Carnes, S. F. 1975. Mollusks from southern Nichupté Lagoon, Quintana Roo, Mexico. Sterkiana 59:21-50; 60:1-40. [Yucatan] Carpenter, P. P. 1863. Review of Prof. C.B. Adams’s ‘Catalogue of the Shells of Panama’ from the type specimens. Proceedings of the Zoological Society of London 1863(June 23):339-369. Cerridwen, S. A. and B. Jones. 1991. Distribution of bivalves and gas- tropods in the Pleistocene Ironshore Formation, Grand Cayman, British West Indies. Caribbean Journal of Science 27(3-4):97-116. Cernohorsky, W.O. 1978. The taxonomy of Caribbean-Atlantic Costel- lariidae (Mollusca: Gastropoda). Records of the Auckland Museum 15:87-109. Chitty, E. 1853. Descriptions of thirty supposed new species and varieties of land and fluviatile shells of Jamaica, with observations on some shells already described. Contributions to Conchology 1(13):1-19. Chitty, E. 1857a. On the Jamaican Cyclotus, and the description of twenty- one proposed new species and eight new varieties of that subgenus from Jamaica. Proceedings of the Zoological Society of London (25):142-157. Chitty, E. 1857b. On Stoastomidae as a family, and on seven proposed new genera, sixty-one new species, and two new varieties from Jamaica. Proceedings of the Zoological Society of London (25):162- 201. Clapp, W. F. 1914. List of land shells from Swan Island, with descriptions of five new species. Nautilus 27(9):97-101, pl. 6. [Honduras] Clark, B. L. and J. W. Durham. 1946. Eocene faunas from the Department of Bolivar, Colombia. Memoirs of the Geological Society of America 16:126 pp., 28 pls. Clark, K. B. and D. De Freese. 1987. Population ecology of Caribbean Ascoglossa (Mollusca: Opisthobranchia): a study of specialized algal herbivores. American Malacological Bulletin 5(2):259-280. [Belize] Clench, W. J. 1936. The Physidae of the West Indies. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 10(5):335- 342, pl. 25. [general] Clench, W. J. 1938. A new species of Oliva from Santo Domingo, with notes on other marine forms. Nautilus 51:109-114, pl. 9. Clench, W. J. 1948. Land shells of Mona Island. Journal de Conch- yliologie 90:269-276, 1 pl. Clench, W. J. 1962a. New land mollusks in the family Camaenidae and Fruticicolidae from Hispaniola. Mus. Argentino Cien. Nat. 8:213- 227. [not seen] Clench, W. J. 1962b. New species of land mollusks from the Reptblica Dominicana. Breviora 173:5 pp., 1 pl. Clench, W. J. 1964. Land and freshwater Mollusca of the Cayman Islands, West Indies. Occasional Papers on Mollusks 2(31):345-380. Clench, W. J. 1970. Land Mollusca of Saba Island, Lesser Antilles. Occasional Papers on Mollusks 3(38):53-60. Clench, W. J. and C. G. Aguayo. 1937. Notes and descriptions of some new land and fresh water mollusks from Hispaniola. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 11(2):61- 76, pl. 7. Clench, W. J. and C. G. Aguayo. 1938. Notes and descriptions of new species of Calliostoma, Gaza, and Columbarium (Mollusca); obtained by the Harvard-Habana [sic] expedition off the coast of Cuba. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 12(5):375-384, pl. 28. Clench, W. J. and C. G. Aguayo. 1939-1941. Notes and descriptions of new deep-water Mollusca obtained by the Harvard-Havana/Habana expedition off the coast of Cuba. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 13(3):189-197, pls. 28-29 (1939, part II); 14(1):77-94, pls. 14-16 (1940, part IID); 15(2):177- 180, pl. 14 (1941, part IV). Clench, W. J. and C. G. Aguayo. 1950. Nuevos helicinidos de la provincia de Oriente, Cuba. Revista de la Sociedad Malacélogica “Carlos de la Torre” 7(2):61-66. Clench, W. J. and C. G. Aguayo. 1951a. The Cuban genus Jeanneretia. Revista de la Sociedad Malacoélogica “Carlos de la Torre” 7(3):81- 92, pls. 13-14. Clench, W. J. and C. G. Aguayo. 1951b. Some new cerionids from Cuba. Revista de la Sociedad Malacélogica “Carlos de la Torre” 8(2):69-82, pls. 10-11. 242 AMER. MALAC. BULL. 10(2) (1993) Clench, W. J., C. G. Aguayo, and R. D. Turner. 1947-1948. The West- Indian marine shells with some remarks. A manuscript printed for circulation among collectors by H. Krebs, 1864. A republication, with remarks and a brief account of the life of H. Krebs. Revista de la Sociedad Malacoélogica “Carlos de la Torre” 5(1):23-40, 5(2):59-80, 5(3):91-116 [1947]; 6(1):11-43, 6(2):44-48, 1 pl. [1948]. [Original printed by W. Laubs Widow & Chr. Jorgensen, Nykjobing, Falster] Clench, W. J. and M. K. Jacobson. 1968. Monograph of the Cuban genus Viana (Mollusca: Archaeogastropoda: Helicinidae). Breviora 298:25 pp., 4 pls. Clench, W. J. and M. K. Jacobson. 1970. The genus Priotrochatella (Mollusca: Helicinidae) of the Isle of Pines and Jamaica, West Indies. Occasional Papers on Mollusks 3(39):61-80, pls. 17-21. Clench, W. J. and M. K. Jacobson. 1971a. Monograph of the Cuban gen- era Emoda and Glyptemoda (Mollusca: Archaeogastropoda: Hel- icinidae). Bulletin of the Museum of Comparative Zoology 141(3):99-130, 7 pls. Clench, W. J. and M. K. Jacobson. 1971b. Monograph of the Cuban gen- era Calidviana, Ustronia, Troschelviana, and Semitrochatella (Mollusca: Archaeogastropoda: Helicinidae) in Cuba. Bulletin of the Museum of Comparative Zoology 141(7):403-463, 8 pls. Clench, W. J. and R. D. Turner. 1950. Edward Chitty, with a bibliography and a catalogue of his species of Jamaica land mollusks. Occasion- al Papers of the Museum of the Institute of Jamaica 1:12 pp., 1 pl. Cockerell, T. D. A. 1894. A list of the Brachiopoda, Pelecypoda, Ptero- poda and Nudibranchia of Jamaica, living and fossil. Nautilus 7(9):103-107, 7(10):113-118. Cockerell, T. D. A. and R. R. Larkin. 1894. On the Jamaican species of Veronicella. Journal of Malacology 3(2):23-30. Colin, P. I. 1978. Caribbean Reef Invertebrates and Plants; a field guide to the invertebrates and plants occurring on coral reefs of the Caribbean, the Bahamas and Florida. T. F. H. Publishers, Inc., Neptune City, New Jersey, 512 pp. Conde, V. 1966. Studies on the Ecology and Distribution of the Marine Shelled Mollusca of Barbados. Master's Thesis, McGill University, Montreal, Quebec, Canada, 378 pp. [not seen] Cooke, C. W. 1919. Tertiary mollusks from the Leeward Islands and Cuba. In: Contributions to the Geology and Paleontology of the West Indies, T.W. Vaughan, ed., Carnegie Institute of Washington Publication 291:103-156, pls. 1-16. Cooke, C. W., J. Gardner, and W. P. Woodring. 1943. Correlation of the Cenozoic formations of the Atlantic and Gulf coastal plain and the Caribbean region. Bulletin of the Geological Society of America 54:1713-1723, chart. [general] Coomans, H. E. 1957. De verspreiding van de West-Indische mariene Gastropoda. Correspondentieblad van de Nederlandse Mala- cologische Vereniging 71:688-690. [general] Coomans, H. E. 1958. A survey of the littoral Gastropoda of the Nether- lands Antilles and other Caribbean Islands. Studies on the Fauna of Curacao and other Caribbean Islands 8(31):42-111, pls. 1-16, (Netherlands Antilles, general] Coomans, H. E. 1961. Shell collecting in the Netherlands Antilles [abstract]. Annual Reports of the American Malacological Union 1960:21-22. Coomans, H. E. 1963a. The marine Mollusca of St. Martin, Lesser Antilles, collected by H. J. Krebs. Studies on the Fauna of Curacao and other Caribbean Islands 16(72):59-87. Coomans, H. E. 1963b. The marine Mollusca of Sint [sic] Martin, Lesser Antilles, especially from the French part. Journal de Conchyliologie 103(3):113-186. Coomans, H. E. 1964a. The genus Conus in the southern Caribbean [ab- stract]. Bulletin of the American Malacological Union 1963:9-10. Coomans, H. E. 1964b. De eerste schelpenfauna van Curacao, in Simon’s beschrijving van dit eiland. Nieuwe West-Indische Gids 43:195- 210. Coomans, H. E. 1965. Shells and shell objects from an indian site on Magueyes Island, Puerto Rico. Caribbean Journal of Science 5(1-2):15-23. Coomans, H. E. 1967. The non-marine Mollusca of St. Martin (Lesser Antilles). Studies on the Fauna of Curacao and other Caribbean Islands 24(94):118-145. Coomans, H. E. 1974. Life and malacological work of Hendrik Elingsz van Rijgersma (1835-1877), a Dutch physician and scientist on St. Martin, Netherlands Antilles. Bijdragen tot de Dierkunde 44(2):115-214. Corea, L. F. 1934. Reports on the collections obtained by the first Johnson-Smithsonian Deep-Sea Expedition to the Puerto Rican Deep; New marine mollusks. Smithsonian Miscellaneous Collec- tions 91(16):9 pp., 3 pls. Cosel, R. von. 1973. Lista preliminar de los moluscos de la Cienaga Grande de Santa Marta (Colombia). Mitteilungen aus dem Institut Colombo-Alemdn Invest. Cient., Santa Marta 7:47-56. Cosel, R. von. 1976. Contribucién al conocimiento del género Voluta Linné, 1758 (Prosobranchia) en la Costa del Caribe de Colombia. Mitteilungen aus dem Institut Colombo-Alemdn Invest. Cient., Santa Marta 8:83-104, 5 pls. Cosel, R. von. 1978. Die Gattung Polymesoda Rafinesque 1820 an der Nordkiiste Siidamerikas (Bivalvia: Corbiculidae). Archiv fiir Molluskenkunde 108(1977)(4/6):201-213, pls. 11-13. [Colombia] Cosel, R. von. 1987. Moluscos de la region de la Cienaga Grande de Santa Marta (Costa del Caribe de Colombia). Anales del Instituto de Investigaciones Marinas de Punta de Betin 15-16(1986):79-370. Cossmann, M. 1913. Etude comparative de fossiles Miocéniques recueil- lis a la Martinique et a Il’Isthme de Panama. Journal de Conch- yliologie 61:1-64, pls. 1-5. Cox, L. R. 1941. Lamellibranchs from the White Limestone of Jamaica. Proceedings of the Malacological Society of London 24(4):135-144, pls. 8-10. Creswell, L. and M. Davis. 1991. Queen conch: the well-bred queen of the Caribbean. World Aquaculture 22(1):28-29,34-38,40. [This issue also contains other articles on mariculture in the Caribbean.] Criales, M. M. 1984. Shrimps associated with coelenterates, echinoderms, and molluscs in the Santa Marta region, Colombia. Journal of Crustacean Biology 4(2):307-317. Crosse, H. 1865. Description d’espéces nouvelles de la Guadeloupe. Journal de Conchyliologie 13(1):27-38, pl. 1. Crosse, H. 1890a. Faune malacologique terrestre et fluviatile de l’ile de Cuba. Journal de Conchyliologie 38:173-335, pls. 4-6. Crosse, H. 1890b. Faune malacologique terrestre et fluviatile de l’ile de la Trinité (Antilles). Journal de Conchyliologie 38:35-65, pl. 2. Crosse, H. 1891a. Catalogue de mollusques terrestres et fluviatiles de I’ile de Saint-Domingue. Journal de Conchyliologie 39:95-194. Crosse, H. 1891b. Distribution géographique et caractéres de la faune malacologique terrestre et fluviatile de Saint-Domingue. Journal de Conchyliologie 39:195-211. Crosse, H. 1891c. Faune malacologique terrestre et fluviatile de l’ile de Saint-Domingue. Journal de Conchyliologie 39:69-95. Crosse, H. 1892. Faune malacologique terrestre et fluviatile de l’ile de Portorico. Journal de Conchyliologie 40:5-71. Crosse, H. and T. Bland. 1873. Description de mollusques nouveaux provenant de Curacao et de Sainte-Lucie (Antilles). Journal de Conchyliologie 21:40-44. Cubit, J. and S. Williams. 1983. The invertebrates of Galeta Reef MIKKELSEN ET AL.: BIBLIOGRAPHY OF CARIBBEAN MALACOLOGY 213 (Caribbean Panama): a species list and bibliography. Atoll Research Bulletin 269:45 pp. Dall, W. H. 1878. Preliminary report on the Mollusca of the expedition. In letter no. 2 from A. Agassiz to C.P. Patterson on dredging opera- tions of the U.S. Coast Survey Steamer “Blake’”..., pp. 60-62. Bulletin of the Museum of Comparative Zoology 5(6):55-64. Dall, W. H. 1880. Reports on the results of dredging, under the supervi- sion of Alexander Agassiz, in the Gulf of Mexico and in the Caribbean Sea, 1877-78, by the United States Coast Survey Steamer “Blake,” Lieutenant-Commander C.D. Sigsbee, U.S.N., command- ing. 5. General conclusions from a preliminary examination of the Mollusca. Bulletin of the Museum of Comparative Zoology 6(3):85-93. Dall, W. H. 1881. Reports on the results of dredging, under the supervi- sion of Alexander Agassiz, in the Gulf of Mexico, and in the Caribbean Sea, 1877-79, by the United States Coast Survey Steamer “Blake,” Lieutenant-Commander C.D. Sigsbee, U.S.N., and Commander J.R. Bartlett, U. S. N., commanding. XV. Preliminary report on the Mollusca. Bulletin of the Museum of Comparative Zoology 9(2):33-144. Dall, W. H. 1885. List of marine Mollusca comprising the Quaternary fos- sils and Recent forms from American localities between Cape Hatteras and Cape Roque including the Bermudas. United States Geological Survey, Bulletin 24:336 pp. Dall, W. H. 1886-1889. 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., and Commander J.R. Bartlett, U.S.N., commanding. XXIX. Report on the Mollusca. Bulletin of the Museum of Comparative Zoology 12(6):171-318, pls. 1-9 (1886, Part I, Brachiopoda and Pelecypoda); 18:1-492, pls.10-40 (1889, Part II, Gastropoda and Scaphopoda). Dall, W. H. 1888. Gastropods and lamellibranchs. Jn: Three cruises of the United States Coast and Geodetic Survey Steamer “Blake” in the Gulf of Mexico, in the Caribbean Sea, and along the Atlantic coast of the United States, from 1877 to 1880, by A. Agassiz, pp.62-76. Bulletin of the Museum of Comparative Zoology, Harvard College 14:xxiii + 314 pp., 1 map; 15:(iv +) 220 pp. Dall, W. H. 1890. Scientific results of explorations by the U.S. Fish Commission Steamer Albatross. Preliminary report on the collec- tion of Mollusca and Brachiopoda obtained in 1887-88. Proceedings of the United States National Museum 12:219-362, pls. 5-14. Dall, W. H. 1896. The mollusks and brachiopods of the Bahama Expedition of the State University of Iowa. Bull. Lab. Nat. Hist., State University of Iowa 4(1):12-27, pl. 1. [also Cuba] Dall, W. H. 1897. Synopsis of the Pinnidae of the United States and the West Indies. Nautilus 11(3):25-26. Dall, W. H. 1899. Synopsis of the Solenidae of North America and the Antilles. Proceedings of the United States National Museum 22(1185):107-112. Dall, W. H. 1912a. New species of fossil shells from Panama and Costa Rica, collected by D.F. MacDonald. Smithsonian Miscellaneous Collections 59(2){publ. 2077]:1-10. Dall, W. H. 1912b. New species of landshells from the Panama Canal Zone. Smithsonian Miscellaneous Collections 59(18)[publ. 2134]:1- 3, pls. 1-2. Dall, W. H. and C. T. Simpson. 1901. The Mollusca of Porto Rico. United States Fish Commission Bulletin 1(1900):351-524, pls. 53-58. Daniel, H. 1941. Apuntes sobre algunos moluscos columbianos. Revista Colombiana de Ciencias Exactas, Fisicas y Naturales 4:372-380. {not seen] Dautzenberg, P. 1900. Croiséres du yacht Chazalie dans |’ Atlantique. Mollusques. Memoires de la Société Zoologique de France 13:145- 265, pls. 9-10. [general] Davis, C. C. 1967. Emergence of veliger larvae from eggs in gelatinous masses laid by some Jamaican marine gastropods. Malacologia 5(2):299-309. De Jong, K. M. and H. E. Coomans. 1988. Marine Gastropods from Curacao, Aruba and Bonaire. E. J. Brill, Leiden, 261 pp., 47 pls. De Jong, K. M. and I. Kristensen. 1965. Gegevens over mariene gastropo- den van Curagao. Correspondentieblad van de Nederlandse Mala- cologische Vereniging Supplement 1965 (Rivon Communication 218):1-56, 6 unnum. pls. De Jong, K. M. and I. Kristensen. 1968. Gegevens over de Moll. van Curagao uifgezonderd de Mariene Gastropoden. Rivon Mededeling 276:1-44, pls. 1-4. Deshayes, G. P. 1857. Note sur différents mollusques de la Guadeloupe, envoyés par M. Schramm. Journal de Conchyliologie 6:137-143. Desjardin, M. 1949. Les Rissoina de Vile de Cuba. Journal de Conch- yliologie 89(4):193-208, pls. 9-10. Diaz, J. M. 1990. Malacofauna subfosil y reciente de la Bahia de Portete, caribe Colombiano, con notas sobre algunos fosiles del Terciario. Boletin Ecotropica (23):1-22. Diaz, J. M. 1991. Las especies de Conus (Mollusca: Gastropoda: Conidae) en la region de Santa Marta, caribe Colombiano, con notas sobre su ecologia y distribucion. Anales del Instituto de Investigaciones Marinas de Punta de Betin 19-20(1989-1990):35-58. Diaz, J. M., L. A. Escobar, and L. E. Velasquez. 1992. Reef associated molluscan fauna of the Santa Marta area, Caribbean coast of Colombia. Anales del Instituto de Investigaciones Marinas de Punta de Betin 19-20(1989-1990):173-196. Diaz, J. M. and K. -J. Gétting. 1986. Mollusken-Geminschaften der Bahia de Nenguange (Karibik, Kolumbien). Helgolander Meeresunter- suchungen 40:279-308. Diaz, J. M. and K. -J. Gotting. 1988. Die Mollusken der Bahia de Nenguange (Karibik, Kolumbien) und ihre zoogeographischen Beziehungen. Zoologische Jahrbiicher, Abteilung fiir Systematik, Okologie und Geografie der Tiere 115(2):145-170. d’Orbigny, A. 1841-1853. Mollusques. In: Histoire Physique, Politique et Naturelle de l’ile de Cuba, by Ramon de la Sagra. Paris. Vol. 1 (pp. 1-208, 1841; 209-264, 1842), Vol. 2 (pp. 1-112, 1842; 113- 380, 18537), Atlas (pls. 1-28, 1842) [Spanish edition Historia Fisica Politica y Natural de la Isla de Cuba (segunda parte, historia natural), 5:376 pp, 1845, Paris, Madrid.] d’Orbigny, A. 1854. List of the Shells of Cuba in the Collection of the British Museum, Collected by M. Ramon de la Sagra. British Museum (Natural History), London, 48 pp. Dreyfuss, R. 1953. Les planorbes de la Martinique. Bulletin de la Société Francaise d’Histoire Naturelle des Antilles 2:41-45. [not seen] Duque, F. 1979. Malacofauna de aguas someras y substrato rocoso en la Bahia de Nenguange. Boletin Museo del Mar, Universidad de Bogota 9:3-29. [Colombia] Eddison, G. G. 1964. Water Isle shelling. New York Shell Club Notes 105:6-7. [Virgin Islands] Edmunds, M. 1964. Eolid Mollusca from Jamaica, with descriptions of two new genera and three new species. Bulletin of Marine Science of the Gulf and Caribbean 14:1-32. Edmunds, M. and H. Just. 1983. Eolid nudibranchiate Mollusca from Barbados. Journal of Molluscan Studies 49:185-203. Edmunds, M. and H. Just. 1985. Dorid, dendronotid and arminid nudi- branchiate Mollusca from Barbados. Journal of Molluscan Studies 51:52-63. 274 AMER. MALAC. BULL. 10(2) (1993) Edwards, C. E. 1977. A layman looks at Caribbean chitons. Of Sea and Shore 8(1):12-14. Ekdale, A. A. 1972. Ecology and Paleoecology of Marine Invertebrate Communities in Calcareous Substrates, Northeast Quintana Roo, Mexico. Master's Thesis, Rice University, Houston, Texas, 159 pp. [not seen] Ekdale, A. A. 1974. Marine molluscs from shallow-water environments (0 to 60 meters) off the northeast Yucatan coast, Mexico. Bulletin of Marine Science 24(3):638-668. Emerson, W. K. 1952. The scaphopod mollusks collected by the first Johnson-Smithsonian Deep-sea Expedition. Smithsonian Miscel- laneous Collections 117(6):1-14, pl. 1. [Puerto Rico] Engel, H. 1925. Westindische opisthobranchiate Mollusken. [I.] Bijdragen tot de Dierkunde 24:33-80. [Curacao] Engel, H. 1927. Westindische opisthobranchiate Mollusken. [II.] Bijdragen tot de Dierkunde 25:83-122. [Curacao] Engel, H. (with contributions by P. Wagenaar Hummelinck). 1936. Zoologische Ergebnisse einer Reise nach Bonair, Curagao und Aruba im Jahre 1930. No. 23. Ueber westindische Aplysiidae und Verwandten anderer Gebiete. Capita Zoologica 8(1):1-76. [Neth- erlands Antilles, general] Ernst, A. 1876. Enumeracion Sistemdtica de las Especies de Moluscos Terrestres y de Agua Dulce, Halladas Hasta Ahora in los Alrededores de Caracas y Demas Partes de la Republica: Estudios Sobre la Flora y Fauna de Venezuela. Caracas, pp.225-230. {Reprinted in Revista de la Sociedad Malacélogica “Carlos de la Torre” 6(2):73-82, 1948] Espinosa, J. 1984. Adiciones al género Diodora (Mollusca: Archaeo- gastropoda) en Cuba. Poeyana 270:1-22. Espinosa, J. 1987. La familia Cryptoplacidae (Mollusca: Polyplacophora) en Cuba. Garciana 3:2 unnumbered pp. Espinosa, J. 1989. El género Cyphoma (Mollusca: Mesogastropoda) en Cuba. Poeyana 377:1-10. Espinosa, J. and R. Fernandez Garcés. 1988. Primer registro de la familia Mathildidae (Mollusca: Gastropoda) para la fauna viviente de Cuba. Miscelanea Zoologica, La Habana 40:3-4. Espinosa, J. and R. Fernandez Garcés. 1989. Primer registro de la familia Stenacmidae (Mollusca: Gastropoda) para las Antillas. Miscelanea Zoologica, La Habana 43:2-3. [Cuba] Espinosa, J. and R. Fernandez Garcés. 1990a. El género Daphnella (Mol- lusca: Neogastropoda) en Cuba. Descripcién de nuevas especies. Poeyana 396:1-16. Espinosa, J. and R. Fernandez Garcés. 1990b. La familia Epitoniidae (Mollusca: Gastropoda) en Cuba. Poeyana 404:1-13. Etayo Serna, F. 1979. La fauna de moluscos del paleoceno de Colombia. Boletin de Geologia 13(27):5-5S. Eudes-Deslongchamps, J. A. 1859. Catalogue des cirrhipédes, des mol- lusques et des rayonnés recueillis par M. E. Déplanche, chirurgien auxiliare de la marine impériale, pendant la campagne de I’aviso a vapeur le Rapide, années 1854-55-56. Bulletin de la Société Linnéene de Normandie 4:1-50. [general] Eva, A. N. 1980. Pre-Miocene seagrass communities in the Caribbean. Palaeontology 23(1):231-236. [general] Eyerdam, W. J. 1961. An excursion to Lake Miragoane, Haiti. Nautilus 75(2):71-75. Faber, M. J. 1988. Studies on West Indian marine mollusks. 13. The mala- cological taxa of Gordon W. Nowell-Usticke. De Kreukel 24(4- 5):67-102. [St. Croix] Faber, M.J. 1990. Studies on West Indian marine mollusks 18. A review of the genus Pyramidelloides (Gastropoda Prosobranchia: Eulimi- dae) in the West Indies, with description of two new species. Basteria 54:105-113. Ferguson, F. F. and K. W. Buckmire. 1974. Notes on the freshwater mol- luscs of Grenada, British West Indies. Caribbean Journal of Science 14(3-4):147-148. Ferguson, F. F. and C. S. Richards. 1963. Fresh water mollusks of Puerto Rico and the U.S. Virgin Islands. Transactions of the American Microscopical Society 82:391-395. Ferreira, A. J. 1978. The chiton species described by C.B. Adams, 1845, from Jamaica. Bulletin of Marine Science 28(1):81-91. Ferreira, A. J. 1985. Chiton (Mollusca: Polyplacophora) fauna of Bar- bados, West Indies, with the description of a new species. Bulletin of Marine Science 36(1):189-219. Fischer, P. and H. Crosse. 1870-1900. Etudes sur les Mollusques Ter- restres et Fluviatiles du Mexique et du Guatemala. Mission scien- tifique au Mexique et dans 1’ Amérique Centrale. Recherches zoo- logiques, pt. 7, vols. I (702 pp.) and II [731 pp., atlas (72 pl.]. Paris. [Yucatan, Guatemala] Flores, C. 1964. Contribucion al conocimiento del género Strombus Linnaeus, 1758, (Mollusca: Mesogastropoda), en las aguas costa- neras de Venezuela. Memorias de la Sociedad de Ciencias Natur- ales La Salle 24:261-276. Flores, C. 1966. La familia Cassididae (Mollusca: Mesogastropoda) en las costas nororientales de Venezuela. Boletin del Instituto de Ocean- ografico, Universidad de Oriente 5(1-2):7-37. Flores, C. 1968. Algunos gastropodos de las Islas Las Aves, Venezuela, y su distribucién. Boletin del Instituto de Oceanogrdfico, Univers- idad de Oriente 7:73-106. Flores, C. 1973. La familia Littorinidae (Mollusca: Mesogastropoda) en las aguas costeras de Venezuela. Boletein del Instituto de Ocean- ogrdfico, Universidad de Oriente 12(1):3-22. Fluck, W. H. 1900. Shell collecting on the Mosquito Coast. Nautilus 14(8):94. [Nicaragua] Fluck, W. H. 1901. Correspondence [from Nicaragua]. Nautilus 15(4):48. Fluck, W. H. 1905-1906. Shell-collecting on the Mosquito Coast of Nicaragua. Nautilus 19(1):8-12 (1905, part I); 19(2):16-19 (1905, part II); 19(3):32-34 (1905, part III); 19(5):55-57 (1905, part IV); 19(7):78-80 (1905, part V); 20(1):1-4 (1906, part VI). Folin, L. de. 1867-1871. Rade de la Pointe-a-Pitre. Nouveau supplément aux mollusques de la Pointe-a-Pitre. In: Les Fonds de la Mer, by L. de Folin and Périer, 1:17-23, 263-265. Paris. [Haiti; not seen] Folin, L. de. 1879. Quelques Caecidae des Antilles. In: Les Fonds de la Mer, by L. de Folin and Périer, 4:14-15. Paris. [general; not seen] Forcart, L. 1950. Von Dr. H. P. Schaub auf der Insel Cuba gesammelte Land- und Siisswasserschnecken. Archiv fiir Molluskenkunde 79(1/3):67-72. Freire, J. A. and P. Alayo. 1946. Excursiones a Casa Azul. Revista de la Sociedad Malacélogica “Carlos de la Torre” 4(2):65-69. [Cuba] Gabb, W. M. 1871. Notes on the geology of Santa Domingo. American Journal of Science and Arts 1:1-4. Gabb, W. M. 1873. [Notes on the] Topography and geology of Santo Domingo. Transactions of the American Philosophical Society, new series, 15:49-259, 2 maps. Gabb, W. M. 1875. Notes on West Indian fossils. The Geological Maga- zine, new series, 2:544-545. [Dominican Republic] Gabb, W. M. 1881a. Descriptions of Caribbean Miocene fossils. Journal of the Academy of Natural Sciences of Philadelphia series 2, 8:337- 348, pls. 44-45. [general] Gabb, W. M. 1881b. Descriptions of new species of fossils from the Pliocene clay beds between Limén and Moen, Costa Rica, together with notes on previously known species from there and elsewhere in the Caribbean area. Journal of the Academy of Natural Sciences MIKKELSEN ET AL.: BIBLIOGRAPHY OF CARIBBEAN MALACOLOGY pA of Philadelphia series 2, 8:349-380, pls. 45-47. Garcia, E. F. 1988. Observations on the Voluta demarcoi complex and the genus Voluta in the Caribbean. Louisiana Malacological Society, Occasional Papers 1:8 pp. [Honduras, general] Garcia, M. T. and A. A. Luque. 1986. Contribucion al conocimiento de los gaster6podos prosobranquios del la Isla de la Juventud y del archip- iélago de los Canarreos (Cuba). Revista de Investigaciones Marinas 7(2):31-52. Garcia-Rios, C. 1983. Comunidades de Gastrépodos Asociados a Macro- algas Marinas Comunes en el Suroeste de Puerto Rico. Master's Thesis, Department of Marine Sciences, University of Puerto Rico, Mayaguez, vii + 65 pp. [not seen] Garcia-Rios, C. 1988. The Chitons of Puerto Rico (Los Quitones de Puerto Rico; guia para la identificacion de organismos marinos de Puerto Rico). University of Puerto Rico Sea Grant Program, 46 pp. [not seen] Gerth, H. 1951. Antillen-Molukken, zwei Inselbégen, ein Vergleich des geologischen Baues und der Schwereanomalie. Geologischen Rundschau 39(1):273-284, 2 maps. Gertman, R. L. 1969. Cenozoic Typhinae (Mollusca: Gastropoda) of the western Atlantic region. Tulane Studies in Geology and Paleon- tology 7(4):143-191, pls. 1-8. Gibson-Smith, J. 1973. The genus Voluta (Mollusca: Gastropoda) in Venezuela, with description of two new species. Geos 20:65-73, 3 pls. Gibson-Smith, J. and W. Gibson-Smith. 1971. Una coleccion de moluscos de Isla de Aves, Venezuela. VJ Conferencia Geologica del Caribe, Porlamar - Edo. Nueva Esparta, Venezuela:\-29. Gibson-Smith, J. and W. Gibson-Smith. 1974. The genus Strombina (Mollusca: Gastropoda) in Venezuela, with descriptions of a new Recent and some fossil species. Boletin Informativo, Asociacion Venezolana de Geologia, Mineria y Petrdleo 17(4-6):49-84, pls. 1-4. Gibson-Smith, J. and W. Gibson-Smith. 1979. The genus Arcinella (Mollusca: Bivalvia) in Venezuela and some associated faunas. Geos 24:11-32, 3 pls. Gibson-Smith, J. and W. Gibson-Smith. 1982a. Lucina s.s. (Mollusca: Bivalvia) in the western Atlantic: a reappraisal. Veliger 25(2):139- 148, 2 pls. Gibson-Smith, J. and W. Gibson-Smith. 1982b. The subfamily Mel- ampinae (Pulmonata: Basommatophora) in Venezuela, with descriptions of two new species. Nautilus 96(3):116-120. Gibson-Smith, J. and W. Gibson-Smith. 1983. Neogene melongenid gas- tropods from the Paraguana Peninsula, Venezuela. Eclogae Geologicae Helvetiae 76(3):719-728. Gladfelter, W. B. 1974. Annotated classification of marine organisms common around St. Croix. In: Guidebook to the Geology and Ecology of Some Marine and Terrestrial Environments, St. Croix, U.S. Virgin Islands, H.G. Multer and L.C. Gerhard, comps. & eds., pp. 21-30. West Indies Laboratory, Fairleigh Dickinson University, Special Publication 5. Gloyne, C. P. 1872. Notes sur le mode de station, les moeurs et les habi- tudes des Mollusques terrestres de la Jamaique, avec la description d’une espéce nouvelle. Journal de Conchyliologie 20:26-47. Gloyne, C. P. 1875. Note supplémentaires sur les mollusques terrestres de la Jamaique. Journal de Conchyliologie 23(2):115-126. Glynn, P. W. 1968a. Ecological studies on the associations of chitons in Puerto Rico, with special reference to sphaeromid isopods. Bulletin of Marine Science 18(3):572-626. Glynn, P. W. 1968b. Mass mortalities of echinoids and other reef flat organisms coincident with midday, low water exposures in Puerto Rico. Marine Biology 1(3):226-243. Glynn, P. W. 1970. On the ecology of the Caribbean chitons Acanthopleura granulata Gmelin and Chiton tuberculatus Linné: density, mortality, feeding, reproduction, and growth. Smithsonian Contributions to Zoology 66:21 pp. [Panama, Puerto Rico] Glynn, P. W., L. R. Almodovar, and J. G. Gonzalez. 1964. Effects of Hurricane Edith on marine life in La Parguera, Puerto Rico. Caribbean Journal of Science 4(2-3):335-345. Gotting, K. -J. 1973. Die Polyplacophora der karibischen Kiiste Kolum- biens. Archiv fiir Molluskenkunde 103(4/6):243-261, pls. 8-11. Gitting, K. -J. 1978. Lista preliminar de los caracoles terrestres de la region septentrional de Colombia. Anales del Instituto de In- vestigaciones Marinas de Punta de Betin 10:101-110. Golley, F. B. 1960. Ecologic notes on Puerto Rican Mollusca. Nautilus 73(4):152-155. Gomez, J. D., M. Vargas, and E. A. Malek. 1986. Freshwater mollusks of the Dominican Republic. Nautilus 100(4):130-134. Gomez P., L. D. and C. E. Valerio G. 1971. Lista preliminar illustrada de los moluscos foésiles de la formaci6n Rio Banano (Mioceno), Limon, Costa Rica. Instituto Geografico Nacional [Costa Rica], Informe Semestral, Enero a Junio, 1971:43-62. Gonzalez, A. R. and C. Flores. 1972. Nota sobre los generos Thais Roeding, Purpura Brugiere [sic], y Murex Linnaeus (Neo- gastropoda: Muricidae) en las aguas costeras de Venezuela. Boletin del Instituto de Oceanografico, Universidad de Oriente 11(2):67-82. Gonzalez C., L. A. and D. Princz. 1979. Moluscos heteropodos y pteropo- dos de la zona central del caribe Venezolano. Boletin del Instituto de Oceanografico, Universidad de Oriente 18(1-2):99-107. Gonzalez, M. A., E. A. Chavez, G. de la Cruz, and D. Torruco. 1991. Patrones de distribuci6n de gasterdpodos y bivalvos en la peninsula de Yucatan, Mexico. Ciencias Marina 17(3):147-172. Goodfriend, G. A. 1983. Late Quarternary land snails from the north coast of Jamaica [abstract]. American Malacological Bulletin 1(1):99- 100. Goodfriend, G. A. 1986. Pleistocene land snails from Sheep Pen Cave in the Cockpit Country of Jamaica. In: Proceedings of the 8th Inter- national Malacological Congress, Budapest, 1983, L. Pintér, ed., pp. 87-90. Hungarian Natural History Museum, Budapest, xviii + 342 pp. Goodfriend, G. A. and R. M. Mitterer. 1988. Late Quaternary land snails from the north coast of Jamaica: local extinctions and climatic change. Palaeogeography, Palaeoclimatology, Palaeoecology 63:293-311. Goodfriend, G. A. and R. M. Mitterer. 1993. A 45,000-year record of a tropical lowland biota: the land snail fauna from cave sediments at Coco Ree, Jamaica. Geological Society of America Bulletin 105(1):18-29. Goodrich, C. and H. van der Schalie. 1937. Mollusca of Petén and North Alta Vera Paz, Guatemala. Miscellaneous Publications of the Museum of Zoology, University of Michigan 34:1-5SO, 1 pl., 1 map. Gould, A. A. 1844. Descriptions and notices of some of the land shells of Cuba. Proceedings of the Boston Society of Natural History 4(4):485-498, pl. 24. Gray, J. E. 1854. List of the Shells of Cuba in the Collection of the British Museum, Collected by M. Ramon de la Sagra, Described by Prof. Alcide d’Orbigny, in the “Histoire de l’Ile de Cuba.” British Museum, London, 48 pp. Guilding, L. 1826. Mollusca caribbaeana. The Zoological Journal 2(8):437-444, pls. XVI bis Suppl., XVII bis Suppl., XIV. Guilding, L. 1828. Observations on the zoology of the Caribaean islands. The Zoological Journal 3(11):403-408 [non Mollusca]; 3(12):527- 544, pls. XV, XXVII Supp., XXVI Supp., XXVIII Supp.; 4(14):164-175. 276 AMER. MALAC. BULL. 10(2) (1993) Gundlach, D. J. 1856. Molluscorum species novae y excursi6n al Pan de Guajaibon. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 2:13-23. [Cuba; not seen] Gundlach, D. J. 1857a. Beschreibung neuer Schnecken aus dem west- lichen Theile von Cuba. Malakozoologische Blatter 3:38-42. Gundlach, D. J. 1857b. Malakologische Notizen aus dem Norden und Westen von Cuba. Malakozoologische Blétter 4:41-47. Gundlach, D. J. 1883. Apuntes para la fauna Puerto-Riquefia. Partes cuar- to y quinta. Moluscos. Anales de la Sociedad Espanola de Historia Natural, Madrid 12:5-58, 441-484. Gunter, G. 1951. The species of oysters of the Gulf, Caribbean and West Indian region. Bulletin of Marine Science of the Gulf and Caribbean 1(1):40-45. Guppy, R. J. L. 1866a. Catalogue of the land and fresh water Mollusca of Trinidad. Proceedings of the Scientific Association of Trinidad 1866:2-27. [not seen] Guppy, R. J. L. 1866b. On the relations of the Tertiary formations of the West Indies. Quarterly Journal of the Geological Society of London 22:570-590, pl.26. [general] Guppy, R. J. L. 1866c. On the terrestrial and fluviatile Mollusca of Trinidad. Annals and Magazine of Natural History (3) 17:42-56. Guppy, R. J. L. 1866d. On the Tertiary Mollusca of Jamaica. Quarterly Journal of the Geological Society of London 22:281-294, pls. 16-18. Guppy, R. J. L. 1867a. Notes on West Indian geology, with remarks on the existence of an Atlantis in the early Tertiary period; and descrip- tions of some new fossils, from the Caribean [sic] Miocene. The Geological Magazine 4:496-501. [general] Guppy, R. J. L. 1867b. On the Tertiary fossils of the West Indies with especial reference to the classification of the Kainozoic rocks of Trinidad. Proceedings of the Scientific Association of Trinidad 1(3):145-176. [Reprinted in Harris, 1921:24-55 (172-203)] Guppy, R. J. L. 1868a. On the terrestrial Mollusca of Dominica and Grenada; with an account of some new species from Trinidad. Annals and Magazine of Natural History, series 4, 1:429-442. Guppy, R. J. L. 1868b. Further additions to the catalogue of the land and freshwater Molluska [sic] of Trinidad. Proceedings of the Scientific Association of Trinidad 1(5):237-245. Guppy, R. J. L. 1869. Notice of some new marine shells found on the shores of Trinidad. Proceedings of the Scientific Association of Trinidad 1(7):366-369. Guppy, R. J. L. 1871. Notes on some new forms of terrestrial and fluvi- atile Mollusca found in Trinidad. American Journal of Conchology 6(4):306-311, pl. 17. Guppy, R. J. L. 1872. Third series of additions to the catalogue of the land and freshwater Molluska [sic] of Trinidad; with a revised list of all the species. Proceedings of the Scientific Association of Trinidad 2(1):17-25. Guppy, R. J. L. 1873a. On new species of bivalve Molluska [sic] found at Cumana, Venezuela. Proceedings of the Scientific Association of Trinidad 2:90-92, pl. 3. [Reprinted in Harris, 1921:73-75 (221-223), pl. 1 (5)] Guppy, R. J. L. 1873b. On some new Tertiary fossils from Jamaica. Proceedings of the Scientific Association of Trinidad 2(2, pt. 10):72-88, pls. 1-2. [Reprinted in Harris, 1921:56-72 (204-220)] Guppy, R. J. L. 1874. On the Westindian Tertiary fossils. The Geological Magazine, new series, 1:404-411, 434-446, pls. 16-18. [Trinidad, general] Guppy, R. J. L. 1875a. Notice of some marine shells found on the shores of Trinidad. Annals and Magazine of Natural History series 4, 15:50-52, pl.7. Guppy, R. J. L. 1875b. Supplement to the paper on West Indian Tertiary fossils. The Geological Magazine, new series, 2:41-42. [Reprinted on 1 unnumbered page] [Jamaica, general] Guppy, R. J. L. 1876. On the Miocene fossils of Haiti. Quarterly Journal of the Geological Society 32:516-532, pls. 28-29. Guppy, R. J. L. 1877. First sketch of a marine invertebrate fauna of the Gulf of Paria and its neighbourhood. Part 1, Molluska [sic]. Proceedings of the Scientific Association of Trinidad 2(3):134-157. [Trinidad] Guppy, R. J. L. 1881. On the land-shells of S. Vincent, W.I. Proceedings of the Scientific Association of Trinidad 2(4):163-166. Guppy, R. J. L. 1882. On the Recent and Tertiary species of Leda and Nucula found in the Westindies; with notices of Westindian shells. Proceedings of the Scientific Association of Trinidad 2(4, pt. 12):168-180, pl. 7. [Reprinted in Harris, 1921:89-101 (237-249), pl.5 (5)] [general] Guppy, R. J. L. 1893. The land and freshwater Mollusca of Trinidad. Journal of Conchology 7:210-231. Guppy, R. J. L. 1910. On a collection of fossils from Springvale, near Couva, Trinidad. Agricultural Society of Trinidad and Tobago, Society Paper 440:1-15. [Reprinted in Harris, 1921:144-157 (292- 305)] Guppy, R. J. L. 1911. Fossils from Springvale, near Couva, Trinidad - sec- ond report. Agricultural Society of Trinidad and Tobago, Society Paper 454:1-10, pls.1-2. [Reprinted in Harris, 1921:158-166 (306- 314), pls.8-9 (12-13)] Guppy, R. J. L. 1913. Observations on the geology of Martinique with notes on fossils from Trinidad and Venezuela. Agricultural Society of Trinidad and Tobago, Society Paper 549:1-5. [Reprinted in Harris, 1921:188-192 (336-340)] Guppy, R. J. L. and W. H. Dall. 1896. Descriptions of Tertiary fossils from the Antillean region. Proceedings of the United States National Museum 19(1110):303-331, pls. 27-30. Guyard, A. and J.-P. Pointier. 1979. Faune malacologique dulgaquicole et vecteurs de la schistosome intestinale en Martinique (Antilles frangaises). Annales de Parasitologie 54:193-205. [not seen] Guyard, A., J.-P. Pointier, A. Theron, and A. Gilles. 1982. Mollusques hotes intermediares de la schistosome intestinale dans les petites Antilles: hypotheses sur le role de Biomphalaria glabrata et B. straminea en Martinique. Malacologia 22(1-2):103-107. Haas, F. 1941. Malacological notes - II. Marine shells from the Mandel Caribbean Expedition. Zoological Series of Field Museum of Natural History 24(17):173-174. [Cuba, Honduras, Yucatan] Haas, F. 1960. Caribbean land molluscs: Vertiginidae. Studies on the Fauna of Curacao and other Caribbean Islands 10(41):1-17, pls. 1-5. Haas, F. 1962. Caribbean land molluscs: Subulinidae and Oleacinidae. Studies on the Fauna of Curadao and other Caribbean Islands 13(58):49-60, pls. 7-11. Haas, F. and A. Solem. 1960. Non-marine mollusks from British Honduras. Nautilus 73(4):129-131. Haas, O. 1942. Miocene Mollusca from Costa Rica. Journal of Paleon- tology 16(3):307-316. Harris, G. D. 1926. Notes on the paleontology [of Trinidad]. Jn: The Geology of the Island of Trinidad, B.W.1., G.A. Waring, ed., pp .87- 112, pls. 16-20. Johns Hopkins University Studies in Geology 7. Harry, H. W. 1950. Studies on the nonmarine Mollusca of Yucatan. Occasional Papers of the Museum of Zoology, University of Michigan 524:34 pp. Harry, H. W. 1962. A critical catalogue of the nominal genera and species of neotropical Planorbidae. Malacologia 1(1):33-53. Harry, H. W. 1964. The foreign freshwater snails now established in Puerto Rico [abstract]. Annual Reports of the American Mala- MIKKELSEN ET AL.: BIBLIOGRAPHY OF CARIBBEAN MALACOLOGY 20) cological Union 1964:4-5. Harry, H. W. and B. Hubendick. 1964. The freshwater pulmonate Mollusca of Puerto Rico. Géteborgs Kungl. Vetenskaps-och Vitter- hets-samhdlles Handlingar series B, 9(5):1-77. [not seen] Hemmen, J. and C. Hemmen. 1979. Beitrag zur Kenntnis der Meeresmollusken-Fauna der Karibischen See. Grenada. Jahr- biicher des Nassauischen Vereins fiir Naturkunde 104:137-172. [Grenada] Henderson, J. B., Jr. 1894. List of shells collected in Jamaica. Nautilus 8:1-5, 19-24, 31-33. Henderson, J. B., Jr. 1916. A list of the land and fresh-water shells of the Isle of Pines. Annals of the Carnegie Museum 10(3-4):315-324. [Cuba] Hermes, J. J. 1945. Geology and Paleontology of East Camaguey and West Oriente, Cuba. Acad. Thesis, Utrecht, 75 pp., 5 pls., 1 map. {rudists only] Herrera, A. and J. Espinosa. 1988. Caracteristicas de la fauna de bivalvos en la bahia de Cardenas. Reporte de Investigacion del Instituto de Oceanologia (17):21 pp., 5 tab., 11 figs. [42 pp. total] [Cuba] Herrera, E. 1945. Moluscos colectados en La Julia, Zulueta. Revista de la Sociedad Malacélogica “Carlos de la Torre” 3(1):1-4. [Cuba] Hidalgo, E. 1956. Algunos moluscos de la Isla de Cozumel, Quintana Roo, Mexico. Acta Zoologica Mexicana 1(10):1-24, pls. 1-6. Higgins, H. H. 1876. Mollusca of the Argo Expedition to the West Indies, 1876. Free Public Museum of Liverpool, Museum Report 1:20 pp., 1 pl. [general; not seen] Hinkley, A. A. 1920. Guatemala Mollusca. Nautilus 34(2):37-55. Hjalmarson, J. and L. Pfeiffer. 1858. Beitrage zur Fauna von Westindien. Malakozoologische Blatter 5:135-155, pls. 2-3. [Santo Domingo] Hodson, F. 1926. Venezuelan and Caribbean turritellas; with a list of Venezuelan type stratigraphic localities. Bulletins of American Paleontology 11(45):1-50 (171-220), pls. 1-30 (5-34). [Venezuela, general] Hodson, F. and H. K. Hodson.1931 Some Venezuelan mollusks. Bulletins of American Paleontology 16(59):1-94, pls. 1-24 (part 1); 16(60):1- 38 (95-132), pls. 1-12 (25-36) (part 2). Hodson, F., H. K. Hodson, and G. D. Harris. 1927. Some Venezuelan and Caribbean mollusks. Bulletins of American Paleontology 13(49):i- viii, 1-160, pls. 1-40. Hoskins, C. W. 1964. Molluscan biofacies in calcareous sediments, Gulf of Batabano, Cuba. Bulletin of the American Association of Petroleum Geologists 48(10):1680-1704, pls. 1-2. Houbrick, J. R. 1968. A survey of the littoral marine mollusks of the Caribbean coast of Costa Rica. Veliger 11(1):4-23. Hubbard, B. 1920. Tertiary Mollusca from the Lares District, Porto Rico. New York Academy of Sciences, Scientific Survey of Porto Rico and the Virgin Islands 3(2):79-164, pls. 1-25. Hubendick, B. 1961. Studies on Venezuelan Planorbidae. Géteborgs Kungl. Vetenskaps-och Vitterhets-samhdlles Handlingar series B, 8(9):1-50. Humfrey, M. 1975. Sea Shells of the West Indies: a Guide to the Marine Molluscs of the Caribbean. Taplinger Publishing Company, New York, 351 pp., 32 pls. [Jamaica] Hummelinck, P. Wagenaar. 1940a. Land and freshwater mollusks from the smaller Venezuelan islands. Archives Néerlandaises de Zoologie 4:352-354. Hummelinck, P. Wagenaar. 1940b. Mollusks of the genera Cerion and Tudora. Studies on the Fauna of Curacao, Aruba, Bonaire and the Venezuelan Islands 2(5):43-82, pl. 5. Hummelinck, P. Wagenaar. 1940c. Studies on the fauna of Curagao, Aruba, Bonaire and the Venezuelan islands. No. 1. Studies on the Fauna of Curacao and other Caribbean Islands 1:1-57, 8 pls. Hummelinck, P. Wagenaar. 1940d. A survey of the mammals, lizards and mollusks. Studies on the Fauna of Curacao, Aruba, Bonaire and the Venezuelan Islands 1(1):59-108, pls. 9-16 [Lesser Antilles, general] Hummelinck, P. Wagenaar. 1940e. Zoogeographical remarks. Studies on the Fauna of Curacao, Aruba, Bonaire and the Venezuelan Islands 1(3):109-130. [general] Hummelinck, P. Wagenaar. 1980. Caribbean land molluscs: Cerion in the Cayman Islands. Studies on the Fauna of Curacao and other Caribbean Islands 61(187):1-67, pls. 1-18. Hummelinck, P. Wagenaar. 1990, About the malacological subdivision of Curagao; a review. Bijdragen tot de Dierkunde 60(3-4):181-187. Imlay, R. W. 1942. Late Jurassic fossils from Cuba and their economic significance. Bulletin of the Geological Society of America 53:1417- 1478, pls. 1-12. Imlay, R. W. 1944. Correlation of the Cretaceous formations of the Greater Antilles, Central America and Mexico. Bulletin of the Geological Society of America 55:1005-1045, pls. 1-3. Ingram, W. M. 1939. New fossil Cypraeidae from the Miocene of the Dominican Republic. Bulletins of American Paleontology 24(85):1- 14 (327-340), pl. 1 (22). Ingram, W. M. 1947. New fossil Cypraeidae from Venezuela and Colombia. Bulletins of American Paleontology 31(121):1-12 (127- 136), pls. 1-2 (8-9). Ingram, W. M. 1948. New fossil Cypraeidae from the Miocene of Florida and Colombia. Proceedings of the California Academy of Sciences series 4, 26(6):125-133, pl. 2. Jackson, J. B. C. 1972. The ecology of the molluscs of Thalassia commu- nities, Jamaica, West Indies. II. Molluscan population variability along an environmental stress gradient. Marine Biology 14(4):304- 337. Jackson, J. B. C. 1973. The ecology of molluscs of Thalassia communi- ties, Jamaica, West Indies. I. Distribution, environmental physiolo- gy, and ecology of common shallow-water species. Bulletin of Marine Science 23(2):313-350. Jackson, J. B. C. and P. Jung. 1992. Molluscan diversification and extinc- tion on opposite sides of the Isthmus of Panama [abstract]. Jn: Fifth North American Paleontological Convention, Abstracts and Programs, Field Museum of Natural History, Chicago, June 28 - July 1, 1992, Paleontological Society Special Publication 6:144. Jackson, J. B. C., P. Jung, A. G. Coates, and L. S. Collins. 1993. Diversity and extinction of tropical American mollusks and emergence of the Isthmus of Panama. Science 260:1624-1626. Jacobson, M. K. 1956. Malacological notes from western Cuba. Nautilus 70(1):1-3. Jacobson, M. K. 1968. The land Mollusca of St. Croix, Virgin Islands. Sterkiana 32:18-28. Jacobson, M. K. 1970. The land and freshwater gastropods collected in Cuba by M.G. Rutten. Bulletin Zoologisch Museum, Universiteit van Amsterdam 1(15):221-233. Jacobson, M. K. 1972. Malacology in Cuba. Sterkiana 45:37-42. Jacobson, M. K. 1974. Charles Wright (1811-1885) in Cuba as revealed by his letters. Sterkiana 53:1-6, [1 pl.]. Jacobson, M. K. and K. J. Boss. 1973. The Jamaican land shells described by C.B. Adams. Occasional Papers on Mollusks 3(47):305-519, pls. 54-91. Jarvis, P. W. 1902a. Notes on the distribution of the Pleurodonte acuta group. Nautilus 15(12):137-141. [Jamaica] Jarvis, P. W. 1902b. Notes on the distribution of the Pleurodonte sinuata group. Nautilus 16:1-4. [Jamaica] Jarvis, P. W. 1903. Distribution of Jamaican species of Colobostylus. Nautilus 17:62-65. 278 AMER. MALAC. BULL. 10(2) (1993) Jaume [Garcia], M. L. 1945a. Excursion malacologica al Pan de Guajaibén. Revista de la Sociedad Malacologica “Carlos de la Torre” 3(2):51-62, pls. 7-8. [Cuba] Jaume [Garcia], M. L. 1945b. Excursion malacologica al Pan de Guajai- boén (continuacién). Revista de la Sociedad Malacélogica “Carlos de la Torre” 3(3):73-83. [Cuba] Jaume [Garcia], M. L. 1945c. El género Oxynoe en Cuba occidental. Revista de la Sociedad Malacélogica “Carlos de la Torre” 3(1):19- 25, pl. 2. Jaume [Garcia], M. L. 1946. Moluscos marinos litorales del Cabo Catoche, Yucatan, México. Revista de la Sociedad Malacologica “Carlos de la Torre” 4(3):95-110. Jaume [Garcia], M. L. 1952. Moluscos terrestres de “El Palenque”, Matanzas. Circulares del Museo y Bibliotheca de Malacologia de la Habana Jan.-Feb. 1952:477-482. [Cuba] Jaume [Garcia], M. L. 1975a. Bibliografia de Miguel L. Jaume Garcia. Ciencias (La Habana, Cuba), series 4 (Ciencias Biol6égicas) (50):1- 47. [extensive bibliography of Jaume’s work, 1927-1974, mostly on Cuba] Jaume [Garcia], M. L. 1975b. Catalogo de la fauna Cubana - XXXVII. Catalogo de los moluscos terrestres cubanos de genero Cerion (Mollusca - Pulmonata - Ceriidae [sic]) (con una bibliografia gener- al). Ciencias (La Habana, Cuba), series 4 (Ciencias Biol6dgicas) (51):1-47. Jaume [Garcia], M. L. and P. Borro. 1946. Novedades en moluscos mari- nos cubanos. Revista de la Sociedad Malaceologica “Carlos de la Torre” 4(1):13-22, pl.2. Jaume [Garcia], M. L. and A. de la Torre. 1972. Los Urocoptidae de Cuba (Mollusca-Pulmonata). Circulares del Museo y Biblioteca de Zoologia de la Habana, pp.1526-1649. [Reprinted in Ciencias Biol6gicas, Universidad de la Habana series 4, 53:1-122 (1976)]} Jaume [Garcia], M. L. and I. Pérez Farfante. 1942. Moluscos pleistoceni- cos de la Zona Franca de Matanzas. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 16(1):37-44. [Cuba] Jaume [Garcia], M. L. and L. Sanchez de Fuentes. 1943. Revision de los moluscos cubanos del género Cryptelasmus. Revista de la Sociedad Malacélogica “Carlos de la Torre” 1(2):41-49, pl. 7. Jaume [Garcia], M. L. and H. Sarastia. 1943. Notas sobre moluscos mari- nos cubanos. Revista de la Sociedad Malacélogica “Carlos de la Torre” 1(2):52-61, pl. 9. Johnson, R. I. 1981. Recent and fossil Unionacea and Mutelacea (freshwa- ter bivalves) of the Caribbean Islands. Occasional Papers on Mollusks 4(60):269-288. Jones, D. S. and P. F. Hasson. 1985. History and development of the marine invertebrate faunas separated by the Central American isth- mus. In: The Great American Biotic Interchange, F. G. Stehli and S. D. Webb, eds., pp. 325-355. Plenum Publishing Company. Jung, P. 1965. Miocene Mollusca from the Paraguana Peninsula, Ven- ezuela. Bulletins of American Paleontology 49(223):385-652, pls. 50-79. Jung, P. 1968. Fossil Pleurotomaria and Haliotis from Barbados and Carriacou, West Indies. Eclogae Geologicae Helvetiae 61(2):593- 605. [Barbados, Grenadines] Jung, P. 1969a. Miocene and Pliocene mollusks from Trinidad. Bulletins of American Paleontology 55(247):289-657, pls. 13-60. Jung, P. 1969b. A Pliocene molluscan faunule from Trinidad. Tulane Studies in Geology and Paleontology 7(1-2):85-89. Jung, P. 1971. Fossil mollusks from Carriacou, West Indies. Bulletins of American Paleontology 61(269):147-262, pls. 1-21. [Grenadines] Jung, P. 1972. Mollusks from the White Limestone group of Jamaica. V/ Conferencia Geologica del Caribe, Margarita, Venezuela, Memorias 1972:465-468. Jung, P. 1973. Pleistocene pteropods - Leg 15, Site 147, Deep Sea Drilling Project. In: Initial Reports of the Deep Sea Drilling Project, by N. T. Edgar, J. B. Saunders, et al., Washington, United States Government Printing Office, 15:753-767, pls.1-5. [Ven- ezuela] Jung, P. 1974. Eocene mollusks from Curacao, West Indies. Verhandlungen der Naturforschenden Gesellschaft in Basel 84(1):483-500, pls. 1-7. Jung, P. 1975. Quaternary larval gastropods from Leg 15, Site 147, Deep Sea Drilling Project. Preliminary Report. Veliger 18(2):109-125, 19 pls. [Venezuela] Jung, P. 1986. Neogene paleontology in the northern Dominican Republic. 2. The genus Strombina (Gastropoda: Columbellidae). Bulletins of American Paleontology 90(324):5-42, pls. 1-14. Jung, P. 1987. Giant gastropods of the genus Campanile from the Caribbean Eocene. Eclogae Geologicae Helvetiae 80(3):889-896, pls. 1-3. Jung, P. and R. E. Petit. 1990. Neogene paleontology in the northern Dominican Republic. 10. The family Cancellariidae (Mollusca: Gastropoda). Bulletins of American Paleontology 98(334):87-144, pls.15-29, fold-out. Kaas, P. 1972. Polyplacophora of the Caribbean region. Studies on the Fauna of Curacao and other Caribbean Islands 41(137):1-162, pls.1-9. [general] Kaufmann, R. and K. -J. Gétting. 1970. Prosobranchia aus dem Litoral der karibischen Kiiste Kolumbiens. Helgoldnder Wissenschaftliche Meeresuntersuchungen 21:333-398. Keen, A. M. 1976. Pacific outposts of the Tertiary Caribbean province abstract]. Bulletin of the American Malacological Union 1975:46. Keen, A. M. and T. F. Thompson. 1946. Notes on the Gatun formation (Miocene), Panama Canal Zone [abstract]. Bulletin of the Geological Society of America 57:1260. Kemperman, Th. C. and H. E. Coomans. 1984. Studies on West Indian marine molluscs, 1. Risomurex mosquitensis, a new Caribbean species, with remarks on the status of the genus Risomurex (Gastropoda: Muricidae). Bulletin Zoologisch Museum, Universiteit van Amsterdam 10(1):1-7. [Costa Rica] Knudsen, J. 1982. Anomalodesmata (Mollusca, Bivalvia) from Saba Bank, the Caribbean region. Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen series C, 85(1):121-146. [Saba] Kobelt, W. 1880. Die geographische Verbreitung der Mollusken (Schluss). III. Die Inselfaunen. Jahrbiicher der Deutschen Malakozoo- logischen Gesellschaft 7:241-286. [general] Kojumdgieva, E. and A. de la Torre. 1986. Paleogene mollusks from cen- tral Cuba and their paleobiogeographic significance. In: Con- tributions of Bulgarian Geology, 27th International Geological Congress, Moscow, 1984, pp. 16-20. Drshavno Publishing House “Technika,” Sofia. [not seen] Kornicker, L. S., F. Bonet, R. Cann, and C. M. Hoskin. 1959. Alacran Reef, Campeche Bank, Mexico. Publications of the Institute of Marine Science, University of Texas 6:1-22, 1 map. [Yucatan] Krebs, H. 1864. See Clench, Aguayo and Turner, 1947-1948. Krebs, H. 1866. Remarks on some species of West Indian marine shells in the cabinet of Amherst College, Mass. Annals of the Lyceum of Natural History of New York 8(1867):394-398. [general] Krebs, H. 1873. On the Westindian [sic] Rissoinae. Proceedings of the Scientific Association of Trinidad 2(2):46-56. Kruckow, T. 1982. Some aspects of the development of Paleocaribbean molluscan faunas. Brenesia 19/20:353-357. Kugler, H. G. 1938. The Eocene of the Soldado Rock near Trinidad. Boletin de Geologia y Mineria (English edition) 2(2-4):1-24, map. Kugler, H. G., P. Jung, and J. B. Saunders. 1984. The Joes River For- MIKKELSEN ET AL.: BIBLIOGRAPHY OF CARIBBEAN MALACOLOGY 279 mation of Barbados and its fauna. Eclogae Geologicae Helvetiae 77(3):675-705. Lamy, E. 1929. Notes sur quelques lamellibranches de la Martinique. Bulletin du Muséum National d'Histoire Naturelle 2e serie, 1(3):201-208. Lewis, J. B. 1960. The fauna of rocky shores of Barbados, West Indies. Canadian Journal of Zoology 38:391-435. Lewis, J. B. 1965. A preliminary description of some marine benthic com- munities from Barbados, West Indies. Canadian Journal of Zoology 43:1049-1074. Lewis, J. B. and A. G. Fish. 1969. Seasonal variations of the zooplankton fauna of surface waters entering the Caribbean Sea at Barbados. Caribbean Journal of Science 9(1-2):1-24. [pteropods] Lipe, E. R. and R. T. Abbott. 1991. Living Shells of the Caribbean and Florida Keys. American Malacologists, Melbourne, Florida, 80 pp. Lorié, J. 1889. Fossile Mollusken von Curacao, Aruba, und der Kueste von Venezuela. Sammlungen des Geologischen Reichsmuseums in Leiden series 2, 1:111-149, 2 pls. [not seen] Lozet, J. B. and C. Pétron. 1977. Shells of the Caribbean. Publications- Elysées, France, 138 pp. Lyons, W. G. 1988. A review of Caribbean Acanthochitonidae (Mollusca: Polyplacophora) with descriptions of six new species of Acantho- chitona Gray, 1821. American Malacological Bulletin 6(1):79-114. MacDonald, D. F. 1919. The sedimentary formations of the Panama Canal Zone, with special reference to the stratigraphic relations of the fos- siliferous beds. Bulletin of the United States National Museum 103:525-545, pls. 153-154. Macsotay, O. and W. Scherer. 1972. Clasificacion numerica de las Turritellidae (moluscos) del Oligoceno al Holoceno, correspondi- entes a la Paleo-provincia del Caribe. Memoria IV Congreso Geolégico Venezolano, Tomo III, Boletin de Geologia, Publicacién Especial 5:1715-1734. Maes, V. O. 1983. Observations on the systematics and biology of a turrid gastropod assemblage in the British Virgin Islands. Bulletin of Marine Science 33(2):305-335. Malcolm, E. 1964. Eolid Mollusca from Jamaica, with descriptions of two new genera and three new species. Bulletin of Marine Science of the Gulf and Caribbean 14(1):1-32. Malek, E. A. 1969. Studies on “tropicorbid” snails (Biomphalaria: Planorbidae) from the Caribbean and Gulf of Mexico areas, includ- ing the southern United States. Malacologia 7(2-3):183-209. Manjarres, G. A. 1979. Observaciones sobre los moluscos e inventario de las especies en las Islas del Rosario. Anales del Instituto de Investigaciones Marinas de Punta de Betin 11:35-38. [Cuba] Mansfield, W. C. 1925. Miocene gastropods and scaphopods from Trinidad, British West Indies. Proceedings of the United States National Museum 66(22):1-65, pls. 1-10. Marcus, Ev. d. B. -R. 1976. Opisthobranchia von Santa Marta, Colombia. Studies on the Neotropical Fauna and Environment 11:119-150. Marcus, Ev. d. B. -R. 1977. An annotated checklist of the western Atlantic warm water opisthobranch molluscs. Journal of Molluscan Studies suppl. 4:1-22. Marcus, Ev. d. B. -R. and H. P. I. Hughes. 1974. Opisthobranch mollusks from Barbados. Bulletin of Marine Science 24(3):498-532. Marcus, Ev. d. B. -R. and Er. Marcus. 1960. Opisthobranchs from American Atlantic warm waters. Bulletin of Marine Science of the Gulf and Caribbean 10:129-203. Marcus, Ev. d. B.-R. and Er. Marcus. 1962. Opisthobranchs from Florida and the Virgin Islands. Bulletin of Marine Science 12(3):450-488. Marcus, Ev. d. B. -R. and Er. Marcus. 1963. Opisthobranchs from the Lesser Antilles. Studies on the Fauna of Curacao and other Caribbean Islands 19(79):1-76. [primarily Netherlands Antilles] Marcus, Ev. d. B. -R. and Er. Marcus. 1967. Opisthobranchs from the southwestern Caribbean Sea. Bulletin of Marine Science 17(3):597- 628. [Colombia, Panama] Marcus, Ev. d. B. -R. and Er. Marcus. 1970. Opisthobranchs from Curacao and faunistically related regions. Studies on the Fauna of Curacao and other Caribbean Islands 33(122):1-129. Marks, J. G. 1952. Especies vivientes de moluscos que se encuentran en las formaciones Terciarias de Venezuela. Acta Cientifica Venezolana 3(4):135-136. Martens, E. von. 1873. Die Binnenmollusken Venezuela’s. Festschr. Ges. Naturfreunde Berlin pp.157-225, pls. 1-2. Martens, E. von. 1877. Land- und Siisswasser-Schnecken von Puertorico. Jahrbuch der Deutschen Malakologischen Gesellschaft 4(4):340- 367, pl. 12. Martens, E. von. 1890-1901. Land and Freshwater Mollusca. Biologia Centrali-Americana. London, xxviii + 706 pp., 44 pls. Martin-Kaye, P. H. A. 1951. Sorting of lamellibranch valves on beaches in Trinidad, B.W.I. The Geological Magazine 88(6):432-434. Matley, C. A. 1926. The geology of the Cayman Islands (British West Indies) and their relation to the Bartlett Trough. Quarterly Journal of the Geological Society of London 82:352-386. Mattox, N. T. 1948. Observations on some uses for marine gastropods in Puerto Rico. Turtox News 26(7):2 pp. Maury, C. J. 1912a. A contribution to the paleontology of Trinidad. Journal of the Academy of Natural Sciences of Philadelphia series 2, 15:25-112, pls. 5-13. Maury, C. J. 1912b. A contribution to the paleontology of Trinidad. Proceedings of the Academy of Natural Sciences of Philadelphia 64:132-134. Maury, C. J. 1917. Santo Domingo type sections and fossils. Bulletins of American Paleontology 5(29):1-251 (165-415), pls. 1-39 (27-65) (part I, Mollusca); 5(30):i-111, 3-43 (419-459), pls. 1-3 (66-68), 1 table (part II, Stratigraphy). Maury, C. J. 1919. On the correlation of Porto Rican Tertiary formations with other Antillean and mainland horizons. American Journal of Science 48:209-215. Maury, C. J. 1920. Tertiary Mollusca from Porto Rico. New York Academy of Science, Scientific Survey of Porto Rico and the Virgin Islands 3(1):1-77, pls. 1-9. Maury, C. J. 1925. A further contribution to the paleontology of Trinidad (Miocene horizons). Bulletins of American Paleontology 10(42):1- 250 (153-402), pls. 1-43 (12-54). Mazé, H. 1874. Catalogue des coquilles terrestres et fluviatiles recueillies, a la Martinique, en 1873. Journal de Conchyliologie 22:158-173. Mazé, H. 1883. Catalogue révisé des mollusques terrestres et fluviatiles de la Guadeloupe et de ses dépendances. Journal de Conchyliologie 31:5-54, pls. 1-2. Mazé, H. 1890. Supplément au catalogue révisé des mollusques terrestres et fluviatiles de la Guadeloupe et de ses dépendances. Journal de Conchyliologie 38:19-34. McGinty, T. L. 1962. Caribbean marine shells. Nautilus 76:39-44. McLean, R. A. 1936. A list of bivalves from northern Cuba. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 10(1): 39-42. McLean, R. A. 1951. The Pelecypoda or bivalve mollusks of Porto Rico and the Virgin Islands. New York Academy of Science, Scientific Survey of Porto Rico and the Virgin Islands 17(1):1-183, 26 pls. Merian, P. 1844. Versteinerungen von Jamaika und Antigua. Bericht ueber die Verhandlungen der Naturforschenden Gesellschaft in Basel 6:63-64. Mestey-Villamil, V. 1979. Estudio Biosistematico de la Familia Neritidae (Mollusca: Archaeogastropoda) en Puerto Rico. Master's Thesis, 280 AMER. MALAC. BULL. 10(2) (1993) Department of Biology, University of Puerto Rico, Rio Piedras, xvili + 194 pp. [not seen] Mestey-Villamil, V. 1980. Checklist of the classes Gastropoda and Pelecypoda from Cayo Icacos, Puerto Rico. Science-Ciencia 7(1):19-24. [not seen] Mestey-Villamil, V. 1982. Los moluscos comestibles en Puerto Rico. Memorias del Tercer Simposio Sobre la Fauna de Puerto Rico, Departamento de Biologia, Colegio Universitario de Puerto Rico, Humacao 1982(Sept.):49-60. Meyer, K. B. 1977. Dorid nudibranchs of the Caribbean coast of the Panama Canal Zone. Bulletin of Marine Science 27(2):299-307. Michelson, E. H. 1953. Some remarks concerning the types of C. B. Adams’ West Indian Planorbidae. Occasional Papers of the Institute of Jamaica 9:1-4, pls. 1-2 [Jamaica] Miller, A. I. 1989. Lateral mixing and spatial resolution in molluscan assemblages of Smuggler’s Cove, St. Croix, U.S.V.I. Jn: Terrestrial and marine Geology of St. Croix, U.S. Virgin Islands, D.K. Hub- bard, ed., West Indies Laboratory, Teague Bay, St. Croix, Special Publication 8:129-134. [not seen] Miller, A. K. and M. L. Thompson. 1937. Beitraége zur Kenntnis tropisch- amerikanischer Tertiirmollusken. VI. Some Tertiary nautiloids from Venezuela and Trinidad. Eclogae Geologicae Helvetiae 30(1):59-73, pls. 7-10. [not seen] Miller, W. B. and E. Naranjo-Garcia. 1991. Familial relationships and bio- geography of the western American and Caribbean Helicoidea (Mollusca: Gastropoda: Pulmonata). American Malacological Bulletin 8(2):147-153. Mitchell-Tapping, H. J. 1979. The Caecidae (Gastropoda: Rissoacea) of Water Island, U. S. Virgin Islands, with a new species. Nautilus 93(2-3):103-105. Morch, O. A. L. 1859. Beitriige zur Molluskenfauna Central-Amerika’s. Malakozoologische Blatter 6:102-126. Morch, O. A. L. 1860. Beitrage zur Molluskenfauna Central-Amerika’s [continued]. Malakozoologische Blatter 7:66-106, 170-213. Morch, O. A. L. 1863. Contributions 4 la faune malacologique des Antilles danoises. Journal de Conchyliologie 11:21-43. [St. Thomas, St. Croix] March, O. A. L. 1874. Synopsis familiae scalidarum Indiarum occidentali- um. Oversigt over vestindiens scalarier. Videnskabelige Med- delelser fra Dansk Naturhistorisk Forening i Kjébenhavn (3) 6(17):250-268. [Also published separately with dual pagination - original and pp.1-19] [general] Morch, O. A. L. 1875-1877. Synopsis molluscorum marinorum indiarum occidentalium imprimis insularum danicarum. Malakozoologische Bldtter 22:142-184 (1875); 23:87-143 (1876); 23:45-58 (1876); 24:14-66 (1877); 24:93-123 (1877). March, O. A. L. 1876. A descriptive catalogue of the Scalidae of the West India islands. Journal of the Academy of Natural Sciences of Philadelphia series 2, 8(2):189-207, pl. 29. Morch, O. A. L. 1878. Catalogue of West-India Shells in the Collection of Dr. C.M. Poulsen. Bianco Luno, Copenhagen, 16 pp. Mongin, D. 1968. Les pectinides du Miocene de la Guadeloupe (Antilles francaises). Bulletins of American Paleontology 54(245):471-510, pls. 40-50. Moolenbeek, R. G. and M. J. Faber. 1984. Studies on West Indian marine molluscs: 3. A new gastropod genus and species from Bonaire, Netherlands Antilles. Studies on the Fauna of Curagao and Other Caribbean Islands 67(206):98-103. Moolenbeek, R. G. and M. J. Faber. 1989. Studies on West Indian marine mollusks. 10. The genus Euchelus (Prosobranchia: Trochidae) in the West Indies. Studies in honour of Dr. Pieter Wagenaar Hum- melinck, Foundation for Scientific Research in Surinam and the Netherlands Antilles, Amsterdam 123:217-226. [not seen] Moore, D. R. 1972. Ecological and systematic notes on Caecidae from St. Croix, U. S. Virgin Islands. Bulletin of Marine Science 22(4):881- 899. Moore, D. R. 1973. Mollusks from a small land-locked Mexican lagoon. Bulletin of the American Malacological Union 1972:5-6. [Cozumel] Moore, D. R. 1977. From the first cruise of the “Blake” to the present day - one hundred years of progress in the study of Gulf and Caribbean marine molluscs. Bulletin of the American Malacological Union 1977:63-68. Moore, J. C. 1850. On some Tertiary beds in the island of San Domingo: from notes by J.S. Heniker, esq., with remarks on the fossils. Quarterly Journal of the Geological Society of London 6:39-53, pls. 9-10. [includes “Descriptions of new species of fossil shells found by J. S. Heniker, esq.” by G. B. Sowerby, pp.44-53, pls. 9-10] Moore, J. C. 1853. Notes on the fossil Mollusca and fish from San Domingo. In: On some Tertiary deposits in San Domingo, by T. S. Heneken, pp. 129-132. Quarterly Journal of the Geological Society 9:115-134. Moore, J. C. 1863. On some Tertiary shells from Jamaica. Quarterly Journal of the Geological Society 19:510-513. Morelet, A. 1849-1851. Testacea Novissima Insulae Cubanae et Americae Centralis. Paris, 1:31 pp., 11:30 pp. [Cuba, general] Morris, P. A. 1973. A Field Guide to Shells of the Atlantic and Gulf Coasts and the West Indies (3rd ed. by W.C. Clench). Houghton Mifflin Co., Boston, xxviii + 330 pp. Muchmore, W. B. 1993. List of terrestrial invertebrates of St. John, U. S. Virgin Islands (exclusive of Acarina and Insecta), with some records of freshwater species. Caribbean Journal of Science 29(1-2):30-38. Mulleried, F. K. G. 1951. Paquidontos nuevos del Cretacio superior de Cuba. Revista de la Sociedad Malacélogica “Carlos de la Torre” 8(2):83-92, pl. 12. Nicol, D. 1945. Restudy of some Miocene species of Glycymeris from Central America and Colombia. Journal of Paleontology 19(6):622- 624, pl. 85. [Colombia, Costa Rica, Panama, general] Noblet, G. P. and R. T. Damian. 1991. On the status of Biomphalaria and Thiara snails and the threat of schistosomiasis and paragonimiasis on Dominica, West Indies. Journal of Medical and Applied Mal- acology 3:1-6. Nowell-Usticke, G. W. 1956. Marine shells from Altona Lagoon, St. Croix, Virgin Islands. Nautilus 70(4):113-116. Nowell-Usticke, G. W. 1959a. A Check List of the Marine Shells of St. Croix, U.S. Virgin Islands, with Random Notations. Lane Press, Burlington, Vermont, 90 pp., 4 pls. Nowell-Usticke, G. W. 1959b. Dredging at St. Croix. New York Shell Club Notes 54:4-S. Nowell-Usticke, G. W. 1961a. The Ham Bluff splash pool zone at St. Croix. New York Shell Club Notes 71:2-S. Nowell-Usticke, G. W. 1961b. A holiday shelling on St. Croix. New York Shell Club Notes 77:4-5. Nowell-Usticke, G. W. 1961c. A trip to the Isle St. Barthélemy. New York Shell Club Notes 72:5-6. Nowell-Usticke, G. W. 1962. Shelling in the Dutch Antilles. New York Shell Club Notes 86:6-7. Nowell-Usticke, G. W. 1963a. Dredging operations in Christiansted Harbor - 1962-1963. New York Shell Club Notes 94:4. [St. Croix] Nowell-Usticke, G. W. 1963b. Southgate Pond. New York Shell Club Notes 89:8. [St. Croix] Nowell-Usticke, G. W. 1964. Shelling in Trinidad. New York Shell Club Notes 102:5. MIKKELSEN ET AL.: BIBLIOGRAPHY OF CARIBBEAN MALACOLOGY 281 Nowell-Usticke, G. W. 1968. Caribbean Cones from St. Croix and the Lesser Antilles. Livingston Publishing Company, Narberth, Pennsylvania, 31 pp., 4 pls. [St. Croix, Lesser Antilles] Nowell-Usticke, G. W. 1969. A Supplementary Listing of New Shells (illustrated). To be added to the check list of the marine shells of St. Croix. Privately published, 32 pp., 6 pls. Nowell-Usticke, G. W. 1971. A Supplementary Listing of New Shells (illustrated). Revised edition to be added to the check list of the marine shells of St. Croix. Privately published, 32 pp., 6 pls. Nutting, C. C. 1919. Barbados-Antigua Expedition. Narrative and prelimi- nary report of a zoological expedition from the University of Iowa to the Lesser Antilles under the auspices of the Graduate College. University of Iowa Studies in Natural History 8(3):1-274, pls. 1-50. {Lesser Antilles, Puerto Rico] Old, W. E. 1979. The Usticke Collection. New York Shell Club Notes 249:3-4. [St. Croix] Olsson, A. A. 1922. The Miocene of northern Costa Rica, with notes on its general stratigraphical relations. Bulletins of American Paleontology 9(39):179-340 (1-168) (part 1); 9(39):341-481 (169- 309), pls. 4-35 (1-32) (part 2, Class Pelecypoda). Olsson, A. A. 1942. Tertiary deposits of northwestern South America and Panama. Proceedings of the Eighth American Sci. Congress 4:231- 287. [Panama, Colombia, Venezuela] Olsson, A. A. 1967a. Pustularias (Jenneria) in the American Neogene. Notulae Naturae 403:1-13, pls. 1-2. [general] Olsson, A. A. 1967b. Some Tertiary Mollusks from South Florida and the Caribbean. Paleontological Research Institute, Ithaca, New York, 61 pp., 9 pls. [Reprinted by P.R.I., 1993, Special Publication 19:13- 75, pls. 1-9] Olsson, A. A. 1972. Origin of the existing Panamic molluscan biotas in terms of their geological history and their separation by the isthmi- an land barrier. Bulletin of the Biological Society of Washington 2:117-123. [Panama] Olsson, A. A. and T. L. McGinty. 1958. Recent marine mollusks from the Caribbean coast of Panama with the description of some new gen- era and species. Bulletins of American Paleontology 39(177):1-58, pls. 1-5. Ortiz-Corps, E. [1985]. An Annotated Checklist of the Recent Marine Gastropoda (Mollusca) from Puerto Rico. University of Puerto Rico, Humacao University College, Humacao, ii + 220 pp. [date teste Zoological Record for 1986]. Palmer, K. v. W. 1923. Foraminifera and a small molluscan fauna from Costa Rica. Bulletins of American Paleontology 10(40):1-18, pls. 1-2. Palmer, K. v. W. 1938. Neocene spondyli from the southern United States and tropical America. Paleontographica Americana 2(8):1-18, pls. 1-3. Palmer, K. v .W. 1945. Fossil fresh-water Mollusca from the state of Monagas, Venezuela. Bulletins of American Paleontology 31(118):1-34, pls. 1-3. Palmer, K. v. W. 1947. Manuscript notes by P.P. Carpenter on d’Orbigny’s series of Cuban shells in the British Museum. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 19(1):89-104. Palmer, K. v. W. 1967. A comparison of certain Eocene molluscs of the Americas with those of the Western Tethys. Systematics Assoc- lation Publication 7:183-193. [general] Paul, C. R. C. 1982. The Jamaican land snail genera Geoscala and Simplicervix (Pulmonata: Urocoptidae). Journal of Conchology 31(2):101-127. Paul, C. R. C. 1983. The comparative anatomy and systematics of Jamaican Urocoptidae (Mollusca: Pulmonata). Zoological Journal of the Linnean Society 78(1):1-28. Percharde, P. L. 1968. Notes on distribution and underwater observations on the molluscan genus Strombus as found in the waters of Trinidad and Tobago. Caribbean Journal of Science 8(1-2):47-55, 2 maps. Percharde, P. L. 1970. Further underwater observations on the molluscan genus Strombus Linne as found in the waters of Trinidad and Tobago. Caribbean Journal of Science 10(1-2):73-81, pls. 1-3. Percharde, P. L. 1982. A comparison of the Strombus (Mollusca) colonies, of two southern Caribbean islands - Trinidad and Grenada. Caribbean Journal of Science 18(1-4):35-39. Perera, G., R. Sanchez, M. Yong, J. R. Ferrer, and O. Amador. 1986. Ecology of some freshwater pulmonates from Cuba. Malacological Review 19:99-104. Perera, G. and M. Yong. 1984. The influence of some abiotic factors on the distribution of freshwater mollusks on the Isle of Youth (Isle of Pines), Cuba. Walkerana 2(7):131-139. Pérez Farfante, I. 1940. Adiciones a la lista de moluscos Cubanos. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 14(1):69-73, pl. 13. Pérez Farfante, I. 1942. Moluscos de la regi6n de Camoa y Somorrostro y sus condiciones de vida. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 16(1):45-56. [Cuba] Perrilliat, M. d. C. 1981. Malacofauna miocenica de Mexico y su signifi- cacion paleogeografica. Annais II Congresso Latino-Americano Paleontologia, Porto Alegre, April 1981:581-590. Petit de la Saussaye, S. 1851. Catalogue des coquilles trouvées a I’ile de la Guadeloupe, par M. Beau. Journal de Conchyliologie 2:422-430. Petit de la Saussaye, S. 1853. Supplément au catalogue des coquilles trou- vées a l’ile de la Guadeloupe, par M. Beau. Journal de Conch- yliologie 4:413-419. Petit de la Saussaye, S. 1856. Deuxiéme supplément au catalogue des coquilles trouvées a la Guadeloupe. Journal de Conchyliologie 5:149-158. Petuch, E. J. 1976. An unusual molluscan assemblage from Venezuela. Veliger 18(3):322-325. Petuch, E. J. 1979. A new species of Siphocypraea (Gastropoda: Cypraeidae) from northern South America with notes on the genus in the Caribbean. Bulletin of Marine Science 29(2):216-225. Petuch, E. J. 1981a. A relict Neogene caenogastropod fauna from northern South America. Malacologia 20(2):307-347. [Colombia, Ven- ezuela] Petuch, E. J. 1981b. A volutid species radiation from northern Honduras, with notes on the Honduran Caloosahatchian Secondary Relict Pocket. Proceedings of the Biological Society of Washington 94(4):1110-1130. Petuch, E. J. 1982. Geographical heterochrony: comtemporaneous [sic] coexistence of Neogene and Recent molluscan faunas in the Americas. Palaeogeography, Palaeoclimatology, Palaeoecology 32:277-312. Petuch, E. J. 1987. New Caribbean Molluscan Faunas. Coastal Education Research Foundation, Charlottesville, Virginia, 154 + 4 pp. Petuch, E. J. 1988. Neogene History of Tropical American Mollusks. Coastal Education Research Foundation, Charlottesville, Virginia, 217 pp. Petuch, E. J. 1990. A new molluscan faunule from the Caribbean coast of Panama. Nautilus 104(2):57-71. Pfeiffer, L. 1839. Bericht tiber die Ergebnisse meiner Reisse nach Cuba im Winter 1838-1839. Archiv fiir Naturgeschichte, Sth year, 1:346-358. Pfeiffer, L. 1840. Uebersicht der im Januar, Februar und Marz 1839 auf Cuba gesammelten Mollusken. Archiv fiir Naturgeschichte, 6th year, 1:250-261. Pfeiffer, L. 1846. Descriptions, by Dr. L. Pfeiffer, of new species of land- 282 AMER. MALAC. BULL. 10(2) (1993) shells from Jamaica, collected by Mr. Gosse. Proceedings of the Zoological Society of London 1845:137-138. Pfeiffer, L. 1856. Verzeichniss der bisher bekannt gewordenen gedeckel- ten Landschnecken von Cuba. Malakozoologische Bldtter 3:118- 150. Pfeiffer, L. 1858a. Fernere Nachrichten tiber die Mollusken-Fauna der Insel Cuba. Malakozoologische Blatter 5:1-12. Pfeiffer, L. 1858b-1864. Zur Molluskenfauna von Cuba. Malako- zoologische Blétter 5:37-49, 173-196 (1858); 6:66-102 (1860) [enti- tled “Zur Molluskenfauna der Insel Cuba’’]; 7:9-32 [continued, 1860, with J. Gundlach]; 9:1-10 [continued, 1862]; 9:121-133 [con- tinued, 1862]; 10:180-199 [continued, 1863]; 10:245-248 [contin- ued, 1863]; 11:2-19 [continued, 1864]; 11:102-109 [continued, 1864]. Pflug, H. D. 1961. Mollusken aus dem Tertiar von Santo Domingo. Acta Humboldtiana, Series Geologica et Paleontologica 1:i-vi, 1-107, pls. 1-26. Piaget, J. 1914. Quelques mollusques de Colombie. /n: Voyage d’Explor- ation Scientifique en Colombie, by O. Fuhrmann and E. Mayor, pp.253-269, pls. 9-10. Mémoires de la Société Neuchdateloise des Sciences Naturelles 5:1-1090. Pickford, G. E. 1950. The Octopoda of the Oxford University Cayman Expedition. Proceedings of the Malacological Society of London 28:139-144. Pilsbry, H. A. 1891. Land and fresh-water mollusks collected in Yucatan and Mexico. Proceedings of the Academy of Natural Sciences of Philadelphia (2):310-334, pls. 14-15. Pilsbry, H. A. 1910a. Land Mollusca of the Panama Canal Zone. Proceedings of the Academy of Natural Sciences of Philadelphia 62:502-509, pl. 37. Pilsbry, H. A. 1910b. A new Haitian Oligocene horizon. Proceedings of the Academy of Natural Sciences of Philadelphia 62:487-489. Pilsbry, H. A. 1911. Scaphopoda of the Jamaican Oligocene and Costa Rican Pliocene. Proceedings of the Academy of Natural Sciences of Philadelphia 63:165-169. Pilsbry, H. A. 1912. Land shells of Carthagena [sic], Colombia. Nautilus 26:84. Pilsbry, H. A. 1920a. Costa Rican land and freshwater mollusks. Proceedings of the Academy of Natural Sciences of Philadelphia 72:2-10. Pilsbry, H. A.1920b. Mollusca from Central America and Mexico. Proceedings of the Academy of Natural Sciences of Philadelphia 71(1919):212-223, pl. 11. [Mexico, Guatemala, general] Pilsbry, H. A. 1920c. Mollusca from Central America and Mexico [contin- ued from Pilsbry, 1920b]. Proceedings of the Academy of Natural Sciences of Philadelphia 72:195-201. [Panama, Guatemala, Mexico, Nicaragua, Costa Rica] Pilsbry, H. A. 1922. Revision of W. M. Gabb’s Tertiary Mollusca of Santo Domingo. Proceedings of the Academy of Natural Sciences of Philadelphia 73(2):305-435, pls. 16-47. Pilsbry, H. A. 1926a. Costa Rican land shells collected by A. A. Olsson. Proceedings of the Academy of Natural Sciences of Philadelphia 78:127-133, pl. 11. Pilsbry, H. A. 1926b. The land mollusks of the Republic of Panama and the Canal Zone. Proceedings of the Academy of Natural Sciences of Philadelphia 78:57-126, pls. 9-10. Pilsbry, H. A. 1927. Revision of the Ampullariidae of Jamaica and Cuba. Proceedings of the Academy of Natural Sciences of Philadelphia 79:247-253, pl. 21. Pilsbry, H. A. 1929. Studies on West Indian mollusks. Proceedings of the Academy of Natural Sciences of Philadelphia 80(1928):581-606, pls. 28-30 ({part I], The genus Zachrysia); 81:449-467, pls. 12-16 (part II, The locomotion of Urocoptidae and descriptions of new forms). [Cuba] Pilsbry, H. A. 1930a. Results of the Pinchot South Sea Expedition, — I. Land mollusks of the Caribbean islands, Grand Cayman, Swan, Old Providence, and St. Andrew. Proceedings of the Academy of Natural Sciences of Philadelphia 82:221-261, pls. 15-19. [Cayman Islands, Honduras, general] Pilsbry, H. A. 1930b. Results of the Pinchot South Sea Expedition. — II. Land mollusks of the Canal Zone, the Republic of Panama, and the Cayman Islands. Proceedings of the Academy of Natural Sciences of Philadelphia 83:339-354, pls. 28-30. Pilsbry, H. A. 1933. Santo Domingan land mollusks collected by Daniel C. Pease, 1932, and by A. A. Olsson, 1916. Proceedings of the Academy of Natural Sciences of Philadelphia 85:121-162, pls. 6-11. Pilsbry, H. A. 1942a. Land Mollusca of the Cayman Islands collected by the Oxford University Biological Expedition, 1938. Nautilus 56(1):1-9, pl. 1. Pilsbry, H. A. 1942b. Helicodiscus in the West Indies. Nautilus 56(2):55- 57. [Cuba] Pilsbry, H. A. 1946. The type specimens of C.B. Adam’s Jamaican species of Vitrinella. Notulae Naturae 162:1-5, 1 pl. Pilsbry, H. A. 1949. Land mollusks of Cayman Brac. Nautilus 63(2):37- 48, pls. 3-4. Pilsbry, H. A. and C. G. Aguayo. 1933. Marine and freshwater mollusks new to the fauna of Cuba. Nautilus 46(4):116-123. Pilsbry, H. A. and A. P. Brown. 1910. The Mollusca of Mandeville, Jamaica, and its environs. Proceedings of the Academy of Natural Sciences of Philadelphia 62:510-535. Pilsbry, H. A. and A. P. Brown. 1912. The land Mollusca of Montego Bay, Jamaica; with notes on the land Mollusca of the Kingston region. Proceedings of the Academy of Natural Sciences of Philadelphia 63:572-588, pl. 43. Pilsbry, H. A. and A. P. Brown. 1914. List of land and fresh-water mol- lusks of Antigua. Proceedings of the Academy of Natural Sciences of Philadelphia 66:429-431. Pilsbry, H. A. and A. P. Brown. 1917. Oligocene fossils from the neigh- borhood of Cartagena, Colombia, with notes on some Haitian species. Proceedings of the Academy of Natural Sciences of Philadelphia 69:32-41, pls. 5-6. Pilsbry, H. A. and G. H. Clapp. 1902. New land shells from the Santa Marta mountains, Colombia. Nautilus 15(12):133-137, pl. 7. Pilsbry, H. A. and C. W. Johnson. 1917. New Mollusca of the Santo Domingan Oligocene. Proceedings of the Academy of Natural Sciences of Philadelphia 69:150-202. Pilsbry, H. A. and T. L. McGinty. 1939. Small Cassididae of Florida and the West Indies. Nautilus 52(3):73-77, pl. 5. Pilsbry, H. A. and A. A. Olsson. 1954. Miocene land shell fossils from the Dominican Republic. Notulae Naturae 266:1-4. Pilsbry, H. A. and B. Sharp. 1898. Scaphopoda of the San Domingo Tertiary. Proceedings of the Academy of Natural Sciences of Philadelphia 49:465-476, pls. 10-11. Pilsbry, H. A. and E. G. Vanatta. 1928. Land shells of Tortuga Island, Haiti, and a new Haitian Oleacina. Proceedings of the Academy of Natural Sciences of Philadelphia 80:475-478, pl. 27. Piplani, S. 1977. More on Caribbean chitons. Of Sea and Shore 8(3):192- 193, 197-198. Pittier, H. 1890. Apuntamientos para la historia natural de Costa Rica. Mollusca. Anales del Instituto Fisico-geogradfico y del Museo Nacional de Costa Rica 3:123-126. [not seen] Poey, F. 1851-1861. Memorias Sobre la Historia Natural de la Isla de Cuba, Acompanadas de Sumarios Latinos y Extractos en Frances. Habana, Vol. 1 [463 pp., 34 pls.], Vol. 2 [442 pp., 19 pls.]. MIKKELSEN ET AL.: BIBLIOGRAPHY OF CARIBBEAN MALACOLOGY 283 Poinar, G. O. and B. Roth. 1991. Terrestrial snails (Gastropoda) in Dominican amber. Veliger 34(3):253-258. Pointier, J. -P. 1974. Faunae malacologique dulgaquicole de I’ile de la Guadeloupe (Antilles frangaises). Bulletin du Muséum National @’Histoire Naturelle 3e series, (235 [Zoologie, 159]):905-933, pls. 1-3. Pointier, J. -P. 1976. Répartition locale et biogéographie des mollusques dulgaquicoles de l’ile de la Guadeloupe (Antilles frangaises). Malacological Review 9:85-103. Potter, G. E. 1946. Some marine gastropod shells from Puerto Rico and Mona Island. Proceedings and Transactions of the Texas Academy of Science 29:80-84. [not seen] Prentice, M. A. 1980. Schistosomiasis and its intermediate hosts in the Lesser Antillean islands of the Caribbean. Bulletin of the Pan American Health Organization 14:258-268. [general; not seen] Princz, D. 1973. Moluscos gastrépodos y pelecipodos del Estado Nueva Esparta, Venezuela. Memorias de la Sociedad de Ciencias Naturales La Salle 33:169-222, 4 pls. Princz, D. 1980. Los moluscos marinos del Golfo de Venezuela. Estacion de Investigaciones Marinas de Margarita, Fundacion La Salle de Ciencias Naturales Contribuci6n 78:51-76. Princz, D. 1982. Lista y bibliografia de los gastr6podos marinos vivos de los mares de Venezuela, Trinidad e Islas de Sotavento. Boletin de la Sociedad Venezolana de Ciencias Naturales 37:103-147. [not seen] Princz, D. 1983. Taxonomia y ecologia de los micromoluscos bentonicos representativos del Golfo de Venezuela. Memorias de la Sociedad de Ciencias Naturales La Salle 43(120):41-58. Princz, D. 1987. A first account of the marine mollusks of La Blanquilla Island, Venezuela. Malacology Data Net 2(1-2):5-14. Princz, D. and L. A. Gonzalez C. 1981. Los moluscos holoplanctonicos del Golfo de Venezuela y una interpretacion de su funcion como indicadores ecologicos en la zona centro occidental del caribe Venezolano. Boletin del Instituto de Oceanogrdafico, Universidad de Oriente 20(1-2):85-90. Princz, D. and A. Gonzalez de Pacheco. 1981. Los moluscos marinos del Parque Nacional Laguna de la Restinga, Isla de Margarita, Venezuela. Estaci6én de Investigaciones Marinas de Margarita, Contribucién 98:1-32 + 5 unnum. pp., 10 unnum. pls. Radwin, G. E. 1969. A Recent molluscan fauna from the Caribbean coast of southeastern Panama. Transactions of the San Diego Society of Natural History 15(14):229-236. Ramirez, R. 1950. Descripcioén de algunos moluscos del Mioceno del Valle del Cibao de la Reptiblica Dominicana. Publicaciones de la Universidad de Santo Domingo series 4 (1949), 70(1):3-39, pls. 1-7. Ramirez, R. 1956. Paleontologia Dominicana. Publicaciones de la Universidad de Santo Domingo series 4, 103(2):1-26, pls. 1-8. Ramsden, C. T. 1914. Notes on some land shells of eastern Cuba. Nautilus 28(5):49-51, pl .2. Read, K. R. H. 1964. Ecology and environmental physiology of some Puerto Rican bivalve molluscs and a comparison with boreal forms. Caribbean Journal of Science 4(4):459-465. Rees, W. J. 1950. Notes on Cephalopoda from the Caribbean. Pro- ceedings of the Malacological Society of London 28(2/3):107-114, pls. 10-14. Rehder, H. A. 1942. Some new land shells from Costa Rica and Panama. Journal of the Washington Academy of Science 32(11):350-352. Rehder, H. A. 1943. New marine mollusks from the Antillean region. Proceedings of the United States National Museum 93:187-203, pls. 19-20. Rehder, H. A. 1954. Mollusks. In: Gulf of Mexico, its origin, waters, and marine life. United States Department of the Interior, Fish and Wildlife Service, Fishery Bulletin 89:469-474. [definition of Caribbean Province] Rehder, H. A. 1962a. Contribucién al conocimiento de los moluscos mari- nos del archipielago de Los Roques y la Orchila. Memorias de la Sociedad de Ciencias Naturales La Salle 22(62):116-138. [Venezuela] Rehder, H. A. 1962b. The Pleistocene mollusks of Grand Cayman Island with notes on the geology of the island. Journal of Paleontology 36(3):583-585. Rehder, H. A. 1966. The non-marine mollusks of Quintana Roo, Mexico with the description of a new species of Drymaeus (Pulmonata: Bulimulidae). Proceedings of the Biological Society of Washington 79:273-296. Rehder, H. A. 1981. The Audobon Society Field Guide to North American Seashells. Alfred A. Knopf, New York, 894 pp., color pls. Rehder, H. A. and R. T. Abbott. 1951. Some new and interesting mollusks from the deeper waters of the Gulf of Mexico. Revista de la Sociedad Malacologica “Carlos de la Torre” 8(2):53-66, pls. 8-9. [Yucatan] Rice, W. H. and L. S. Kornicker. 1962. Mollusks of Alacran Reef, Campeche Bank, Mexico. Publications of the Institute of Marine Science, University of Texas 8:366-403, pls. 1-9. Rice, W. H. and L. S. Kornicker. 1965. Mollusks from the deeper waters of the northwestern Campeche Bank, Mexico. Publications of the Institute of Marine Science, University of Texas 10:108-172, pls. 1-16. Richards, C. S. 1964. Puerto Rican species of Tropicorbis and Drepanotrema; comparison with Australorbis glabratus and other planorbids. Malacologia 2:105-129. Richards, H. G. 1933. A conchological expedition to Cuba. Proceedings of the Pennsylvania Academy of Science 7:167-172. Richards, H. G. 1935. Pleistocene mollusks from western Cuba. Journal of Paleontology 9(3):253-258, pl. 25. Richards, H. G. 1937. Land and freshwater mollusks from the island of Cozumel, Mexico, and their bearing on the geological history of the region. Proceedings of the American Microscopical Society 77:249- 262, pls. 1-4. Richards, H. G. 1938. Land mollusks from the island of Roatan, Honduras. Proceedings of the American Philosophical Society 79(2):167-178, pls. 1-3. Richards, H. G. 1939. Land mollusks from Corn Island, Nicaragua. Proceedings of the American Philosophical Society 81(1):29-36. Richards, H. G. 1955. The geological history of the Cayman Islands. Notulae Naturae 284:1-11. Richards, H. G. and P. Wagenaar Hummelinck. 1940. Land and freshwater mollusks from Margarita Island, Venezuela. Notulae Naturae 62:1-16. Righi, G. 1968. On the radulae and spines of some Polyplacophora and Archaeogastropoda from Curagao. Studies on the Fauna of Curacao and other Caribbean Islands 25(99):73-82. Robart, G., G. Mandahl-Barth, and C. Ripert. 1976. Inventaire, repartition geographique et ecologie des mollusques dulgaquicoles d’ Haiti (Caraibes). Haliotis 8:159-171 bis. Robertson, R. 1963. The mollusks of British Honduras. Proceedings of the Philadelphia Shell Club 1(7):15-20. Robinson, D. G. 1991. The Systematics and Paleoecology of the Prosobranch Gastropods of the Pleistocene Moin Formation of Costa Rica. Doctoral dissertation, Tulane University, New Orleans, Lousiana, 749 pp. [not seen] Robinson, D. G. 1993. The zoogeographic implications of the prosobranch gastropods of the Moin Formation of Costa Rica. American Malacological Bulletin 10(2):251-255. Robinson, D. G. and J. M. Montoya. 1988. Los moluscos marinos de la 284 AMER. MALAC. BULL. 10(2) (1993) costa Atlantica de Costa Rica. Revista de Biologia Tropical 35:375- 400. Rodriguez, G. 1959. The marine communities of Margarita Island, Venezuela. Bulletin of Marine Science of the Gulf and Caribbean 9(3):237-280. Rolan, E. 1992. The family Omalogyridae G. O. Sars, 1878 (Mollusca, Gastropoda) in Cuba with description of eight new species. Apex 7(2):35-46. [not seen] Rolan, E. and J. Espinosa. 1992a. La familia Cerithiopsidae H. y A. Adams, 1853 (Mollusca, Gastropoda), en la Isla de Cuba. 1. El genero Retilaskeye Marshall, 1978. Publicagoes Ocasionis da Sociedade Portuguesa de Malacologia (16):39-43, 1 pl. [not seen] Rolan, E. and J. Espinosa. 1992b. La familia Cerithiopsidae H. y A. Adams, 1853 (Mollusca, Gastropoda), en la Isla de Cuba. 2. El genero Horologica Laseron, 1956. Publicagoes Ocasionis da Sociedade Portuguesa de Malacologia (16):45-49, 1 pl. [not seen] Rolan, E., J. Espinosa, and R. Fernandez-Garcés. 1990. Los generos Sansonia y Mareleptopoma en la Isla de Cuba (Mollusca, Gastro- poda). Thalassas 8:85-91. Rosenberg, G. 1993. A database approach to studies of molluscan taxono- my, biogeography and diversity, with examples from western Atlantic marine gastropods. American Malacological Bulletin 10(2):257-266. Rosewater, J. 1975. Mollusks of Gatun Locks, Panama Canal. Bulletin of the American Malacological Union 1974:42-43. Rosewater, J. 1976. Some results of the National Museum of Natural History-Smithsonian Tropical Research Institute survey of Panama 1971-1975. Bulletin of the American Malacological Union 1975: 48-50. Riitzler, K. and I. G. Macintyre, eds. 1982. The Atlantic barrier reef ecosystem at Carrie Bow Cay, Belize, I: structure and communities. Smithsonian Contributions to Marine Science 12:xiv + 539 pp. [col- lection of papers, see esp. those by Riitzler and Macintyre, Young and Young, Waller and Macintyre] Rush, W. H. 1891. List of species collected on the islands of St. Thomas, St. Kitts, Barbados, Jamaica, and at Pensacola, Florida, with prefac- tory notes. Nautilus 5(6):65-70. Russell, H. D. 1941. The Recent mollusks of the family Neritidae of the western Atlantic. Bulletin of the Museum of Comparative Zoology at Harvard College 88(4):347-404, pls. 1-7. Russell Hunter, W. 1955. Endemicism in the snails of Jamaica. Glasgow Naturalist 17(pt.4, 1954):173-183. Rutsch, R. 1934a-1942a. Beitrige zur Kenntnis tropisch-amerikanischer Tertiarmollusken. Eclogae Geologicae Helvetiae 27(1):1-9 (1934, part I, Angebliche Rudisten aus dem Tertiar von Trinidad [Brit. Westindien]; 27(2):299-326, pl. 8 (1934, part II, Pteropoden und Heteropoden aus dem Miocaen von Trinidad [Brit. Westindien]); 35(2):213-223, pl. 7 (1942, part VII, Larkinien (Arcidae) aus dem Jungtertiér von Trinidad [B.W.I.]). Rutsch, R. 1934b. Die Gastropoden aus dem Neogene der Punta Gavilan in Nord-Venezuela. Abhandlungen der Schweizerischen Pala- ontologischen Gesellschaft 54(3):1-88, pls. 1-7 (part 1); 55(1):89- 169, pls. 8-9 (part 2). Rutsch, R. 1940. The Molluscs from the Manzanilla Beds (Middle Miocene) of Trinidad (B.W.I.). Private report (fide Woodring, 1966). [not seen] Rutsch, R. 1942b. Die Mollusken der Springvale-Schichten (Obermiocaen) von Trinidad (Britisch-West-Indien). Verhand- lungen der Naturforschenden Gesellschaft in Basel 54:96-182, pls. 3-9. Rutsch, R. 1942c. Preliminary Report on the Molluscs from the Matura Beds of Trinidad (B.W.1.). Private report (fide Woodring, 1966). (not seen] Rutsch, R. 1943. Die Paleocaen-Mollusken der Inseln Trinidad und Soldado Rock. Eclogae Geologicae Helvetiae 36:139-192. [not seen] Salisbury, A. E. 1953. Mollusca of the University of Oxford Expedition to the Cayman Islands in 1938. Proceedings of the Malacological Society of London 30:39-54, pls. 7-8. Sanchez Roig, M. 1926. La fauna Cretdcica de la regién central de Cuba. Memorias de la Sociedad Cubana de Historia Natural “Felipe Poey” 7:93-102, 10 pls. [not seen] Sanchez Roig, M. 1948. Nuevos moluscos de Cuba. Revista de la Socie- dad Malacologica “Carlos de la Torre” 6(2):67-72. Sanchez Roig, M. 1949. Nuevos moluscos de Cuba. Revista de la Sociedad Malacologica “Carlos de la Torre” 6(3):85-88, pl. 3. Sanchez Roig, M. 1951a. Nuevos moluscos del género Liguus. Revista de la Sociedad Malacélogica “Carlos de la Torre” 8(1):23-38, pls. 4- 5. [Cuba] Sanchez Roig, M. 1951b. Nuevos moluscos terrestres cubanos. Revista de la Sociedad Malacologica “Carlos de la Torre” 7(3):95-120, pls. 15-19. Sander, F. and C. M. Lalli. 1982. A comparative study of mollusk commu- nities on the shelf-slope margin of Barbados, West Indies. Veliger 24(4):309-318, 1 pl. Sarastia, H. 1943. Vocabulario malacologico de voces cubanas. Revista de la Sociedad Malacologica “Carlos de la Torre” 1(2):62-68. Sarastia, H. 1944. Los moluscos de los jardines cubanos. Revista de la Sociedad Malacélogica “Carlos de la Torre” 2(2):59-63. Sarastia, H. 1970. Prosobranquios marinos nuevos para la fauna de Cuba (Mollusca: Gastropoda). Poeyana serie A, (72):1-20, 4 pls. Sarastia, H. and J. Espinosa. 1979. El género Muricopsis (Mollusca: Neogastropoda) en Cuba. Poeyana (193):1-6. Sarasa, H. and J. Espinosa. 1984. Contribuci6n al conocimiento del orden Neogastropoda (Mollusca: Prosobranchia) en Cuba. Poeyana (273):1-18, 6 pls. Saunders, J. B., P. Jung, and B. Biju-Duval. 1986. Neogene paleontology in the northern Dominican Republic. 1. Field surveys, lithology, environment, and age. Bulletins of American Paleontology 89(323):1-79, pls. 1-9, tables 1-4, appendices 1-5. Schenck, H. G. 1935. Beitrage zur Kenntnis tropisch-amerikanischer Tertiérmollusken. III. Nuculid pelecypods of the genus Acila in the Tertiary of Venezuela, northern Colombia, and Trinidad. Eclogae Geologicae Helvetiae 28(2):503-510. Schilder, F. A. 1939. Cypraeacea aus dem Tertiar von Trinidad, Venezuela und den Antillen. Abhandlungen der Schweizerischen Pala- ontologischen Gesellschaft 62:1-35. [Trinidad, Venezuela, general] Schramm, A. 1867. Catalogues des Coquilles et des Crustacés de la Guadeloupe Envoyés a |l’Exposition Universelle de 1867. \m- primerie du Gouvernement, Basse-Terre, 27 pp. [Also 2nd ed., 1869, Catalogue des Coquilles et des Crustacés de la Guadeloupe Envoyés a l’Exposition Universelle de 1867 par l’Administration de la Colonie] Scott, P. J. B. 1987. Associations between corals and macro-infaunal in- vertebrates in Jamaica, with a list of Caribbean and Atlantic coral associates. Bulletin of Marine Science 40(2):271-286. [Bivalvia] Seamon, N. and E. Seamon. 1967. Collecting in the Netherlands Antilles. New York Shell Club Notes 134:3-S. Senn, A. 1940. Paleogene of Barbados and its bearing on history and structure of Antillean-Caribbean region. Bulletin of the American Association of Petroleum Geologists 24(9):1548-1610. [Barbados, general] Sharff, R. F. 1922. On the origin of the West Indian fauna. Dierk Amsterdam Bijdr., Weber Feestbundel, pp. 62-72. [not seen] MIKKELSEN ET AL.: BIBLIOGRAPHY OF CARIBBEAN MALACOLOGY 285 Shuster, C. N. Jr. and W. T. Bode. 1961. Observations on Vasum globulus nuttingi, with comments on other Caribbean vase shells. Nautilus 75(1):1-7. [Antigua, Barbuda] Shuttleworth, R. J. 1853. Uber den Bau der Chitoniden, mit Aufzahlung der die Antillen und die Canarischen Inseln bewohnenden Arten. Mittheilungen der Naturforschenden Gesellschaft in Bern (286- 291):169-207. [general] Shuttleworth, R. J. 1854a. Beitraige zur naheren Kenntniss der Land- und Stisswasser-Mollusken der Insel Portorico. Mittheilungen der Naturforschenden Gesellschaft in Bern (314-316):33-56, (321- 322:)89-103. Shuttleworth, R. J. 1854b. Catalogue of the terrestrial and fluviatile shells of St. Thomas, West Indies. Annals of the Lyceum of Natural History of New York 6(1858):68-73. Shuttleworth, R. J. 1856. Description de nouvelles espéces. Premiére décade: espéces nouvelles pour la faune des Antilles. Journal de Conchyliologie 5:168-175. Simpson, C. T. 1895. Distribution of the land and fresh-water mollusks of the West-Indian region, and their evidence with regard to past changes of land and sea. Proceedings of the United States National Museum 17(1894):423-450, pl. 16. Simroth, H. 1914. Beitrag zur Kenntnis der Nacktschnecken Columbiens zugleich eine Uebersicht tiber die neotropische Nacktschnecken- Fauna tiberhaupt. Jn: Voyage d’Exploration Scientifique en Colombie, by O. Fuhrmann and E. Mayor, pp. 270-341, pls.11-14. Mémoires de la Société Neuchdateloise des Sciences Naturelles 5:1- 1090. [Colombia, general] Sleurs, W. 1989. A zoogeographical analysis of the rissoinine fauna of the eastern Pacific with special reference to a comparison with the Caribbean fauna and with a checklist of the eastern Pacific Rissoininae Stimpson, 1865 (Mollusca: Gastropoda). Annales de la Société Royale Zoologique de Belgique 119(2):155-164. Smith, E. A. 1896. A list of the land and freshwater Mollusca of Trinidad. Journal of Conchology 8(7):231-251, pl.8. Smith, E. A. 1898. On some land shells from Trinidad. Journal of Conchology 9(1):27-29. Snyder, N. F. R. and H. W. Kale, III. 1983. Mollusk predation by snail kites in Colombia. Auk 100:93-97. Sohl, N. F. 1992. Upper Cretaceous gastropods (Fissurellidae, Haliotidae, Scissurellidae) from Puerto Rico and Jamaica. Journal of - Paleontology 66(3):414-434. Sohl, N. F. and E. G. Kauffman. 1964. Giant Upper Cretaceous oysters from the Gulf Coast and Caribbean. United States Geological Survey, Professional Paper 483-H:1-22, pls. 1-5. [general] Solem, A. 1961. A preliminary review of the pomatiasid land snails of Central America (Mollusca, Prosobranchia). Archiv fiir Mollus- kenkunde 90(4/6):191-213, pls. 10-12. Sowerby, G. B. (II) 1850. See J. C. Moore, 1850. Stanley, S. M. 1986. Anatomy of a regional mass extinction: Plio- Pleistocene decimation of the western Atlantic bivalve fauna. Palaios 1:17-36. [general] Starmiihlner, F. 1984. Occurrence, longitudinal distribution and geograph- ical range of the fresh- and brackish water molluscs of the Lesser Antillean islands (Guadeloupe, Dominica and Martinique). Soosiana 12:83-102. StarmiihlIner, F. and Y. Therezien. 1983. Résultats de la mission hydrobi- ologique austro-francaise de 1979 aux jles de la Guadeloupe, de la Dominique et de la Martiniques (Petites Antilles). Annalen des Naturhistorischen Museums in Wien 85/B:171-218, pls. 1-4 (Partie I A: Rapport préliminaire: Introduction, méthodes et situation générale de I’Ile de la Guadeloupe (géologie, géographie, climat et végétation). Description des stations et remarques générales sur la distribution de la faune dans les eaux continentales de I’Ile de la Guadeloupe.); 85/B:219-262, pls. 1-6 (Partie I B: Rapport prélimi- naire: Introduction, méthodes et situation générale de I’Ile de la Dominique et de I’Ile de la Martinique (géologie, géographie, cli- mat et végétation). Description des stations et remarques générales sur la distribution de la faune dans les eaux coourantes de 1’ Ile de la Dominique et de la Martinique.) [Guadeloupe, Dominica, Martinique] Strebel, H. 1873-1882. Beitrag zur Kenntnis der Fauna mexikanischer Land- und Siisswasser-Conchylien. Unter [besonderer] Beriick- sichtigung der Fauna angrenzender Gebiete. Abhandlungen aus dem Gebiete der Naturwissenschaften Herausgegeben von dem Naturwissenschaftlichen Verein in Hamburg 6:1-69, pls. 1-9 (1873, Theil I); 1-58, 2 pls., pls. 1-13 (1875, Theil II); 1-51, 2 pp. index, pls. 1-22 (1878, Theil III); 1-112, pls. 1-15 [1880, Theil IV (with G. Pfeiffer)]; 1-144, pls. 1-19 [1882, Theil V (with G. Pfeiffer)]. (Mexico, general] Sutton, F. A. 1946. Geology of Maracaibo Basin, Venezuela. Bulletin of the American Association of Petroleum Geologists 30(10):1621- 1739, pls. 2-9 (appendix by A. N. Dusenbury, pp.1739-1741). Sutty, L. 1986. Seashell Treasures of the Caribbean. E. P. Dutton, New York, 128 pp. [Lesser Antilles, Windward Islands] Talavera, F. G. and D. Princz. 1984. Marginella lasallei spec. nov. y revi- sion de la familia Marginellidae en el Mar Venezolano. Bolletino Malacologico 20(9-12):273-282. Tate, R. 1870. On the land and fresh-water Mollusca of Nicaragua. American Journal of Conchology new series, 5(3):151-162, pl. 16. Tello, J. 1975. Catdlogo de la Fauna Venezolana, Vol. 8, Mollusca, pp. 33-163. Sociedad Venezolana Ciencias Naturales. [not seen] Thiele, J. 1910. Molluskenfauna Westindiens. Zoologische Jahrbiicher suppl. 11(2):109-132, pl. 9. Thomé, J. W. 1989. Annotated and illustrated preliminary list of the Veronicellidae (Mollusca: Gastropoda) of the Antilles, and Central and North America. Journal of Medical and Applied Malacology 1:11-28. Thompson, F. G. 1957. A collection of mollusks from northern Venezuela. Occasional Papers of the Museum of Zoology, University of Michigan 591:1-10, pls. 1-2. Thompson, F. G. 1976. Land snails from Monito Island, West Indies. Nautilus 90(4):152-153. Thompson, F. G. 1987. A review of the land snails of Mona Island, West Indies. Bulletin of the Florida State Museum, Biological Sciences 31(2):69-106. Thompson, T. E. 1977a. Hunting for nudibranchs in the Caribbean Sea. Journal of Molluscan Studies 42(1976):451. [Jamaica] Thompson, T. E. 1977b. Jamaican opisthobranch molluscs I. Journal of Molluscan Studies 43:93-140. Thompson, T. E. 1980. Jamaican opisthobranch molluscs II. Journal of Molluscan Studies 46:74-99. Tomlin, J. R. le B. 1929. The Mollusca of the ‘St. George’ expedition. (II). The West Indies. Journal of Conchology 18(11):307-310. Torre, A. de la. 1929. New Cuban Urocoptidae. Proceedings of the Academy of Natural Sciences of Philadelphia 81:443-448, pls. 12,13,15. Torre, A. de la. 1952. Datos sobre los moluscos del “Cafetal Fundador de Canimar” en Matanzas y sobre la llegada a Cuba, en 1839, de los sabios alemanes Gundlach, Pfeiffer y Otto. Revista de la Sociedad Malacologica “Carlos de la Torre” 9(1):9-25, 4 pls. Torre, A. de la. 1954. Nuevos moluscos cubanos de los géneros Cerion y Chondropoma. Revista de la Sociedad Malacologica “Carlos de la Torre” 9(2):39-46, pl .S. Torre, A. de la. 1960. Caribbean species of Truncatella. Nautilus 286 AMER. MALAC. BULL. 10(2) (1993) 73(3):79-88. [Cuba, general] Torre, A. de la and P. Bartsch. 1938. The Cuban operculate land shells of the subfamily Chondropominae. Proceedings of the United States National Museum 85:193-403, pls. 7-39. Torre, A. de la and P. Bartsch. 1941. The Cuban operculate land mollusks of the family Annulariidae, exclusive of the subfamily Chondropominae. Proceedings of the United States National Museum 89(3096):131-385, pls. 9-57. Torre, C. de la and P. Bartsch. 1942. The cyclophorid mollusks of Cuba. In: The Cyclophorid Operculate Land Mollusks of America, by C. de la Torre, P. Bartsch, and J. P. E. Morrison. Bulletin of the United States National Museum 181(pt.1):3-42, pls. 1-8. Torre, A. de la and J. B. Henderson. 1921. New mollusks from Camaguey and Santa Clara Provinces, Cuba. Proceedings of the United States National Museum 59(2369):247-267, pls. 38-42. Toula, F. 1909-1911. Eine/Die jungtertiare Fauna von Gatun am Panama- Kanal. Jahrbiicher der Kaiserlich-Kéniglichen Geologischen Reich- sanstalt 58(4):673-760, pls. 25-28 (1909, I. Teil); 61(3-4):487-530, pls. 30-31(I-II) (1911, Il. Teil). Trechmann, C. T. 1923. The Yellow Limestone of Jamaica and its Mol- lusca. The Geological Magazine 60(710):337-367, pls. 14-18. Trechmann, C. T. 1924. The Carbonaceous Shale or Richmond Formation of Jamaica. The Geological Magazine 61:2-19, pls. 1-2. Trechmann, C. T. 1925a. The northern range of Trinidad. The Geological Magazine 62:544-551, pl. 25. Trechmann, C. T. 1925b. The Scotland beds of Barbados. The Geological Magazine 62(737):48 1-504, pls. 21-24. Trechmann, C. T. 1929. Fossils from the Blue Mountains of Jamaica. The Geological Magazine 66(785):481-491, pl. 18. Trechmann, C. T. 1930. The Manchioneal beds of Jamaica. The Geo- logical Magazine 67(791):199-218, pls. 12-13. Trechmann, C. T. 1934. Tertiary and Quaternary beds of Tobago, West Indies. The Geological Magazine 71:481-493, pls. 24-25, map. Trechmann, C. T. 1935a. Fossils from the Northern Range of Trinidad. The Geological Magazine 72:166-175, pl. 5. Trechmann, C. T. 1935b. The geology and fossils of Carriacou, West Indies. The Geological Magazine 72:529-555, pls. 20-22. [Grenadines] Treece, G. D. 1980. Bathymetric records of marine shelled Mollusca from the northeastern shelf and upper slope of Yucatan, Mexico. Bulletin of Marine Science 30(3):552-570. Turner, R. D. 1955. Scaphopods of the Atlantis dredgings in the Western Atlantic with a catalogue of the scaphopod types in the Museum of Comparative Zodlogy. Papers in Marine Biology and Ocean- ography, 1955:309-320. [Cuba, general] Turner, R. D. 1958. The genus Hemitrochus in Puerto Rico. Occasional Papers on Mollusks 2(22):153-178. Turner, R. D. 1960. Land shells of Navassa Island, West Indies. Bulletin of the Museum of Comparative Zoology 122(5):233-244, pls. 1-7. Vanatta, E. G. 1912. Pleistocene fossils from eastern Cuba. Nautilus 26:69. Vanatta, E. G. 1914. [Pleistocene shells from near Sierra Nueva, Santo Domingo.] Nautilus 27(10):120. van Benthem Jutting, T. 1927. Marine molluscs of the island of Curacao. Bijdragen tot de Dierkunde 25:1-36. van Benthem Jutting, W. S. S. 1934. Enkele beschouwingen over de week- dierfauna van Curacao. Natuur en Mensch 54:34-36. [not seen] van Benthem Jutting, W. S. S. 1945. Quarternary shells from several Venezuelan islands and from the north coast of South America. Verhandelingen van Het Geologisch-Mijnbouwkundig Genootschap voor Nederland en Kolonién, Geologische Serie 14:71-83. [Curagao, Venezuela] van der Schalie, H. 1948. The land and fresh-water mollusks of Puerto Rico. Miscellaneous Publications, Museum of Zoology, University of Michigan 70:1-134, pls. 1-14. van Kuyp, E. 1949. Planorbidae records on the Netherlands Antilles. American Journal of Tropical Medicine 29:259-261. [not seen] van Kuyp, E. 1951. Notes on the Planorbidae in the Netherlands Antilles. Docum. Neerl. Indon. Morc. Trop. 3:283-284. [not seen] van Regteren Altena, C. O. 1941. Old Tertiary Mollusca from Curagao. Proceedings, Nederlandse Akademie van Wetenschappen 44(10): 1234-1238, 1 pl. van Regteren Altena, C. O. 1961. The Mollusca from the limestone of Brimstone Hill, St. Kitts, and Sugar Loaf and White Wall, St. Eustatius, Lesser Antilles. Proceedings, Koninklijke Nederlandse Akademie van Wetenschappen series B, 64(2):288-304, 1 pl. van Winkle, K. 1919. Remarks on some new species from Trinidad. Bulletins of American Paleontology 8(33):19-27, pl. 3. Vaughan, T. W. 1915-1920. Study of the stratigraphic geology and of the fossil corals and associated organisms in several of the smaller West Indian islands. Carnegie Institute of Washington Yearbook 13(1914):358-360 (1915); 14(1915):368-373 (1916); 15(1916):359 (1917); 16(1917):319 (1918); 18(1919):345-346 (1920). [Lesser Antilles] Vaughan, T. W. 1919. The biologic character and geologic correlation of the sedimentary formations of Panama in their relation to the geo- logic history of Central America and the West Indies. Contributions to the geology and paleontology of the Canal Zone, Panama, and geologically related areas in Central America and the West Indies. Bulletin of the United States National Museum 103: i-iv, 547-612. [Panama, Cuba, Lesser Antilles, general] Vendryes, H. 1899. Systematic catalogue of the land and fresh-water shells of Jamaica. Journal of the Institute of Jamaica 2(6):590-607. Venmans, L. A. W. C. 1963. Caribbean land molluscs: Streptaxidae. Studies on the Fauna of Curacao and other Caribbean Islands 14(61):41-76, pls. 2-5. Vermeij, G. J. i973. West Indian molluscan communities in the rocky intertidal zone: a morphological approach. Bulletin of Marine Science 23(2):351-386. Vermeij, G. J. and E. J. Petuch. 1986. Differential extinction in tropical American molluscs: endemism, architecture, and the Panama land bridge. Malacologia 27(1):29-41. Vermeij, G. J. and G. Rosenberg. 1993. The tropical Atlantic as donor and recipient region for invading species. American Malacological Bulletin 10(2):181-194. Vernhout, J. H. 1914. The land and freshwater-molluscs of the Dutch West-Indian islands. Notes from the Leyden Museum 36(3-4):177- 189. Verrill, A. H. 1950. New marine mollusks from Dominica, B. W. I. Nautilus 63(4):126-128, pls. 9-10. Vokes, E. H. 1963-1992. Cenozoic Muricidae of the western Atlantic region. Tulane Studies in Geology and Paleontology 1(3):95-123, pls. 1-4 (1963, part I); 3(4):181-204, pls. 1-3 (1965, part II); 5(3):133-166, pls. 1-6 (1967, part IID; 6(3):85-126, pls. 1-8 (1968, part IV); 8(1):1-50, pls. 1-7 (1970, part V); 11(3):121-162, pls. 1-6 (1975, part VI); 12(1):45-46 (1976); 12(3):101-132, pls. 1-8 (1976, part VII); 23(1-3):1-96, pls. 1-12 (1990, part VIII); 25(1-3):1-108, pls. 1-20 (1992, part IX). Vokes, E. H. 1979. The age of the Baitoa Formation, Dominican Republic, using Mollusca for correlation. Tulane Studies in Geology and Paleontology 15(4):105-116, 2 pls. Vokes, E. H. 1983. Additions to the Typhinae (Gastropoda: Muricidae) of the Gatun Formation, Panama. Tulane Studies in Geology and Paleontology 17(4):123-130, pl. 1. Vokes, E. H. 1989. Neogene paleontology in the northern Dominican Republic. 8. The family Muricidae (Mollusca: Gastropoda). MIKKELSEN ET AL.: BIBLIOGRAPHY OF CARIBBEAN MALACOLOGY 287 Bulletins of American Paleontology 97(332):5-94, bibl. pp. 130- 141, pls. 1-12 + pp. 142-153, index. pp. 161-181. Vokes, H. E. 1938. Upper Miocene Mollusca from Springvale, Trinidad, British West Indies. American Museum Novitates 988:1-28. Vokes, H. E. 1983. Distribution of shallow-water marine Mollusca, Yucatan Peninsula, Mexico. National Geographic Society Research Reports 15:715-723. Vokes, H. E. 1989. Neogene paleontology in the northern Dominican Republic. 9. The family Cardiidae (Mollusca: Bivalvia). Bulletins of American Paleontology 97(332):95-141, pls. 13-19 + pp. 154- 160, index. pp. 161-181. Vokes, H. E. and E. H. Vokes. 1984. Distribution of shallow-water marine Mollusca, Yucatan Peninsula, Mexico. Mesoamerican Ecology Institute, Monograph 1, Middle American Research Institute, Publication 54(1983):i-viii, 1-183, pls. 1-50. Vokes, H. E. and E. H. Vokes. 1992. Neogene paleontology in the north- ern Dominican Republic 12. The genus Spondylus (Bivalvia: Spondylidae). Bulletins of American Paleontology 102(339):5-13. Voss, G. L. 1955. The Cephalopoda obtained by the Harvard-Havana expedition off the coast of Cuba in 1938-39. Bulletin of Marine Science of the Gulf and Caribbean 5(2):81-115. Voss, G. L. 1958. The cephalopods collected by the R/V “Atlantis” during the West Indian cruise of 1954. Bulletin of Marine Science of the Gulf and Caribbean 8(4):369-389. [Puerto Rico, Virgin Islands] Voss, G. L. 1968. Octopods from the R.V. Pilsbry southestern Caribbean cruise, 1966, with a description of a new species, Octopus zonatus. Bulletin of Marine Science 18(3):645-659. [Colombia, general] Voss, G. L. 1971. The cephalopod resources of the Caribbean Sea and adjacent regions. Symposium on Investigations and Resources of the Caribbean and Adjacent Regions. Papers on Fishery Resources, FAO, ROME, 1971:307-323. [not seen] Voss, G. L., L. Opresko, and R. Thomas. 1973. The potentially commer- cial species of Octopus and squid of Florida, the Gulf of Mexico and the Caribbean area. [University of Miami] Sea Grant Field Guide Series, 2:vi + 33 pp. Vries, W. de. 1974. Caribbean land molluscs: notes on Cerionidae. Studies on the Fauna of Curacao and other Caribbean Islands 45(146):81-117, pls. 8-14, figs.95-104. [Netherlands Antilles] Wade, B. A. 1972. A description of a highly diverse soft-bottom commu- nity in Kingston Harbour, Jamaica. Marine Biology 13(1):57-69. Waller, T. R. 1993. The evolution of “Chlamys” (Mollusca: Bivalvia: Pectinidae) in the tropical western Atlantic and eastern Pacific. American Malacological Bulletin 10(2):195-249. Warmke, G. L. 1960. Seven Puerto Rico cones: notes and radulae. Nautilus 73(4):119-124. Warmke, G. L. and R. T. Abbott. 1961. Caribbean Seashells. Livingstone Publishing Company, Narberth, Pennsylvania, 348 pp., 44 pls. [Reprinted in 1975 by Dover Publications, New York] [Puerto Rico, general] Warmke, G. L. and L. R. Almodévar. 1963. Some associations of marine mollusks and algae in Puerto Rico. Malacologia 1(2):163-177. Warmke, G. L. and D. S. Erdman. 1963. Records of marine mollusks eaten by bonefish in Puerto Rican waters. Nautilus 76(4):115-120. Weber, J. A. 1961. Marine shells of Water Island, Virgin Islands. Nautilus 75(2):55-60. Weisbord, N. E. 1926. Notes on marine mollusks from the Yucatan Peninsula, Mexico. Nautilus 39:81-87. Weisbord, N. E. 1929. Miocene Mollusca of northern Colombia. Bulletins of American Paleontology 14(54):1-74, pls. 1-9. Weisbord, N. E. 1962. Late Cenozoic gastropods from northern Venezuela. Bulletins of American Paleontology 42(193):1-672, pls.1-48. Weisbord, N. E. 1964a. Late Cenozoic pelecypods from northern Venezuela. Bulletins of American Paleontology 45(204):1-564, pls. 1-59. Weisbord, N. E. 1964b. Late Cenozoic scaphopods and serpulid poly- chaetes from northern Venezuela. Bulletins of American Paleontology 47(214):111-203, pls. 16-22. Welch, d’A. A. 1929. Some operculate snails from northwestern Camaguey, Cuba. Nautilus 42(3):98, pl. 5. Welch, d’A. A. 1934. New Cuban land shells from Oriente and Camaguey Provinces. Nautilus 47(3):104-108, 130-135, pl. 11. Wells, F. E. 1975. Comparison of euthecosomatous pteropods in the plankton and sediments off Barbados, West Indies. Proceedings of the Malacological Society of London 41(6):503-509. Wetherbee, D. K. and W. J. Clench. 1987. Catalog of the Terrestrial and Fluviatile Mollusk Fauna of Hispaniola, and an History of Early Hispaniolan Malacology. Privately published, 89 + v pp. [not seen] Wheeler, H. E. 1913. A collector in western Cuba and the Isle of Pines. Nautilus 26:99-108, 111-114. Williams, E. H. Jr., I. Clavijo, J. J. Kimmel, P. L. Colin, C. D. Carela, A. T. Bardales, R. A. Armstrong, L. Williams, R. H. Boulon, and J. R. Garcia. 1983. A checklist of marine plants and animals of the south coast of the Dominican Republic. Caribbean Journal of Science 19(1-2):39-53. Williams, P. 1985. Remarks on Nowell-Usticke’s Caribbean shells. Shells and Sea Life 17(10):219-221. [St. Croix] Woodlock, E. 1963. Shelling at St. John, Virgin Islands. New York Shell Club Notes 95:5-7. Woodlock, E. 1964. Shelling at St. John, Virgin Islands, 1964. New York Shell Club Notes 106:6-7. Woodlock, E. 1965. Caribbean shelling. New York Shell Club Notes 115:4-6. [St. John, Virgin Islands] Woodlock, E. 1966. Return to St. John. New York Shell Club Notes 127:7. Woodlock, E. 1966. Shelling at Grand Cayman, B. W. I. New York Shell Club Notes 127:5-7. Woodlock, E. 1968. Shelling on Little Cayman Island, B. W. I. New York Shell Club Notes 139:5-7. Woodring, W. P. 1923. Tertiary mollusks of the genus Orthaulax from the Republic of Haiti, Porto Rico and Cuba. Proceedings of the United States National Museum 64(1):1-12, pls. 1-2. Woodring, W. P. 1924. West Indian, Central American and European Miocene and Pliocene mollusks. Bulletin of the Geological Society of America 35:867-886. [general] Woodring, W. P. 1925-1928. Contributions to the geology and paleontol- ogy of the West Indies. Miocene molluscs from Bowden, Jamaica. Carnegie Institution of Washington Publication 366:222 pp., 28 pls. (1925, [part I], Pelecypods and scaphopods); 385:564 pp., 40 pls. (1928, part II, Gastropods and discussion of results). [Jamaica, Cuba, Haiti, Puerto Rico] Woodring, W. P. 1957-1982. Geology and paleontology of Canal Zone and adjoining parts of Panama. Geology and description of Tertiary mollusks. United States Geological Survey, Professional Paper 306:759 pp., 124 pls. [part A, pp.i-iv, 1-145, pls. 1-23 (1957); part B, pp.iii, 147-239, pls. 24-38 (1959); part C, pp.iii, 241-297, pls. 39-47 (1964); part D, pp.iii, 299-452, pls. 48-66 (1970); part E, pp.iii, 453-539, pls. 67-82 (1973); part F, pp.ili-iv, 541-759, pls. 83- 124 (1982)] Woodring, W. P. 1959. Tertiary Caribbean molluscan faunal province. International Oceanographic Congress, American Association for the Advancement of Science 1959:299-300. [general; not seen] Woodring, W. P. 1961. Oligocene and Miocene in the Caribbean region. Transactions of the Second Caribbean Geological Conference, Mayagiiez, Puerto Rico, 4-9 January 1960, pp. 27-32. [general] 288 AMER. MALAC. BULL. 10(2) (1993) Woodring, W. P. 1965. Endemism in middle Miocene Caribbean mollus- can faunas. Science 148(3672):961-963. Woodring, W. P. 1966. The Panama land bridge as a sea barrier. Pro- ceedings of the American Philosophical Society 110(6):425-433. Woodring, W. P. 1974. The Miocene Caribbean Faunal Province and its subprovinces. Verhandlungen der Naturforschenden Gesellschaft Basel 84(1):209-213. Woodring, W. P. 1978. Conexiones terrestres entre Norte y Sudamerica. X. Distribution of Tertiary marine molluscan faunas in southern Central America and northern South America. Boletin del Instituto de Geologia, Universidad National de Auton. México 101:153-165. Woodring, W. P. and T. F. Thompson. 1949. Tertiary formations of Panama Canal Zone and adjoining parts of Panama. Bulletin of the American Association of Petroleum Geologists 33(2):223-247. Work, R. C. 1969. Systematics, ecology, and distribution of the mollusks of Los Roques, Venezuela. Bulletin of Marine Science 19(3):614- 711. Wurtz, C. B. 1950. Results of the Catherwood-Chaplin West Indies Expedition, 1948. Part IV. Land snails of North Cat Cay (Bahamas), Cayo Largo (Cuba), Grand Cayman, Saint Andrews and Old Providence. Proceedings of the Academy of Natural Sciences of Philadelphia 102:95-110, pl. 2. Yong, M. and G. Perera. 1984. A preliminary study of the freshwater mol- lusks of the Isle of Youth (Isle of Pines), Cuba. Walkerana 2(7):121-123. INDEX General and Multiple Locations: Abbott 1957, 1984; Adam 1937; H. B. Baker 1943; Bandel 1975a; Bartsch 1930, 1942; Bayer 1971; Bertsch 1979; Bland 1855, 1861; Bolivar de Carranza and Hidalgo- Escalante 1990; Boone 1925; Borkowski 1975; Breure 1974, 1975; Buehler 1987; Bullock 1974; Cernohorsky 1978; Clench 1936; Clench et al. 1947-1948 [= Krebs 1864]; Colin 1978; Cooke et al. 1943; Coomans 1957, 1958, 1964a; Creswell and Davis 1991; Dall 1878, 1880, 1881, 1885, 1886-1889, 1888, 1890, 1897, 1899; Dautzenberg 1900; Edwards 1977; Engel 1936; Eudes- Delongchamps 1859; Eva 1980; Faber 1990; Folin 1879; Gabb 1881a; E. F. Garcia 1988; Gertman 1969; Gibson-Smith and Gibson-Smith 1982a; Guilding 1826, 1828; Gunter 1951; Guppy 1866b, 1867a, 1874, 1875b, 1882; Guppy and Dall 1896; F. Haas 1960, 1962; Harry 1962; Higgins 1876; Hodson 1926; Hummelinck 1940d, e; Imlay 1944; Johnson 1981; Jones and Hasson 1985; Jung 1987; Kaas 1972; Keen 1976; Kobelt 1880; Krebs 1866, 1873; Kruckow 1982; Lipe and Abbott 1991; Lozet and Pétron 1977; Lyons 1988; Macsotay and Scherer 1972; Malek 1969; Marcus 1977; Marcus and Marcus 1960, 1963; Martens 1890-1901; McGinty 1962; W. B. Miller and Naranjo-Garcia 1991; Mérch 1859, 1860, 1984, 1875-1877, 1876, 1878; Moolenbeek and Faber 1989; D. R. Moore 1977; Morelet 1849-1851; Morris 1973; Nicol 1945; Olsson 1967a, b; Palmer 1938, 1967; Petuch 1979, 1982, 1987, 1988; Pilsbry 1920b, 1930a; Pilsbry and McGinty 1939; Piplani 1977; Prentice 1980; Rees 1950; Rehder 1943, 1954, 1981; Rosenberg 1993; Russell 1941; Schilder 1939; Senn 1940; Sharff 1922; Shuttleworth 1853, 1856; Simpson 1895; Simroth 1914; Sleurs 1989; Sohl and Kauffman 1964; Solem 1961; Stanley 1986; Strebel 1873-1882; Thiele 1910; Thomé 1989; Tomlin 1929; A. de la Torre 1960; Turner 1955; Vaughan 1919; Venmans 1963; Vermeij 1973; Vermeij and Rosenberg 1993; E. H. Vokes 1963- 1992; Voss 1968, 1971; Voss et al. 1973; Waller 1993; Warmke and Abbott 1961; Woodring 1924, 1959, 1961, 1965, 1974, 1978; Wurtz 1950. Cuba (including Isle of Pines [Isla de Pinos]): Aguayo 1932, 1934, 1935, 1938a, b, c, 1943a, b, 1944a-d, 1948, 1949a, 1953, 1959, 1961b- 1963; Aguayo and Borro 1946a, b; Aguayo and A. de la Torre 1951, 1954; Aguayo and Jaume 1934, 1936, 1939, 1945, 1947a, 1947b-1951a, 1951b, 1952, 1953, 1954, 1957-1958; Alayo Dalmau 1960; Alcalde Ledén 1945a-1948, 1945b; Arango y Molina 1878- 1880; Bavay 1922; Bidart and Espinosa 1989; Boone 1925; Boss and Jacobson 1973a, b, 1974, 1975a, b; Clench and Aguayo 1938, 1939-1941, 1950, 1951a, b; Clench and Jacobson 1968, 1970, 1971a, b; Cooke 1919; Crosse 1890a; Dall 1896; Desjardin 1949; d’Orbigny 1841-1853, 1854; Espinosa 1984, 1987, 1989; Espinosa and Fernandez Garcés 1988, 1989, 1990a, b; Forcart 1950; Freire and Alayo 1946; Gould 1844; Gray 1854; Gundlach 1856, 1857a, b; F. Haas 1941; Henderson 1916; Hermes 1945; A. Herrera and Espinosa 1988; E. Herrera 1945; Hoskins 1964; Imlay 1942; Jacobson 1956, 1970, 1972, 1974; Jaume 1945a-c, 1952, 1975a, b; Jaume and Borro 1946; Jaume and A. de la Torre 1972; Jaume and Pérez Farfante 1942; Jaume and Sanchez de Fuentes 1943; Jaume and Sarasta 1943; Kojumdgieva and A. de la Torre 1986; Manjarres 1979; McLean 1936; Morelet 1849-1851; Mulleried 1951; Palmer 1947; Perera and Yong 1984; Perera et al. 1986; Pérez Farfante 1940, 1942; Pfeiffer 1839, 1840, 1856, 1858a, 1858b-1864; Pilsbry 1927, 1929, 1942b; Pilsbry and Aguayo 1933; Poey 1851-1861; Ramsden 1914; H. G. Richards 1933, 1935; Rolan 1992; Rolan and Espinosa 1992a, b; Rolan et al. 1990; Sanchez Roig 1926, 1948, 1949, 1951a, b; Sarasa 1943, 1944, 1970; Sarastia and Espinosa 1979, 1984; A. de laTorre 1929, 1952, 1954, 1960; A. de la Torre and Bartsch 1938, 1941; A. de la Torre and Henderson 1921; C. de la Torre and Bartsch 1942; Turner 1955; Vanatta 1912; Vaughan 1919; Voss 1955; Welch 1929, 1934; Wheeler 1913; Woodring 1923, 1925-1928; Wurtz 1950; Yong and Perera 1984. Cayman Islands: Abbott 1958; Cerridwen and Jones 1991; Clench 1964; Hummelinck 1980; Matley 1926; Pickford 1950; Pilsbry 1930a, b, 1942a, 1949; Rehder 1962b; H. G. Richards 1955; Salisbury 1953; Woodlock 1966, 1968; Wurtz 1950. Jamaica: C. B. Adams 1845, 1846a, b, 1849a-h, 1850a-e, 1851a-d, 1852b; H. B. Baker 1934-1935; Boss 1972; Boss and Jacobson 1975b; Chitty 1853, 1857a, b; Clench and Jacobson 1970; Clench and Turner 1950; Cockerell 1894; Cockerell and Larkin 1894; Cox 1941; Davis 1967; Edmunds 1964; Ferreira 1978; Gloyne 1872, 1875; Goodfriend 1983, 1986; Goodfriend and Mitterer 1988, 1993; Guppy 1866d, 1873b, 1875b; Henderson 1894; Humfrey 1975; Jackson 1972, 1973; Jacobson and Boss 1973; Jarvis 1902a, b, 1903; Jung 1972; Malcolm 1964; Merian 1844; Michelson 1953; J. C. Moore 1863; Paul 1982, 1983; Pfeiffer 1946; Pilsbry 1911, 1927, 1946; Pilsbry and Brown 1910, 1912; Rush 1891; Russell Hunter 1955; Scott 1987; Sohl 1992; T. E. Thompson 1977a, b, 1980; Trechmann 1923, 1924, 1929, 1930; Vendryes 1899; Wade 1972; Woodring 1925-1928. Navassa Island: Turner 1960. Hispaniola: Bartsch 1946; Boss and Jacobson 1973a; Clench 1962a; Clench and Aguayo 1937; Wetherbee and Clench 1987. Haiti: Bartsch 1932; Eyerdam 1961; Folin 1867-1871; Guppy 1876; Pilsbry 1910b; Pilsbry and Vanatta 1928; Robart et al. 1976; Woodring 1923, 1925-1928. Dominican Republic (= Santo Domingo, San Domingo): Aguayo 1949b; Clench 1938, 1962b; Crosse 1891a-c; Gabb 1871, 1873, 1875; Gomez et al. 1986; Hjalmarson and Pfeiffer 1858; Ingram 1939; Jung 1986; Jung and Petit 1990; Maury 1917; J. C. Moore 1850, 1853; Pflug 1961; Pilsbry 1922, 1933; Pilsbry and Johnson 1917; Pilsbry and Olsson 1954; Pilsbry and Sharp 1898; Poinar and MIKKELSEN ET AL.: BIBLIOGRAPHY OF CARIBBEAN MALACOLOGY 289 Roth 1991; Ramirez 1950, 1956; Saunders et al. 1986; Vanatta 1914; E. H. Vokes 1979, 1989; H. E. Vokes 1989; Vokes and Vokes 1992; E. H. Williams et al. 1983. Puerto Rico [including Mona Island and Monito Island]: Aguayo 1961a, 1961b-1963, 1966; Arnow et al. 1963; H. B. Baker 1940, 1941a, b, 1961a, b, 1962a-d; Bartsch 1934a, b; Boss and Jacobson 1973a; Clench 1948; Coomans 1965; Corea 1934; Crosse 1892; Dall and Simpson 1901; Emerson 1952; Ferguson and Richards 1963; Garcia-Rios 1983, 1988; Glynn 1968a, b, 1970; Glynn et al. 1964; Golley 1960; Gundlach 1883; Harry 1964; Harry and Hubendick 1964; Hubbard 1920; Martens 1877; Mattox 1948; Maury 1919, 1920; McLean 1951; Mestey-Villamil 1979, 1980, 1982; Nutting 1919; Ortiz-Corps [1985]; Potter 1946; Read 1964; C. S. Richards 1964; Shuttleworth 1854a; Sohl 1992; F. G. Thompson 1976, 1987; Turner 1958; van der Schalie 1948; Voss 1958; Warmke 1960; Warmke and Abbott 1961; Warmke and Almodovar 1963; Warmke and Erdman 1963; Woodring 1923, 1925-1928. Virgin Islands: Eddison 1964; Ferguson and Richards 1963; Maes 1983; Marcus and Marcus 1962; Maury 1920; McLean 1951; Voss 1958; Weber 1961. St. Croix: Bland 1881; Faber 1988; Gladfelter 1974; Jacobson 1968; A. I. Miller 1989; Mérch 1863; D. R. Moore 1972; Nowell-Usticke 1956, 1959a, b, 1961a, b, 1963a, b, 1968, 1969, 1971; Old 1979; P. Williams 1985. St. John: Muchmore 1993; Woodlock 1963, 1964, 1965, 1966. St. Barthélemy: Nowell-Usticke 1961c. St. Thomas: Bland 1852, 1854; Mérch 1863; Rush 1891; Shuttleworth 1954b. Water Island: Mitchell-Tapping 1979. Lesser Antilles - Leeward Islands: Adam 1957; Cooke 1919; Gerth 1951; Hummelinck 1940d; Nowell-Usticke 1968; Nutting 1919; Vaughan 1919. Antigua: Brown 1913; Brown and Pilsbry 1914; Merian 1844; Nutting 1919; Pilsbry and Brown 1914; Schuster and Bode 1961. Barbuda: Schuster and Bode 1961. Saba: Clench 1970; Knudsen 1982. St. Eustatius: van Regteren Altena 1961. St. Christopher (= St. Kitts): Rush 1891; van Regteren Altena 1961. St. Martin: Coomans 1963a,b, 1967, 1974. Lesser Antilles - Windward Islands: Adam 1957; Gerth 1951; Guyard et al. 1982; Nowell-Usticke 1968; Nutting 1919; Sutty 1986; Vaughan 1915-1920. Barbados: Conde 1966; Edmunds and Just 1983, 1985; Ferreira 1985; Jung 1968; Kidd and Sander 1979; Kugler et al. 1984; Lewis 1960, 1965; Lewis and Fish 1969; Marcus and Hughes 1974; Nutting 1919; Rush 1891; Sander and Lalli 1982; Senn 1940; Trechmann 1925b; Wells 1975. Dominica: Bergh 1890; Guppy 1868a; Noblet and Damian 1991; Starmiihlner 1984; StarmiihIner and Therezien 1983; Verrill 1950. Grenada: Ferguson and Buckmire 1974; Guppy 1868a; Hemmen and Hemmen 1979; Percharde 1982. Guadeloupe: Beau 1858a, b; Crosse 1865; Deshayes 1857; Mazé 1883, 1890; Mongin 1968; Petit de la Saussaye 1851, 1853, 1856; Pointier 1974, 1976; Schramm 1867; StarmiihIner 1984; Starmiihlner and Therezien 1983. Martinique: Beau 1858; Bordaz 1899; Cossmann 1913; Dreyfuss 1953; Guppy 1913; Guyard and Pointier 1979; Lamy 1929; Mazé 1874; Starmiihlner 1984; Starmiihlner and Therezien 1983. 1893, 1910, 1911, 1913; Harris 1926; Jung 1969a, b; Kugler 1938; Mansfield 1925; Martin-Kaye 1951; Maury 1912a, b, 1925; A. K. Miller and Thompson 1937; Nowell-Usticke 1964; Percharde 1968, 1970, 1982; Princz 1982; Rutsch 1934a-1942a, 1940, 1942b, c, 1943; Schenck 1935; Schilder 1939; Smith 1896, 1898; Trechmann 1925a, 1934, 1935a; van Winkle 1919; H.E. Vokes 1938. Venezuela: Almeida 1974; Arocha and Urosa 1982; Arocha et al. 1991; H. B. Baker 1923-1930; Beauperthuy 1967; Bullis 1964; Ernst 1878; Flores 1964, 1966, 1968, 1973; Gibson-Smith 1973; Gibson- Smith and Gibson-Smith 1971, 1974, 1979, 1982b, 1983; Gonzalez and Flores 1972; Gonzalez and Princz 1979; Guppy 1873a, 1913; Hodson 1926; Hodson and Hodson 1931; Hodson et al. 1927; Hubendick 1961; Hummelinck 1940a-c; Ingram 1947; Jung 1965, 1973, 1975; Lorié 1889; Marks 1952; Martens 1873; A. K. Miller and Thompson 1937; Olsson 1942; Palmer 1945; Petuch 1976, 1981a; Princz 1973, 1980, 1982, 1983, 1987; Princz and Gonzalez C. 1981; Princz and Gonzalez de Pacheco 1981; Rehder 1962a; H. G. Richards and Wagenaar Hummelinck 1940; Rodriguez 1959; Rutsch 1934b; Schenck 1935; Schilder 1939; Sutton 1946; Talavera and Princz 1984; Tello 1975; F. G. Thompson 1957; van Benthem Jutting 1945; Weisbord 1962, 1964a, b; Work 1969. Netherlands Antilles: H. B. Baker 1924; Coomans 1958, 1961, 1974; De Jong and Coomans 1988; Engel 1936; Hummelinck 1940b, c; Nowell-Usticke 1962; Seamon and Seamon 1967; van Kuyp 1949, 1951; Vernhout 1914; Vries 1974. Aruba: Lorié 1889. Bonaire: Adam 1937; Moolenbeek and Faber 1984. Curacao: Adam 1937; Bandel 1976e; Coomans 1964b; Crosse and Bland 1873; De Jong and Kristensen 1965, 1968; Engel 1925, 1927; Hummelinck 1990; Jung 1974; Lorié 1889; Marcus and Marcus 1970; Righi 1968; T. van Benthem Jutting 1927; W. van Benthem Jutting 1934, 1945; van Regteren Altena 1941. Colombia (Atlantic): Anderson 1927, 1928, 1929; Bandel 1974a, b, 1975a, 1976a-f, 1984; Bandel and Wedler 1987; Brattstré6m 1980; B. L. Clark and Durham 1946; Cosel 1973, 1976, 1978, 1987; Criales 1984; Daniel 1941; Diaz 1990, 1991; Diaz and Gétting 1986, 1988; Diaz et al. 1991; Duque 1979; Etayo Serna 1979; Garcia and Luque 1986; Gotting 1973, 1978; Ingram 1947, 1948; Kaufmann and Gotting 1970; Marcus 1976; Marcus and Marcus 1967; Nicol 1945; Olsson 1942; Petuch 1981a; Piaget 1914; Pilsbry 1912; Pilsbry and Brown 1917; Pilsbry and Clapp 1902; Schenck 1935; Simroth 1914; Snyder and Kale 1983; Voss 1968; Weisbord 1929. Panama (Atlantic): C. B. Adams 1852a; Aguayo 1949a; Bartsch 1931; Brattstr6m 1985; Brown and Pilsbry 1911-1913a, 1913b; Carpenter 1863; Cossmann 1913; Cubit and Williams 1983; Dall 1912a, b; Glynn 1970; Jackson and Jung 1992; Jackson et al. 1993; Keen and Thompson 1946; MacDonald 1919; Marcus and Marcus 1967; Meyer 1977; Nicol 1945; Olsson 1942, 1972; Olsson and McGinty 1958; Petuch 1990; Pilsbry 1910a, 1920c, 1926b, 1930b; Radwin 1969; Rehder 1942; Rosewater 1975, 1976; Toula 1909-1911; Vaughan 1919; Vermeij and Petuch 1986; E. H. Vokes 1983; Woodring 1957-1982, 1966; Woodring and Thompson 1949. Costa Rica (Atlantic): Bakus 1968; Dall 1912a; Gabb 1881b; Gémez and Valerio 1971; O. Haas 1942; Houbrick 1968; Kemperman and Coomans 1984; Nicol 1945; Olsson 1922; Palmer 1923; Pilsbry 1911, 1920a, c, 1926a; Pittier 1890; Rehder 1942; Robinson 1991, 1993; Robinson and Montoya 1988. St. Lucia: Crosse and Bland 1873. St. Vincent and the Grenadines: Guppy 1881; Jung 1968, 1971; Trechmann 1935b. Trinidad and Tobago: Baboolal et al. 1981; Crosse 1890b; Guppy 1866a, c, 1867b, 1868a, b, 1869, 1871, 1872, 1874, 1875a, 1877, Nicaragua (Atlantic): Bock and Moore 1971; Fluck 1900, 1901, 1905- 1906; H. G. Richards 1939; Tate 1870. Honduras (Atlantic): Ancey 1886; Boone 1925; Britton 1976; Clapp 1914; E. F. Garcia 1988; F. Haas 1941; F. Haas and Solem 1960; Petuch 1981b; Pilsbry 1930a; H. G. Richards 1938; Robertson 290 AMER. MALAC. BULL. 10(2) (1993) Guatemala (Atlantic): Fischer and Crosse 1870-1900; Goodrich and van der Schalie 1937; Hinkley 1920; Pilsbry 1920b, c. Belize: Boone 1925; K. B. Clark and De Freese 1987; Riitzler and Macintyre 1982. Mexico (Yucatan Peninsula, Caribbean coast): F. C. Baker 1891; Bequaert and Clench 1933, 1936, 1938; Bergh 1890; Branson and McCoy 1963; Carnes 1975; Ekdale 1972, 1974; Fischer and Crosse 1870-1900; Gonzalez et al. 1991; F. Haas 1941; Harry 1950; Hidalgo 1956; Jaume 1946; Kornicker et al. 1959; D. R. Moore 1973; Perrilliat 1981; Pilsbry 1891, 1920b, c; Rehder 1966; Rehder and Abbott 1951; Rice and Kornicker 1962, 1965; H. G. Richards 1937; Strebel 1873-1882; Treece 1980; H. E. Vokes 1983; Vokes and Vokes 1984; Weisbord 1926. ACKNOWLEDGMENTS For assistance in literature acquisition, we are indebted to Librarians Kristen Metzger and Marilyn Morris and volunteer Marguerite Repass, all of Harbor Branch Oceanographic Institution. Dr. Gary Rosenberg (Academy of Natural Sciences, Philadelphia) offered a number of useful additions to the manuscript. This is Harbor Branch Oceanographic Institution Contribution no. 970 and contribution no. 335 of the Smithsonian Marine Station at Link Port. LITERATURE CITED Bieler, R. and A. R. Kabat. 1991. Malacological journals and newsletters, 1773-1990. Nautilus 105(2):39-61. Harris, G. D. 1921. A reprint of the more inaccessible paleontological writings of Robert John Lechmere Guppy. Bulletins of American Paleontology 8(35):i-iv, 1-198 (149-346), pls. 1-10 (5-14). Jacobson, M. K., comp. 1971. Index to the Revista de la Sociedad Malacélogica ‘Carlos de la Torre’ volumes 1-9, 1943-1954 (all vol- umes issued). Sterkiana 44:1-44. Date of manuscript acceptance: 16 July 1993 BOOK REVIEW Field Guide to Freshwater Mussels of the Midwest Illinois Natural History Survey Manual 5. Kevin S. Cummings and Christine A. Mayer. Illinois Natural History Survey, Champaign, Illinois. 1992. 194 pp. $15.00 This manual, intended for aquatic biologists and amateur naturalists, fills a recognized void in practical identification guides to the freshwater mussels. Current guides to mussels are of limited value to frustrated novices with bags of shells not readily identified by line drawings in the national and state-level keys available, but difficult to procure. The nearly pocket-sized guide provides coverage of 78 species arranged in systematic order. Using the assumption that nearly all midwestern mussels can be iden- tified by written descriptions and especially color pho- tographs, with concurrence from range maps, the authors provide a key to subfamilies followed by species accounts, with nomenclature largely following Turgeon et al. (1988). The matching of valves in hand to those in color plates pro- vides a simple and largely effective means to identify most species likely to be collected. Problematic taxa, such as pigtoes of various sizes and rivers of origin, will continue to befuddle even the most astute conchologists. Species accounts consist of one-page narratives arranged as follows: common and scientific name, other common names, key characters, similar species, descrip- tion, habitat and status. The opposing page contains a color photo of the external valve and a range distribution map for the Midwest. Photos depict representative specimens, with sexually dimorphic species represented by both sexes. The 294 pictorial guide will facilitate comparison of collected valves to identified species; a conchological ‘‘match the catch”. Inclusion of key characters and a list of look-alikes makes this guide very user friendly to all collectors. Because the range maps portray historic distributions only, they are marginally useful to nonspecialists. For example, the winged mapleleaf probably consists of a single popula- tion in the St. Croix River, Wisconsin, yet the map shows it to be widespread in seven states. Additional accounts on fingernail clams, asian clam, and the zebra mussel cover the other bivalves likely to be collected with unionids. The text's education message on status of this mol- luscan family in the Midwest is clear; more than 50% of species are threatened or endangered throughout their range or in various states where they reside. The colorful and conspicuous native mussels in all rivers are in deep trouble because of habitat loss and degradation, and on a collision course with the invading zebra mussel. This reasonably priced field guide is a valuable diagnostic and instructional tool for all mussel enthusiasts, and it will hopefully serve as a model for any future state or regional keys. —Richard J. Neves Virginia Tech. Blacksburg, Virginia AMERICAN MALACOLOGICAL UNION FINANCIAL REPORT 1992 Income and Expenses Income: TO9O Ge 1991: Dues (all cate Orie) 5 .ilecs.scucssueiesees oye des rae ete cee ee ue eee ee weet re $ 60.00 [992 Dues: (all Cates Ones) 5... 2acoecs tee cess ce sag season ricdeienee ose eee 5,801.00 1993 Dues:(all Cate mOries ) ces. csch st cithee tat, ensaaste ev eatscs Navecaceauies caveman ete cei gep neeee eee 2,723.00 Bulletin Subscriptions = VOlwme ® ..ssisessscist ascdiensear eee tetera etme anki 134.00 Bulletin Subscripuions:= VOlMMe: 9... i sasecsvcovcs vesstuasesccas ssa eyensnsene dsekcaseapseenvaivaeaeee eee ee 337.00 Bulletin Subscriptions = Volume 10 iis..40.020065 2 ssiuaaiaasstieceones anette eee ae ee 1,954.00 Bulletin, Subscnpuons = VOLUME UY sosccsecadceusecesnceesseetee be ralieseess.c cat occeecaceleteos cater egasee cue asec ee oe! 645.00 Bulletin back issues Gc Supplements scsscdcsci cacacts .cdshjoccescte sh ckasnczatestetceuths. ceo ayeasies siden aeueree ee aes ene 1,153.65 Page charges and reprints: (Vol. 901) (2) issccssssteissncsvacastoartecasnesinystapssseecsssarvautieauetanniczaccesa se eee, 2,726.24 MEMEF PUD] CALLOMS ac sss shiistcaciisncavionsinneas mesaadetemes euatentensynssaneuutensderveanesie to pennidvanetavnetneedtuagaay uanetysuvieeoranuneceeteenekeas 187.00 PSHE SALES x ced.stswess eos sie acatesvacostcussd esa ast oadesustens cuteceucsateaspenuas tat assistive stuaueedecdidyiiee nazar cemsteraet se erecree eee rae 28.00 Stadent Aware Donation 2s cates osc cssceee aise atioe cy eves boahec aecennaece oaceoeeee a atece etre see caeecneeee nan ee ene 250.00 PRUECATON 5255s ss eas actagdn oh oa ccs ise snthoal sas susan ade val gsousienesz scameics hea aneotena Seehe tie eee eeieereese cate ate eee eee eee ee 608.00 Interest (all ACCOUNTS) fe ceceecc lesen ces vaveresce st iaaccdeed edbdzscas Seciveevis coe tivsoseek es oe ae 5,926.45 MiISCellAmeOUS BICORNE «geeks foc carcasses cae tanta tain ues ban ua desaphiecans Cvs auseaan ts tanee cosine toaah nes ee meaacceeee ate eee 156.87 Symposiim Fund Don attons siesessstesacies veniiscwtasaidtecinast salah adakecetetaas sesawsn davai gueiuesacuaeoraseseneventheceenettemmeseetes 459.00 TOT AM eos eee saat ee $ 23,149.21 Expenses: OTEICE EXPENSES xb sessccconanseckateainicle tess sescatscitabd cc asias a hays taendae a ecaet isle ec ee cecewe anes ee teen eee 407.45 PUMA Bassa costa cocsneuiaaseagentcavensoesoveboootiennastioyiadevsnanitvpasdunadvehadueancsvnentvean sevbavuubieoandiatecdseotearegeyaeeysDepesseeegtenseuee ts 421.09 Telephone amd) POStA we ius. cctce cc sccepes sch tas sv pcos Saaee ees cota g eaaeceeeeeeeea aac sndeegce eceantecnescsta canta sat tee neeee een 688.23 Dues in other Organ ZatlOns szsicseces cs sscncccaecesssissugustu ceecisees consivesessosaaseeusesa savdnae saudeasveasbanuaanessosossaueceucasense meas 265.00 California Franchise Fee p2 0.001; * - 0.05 > p > 0.01; NS - p = 0.05 [not significant]. (B, burrowing orientation; D, umbo downstream, U, umbo upstream). Mean in grams. Sculptured Emo | -1.5923 | mean Eee eee : 6.1571 9.2466 20.7972 H 1.5493 -— tt Cen a FT | 20.7972 | 22.3176 14.5324 -3.2308 | 2.9017 8.7393 2.4071 4.6584 -8.0650 -7.6930 i sa {| 2.6518 -9.2409 -9.1769 -4.4795 | 0.4441 4.1453 7.2317 Smoothed Pp} | -10.1987 | -10.3281 | -5.7957 | -2.3737 4.4142 co 12.0377 | 3.4278 | 9.4788 | 11.0801 | 12.5642 8.0620 — 7.1761 Lee 3.3916 37.4 4 AMER. MALAC. BULL. 11(1) (1994) Table 2. Comparison of burrowing and anchoring drag in sculptured and smoothed shells of Obliquaria reflexa in sand and sand/gravel mixture. Explanation as in Table 1. Sipe Simoathed | Send & gravel | Send Send grovel u_| 6 | ae ee a EAaDresl ae ae 2.7784 | 20.8825 | 21.3674 | 15.9852 | 13.8360 | 8.1172 | 12.3375 | 21.6800 | 21.3384 | 25.1743 | fanst_| 262) asia | yee | sage | oe | ee ee ee ae | 27.6648 | 29.9217 | 21.3789 | 18.6306 | 11.2447 | 15.8441 | -4.6208 | -6.3131 | 8.8523 | -4.1318 | 2.2409 | 1.3828 | 2.5587 | | | 3.8281 | 10.0047 | 9.5982 | 11.1359 | _| 5.5525 | 4.9913 | 6.0290 | | ns -0 | ons | ns | | 43.4 | 22.2 | 246 | 23.0 | am a LB! EB! 28 a Ae Oe Table 3. Comparison of burrowing and anchoring drag in sculptured and smoothed shells of Quadrula pustulosa in sand and sand/gravel mixture. Explanation as in Table 1. Sculptured [send grver Sonar gee Po [uv fs | ED & gravel | 8 | oo Pee ise ee peo am [| > Table 4. Comparison of burrowing and anchoring drag in sculptured and smoothed shells of Amblema plicata in sand and sand/gravel mixture. Explanation as in Table 1. Smoothed 8. | ot | 10.2793 | 12.9917 | 1.1543 | 18.4661 | 12.5064 | , | 10.2549 | | 15.2313 | 10.2549 16.8974 -5.0874 3.1478 poo fe 7.4066 i a NS 6.7195 Fe AOE al 11.4401 ae ee 18.0156 ee oa Seances | Ns | eee | 48.2 | & gravel 9.4808 0.2039 Bee 1 BA x a N U U U U | 43.0 | The angles of the medial anchor and U2-D2 arms tion of 4.9. The U2-D2 anchor has a mean of 35.4° and are grouped in Fig. 4. Both groups have a single well- standard deviation of 2.7. defined mode and a narrow range of actual values. The The values of k derived from the allometric equa- medial anchor has a mean of 77.8° with a standard devia- tion fit to sculptural intervals show that all species exam- WATTERS: UNIONOIDEAN SHELL FORM AND FUNCTION 5 ined possess allometric densing (Fig. 5). Values ranged from 0.23 to 0.50, all well below the isometric value of 1.0. The interval between ribs or knobs in medial anchors or U2-D2 arms is smaller than would be expected by isometric growth. The results of the anti-scouring experiment are shown in Fig. 6. Like the results of Stanley (1981), these are strictly qualitative. It is apparent that the smoothed side was excavated to a greater extent, with more sediment removed revealing more shell than with the unaltered side in both species, even though not all sculpture could be eliminated from the shell of Tritogonia verrucosa. DISCUSSION ADAPTATIONS TO SOFT SUBSTRATA Perhaps the foremost danger to an organism living in or on a soft substratum is sinking (Eagar, 1950). This lia- bility may preclude invasion by large predators and equal- sized competitors, which would sink just as readily. The underlying consideration of a functional model or paradigm for shell form and sculpture for this type of bivalve is fairly unambiguous: increase buoyancy. This may be attained by two means, both of which occur in the freshwater bivalves: increase aspect to substratum and reduce shell weight. This discussion is limited to hypothetical considerations of the paradigm, and does not have the weight of experimental evidence found in other parts of this study. Lea (1829) recognized that some unionoideans pos- sessed a structure in which the shell extends dorsally above the hinge posterior and anterior to the umbo, uniting the left and right valves in an uninterrupted fold. He referred to these as symphynote shells. To separate the valves com- pletely, the shell must be broken along this structure, which 100 T no) B a0) = O x< LU 60 4+ | ” Cc oO £ S 40 | Q op) re) xe 20 + + }_ A O =i Vis T ‘ ti 0 20 40 60 80 100 120 140 160 180 Angle on Shell Disc Fig. 4. Distribution of U2-D2 arm angles (dark shading) and medial anchor angles (light shading) in 5° increments over disc of shell. ® 2 2 2 2 eee a B C D Oo 2 B 3 - 51 1 Anim 1 _" Ac ro) D o S ae k=0.50 & k=0.42 a k=0.23 0 k=0.36 a6 1 20 1 20 1 2 0 1 2 Log Distance from Umbo Fig. 5. Allometric plots of distance between sculpture (medial anchor or U2-D2 arms) and distance from umbo. ™@ , observed values. Diagonal line - expect- ed isometric line (k = 1.0). k, allometric coefficient for observed values. (A, Amblema neisleri; B, Plethobasus cicatricosus, C, Obliquaria reflexa; D, Plectomerus dombeyana). 6 AMER. MALAC. BULL. 11(1) (1994) Fig. 6. Effects of anti-scouring sculpture (left valve) and smoothed (nght valve) in sand. A, Quadrula quadrula; B, Tritogonia verrucosa. Direction of current is from top to bottom. Shell is oriented with umbo at bottom. encases the functional ligament. This structure has arisen independently in many genera: the unionids Lasmigona, Leptodea, Potamilus, some Anodonta, and others; and in the hyriids, e.g. Prisodon, Paxyodon, and Triplodon (Fig. 7). Symphynote species are rare in Recent marine groups, occurring most often in the Pinnidae, Pteriidae, and Mytilidae. In both the Pinnidae and Mytilidae, the sym- phynote condition appears to act as a secondary hinge in these essentially teeth-less groups. Members of the Pteriidae are largely byssate, epifaunal species, often occur- ring in areas of significant wave action. The symphynote “wings” or alae have been shown to act as rudders that enable the bivalve to pivot along its byssus and maintain a minimum aspect into the current (Stanley, 1972). Most symphynote unionoideans are large, unsculp- tured, thin-shelled, and complanate. These are the same characteristics that Thayer (1975) predicted for bivalves in soft substrata in general. The habitat of these freshwater bivalves is often fine sand or mud, with or without a cur- rent. A compressed shell offers less resistance to a current than an inflated one of similar volume in some orientations. It will lie flat on the substratum as the animal reburrows, being held down by the overlying current (pers. obs., Fig. 8). A complanate shape also will act as “snowshoes” on soft substratum. This form is rare in marine groups. The contemporary marine Corculum, a cardiid, has the valves expanded laterally to rest on the substratum surface, although this may be an adaptation for maximizing light to its symbiotic algal colonies (Kawaguti, 1950). The Permian alatoconchids, also laterally expanded, appear to have among the most extreme “snowshoe” form of any bivalve (Yancey and Boyd, 1983). In contrast, the complanateness in unionoideans is a result of peripheral (gnomonic) expan- sion of the valves rather than lateral expansion. Although the alae could also function somewhat as “secondary ligaments” (Owen, 1958; Savazzi and Peiyi, 1992) in unionoideans, many symphynote species possess a fully developed true ligament. However, many symphynote species also have reduced dentition. In these groups, the alae also function as hinge teeth by aligning the valves dur- ing closure and preventing shearing in those forms lacking well-developed dentition. These species often are found in soft substrata where shearing is minimal. Thus, although the hinge no longer fulfills its usual functions (valve align- ment and anti-shearing) in some unionoideans, that func- tion has been exapted to varying degrees by the alae. The loss of dentition is characteristic of unionoideans living in soft sediments, as explained below. Thus symphynote shells may function in two differ- ent ways: 1) as a stabilizing shape in a current preventing the animal from being washed away if dislodged, and to attain a favorable orientation when righted; and 2) as a buoyancy or “snowshoe” device in very soft substrata. The function of alae is here interpreted as a method of increas- ing the degree of complanateness, with little increase in actual volume, while also acting as functional replacements Fig. 7. Examples of Recent symphynote unionoideans. A, Hyriopsis bialata Simpson, 1900 (after Haas, 1969). B, Cristaria plicata (Leach, 1815) (after Clessin, 1873-1876); C, Lasmigona complanata (Barnes, 1823) (after Kuester, 1858-1859). WATTERS: UNIONOIDEAN SHELL FORM AND FUNCTION a) ty ys Fig. 8. Symphynote shape acting to hold shell to substratum during burial. Arrow, direction of water flow (shell modified after Kuester, 1858-1859). for hinge teeth. Reducing the density of the organism relative to the substratum is a straightforward problem. The weight of necessary organ systems scarcely can be diminished, but that of the shell, which is much denser than the soft parts, can be decreased significantly. The relative specific gravity of the whole animal is reduced in species having a thin shell (Eagar, 1977, 1978; Anderson and Ingham, 1978; Ghent et al., 1978; Seed, 1980; Eagar et al., 1984). Three avenues are available to lessen shell weight: decrease shell thickness; eliminate sculpture (heavy concentrations of shell material); or reduce or eliminate hinge teeth (also a major source of shell weight). Combinations of all three are found in soft-sediment unionoideans. Species of Anodonta, Leptodea, Hyriopsis, and Potamilus typically are thin- shelled, unsculptured, and have little or no dentition. These genera also are symphynote. Increasing the aspect of the shell in a dorso-ventral plane to the substratum also will augment buoyancy by increasing the amount of shell surface normal to the sub- strate. Globose shells are more buoyant than non-globose ones in this orientation (Anderson and Ingham, 1978; Ghent et al., 1978; Tevesz and McCall, 1979; Stern, 1983). When used together with thin shells, this can be an effec- tive form in soft substrata, although such shells are suscep- tible to dislodgment by a current over that same substratum (Menard and Boucot, 1951). Few unionoideans have such a shell, among them: Anodonta grandis s.l. Say, 1829, Potamilus capax (Green, 1832), and some species of Toxolasma and Lampsilis (including the Miocene Lampsilis marshalli MacNeil, 1935). These species can be found in Fig. 9. Examples of divaricate sculpture in unionoideans. A, Recent Scabies scobinatus (Lea, 1856) (after Haas, 1969); B, Recent Nodularia fluctigera (Lea, 1859) (after Lea, 1860); C, Recent N. douglasiae (Gray, 1833) (after Lea, 1869); D, Cretaceous “Unio” holmesianus Dyer, 1930 (after White, 1907); E, Cretaceous Nippononaia ryosekiana (Suzuki, 1941) (after Suzuki, 1941); F, Recent Triplodon rugosus Spix, 1827 (after Haas, 1969); G, Recent Quincuncina infucata (Conrad, 1834) (after Burch, 1973); H, Oligocene Diplodon latouri (Pilsbry and Olsson, 1935) (after Parodiz, 1969); I, Cretaceous Proparreysia letsoni (Whitfield, 1907) (after Whitfield, 1907); J, Recent Chevronaias colombiana Olsson and Wurtz, 1951 (after Olsson and Wurtz, 1951); K, Recent Parreysia houngdaranicus (Tapparone-Canefri, 1889) (after Prashad, 1922). 8 AMER. MALAC. BULL. 11(1) (1994) muddy sand or silty environments in quiet water. Similar shell characteristics were found in marine bivalves that inhabit soft substrata (Nichols et al., 1978). The soft substratum modifications of the shell have evolved independently in many unionoidean lines, resulting in convergence of shell shapes. Many members of one group, the anodontine unionoids, have shell forms that seem especially evolved for this habitat. This group has anatomical and reproductive differences that set it apart from other unionoideans (Heard, 1975), a unique parasitic fauna (Vidrine and Bereza, 1977), and could represent an early offshoot of the Unionidae that exploited soft sub- strata. ADAPTATIONS TO HARD SUBSTRATE Divaricate Sculpture: Divaricate sculpture is surface orna- mentation that forms a ““V”-shaped pattern, often combin- ing to form several continuous “V’’’s (Fig. 9). This pattern is relatively easy to model mathematically, but considerably more complicated to explain biologically. Models for divar- icate color patterns have been developed. North (1987) seemed to have given credit for the recognition and simula- tion of this pattern in the coloration of mollusc shells to Meinhardt and Klingler (1987), ignoring the earlier, very similar work of Waddington and Cowe (1969). Zimmer (1992) reported on the models of Fowler et al. (in press) which generated shell shape and color patterns in remark- able detail. Ermentrout et al. (1986) produced a model that also created color patterns similar to those found in mollusk shells. These models may be extended from the deposition of color to the deposition of shell material. Although the area of sculpture may be concentric [or, more properly, gnomonic (see Stasek, 1963)], it cannot be deposited by the same accretionary process that forms concentric or radial ribs. Because the sculpture runs obliquely to the line of shell formation, the area of deposi- tion must migrate laterally. To form “V-,” or as is more common, “W”-shaped ridges, two additional conditions must be met: 1) the initial sculpture must originate at a sin- gle point and usually diverge; and 2) when two ridges con- verge, they end. Nearly all unionoidean sculpture can be derived from a simple divaricate pattern. It seems likely that this pattern originated in the ancestral stock and has been exapt- ed for new functions. The trigonioideans are considered by many workers to be the ancestors of freshwater bivalves (Morris, 1978). Tevesz (1975) and Newell and Boyd (1975) have pointed out the similarities between unionids and Neotrigonia: similar gill ciliary patterns, unfused mantle, prismato-nacreous shells, and filibranch gills. Stanley (1977a: 211) believed that the unionoideans were derived from “relatively unspecialized early representatives” of the trigonioideans. Unionoidean sculpture can be derived from an ancestral divaricate pattern found in many trigonioideans, such as Laevitrigonia, Quadratotrigonia, Litschovitrigonia, and Malagasitrigonia, for example. Sculpture has been shown to be a conservative feature in trigonioideans (Stanley, 1977a), and other trigonioideans display the same derivations of the basic pattern as those found in the unionoideans. However, trigonioideans also possess sculp- ture that is very rare in unionoideans, although common in marine bivalves: concentric and radial ribs. Perhaps only the Triassic Diplodon gregoryi Reeside, 1927, and the Jurassic Plicatounio have true radial ribs, and no Recent unionoidean that I have seen has concentric sculpture [with the doubtful exception of the Recent Lampsilis straminea straminea (Conrad, 1834)]. Carter’s comment that “Castalia ambigua [(Say, 1825)] [is] the only fresh water lamellibranch to possess radial ribs” (1968: 55) was incor- rect. Other Castalia, also Chevronaias, have “radial ribs.” However, these are not true radial ribs but divaricate ribs intersecting the shell margin at nearly right angles. Several of the oldest unionoidean genera have dis- tinctly divaricate sculpture. Sulcatapex from the Lower Cretaceous has divaricate beak sculpture and ribs on the dorsal slope. Nippononaia, also from the Lower Cretaceous, has pronounced divaricate sculpture, as does its potential descendant, Trigonioides. The Triassic Diplodon haroldi Reeside, 1927, is one of earliest to show this sculp- ture clearly. This type of sculpture thus has been evident in freshwater bivalves for at least 120 My. For ease of explanation, a simple notation is used here to identify the parts of the prototype divaricate sculp- ture (Fig. 10). The apices of the pattern are numbered U1- U3 (U = up) from anterior to posterior. The apices point dorsally. The valleys of the pattern similarly are numbered D1-D3 (D = down). Thus the U’s represent points of origi- nation and potential bifurcation, while the D’s are points where converging lines cancel out. A rib (or line of pus- tules) between two points is referred to as U1-D1, for example. Some species appear to possess more divarica- tions than those found in the prototype, but the modifica- Fig. 10. Prototype divaricate sculpture and types of modifications. WATTERS: UNIONOIDEAN SHELL FORM AND FUNCTION ) tions essentially remain the same. The similarities between sculpture are geometric and do not necessarily represent phylogenetic paths. Similar sculpture may have arisen by convergence in unrelated groups. Divaricate sculpture may be geometrically modified in unionoideans in four ways: 1) Asymmetry - all points of U and D may not be equidistant from each other, resulting in long and short U-D segments. 2) Suppression - prototype U-D segments may not be expressed. The U1-D1 arm often is absent. The U3-D3 arm usually is retained. 3) Enhancement - U or D points, rarely both, may be exaggerated into knobs or spines. The enhancement of points often accompanies suppression of arms. 4) Smoothing - the angular nature of the sculpture may become sinusoidal. This typically occurs at D points. Divaricate sculpture in unionoideans often is divid- ed into pustules, or combinations of ribs and pustules, and portions of the pattern may be suppressed. Erosion and injury can make the pattern unobservable. Sculpture also may be modified during ontogeny. It may become less pro- nounced, or completely disappear, during growth. Rarely, the sculptural pattern can change abruptly. This occurs most often with the metamorphosis from nepionic to adult or metaconch. These conditions have the result of obscur- ing the underlying divaricate pattern. However, most speci- mens clearly show divaricate sculpture. In many species the apparent underlying pattern is modified by one or more of the means mentioned above. In Amblema, for example, the U2-D2 rib occupies most of the shell surface and may be the only remnant of the prototype pattern. Obliquaria usually retains only the U3-D3 arm as a series of ribs on the posterior slope (as do many species, see below), and the D2 point as an exaggerated knob. The unionids of the polyphyletic “genus” Canthyria, and juve- nile Unio pictorum (Linné, 1758), have enhanced the D2 point into a spine. Parreysia bhamoensis (Theobald, 1873) also has juvenile spines (von Martens, 1899). Seilacher (1973) found that some shell features seemed to constitute an array of burrowing sculpture, con- sisting of four features that readily can be observed: 1) Cross orientation - sculpture should be normal to burrowing for maximum friction. This includes divaricate sculpture that is only normal to movement at specific inter- vals during rotational burrowing. 2) Frictional asymmetry - the sculpture should pre- sent less friction in the direction of movement normal to it. This results in asymmetrical sculpture, also called “ratchet sculpture.” Cross orientation usually accompanies frictional asymmetry to be functional as burrowing sculpture. Stanley (1969, 1970) observed that there were relatively few marine bivalves with this type of sculpture. 3) Perimeter smoothing - sculpture should be mini- mized at the area of maximum aspect (i.e. the shell disc) to minimize friction during burrowing. Seilacher’s choice of the word “perimeter” potentially is misleading, as his perimeter is not the shell margin but the cross-section of the shell perpendicular to burrowing. 4) Allometric densing - sculpture size should remain the same magnitude throughout the growth of the shell. Typically, sculpture would be expected to increase in size proportional to shell growth. If the sculpture does not change size, then it is decreasing allometrically relative to shell size. The rate of sculptural growth is less than the rate of overall shell growth. This results in sculpture of approxi- mately the same magnitude in juveniles as in large adults. This is thought to be an optimizing modification where the sculpture is most efficient against a particular sediment grain size at a particular sculpture size and spacing. Since the sediment particles do not increase in size as the bivalve grows, it would be advantageous to maintain sculpture at an optimum size, hence the allometric change in sculpture size relative to overall size increase. Allometric densing rarely is perfect as growth aspects typically can be observed. Most unionoidean sculpture does not fit all four of Seilacher’s criteria. Divaricate sculpture often is not orient- ed with burrowing, eliminating feature | of the paradigm. Although divaricate sculpture in some marine groups has been shown to enhance burrowing (Raup, 1969; Stanley, 1969, 1970), in unionoideans that particular orientation is unknown or the magnitude of the sculpture apparently is ineffectual. Frictional asymmetry (feature 2) is common in unionoideans, but is not oriented with burrowing. Furthermore, it is often oriented in opposite directions in different populations and as often the same sculpture may be symmetrical in others. Savazzi (1982) found that some marine species without frictional asymmetry can still have burrowing sculptures, but noted that those species do not burrow by the same method of most marine (or freshwater) forms. The third feature, perimeter smoothing, is replaced by “perimeter roughing” in most unionoideans, where the cross-sectional aspect is often the most highly sculptured region of the shell. Absence of sculpture on the anterior portion of the shell in otherwise sculptured species (anterior smoothing) also is not predicted by this model. Only the criterion of allometric densing (feature 4) is met in any degree in unionoideans (Fig. 5). In adult specimens of many species, the pustules and ribs are of fairly uniform size and interval across the shell (e.g. Amblema, Megalonaias, and Plectomerus). This is not unexpected if allometric densing optimizes sculpture for a particular sedi- ment particle size, for whatever function. Based on Seilacher’s criteria for burrowing sculp- 10 AMER. MALAC. BULL. 11(1) (1994) ture, unionoidean sculpture does not fulfill the conditions. Unlike marine species, burrowing sculpture is not common in unionoideans. It is proposed here that the functions of this sculpture are composite, with some parts acting as anchoring devices during and after burrowing against exca- vation by currents, and other parts as anti-scouring devices. Savazzi and Peiyi (1992) also reached this conclusion. The Medial Anchor: A recurring theme in unionoidean sculpture is the medial anchor, a radiating area of sculpture across the medial disc of the shell. It can be the only sculp- ture on the shell or be part of a suite of composite structures (Figs. 11, 12). The anchor may be a series of pronounced knobs or spines. Often populations of taxa normally having a generalized anchor (see below) will have all sculpture absent except that forming the medial anchor. This sculp- ture is interpreted as an anchoring device during and after burial by a unionoidean against excavation by water cur- rents. The sculpture can be substratum specific given the presence of allometric densing. The broad ribs of Amblema, Megalonaias, Arcidens, and others, also function as anchors, but differ in their derivation from the hypothesized prototype pattern, and will be discussed separately. This structure is widespread in several unrelated genera, and may occur in only a few species in a genus. Examples of unionids with medial anchors include: Obliquaria reflexa, Plethobasus cyphyus, Epioblasma toru- losa torulosa, and Quadrula nodulata. It reaches its most Fig. 11. Diagrammatic representation of a medial anchor. extreme development in the contemporary genus Canthyria, which may represent a polyphyletic group hav- ing convergent sculpture. Fossil species such as Pletho- basus aesopiformis, and Pl. biesopoides Whitfield, 1907, Proparreysia verrucosiformis Russell, 1976, and Pr. pau- cinodosa also possess medial anchors. The prominent row of knobs on the Lower Cretaceous trigonoidean Malagasitrigonia resembles a medial anchor. The function of the medial anchor may be modeled and tested by a simple set of hypotheses: 1) the sculpture should be functional for its purpose; but 2) its correspond- ing detrimental effect on other functions must be mini- mized; 3) the anchoring device should be normal to the direction of potential excavation; and 4) for maximum effect, the sculpture should extend across the greatest area of the shell. Conditions 3 and 4 are structurally correlated in a bivalve shell because the normal axis of condition 3 is the approximate medial radius of growth. In elongated species, condition 4 may be compromised as the height becomes significantly less than the length, but these taxa rarely have a medial anchor. To meet condition 2, the sculp- ture should offer the least resistance in all attitudes other than the anchoring position. This is accomplished by elon- gating the sculpture perpendicular to the anchoring direc- tion. Although this sculpture necessarily increases the drag of non-anchoring functions, its benefits apparently out- weigh the increased friction. We can construct an hypothesis to test this paradigm based upon conditions | and 2, and a separate one for con- dition 3 (4, as stated, may be unmet). If the sculpture acts as an anchor, it should create more drag than a non-sculp- tured species of similar size and weight, or the same shell with the sculpture removed. Furthermore, if condition 2 is to be met, the resistance should be less for orientations other than that of the anchoring attitude. Comparisons of the amount of force (grams) needed to dislodge two unionid species with medial anchors are summarized in Tables 2 and 3. The anchors create drag in anchoring positions, with minimal drag along the direction of burrowing. The sculpture nevertheless produces substan- tial drag when compared with smoothed specimens. There is an indication that the sculpture may be optimized for the natural substratum and in the most common life orientation - with the umbo directed downstream. Condition 3, that the anchoring sculpture should be normal to the direction of excavation, may be tested with two hypotheses. First, if sculpture does not have a function, there is no reason to suppose that sculpture should be con- centrated on any one part of the shell disc more than on any other. If, however, sculpture occurs consistently in one sec- tor and rarely elsewhere in many species, then a functional explanation for its position is suggested. Second, if sculp- WATTERS: UNIONOIDEAN SHELL FORM AND FUNCTION 11 \ Fig, 12. Examples of the medial anchor in unionoideans. A, Recent Canthyria spinosa (Lea, 1836) (after Kuester, 1858-1859); B, Recent Parreysia bhamoensis (after von Martens, 1899); C, Recent Quadrula quadrula (after Lea, 1832); D, Recent Dromus dromas (Lea, 1834) (after Kuester, 1858-1859); E, Recent Epioblasma torulosa torulosa (Rafinesque, 1820); F, Recent Obliquaria reflexa; G, Recent Q. nodulata (Rafinesque, 1820); H, Recent Plethobasus cyphyus (Rafinesque, 1820) (E-H after Burch, 1973); I, Cretaceous P. aesopiformis (Whitfield, 1903) (after Russell, 1976); J, Cretaceous Proparreysia letsoni (after Whitfield, 1907); K, Cretaceous P. paucinodosa Russell, 1976; L, Cretaceous P. biesopoides (Whitfield, 1907) (K-L after Russell, 1976). ture is found concentrated in one region, then to act as an anchor that region should be oriented to maximize drag in the direction of excavation. The angle of the medial anchor sculpture (Fig. 4) was measured for 21 species. The results show a narrow range of anchor angles with only a single, well-defined peak, approximately normal to the shell length. In fact, the angle is closer to being normal to the actual angle of exca- vation because most unionoideans do not typically burrow at right angles into the substratum, but at an angle to it, resulting in the angle of excavation being slightly less than the predicted 90°. The observed offset of 12° may be a compensation for this effect. The sector of the shell possessing the medial anchor corresponds to one of three points of the divaricate proto- type sculpture: the U2, D1 or D2 point, and because of the accretionary growth process, the anchor will be a sector. This necessarily limits medial anchors to shells that are cir- cular or nearly so. As elongation occurs, the U2 and D2 points are skewed posteriorly, removing them from their functional position. This is substantiated by the fact that no elongate species have medial anchors as adults. Species having pronounced medial anchors typical- ly are riverine species. Big river individuals of Dromus dro- mas have a medial anchor, while headwater specimens do not. The big river subspecies Epioblasma torulosa torulosa has a strong medial anchor, but it is weak in the headwater subspecies E. t. rangiana (Lea, 1839) and E. t. gubernacu- lum (Reeve, 1865). This feature, and others discussed below, form a suite of “Big River’ morphological charac- teristics. Although it is easy to explain why these attributes are suited to big rivers, it is difficult to explain why they would not be beneficial in small rivers and creeks. These characteristics will be addressed in a later section. The medial anchor is a refinement of existing sculp- ture (the generalized anchor, below) in response to a habitat characterized by a strong current. Sculpture such as the medial anchor may be retained despite suppression of other sculpture. In some groups, such as Quadrula, the medial anchor occurs along with other sculpture. In Q. pustulosa, 12 AMER. MALAC. BULL. 11(1) (1994) the medial anchor can be present alone or be part of the generalized anchor. In Q. nodulata, the medial anchor is nearly the only sculpture expressed. In the Cretaceous Proparreysia letsoni, both divaricate sculpture and anchor- ing knobs are present. The results suggest that poorly developed medial anchors are less effective than well-developed generalized anchors. The evolutionary shift from a generalized to an effective medial anchor cannot be made gradually without passing through a less efficient phase with decreased selec- tive advantage (unless an additional, unidentified character is synergistic). Well-developed medial anchors such as those on Obliquaria reflexa could arise through major mor- phological shifts. This may explain the relative rarity and scattered occurrence among genera of this sculpture in the unionoideans, and why many of the taxa possessing it are found in either numerically small or monotypic genera. The medial anchor is best considered a more efficient refine- ment of the generalized anchor in which sculpture is retained on the shell disc only in its most functional posi- tion. In some genera, such as Amblema, Megalonaias, Arcidens, a medial anchoring device has taken a form dif- ferent from that found in Obliquaria reflexa, and is derived by a different method from the prototype sculpture (Fig. 13). In these groups, the U2-D2 arm has been extended across most of the shell disc (modification by asymmetry), occasionally with all other sculpture suppressed. In the lat- ter case, the same arm acts as an anchor on the disc of the shell and an anti-scouring device on the posterior slope (anti-scouring sculpture is considered in a later section). The angle of the arm does not meet the paradigm for bur- rowing sculpture, but functions as an anchor, and is contin- gent on the geometry of the hypothetical prototype sculp- ture. To bring an arm into an anchoring position requires that the arm be greatly lengthened. Lengthening moves the U1-D1-U2 arms so far forward that they are eliminated by anterior smoothing. The D2-U3-D3 arms are moved so far posteriorly that the U2-D2 arm may usurp their function on the posterior slope as an anti-scouring device. As with the medial anchor, the angles of the U2-D2 arms for different species (Fig. 4) are very similar to each other (mean = 35.4°; SD = 2.7), suggesting a common function. This U2-D2 sculpture fulfills the same function as the medial anchors, but has arisen by a different route. Clarke (1982) believed that the ridges aid in maintaining the effective orientation of the bivalve in the substratum. He did not state how he envisioned this anchoring function to occur, but proper orientation would be expected as the by- product of anchoring devices. In general, Amblema-like sculpture forms anchors by enhancing divaricate arms while Obliquaria-like sculpture forms anchors by enhanc- ing divaricate points. Clearly, not all riverine species possess such anchor- ing devices. It must be stressed that anchoring devices are but one evolutionary response to a riverine habitat and that other responses have occurred in other groups. Analogously, Seilacher (1973: 451) found sculptured and smooth marine bivalve forms living side by side in soft sub- strate. He concluded that ‘“‘Why the one or the other strategy is chosen in a particular case, remains unknown, particular- ly since smooth and sculptured forms are found living side by side.” The Generalized Anchor: The least functionally special- ized type of anchor is here interpreted as the most primi- Fig. 13. Examples of U2-D2 anchoring sculpture in unionoideans. A, Recent Arcidens confragosus (Say, 1829), B, Recent Amblema neisleri (A-B after Burch, 1973); C, Recent Megalonaias boykiniana (Lea, 1840) (after Simpson, 1900); D, Recent Elliptoideus sloatianus (Lea, 1840) (after Lea, 1842); E, Cretaceous “Margaritana” nebrascensis White, 1883; F, Cretaceous “Unio” belliplicatus Meek, 1864 (E-F after White, 1907); G, Miocene Megalonoidea rugicosta MacNeil, 1935 (after MacNeil, 1935); H, Cretaceous “Unio” gonionotus White, 1883 (after White, 1907). WATTERS: UNIONOIDEAN SHELL FORM AND FUNCTION 13 Fig. 14. Diagrammatic representation of a generalized anchor. tive. It occurs in many unionoids (particularly the amblem- ines) and hyriids, in small groups of often unrelated genera. The generalized anchor is characterized by many pustules, or pustules coalesced into ribs, that are not confined to any yi atin Wie) : hii ‘ae Whi, Gina wet oes gta ce hee Rigor tess specific area of the shell disc, but occur nevertheless in a divaricate pattern (Figs. 14, 15). The pustules and ribs are often symmetric individually, and therefore do not meet cri- teria 2 and 3 of the medial anchor paradigm. The sculpture acts as an anchor in most species, as suggested by Clarke (1982), but would hinder burrowing. It could be that in the absence of synergistic functional effects, the selective advantage of anchors (i.e. the ability to remain buried) is greater than the selective advantage of lacking sculpture (i.e. the ability to rebury efficiently) in many species. This type of anchor may be directional, the most efficient orien- tation being with the umbo downstream, the most prevalent position found in unionoideans in nature. Fossil unionoideans often have a generalized anchor that displays the ancestral divaricate pattern with little alter- ation. Species such as Proparreysia holmesiana, P. maclearni, and P. barnumi (all Dyer, 1930), Tritogonia natosini (McLearn,1929) (see White, 1907; McLearn, 1945; Russell, 1967, 1976), and Diplodon haroldi, all are sculptured heavily with divaricate patterns. Several trigo- nioideans also have this sculpture, e.g. Korobkovitrigonia, Steinmanella, and Litschkovitrigonia. Anti-Scouring Sculpture: In a water current, partially buried obstacles, whether they are stones or bivalves, tend to be excavated by scour. Water flowing near the water/sed- iment interface moves slower than water in overlying lay- ers, Causing a local vortex, which, when it contacts an "by Ise Fig. 15. Examples of the generalized anchor in unionoideans. A, Recent Cyclonaias tuberculata (after Burch, 1973); B, Recent Tritogonia verrucosa (after Kuester, 1858-1859); C, Recent Cyprogenia stegaria (Rafinesque, 1820) (after Burch, 1973); D, Recent Discomya radulosa (Drouét and Chaper, 1892) (after Haas, 1969); E, Cretaceous “Unio” verrucosiformis Whitfield, 1903 (after Whitfield, 1907); F, Recent Quadrula apiculata speciosa (Lea, 1862) (after Lea, 1863). 14 AMER. MALAC. BULL. 11(1) (1994) obstacle, will begin scouring the area around the barrier when a critical velocity is attained (Richardson, 1968). The resulting scour may destabilize the object to the point of dislodgment. Stanley recognized three different types of shell sculpture or form in marine bivalves that function to minimize scour. All three occur in unionoideans. A common sculpture appearing in unionoideans is radiating ribs or corrugations on the dorsal slope (Figs. 16, 17). This feature occurs in a minority of species of other- wise unsculptured genera, such as Elliptio, Lasmigona, and Ptychobranchus, and is nearly always present in sculptured species. Even when the shell is devoid of this sculpture, sometimes, such as in several Elliptio species, the perios- tracum becomes ribbed in this region over the smooth shell. Dorsal ribbing is found first in Cretaceous unionids such as Rhabdotophorus gracilis Russell, 1935. White (1907) illus- trated four: Unio senectus, U. goniambonatus White, 1878, U. gonionotus, and U. aldrichi. Dorsal ribbing acts as an anti-scouring device by breaking the scouring vortex into random currents. Thus, although some currents remove sediment, others deposit them, so that the net result is no scour. The ribs are formed in nearly all cases by U3-D3 arms. In some species the U2- D2 arm extends to the dorsal margin, where they function in the same manner. Dorsal ribbing is often an adult sculp- ture, although sometimes becoming obsolete on large indi- viduals. A maximum critical weight can be reached where anti-scouring sculpture is no longer needed. At the other extreme, nepionic shells are small enough that scouring is not a complication, the individuals being only slightly larg- er than some sediment particles. The main threat here is Fig. 16. Diagrammatic representation of anti-scouring sculpture. dislodgment by virtue of small size, and a premium is placed on anchoring devices, including byssal threads. Anti-scouring sculpture also can show allometric densing, where the magnitude of the sculpture does change isometrically during growth. This could be an optimization for a particular sediment size. Anti-scouring sculpture is most effective with small sediment particles. Although larg- Fig. 17. Examples of anti-scouring sculpture in unionoideans. A, Recent Quadrula rumphiana (Lea, 1852) (after Lea, 1858); B, Recent Lasmigona costata (Rafinesque, 1820) (after Kuester, 1858-1859); C, Recent Schepmania nieuwenhuisi (Schepman, 1892) (after Haas, 1969); D, Cretaceous “Unio” senectus White, 1883; E, Cretaceous “Unio” aldrichi White, 1883 (D-E after White, 1907); F, Recent Pseudodon loomisi Simpson, 1900 (after Haas, 1969). WATTERS: UNIONOIDEAN SHELL FORM AND FUNCTION 15 larger particles are less affected, the interstitial sand and silt renders the surrounding substratum more stable. As ubiquitous as dorsal sculpture, is the presence of a dorsal ridge in unionoideans. Nevertheless, the dorsal ridge is uncommon in marine taxa. The function of both dorsal ribbing and the dorsal ridge in marine groups have been hypothesized by Stanley (1977b, 1981) as anti-scour- ing devices. The dorsal ridge coincides with the U3 point and often with the mantle area joining the inhalant and exhalant siphons. The dorsal ridge does not occur random- ly, but at a point of divarication. Although a dorsal ridge is nearly ubiquitous in modern unionoideans, it seems absent in some species. The Triassic taxa Unio felchii White, 1883, and U. toxonotus White, 1883, appear to lack this feature, as do several species of Recent Elliptio. Thus, the dorsal ridge is not a necessary, structural consequence of valve morphometry, but an adaptive feature that has arisen in most living unionoideans. An additional anti-scouring feature found in unionoideans is the dorsal slope itself. Although sculpture- less, this type of shell-form was hypothesized by Stanley (1975b, 1977b) as an effective deterrent to scouring. The effect is not due to the ridge, but to the constriction of the shell posterior to it. The concave shape of the slope allows scouring currents to flow over it without excavation. This design is most effective in unstreamlined species, where the slope may approximate the sediment surface. Many unstreamlined taxa have pronounced dorsal ridges/slopes: species of Quadrula, Fusconaia, Pleurobema, etc. Anti-scouring adaptations only are necessary in regions having a current. It is significant to note the lack of these devices in most lacustrine taxa such as Anodonta, where appreciable unidirectional currents may not occur. Although Stanley (1981) believed that anti-scouring sculp- ture would result in burial in severe currents, this does not appear to be the case in unionoideans, where riffle species often have dorsal ribs. Umbonal Sculpture: In many taxa, particularly the amblemines, the adult sculpture is directly derived from this nepionic structure (Howard, 1914). More importantly, species unsculptured as adults often have distinct beak sculpture. This implies a function for beak sculpture that may be retained in adults or lost. If the juvenile sculpture persists, the derived adult sculpture is an anchoring device (e.g. Quadrula). In taxa where it is does not persist, the species are often either streamlined forms (Elliptio), or soft substratum forms (Anodonta). There is some evidence that umbonal sculpture acts as anchors in nepionic shells. The beak sculpture is often very coarse relative to the small shell size. The sculpture is oriented as anchoring devices and does not include dorsal ribbing, and therefore does not function as an anti-scouring device. Beak sculpture may be optimized for particle size (particles may be large relative to nepionic shells, hence the coarse sculpture), and oriented to minimize dislodgment. The lack of anti-scouring devices is explained by the very small size of nepionic shells, which presumably would not create appreciable scour. At some critical weight or size, anti-scouring sculpture may become necessary. Clarke (1986) and Yeager et al. (1993) have shown that some unionids may be buried after metamorphosis, and Clarke noted that species that lack umbonal sculpture usu- ally were thick-shelled after transformation. He believed umbonal sculpture to be a predator deterrent. I disagree with this hypothesis, at least as the main function of umbonal sculpture, but I believe his observations lend sup- port for an anchoring function. This would be useful to a buried juvenile unionoidean, and a thick (heavy) shell may be sufficient to prevent dislodgment without requiring anchors. Beak sculpture, like that of adults, is divaricate. In fact, the two types of sculpture should not be considered distinct, but a continuum of sculpture that may change dur- ing ontogeny (Howard, 1914). Dall’s (1895: 523) con- tention that “the sculpture of the tips of the umbones is probably due to the influence of the hooked or serrated edges of the glochidium valves” was incorrect. The geome- try of the umbonal sculpture is independent of whether the glochidium possessed hooks or not. Three general patterns of beak sculpture are found: double-looped, single-looped, and barred. All are derived from the prototype sculpture. The single-looped and barred are here considered as a smoothed double-looped sculpture, rather than as a single, unsuppressed “V.” Big River Morphology: Some amblemines are known to exhibit a clinal change of form from headwaters to the most downstream reaches (Ortmann, 1920; Ball, 1922; Grier and Mueller, 1926; Goodrich and van der Schalie, 1944; Eagar, 1950; Clarke, 1973; Tevesz and Carter, 1980; Clarke, 1982; Stansbery, 1983). Ortmann referred to this as the Law of Stream Distribution. The downstream form of Fusconaia flava, for example, contains a full suite of features adapted to a riverine habitat, while headwater specimens are lacking these features (Fig. 18). Comparison of headwater and “Big River” morphs reveals that significant differences exist between the burrowing and anchoring capacity of F. flava (and presumably other taxa) in these two habitats. The headwater form offers less resistance to burrowing than the downstream form, but the latter is more efficient as an anchor. Downstream features in F: flava include high, pros- ogyrous umbones, a relatively thicker shell, a more globose shape, and a more defined sulcus. The same differences 16 AMER. MALAC. BULL. 11(1) (1994) Fig. 18. Big River morphs of Fusconaia flava (after Call, 1898). A, Big River. B, Headwaters. occur in other species, and the fact that unrelated forms show the same type of clinal variation suggests a common function. The convergence in shapes, and therefore clines, is a common reaction to similar environmental conditions in which the adaptive responses are limited by shell geome- try and an intrinsic sculptural pattern. Infaunal, free-living marine bivalves do not show “Big River” modifications, although Rosenburg (1972) found that a species of venerid was slightly more elongate in these tidal areas. The important consideration here is constancy of current. Open ocean populations may experi- ence no net water movement. In riverine habitats, the direc- tion of the current remains constant, but variable in degree, for perhaps tens of thousands of years or more. I suggest that marine groups have had little opportunity to evolve current-resistant shell forms and sculpture on a large scale, while in unionoideans this has been the paramount adapta- tion. The downstream features are predictably strong cur- rent adaptations. Such changes are most prominent in sculptureless taxa. High umbones, increased shell thick- ness, and a strongly prosogyrous orientation move most of the mass of the shell to the anterior margin. Sculpture is reduced or absent, its function now filled by the prosogy- rous umbones (Stanley, 1977b). The high prosogyrous umbones create a center of mass near the anterior margin on a broad lunule base (Fig. 18) (see also Savazzi and Peiyi, 1992). The post-umbonal region of the shell, often distinctly narrowed by a sulcus, now assumes a laterally flattened, conical, hydrofoil-shape conducive with hydrody- namically pushing the shell toward the substratum. Stanley, in his study of anti-scouring devices in trigoniids, stated that this paradigm was “usually associated with the proso- gyre condition” (1977b: 883). Thus, downstream character- istics function to position the shell such that the anterior margin is directed toward the substratum, the optimum bur- rowing orientation. Stanley (1972) found a similar orienting function for this association of characteristics in marine byssate bivalves. Taxa having “Big River” characteristics occurred at Fig. 19. Examples of Big River characteristics in unionoideans. A, Recent [heringella isocardioides (Lea, 1856) (after Lea, 1857); B, Recent Fusconaia ebena (Lea, 1831) (after Call, 1898); C, Recent Epioblasma sampsoni (Lea, 1861) (after Lea, 1863); D, Recent F. flava (after Baker, 1898); E, Cretaceous “Unio” dawsoni Russell, 1931 (after Russell, 1931); F, Cretaceous “Unio” propheticus White, 1876 (after White, 1907); G, Cretaceous “Unio” proavitus White, 1883 (after White, 1907); H, Pliocene Parunio crassus Ping, 1931 (after Ping, 1931) . WATTERS: UNIONOIDEAN SHELL FORM AND FUNCTION 17 least as far back as the Jurassic (Fig. 19). White (1907) illustrated several examples: Unio stewardi White, 1883, U. endlichi White, 1877, and U. proavitus. U. stewardi, Vetulonaea faberi, and V. whitei Branson, 1935, bear a strong resemblance to contemporary big river Pleurobema, and U. proavitus is virtually identical with big river Epioblasma, such as E. sampsoni. Others, such as U. daw- soni, have no Recent counterparts. As headwater individuals may occur in water with currents equally as strong as downstream ones, it is difficult to explain why upstream populations lack these riverine devices. The shape of some unionoideans approaches in magnitude the most streamlined forms found in marine taxa, but is not related to tube-dwelling or active burrowing beneath the substratum (Watters, 1992). Eagar (1978) and others have suggested that it is the occasional unusually strong currents (i.e. floods) in headwaters that have selected for streamlined forms. DISTRIBUTION OF UNSCULPTURED UNIONOI- DEANS. The presence of sculptured species co-occurring with unsculptured ones is difficult to explain. The distribu- tion of sculpture by river size clearly shows that sculptured taxa tend to occur in big rivers, while unsculptured taxa live in a wider range of river sizes, although mostly found in small systems. Headwater individuals tend to have less anchoring and anti-scouring sculpture than do big water forms, and Lewis and Riebel (1984) showed that unsculp- tured unionids were able to successfully burrow in a variety of substrata, including clay, sand, and gravel. It is likely that most unionoidean speciation has taken place in the iso- lated headwaters of rivers during periods of marine trans- gressions. This is particularly true of southern and eastern unionoidean species, which have experienced a much greater degree of transgression-related isolation than interi- or basin systems (Johnson, 1970, 1972; Sepkoski and Rex, 1974; Butler, 1990). There is evidence that at least the members of the genus Elliptio have speciated at a high rate in these areas relative to the interior basin. Three species occur in the interior basin, but perhaps as many as 31 live in coastal rivers. The largest number of endemic species of unionids occurs in coastal rivers. Most important, except for the three species of the east coast endemic “genus” Canthyria, no east coast species has any pronounced sculp- ture. The Coastal Plain has no representatives of the typi- cally big river, interior basin, sculptured genera such as Amblema or Megalonaias. The most diverse of the sculp- tured unionoidean genera, Quadrula, has no east coast rep- resentatives, reflecting the fact that no large river bisects the Appalachian Divide (except the New River, but that river is interrupted by high falls), and that members of this big river genus did not have the chance to establish themselves in the headwaters of eastern rivers. However, the Coastal Plain has many representatives of interior headwater genera: e.g. Elliptio, Lampsilis, and Villosa. This represents circumstantial but compelling evi- dence that east coast species may be derived primarily from central interior headwater taxa by sharing of habitat or host fishes in the headwaters. Because headwater taxa generally are sculptureless, the derived taxa on the east coast also are sculptureless. Gulf drainage species have sculptured repre- sentatives, but these are generally associated with large interior rivers (Apalachicola River, Alabama River, etc.) or rivers close to the Mississippi River drainage (e.g. Pearl River) from which taxa could have spread through stream capture or lowland exchange of host fishes between ancient big river confluences. This study suggests that shell form and sculpture could be selectively advantageous for different substrata and habitats. However, whether unionoideans actively select a particular substratum is not understood. Tevesz and McCall (1979) and Bailey (1989) showed that one species apparently may move to a preferred substratum, but evi- dence for other species is not available. Many species often are found in a particular substratum, suggesting habitat preferences by them or their host (Harman, 1972; Sickel, 1980). But other studies have not supported this hypothesis (Horn and Porter, 1981), or have found mixed results (Huehner, 1987). Although Tadic (1960) and Tudorancea (1972) reported that unionoideans migrate, other authors suggested that they do not (Isely, 1911, 1914; Young and Williams, 1983). A species of the sculptureless genus Lampsilis has been shown to alter its shell dimensions when reciprocally transplanted between mud and sand to match that of the original population morphology (Hinch et al., 1986), although growth rates seemed controlled geneti- cally (Hinch and Green, 1989). The shell shape of Elliptio complanata (Lightfoot, 1786) was found to be narrower in coarse sand in deep water than in shallow water (Hinch et al., 1989), and the degree of obesity was shown to increase in areas of high productivity (Podemski and Green, 1993) for L. radiata (Gmelin, 1781). Whether different habitats change the amount of sculpture has not been determined. The genetic ability to construct sculptured shells may be lost in many unionoidean lineages, particularly those resulting from speciation in headwaters. The presence or absence of sculpture also may be tied synergistically to a more complicated suit of genetic effects. Both hypotheses are potentially correct, but they have not been explored in this study. The present co-occurrence of sculptured and unsculptured shells in medium sized rivers may indicate that headwater lineages have reinvaded the larger river reaches. Being sculptureless, extant species have adapted other shell-morphology characteristics to this habitat. The 18 AMER. MALAC. BULL. 11(1) (1994) most commonly encountered adaptation is the “Big River Effect.” SUMMARY The highly diverse sculpture found in the Unionoidea could be derived from ancestral trigoniids. In that group the main function of sculpture was as an aid to burrowing, but the sculpture of the unionoideans has been exapted for use as anchoring devices, although still superfi- cially similar. The sculpture forms three types of anchors. The generalized anchor is considered the most primitive. The medial anchor is derived from the selective elimination of some generalized anchor sculpture and the selective enhancment of others. A third anchor is derived from the enhancment of a prototypical divaricate arm, often at the expense of most other sculpture. All anchors function to prevent dislodgement during and after burial, and to main- tain a certain position in the substratum. Sculpture may be optimized for a particular substratum and life orientation. The shells of many unionoideans also have anti- scouring devices that tend to eliminate removal of substra- tum by currents adjacent to the shell. These devices also are found in ancestral trigoniids and include fluted posterior slopes, dorsal ridges, and concave dorsal slopes. They func- tion mainly by retaining sand and silt that result in a more stable adjacent substratum. Unionoideans living in soft substrata have modified the shell by several methods to increase buoyancy. These shells typically are thin, alate, and complanate, with a reduced or absent hinge. All of these factors decrease the relative weight of the animal in the substratum. These shells also function as “snowshoes” on soft sediments, and tend to hold the animal in place if dislodged in a current. A few species have achieved buoyancy by having greatly inflated shells, offering a greater aspect to the substratum. These shells however are unstable in a current. The ability to remain buried through anchors, anti- scouring sculpture, and ponderous, prosogyrous shells was central to speciation in large rivers, where presumably the earliest unionoideans invaded freshwater. But in the head- waters, a premium was placed on the ability to rebury if dislodged by floods. Apparently even anchors do not suf- fice in these conditions. Headwater taxa therefore generally have unsculptured, often streamlined, shells. The gradient of headwater forms to big river ones, the Big River Effect, is the result of these differing adaptations. The greatest force operating on a unionoidean shell to drive morphological changes is the constancy of a unidi- rectional water current, a factor that did not shape the shells and sculpture of unionoidean ancestors. But the move to freshwater rivers necessitated the exaptation of pre-existing sculpture for unique, new problems encountered in riverine habitats. ACKNOWLEDGMENTS Dr. David Stansbery, Dr. Abbot Gaunt, and Dr. Barry Valentine (Department of Zoology, Ohio State University) provided much-needed suggestions on the original draft. Dr. David Stansbery also made available the use of the unionid collection at the Museum of Zoology, Ohio State University. Additional specimens were donated by Robert Anderson (Indiana Department of Natural Resources, Indianapolis), Robert Butler (U.S. Fish and Wildlife Service, Daphne, Alabama), Richard Forrer (Northfield, Ohio), and Dr. Henry McCullagh (St. Petersburg, Florida). I also am grateful for the advice of two anonymous reviewers, and the edi- tor. LITERATURE CITED Anderson, R. V. and R. E. Ingham. 1978. Character variation in mollusc populations of Lampsilis Rafinesque, 1820. Transactions of the Illinois State Academy of Science 71:403-411. Bailey, R. C. 1989. Habitat selection by a freshwater mussel: an experi- mental test. Malacologia 31:205-210. Baker, F. C. 1898. The Mollusca of the Chicago area. Pelecypoda. Bulletin of the Chicago Academy of Sciences Natural History Survey 3(1):1- 130. Ball, G. H. 1922. Variation in freshwater mussels. Ecology 3:93. Burch, J. B. 1973. Freshwater unionoidean clams (Mollusca: Pelecypoda) of North America. U. S. Environmental Protection Agency, Biota of Freshwater Ecosystems Identification Manual (11). 176 pp. Butler, R. S. 1990. Distributional records for freshwater mussels (Bivalvia: Unionidae) in Florida and south Alabama, with zoogeo- graphic and taxonomic notes. Walkerana 3:239-261. Call, R. E. 1898. A descriptive catalogue of the Mollusca of Indiana. 24th Annual Report of the Indiana Department of Geology and Natural Resources:335-535; 78 pls. Carter, R. M. 1968. On the biology and palaeontology of some predators of bivalved Mollusca. Palaeogeography, Palaeoclimatology and Palaeoecology 4:29-65.° Clarke, A. H. 1973. The freshwater molluscs of the Canadian interior basin. Malacologia 13:1-509. Clarke, A. H. 1982. The recognition of ecophenotypes in Unionidae. In: Report of the Freshwater Mussels Workshop, 19-20 May, 1981, A. C. Miller, ed. pp. 28-34. U.S. Army Waterways Experiment Station, Environmental Laboratory; Vicksburg, Mississippi. Clarke, A. H. 1986. The mesoconch: a record of juvenile life in Union- idae. Malacology Data Net 1:21-36. Clessin, S. 1873-1876. Die Gattung Anodonta nebst den iibrigen Najaden mit unvollkommenem Schloss. In: Systematischen Conchylien- Cabinet (2) 9, Martini and Chemnitz, pp. 65-287. Dall, W. H. 1895. Contribution to the Tertiary fauna of Florida with espe- cial reference to the Miocene silex-beds of Tampa and the Pliocene beds of the Caloosahatchie River. Part 3. A new classification of the Pelecypoda. Transactions of the Wagner Free Institute of Science of Philadelphia 3:1-704. Eagar, R. M. C. 1950. Variation in shape of shell with respect to ecologi- cal station. A review dealing with Recent Unionidae and certain species of the Anthracosiidae in Upper Carboniferous times. Proceedings of the Royal Society of Edinburgh, Section B 63:130- 148. Eagar, R. M. C. 1977. Shape of shell in relation to weight of Margaritifera margaritifera (L.) (Bivalvia: Margaritiferidae). Journal of Conch- ology 29:207-218. Eagar, R. M. C. 1978. Shape and function of the shell: a comparison of some living and fossil bivalve molluscs. Biological Reviews of the WATTERS: UNIONOIDEAN SHELL FORM AND FUNCTION 19 Cambridge Philosophical Society 53:169-210. Eagar, R. M. C., N. M. Stone, and P. A. Dickson. 1984. Correlations between shape, weight and thickness of shell in four populations of Venerupis rhomboides (Pennant). Journal of Molluscan Studies 50:19-38. Ermentrout, B., J. Campbell, and G. Oster. 1986. A model for shell pat- terns based on neural activity. The Veliger 28:369-388. Fowler, D. R., H. Meinhardt, and P. Prusinkiewicz. (In press). Modeling Seashells. Proceedings of SIGGRAPH for 1992. Ghent, A. W., R. Singer, and L. Johnson-Singer. 1978. Depth distributions determined with SCUBA, and associated studies of the freshwater unionid clams Elliptio complanata and Anodonta grandis in Lake Bernard, Ontario. Canadian Journal of Zoology 56:1654-1663. Goodrich, C. and H. van der Schalie. 1944. A revision of the Mollusca of Indiana. American Midland Naturalist 32:257-326. Gould, S. J. and E. S. Vrba. 1982. Exaptation - a missing term in the sci- ence of form. Paleobiology 8:4-15. Grier, N. M. and J. F. Mueller. 1926. Further studies in correlation of shape and station in fresh water mussels. Bulletin of the Wagner Free Institute of Philadelphia 1:11-28. Haas, F. 1969. Unionoidea. In: Treatise on Invertebrate Paleontology, Part N, Vol. 1, Mollusca 6. R. C. Moore, ed. pp. N411-N467. Geological Society of America and University of Kansas, Lawrence. Harman, W. N. 1972. Benthic substrata: their effect on fresh-water Mollusca. Ecology 53:271-277. Heard, W. H. 1975. Sexuality and other aspects of reproduction in Anodonta (Pelecypoda: Unionidae). Malacologia 15:81-103. Hinch, S. G. and R. H. Green. 1989. The effects of source destination on growth and metal uptake in freshwater clams reciprocally trans- planted among south central Ontario lakes. Canadian Journal of Zoology 67:855-863. Hinch, S. G., R. C. Bailey, and R. H. Green. 1986. Growth of Lampsilis radiata (Bivalvia: Unionidae) in sand and mud: a reciprocal trans- plant experiment. Canadian Journal of Fisheries and Aquatic Sciences 43:548-552. Hinch, S. G., L. J. Kelly, and R. H. Green. 1989. Morphological variation of Elliptio complanata (Bivalvia: Unionidae) in differing sediments of soft-water lakes exposed to acidic deposition. Canadian Journal of Zoology 67:1895-1899. Holt, E. L. 1942. A new Unio from the Morrison Formation of the Grand River valley, Colorado. Journal of Paleontology 16:459-460; pl. 69. Horn, K. J. and H. J. Porter. 1981. Correlations of shell shape of Elliptio waccamawensis, Leptodea ochracea and Lampsilis sp. (Bivalvia, Unionidae) with environmental factors in Lake Waccamaw, Columbus County, North Carolina. Bulletin of the American Malacological Union for 1981:1-3. Howard, A. D. 1914. Experiments in propagation of fresh-water mussels of the Quadrula group. Report of the Commissioner of Fisheries for the Fiscal Year 1913, Appendix 4 [Document 80] ]:1-52. Huehner, M. K. 1987. Field and laboratory determination of substratum preferences of unionid mollusks. Ohio Journal of Science 87:29-32. Isely, F. B. 1911. Preliminary note on the ecology of the early juvenile life of the Unionidae. Biological Bulletin 20:77-80. Isely, F. B. 1914. Experimental study of the growth and migration of fresh-water mussels. Report of the Commissioner of Fisheries for the Fiscal Year 1913, Appendix 3 [Document 792 ]:1-24. Johnson, R. I. 1970. The systematics and zoogeography of the Unionidae (Mollusca: Bivalvia) of the southern Atlantic slope region. Bulletin of the Museum of Comparative Zoology 140:263-449. Johnson, R. I. 1972. The Unionidae (Mollusca: Bivalvia) of peninsular Florida. Bulletin of the Florida State Museum, Biological Sciences 16:181-249. Kawaguti, S. 1950. Observations on the heart shell, Corculum cardissa (L.) and its associated Zooxanthellae. Pacific Science 4:43-49. Kuester, H. C. 1858-1859. Die Flussperlmuscheln (Unio et Hyria) in Abbildungen nach der Natur, mit Beschreibungen. 318 pp. [In] Martini and Chemnitz, Systematischen Conchylien-Cabinet (2) 9. Lea, I. 1829. Description of a new genus of the family of naiads, including eight species, four of which are new; also the description of eleven new species of the genus Unio from the rivers of the United States; with observations on some of the characters of the naiads. Transactions of the American Philosophical Society, New Series, 3:403-457. Lea, I. 1832. Observations on the genus Unio, together with descriptions of new species in the families Natades, Conche, Colimacea, Lymnena, Melaniana and Peristomiana [1]. Privately printed. 133 Pp.. Lea, I. 1842. Observations on the genus Unio, together with descriptions of new species in the families Naiades, Colimacea, Lymnéna, Melaniana and Peristomiana, 3. Privately printed. 87 pp. Lea, I. 1857. Observations on the genus Unio, together with descriptions of new species, their soft parts, and embryonic forms, in the family Unionidae, 6(1). Privately printed. 41 pp. Lea, I. 1858. Observations on the genus Unio, together with descriptions of new species, their soft parts, and embryonic forms, in the family Unionidae, 6(2). Privately printed. 53 pp. Lea, I. 1860. Observations on the genus Unio, together with descriptions of new species, their soft parts, and embryonic forms, in the family Unionidae, 7(2). Privately printed. 38 pp. Lea, I. 1863. Observations on the genus Unio, together with descriptions of new species, their soft parts, and embryonic forms, in the family Unionidae, and descriptions of new genera and species of the Melanidae, 9. Privately printed. 180 pp. Lea, I. 1869. Observations on the genus Unio, together with descriptions of new species in the family Unionidae, together with descriptions of new species of the Melanide and Paludine, 12. Privately printed. 103 pp. Lewis, J. B. and P. N. Riebel. 1984. The effect of substratum on burrow- ing in freshwater mussels (Unionidae). Canadian Journal of Zoology 62:2023-2025. MacNeil, F. S. 1935. Freshwater mollusks from the Catahoula Sandstone (Miocene) of Texas. Journal of Paleontology 9:10-17. Martens, K. E. von. 1899. Conchologische miscellen HI. Archive fiir Naturgeschichte 1899:27-48. McLearn, F. H. 1945. Revision of the Lower Cretaceous of the western interior of Canada. Canada Department of Mines and Resources, Geological Survey Paper 44-17 (2nd ed.):1-14. Meinhardt, H. and M. Klingler. 1987. A model for pattern formation on the shells of molluscs. Journal of Theoretical Biology 126:63-89. Menard, H. W. and A. J. Boucot. 1951. Experiments on the movement of shells by water. American Journal of Science 249:131-151. Morris, N. J. 1978. The infaunal descendants of the Cycloconchidae: an outline of the evolutionary history and taxonomy of the Heteroconchia, superfamilies Cycloconchacea to Chamacea. Philosophical Transactions of the Royal Society of London, Series B 284:259-275. Newell, N. D. and D. W. Boyd. 1975. Parallel evolution in early trigo- nioidean bivalves. Bulletin of the American Museum of Natural History 154:55-162. Nichols, J., G. T. Rowe, C. H. Clifford, and R. A. Young. 1978. In situ experiments on the burial of marine invertebrates. Journal of Sedimentary Petrology 48:419-425. North, G. 1987. Shell patterns simulated. Nature 328:294,295. Olsson, A. A. and C. B. Wurtz. 1951. New Colombian naiades, with observations on other species. Notulae Naturae (239):1-9. 20 AMER. MALAC. BULL. 11(1) (1994) Ortmann, A. E. 1920. Correlation of shape and station in fresh-water mus- sels (naiades). Proceedings of the American Philosophical Society 59:269-312. Owen, G. 1958. Shell form, pallial attachment and the ligament in the Bivalvia. Proceedings of the Zoological Society of London 131:637- 648. Parodiz, J. J. 1969. The Tertiary non-marine Mollusca of South America. Annals of the Carnegie Museum 40:1-242. Ping, C. 1931. On some new fossil mollusks of Mongolia. Bulletin of the Geological Society of China 10:229-247. Podemski, C. L. and R. H. Green. 1993. Growth and morphology of Lampsilis radiata (Unionidae) in relation to benthic infaunal pro- ductivity in the Bay of Quinte, Lake Ontario. Bulletin of the North American Benthological Society 10:99 [abstract]. Prashad, B. 1922. A revision of the Burmese Unionidae. Records of the Indian Museum 24:91-111,; pl. 2. Raup, D. M. 1969. Bivalve mollusk burrowing aided by discordant shell ornamentation. Science 166:634-635. Richardson, P. D. 1968. The generation of scour marks near obstacles. Journal of Sedimentary Petrology 38:965-970. Rosenburg, G. D. 1972. Correlation of shell structure and chemistry with life habitat of the bivalve Chione undatella Sowerby. 68th Annual Meeting of the Geological Society of America, Programs with Abstracts 4:227. Russell, L. S. 1967. Unionidae from the Cretaceous and Tertiary of Alberta and Montana. Journal of Paleontology 41:1116-1120. Russell, L. S. 1976. Pelecypods of the Hell Creek Formation (Uppermost Cretaceous) of Garfield County, Montana. Canadian Journal of Earth Sciences 13:365-388. Savazzi, E. 1982. Shell sculpture and burrowing in the bivalves Scapharca inaequivalvis and Acanthocardia tuberculata. Stuttgarter Beitrdge zlir Naturkunde, Serie A, Biologie 353:1-12. Savazzi, E. and Y. Peiyi. 1992. Some morphological adaptations in fresh- water bivalves. Lethaia 25:195-209, Seed, R. 1980. Shell growth and form in the Bivalvia. In: Skeletal Growth of Aquatic Organisms. D. C. Rhoads and R. A. Lutz, eds. pp. 23-67. Plenum Press, New York. Seilacher, A, 1973. Fabricational noise in adaptive morphology. Systematic Zoology 22:451-465. Sepkoski, J. J. and M. A. Rex. 1974. Distribution of freshwater mussels: coastal rivers as biogeographic islands. Systematic Zoology 23:165- 188. Sickel, J. B. 1980. Correlation of unionid mussels with bottom sediment composition in the Altamaha River, Georgia. Bulletin of the American Malacological Union for 1980:10-13. Simpson, C. T. 1900. New and unfigured Unionidae. Proceedings of the Academy of Natural Sciences of Philadelphia for 1900:74-86; 5 pls. Stanley, S. M. 1969. Bivalve mollusk burrowing aided by discordant shell ornamentation. Science 166:634-635. Stanley, S. M. 1970. Relation of shell form to life habits in the Bivalvia. Memoirs of the Geological Society of America 125:1-296. Stanley, S. M. 1972. Functional morphology and evolution of byssally attached bivalve mollusks. Journal of Paleontology 46:165-212. Stanley, S. M. 1975a. Why clams have the shape they have: an experimen- tal analysis of burrowing. Paleobiology 1:48-58. Stanley, S. M. 1975b. Adaptive themes in the evolution of the Bivalvia (Mollusca). Annual Review of the Earth and Planetary Sciences 3:361-385, Stanley, S. M. 1977a. Trends, rates, and patterns of evolution in the Bivalvia. In: Patterns of Evolution as Illustrated by the Fossil Record. A. Hallam, ed. pp. 209-250. Elsevier Scientific Publishing Company, Amsterdam. Stanley, S. M. 1977b. Coadaptation in the Trigoniidae, a remarkable fami- ly of burrowing bivalves. Paleontology 20:869-899. Stanley, S. M. 1981. Infaunal survival: alternative functions of shell orna- mentation in the Bivalvia (Mollusca). Paleobiology 7:384-393. Stansbery, D. H. 1983. Some sources of nomenclatorial and systematic problems in unionid mollusks. In: Report of the Freshwater Mussels Workshop, 26-27 October 1982. A. C. Miller, ed. pp. 46- 62. U. S. Army Engineer Waterways Experiment Station, Envir- onmental Laboratory, Vicksburg, Mississippi. Stasek, C. R. 1963. Geometrical form and gnomonic growth in the bivalved Mollusca. Journal of Morphology 112:215-229 Stern, E. M. 1983. Depth distribution and density of fresh water mussels (Unionidae) collected with SCUBA from the lower Wisconsin and St. Croix Rivers, U.S.A. The Nautilus 97: 36-42. Suzuki, K. 1941. A new naiad Unio (Nippononaias) ryosekiana n. sub- gen., and n. sp., from the Lower Cretaceous of Japan. Journal of the Geological Society of Japan 48:410-413. Tadic, A. 1960. Najvazniji predstavnici skoljaka roda Unio iz Save, Dunava i Kopackog Jereza [English translation 1973, The most important representatives of the mussels of the genus Unio from the Sava, the Danube and Kopacko Jezero Lake. Bulletin of the Museum of Natural Sciences, Belgrade, Special Publications 28:1- 84.) Tevesz, M. J. S. 1975. Structure and habits of the ‘living fossil’ pelecypod Neotrigonia. Lethaia 8:321-327. Tevesz, M. J. S. and J. G. Carter. 1980. Environmental relationships of shell form and structure of unionoidean bivalves. In: Skeletal Growth of Aquatic Organisms. D. C. Rhoads and R. A. Lutz, eds. pp. 295-322. Plenum Press, New York. Tevesz, M. J. S. and P. L. McCall. 1979. Evolution of substratum prefer- ence in bivalves (Mollusca). Journal of Paleontology 53:112-120. Thayer, C. W. 1975. Morphologic adaptations of benthic invertebrates to soft substratum. Journal of Marine Research 33:177-189. Tudorancea, C. 1972. Studies on Unionidae populations from the Crapina- Jijila complex of pools (Danube zone liable to inundation). Hydrobiologia 39:527-561. Twain, M. [Clemens, S.] 1874. Memorable midnight experience. Essay. Vidrine, M. F. and D. J. Bereza. 1977. Some considerations and implica- tions of host-specificity studies of unionicolid mite parasites on the systematics of some groups of North American unionoidean fresh- water mussels. Bulletin of the American Malacological Union for 1977:85-86. Waddington, C. H. and R. J. Cowe. 1969. Computer simulation of a mul- luscan [sic] pigmentation pattern. Journal of Theoretical Biology 25:219-225. Watters, G. T. 1992. Some aspects of the functional morphology of the shell of infaunal bivalves (Mollusca). Malacologia 35: 315-342. White, C. A. 1907. The ancestral origin of the North American Unionidae, or fresh-water mussels. Smithsonian Miscellaneous Collections 48:75-88. Whitfield, R. P. 1907. Remarks on and descriptions of new fossil Unionidae from the Laramie clays of Montana. Bulletin of the American Museum of Natural History 23:623-628. Yancey, T. E. and D. W. Boyd. 1983. Revision of the Alatoconchidae: a remarkable family of Permian bivalves. Palaeontology 26:497-520. Yeager, M. M., D. S. Cherry, and R. Neves. 1993. Interstitial feeding behavior of juvenile unionid mussels. Bulletin of the Association of Southeastern Biologists 40:113-114 [abstract]. Young, M. and J. Williams. 1983. Redistribution and local recolonization by the freshwater pearl mussel Margaritifera margaritifera (L.). Journal of Conchology 31:225-234. Zimmer, C. 1992. Shell game. Discover 13(5):38-42. Date of manuscript acceptance: 18 January 1994 Freshwater mussels (Bivalvia: Unionidae) of the Hiwassee River in east Tennessee Paul W. Parmalee and Mark H. Hughes Frank H. McClung Museum and Department of Zoology, University of Tennessee, Knoxville, TN 37996, U.S.A. Abstract. Thirteen native species of freshwater mussels are reported from a 20 km stretch of the Hiwassee River, Polk County, Tennessee. Fusconaia barnesiana (Lea, 1838), Elliptio dilatata (Rafinesque, 1820), Villosa iris (Lea, 1829), and V. trabalis (Conrad, 1834) are the most abundant species present, the last taxon is included on the United States federal list of Endangered and Threatened Wildlife and Plants. Epioblasma florentina walkeri (Wilson and Clark, 1914), also included on the federal list and previously unreported from the Hiwassee River, is present but rare. This section of the Hiwassee is unaf- fected by a power generating dam and appears free of pollutants, thus serving as a refugium for viable populations of several mussel taxa. Inference is made to late prehistoric-early historic Hiwassee River mussel assemblages based on shell samples from aboriginal sites. An approximately 20 km stretch of the Hiwassee River (HR), HR km 85.6-105.6, Polk County, Tennessee, flowing between the Apalachia Dam and the Apalachia Powerhouse, is unique compared with all other Tennessee Valley Authority (TVA) dams and river impoundment sys- tems. The dam, built by the TVA and completed in 1943, is located in Cherokee County, North Carolina, 0.25 km from the Tennessee-North Carolina border. Maximum height of the dam is 45 m, its length 392 m; the lake (Apalachia Lake) is 16 km in length and its area at full pool is 440 ha. Rather than hydroelectric power being generated by tur- bines in the dam, the Hiwassee River (lake water behind the dam) is diverted through a tunnel hewn from rock and connected with sections of welded steel pipe 5.4 m in diam- eter. This conduit extends along the top of the bluff to the Apalachia Powerhouse 20 km downstream. Electric power is generated as reservoir water is released through paired pipes that descend vertically from the bluff to the Powerhouse directly below at the river’s edge. Downstream from Apalachia Dam is a gravel road that parallels the dam; during normal water levels a shallow pool of water forms between the road and base of the dam. Renewed flow of the Hiwassee River between Apalachia Dam and Powerhouse is maintained by a series of small- to medium-sized tributaries including Coker, Loss, Turtle- town, and Wolf creeks and numerous smaller branches. Depth throughout this 20 km stretch of river, situated in a high gradient gorge of the Blue Ridge physiographic province, seldom exceeds | m, and then in only isolated pools. Substrata consist generally of large boulders, cob- bles and gravel interspersed with areas of sand and occa- sionally fine silt. Uplifted ridges of exposed Precambrian sandstones divert the main channel into numerous side channels, and in some stretches this has resulted in marsh- like conditions supporting a variety of herbaceous species along banks and on small islands. Eastern hemlock [7suga canadensis (L.) Carr.], American sycamore (Platanus occi- dentalis L.), hornbeam (Carpinus caroliniana Walter), and rhododendron (Rhododendron maximum L.) are common in the flood plain and on the base of the bluffs, while common alder [Alnus serrulata (Aiton) Willd.] and water willow [Justicia americana (L.) Vahl.] abound on low islands and gravel beaches. Fishes typical of an upper Tennessee River warm-water habitat are found here. Several representative taxa occurring in the Hiwassee River between the Apalachia Dam and Apalachia Powerhouse are indicative of quality river habitat (Hitch and Etnier, 1974). These in- clude cyprinid shiners [Cyprinella galactura (Cope, 1868), Notropis leuciodus (Cope, 1868), and N. spectrunculus (Cope, 1868)], centrarchids [Lepomis auritus (Linné, 1758), L. cyanel- lus Rafinesque, 1819, and Micropterus dolomieui Lacepede, 1802] and darters [Etheostoma blenniodes Rafinesque, 1819, E. simoterum (Cope, 1868), Percina caprodes (Rafinesque, 1818) and P. aurantiaca (Cope, 1868)]. At the lower end of this 20 km stretch at HR km 85.6, the Apalachia Powerhouse discharges a tremendous volume of cold reservoir water into the warm waters of the river. Substratum is scoured and temperatures are lowered. There is an abrupt change in physical parameters of the water and habitat below this point. The lower Hiwassee River becomes impounded by the backwaters of Chickamauga Reservoir (Tennessee River) at approximate- ly HR km 41.6 before it enters its historic confluence with the Tennessee River. American Malacological Bulletin, Vol. 11(1) (1994):21-27 21 22 AMER. MALAC. BULL. 11(1) (1994) METHODS Approximately 100 work-hours were spent col- lecting mussels in the Hiwassee River (HR km 85.6-105.6) during July (15), August (5, 15, 21), September (3, 10) and October (10, 15), 1992. Two specimens of Lampsilis ovata were collected alive; all other live unionids found were replaced in the substratum. The majority of specimens rep- resenting the other 12 taxa reported here was comprised of shells discarded by muskrats [Odontra zibethica (Linné, 1766)] following predation. Shells were left at feeding sta- tions along the banks and on rock islands. Seven river km were surveyed for unionids, but, because of especially suit- able habitat judging by the great abundance of shell at HR km 85.6 and 98.8, most of the collecting time was spent at these two locations. In addition, collections were made in Coker, Spring, Conasauga, and Chestuee creeks, the largest tributaries of the Hiwassee River in Polk County, Tennessee. All archaeological shell discussed in this paper and the majority of specimens collected during this study are housed in the Frank H. McClung Museum. Voucher specimens have been deposited in the Illinois Natural History Survey, Champaign-Urbana, the Illinois State Museum, Springfield, and the Carnegie Museum, Pitts- burgh, Pennsylvania. Unionid taxonomy used in this study is from Turgeon et al. (1988). RESULTS Thirteen species were recovered during this study (Table 1). Numerous stretches of this braided 20-km sec- tion of the river were searched, but most lacked a stable substratum of fine gravel and sand; moderate to swift cur- rent flowing over bedrock prevented the establishment of suitable substrate for mussel populations. Although one locality, HR km 91.7-92.5, appeared to provide especially suitable habitat, only 22 specimens were found. Two other localities, however, HR km 85.6-86.4 (immediately upstream from the Powerhouse) and HR km 98.8, con- tained abundant mussel populations; 1201 and 1401 speci- mens, respectively, were obtained during approximately 40 work-hours of collecting at each locale (Table 2) during July-October 1992. Differences between the size of populations of several species making up the mussel assemblage at these two localities are striking. Shells of Villosa iris comprised 82% of all specimens collected at HR km 98.8 (Fig. 1), while < 2% of the specimens collected at HR km 85.6 were V. iris. Nearly 25% of the mussels found at HR km 85.6 (Fig. 2) were V. trabalis, but at HR km 98.8 this species comprised < 2% of the population. At HR km 85.6, shells of Fusconaia barnesiana comprised 54% of all specimens collected, but only 7% at HR km 98.8. Although differences in abundance of host fish for glochidia at these locations might affect mussel numbers, variations in habitat between these two sections of river are probably the determining factors. While both are charac- terized by a substratum composed of fine gravel, sand and silt, HR km 98.8 has an estimated mean depth of < 0.5 m with numerous channels divided by uplifted shelves of sandstone (Fig. 1). In contrast, HR km 85.6 has an estimat- ed mean depth of 1.0 m; a few pockets in and near the main channel are 1.5-2.0 m deep. This is the widest section of river (Fig. 2), approximately 110 m, with some areas of little or no current. Alasmidonta marginata, Elliptio dilatata, Fus- conaia barnesiana and Lampsilis fasciola occurred at all localities with suitable habitat collected by us, but they var- 4 i ‘ { ieee j ies! é Fig. 1. Braided section of the Hiwassee River (HR km 98.8) ca. 3.5 km dowstream from the Apalachia Dam. Of the eight species identified from this locality, shells of Villosa iris (N = 1,153) comprised 82% of the sam- ple. Fig. 2. Hiwassee River (HR km 85.6), looking upstream, where viable populations of Villosa trabalis occurred, and where 17 specimens of Epioblasma florentina walkeri were collected. Turbulent flow in fore- ground is discharge water from the Apalachia Powerhouse. PARMALEE AND HUGHES: MUSSELS OF THE HIWASSEE RIVER 7 Table 1. Archaeological, pre-impoundment and 1992-1993 records of freshwater mussel taxa from the Hiwassee River, Polk County, Tennessee. Species 1992-1993 Actinonaias ligamentina (Lamark, 1819), mucket Alasmidonta marginata Say, 1818, elktoe Alasmidonta viridis (Rafinesque, 1820), slippershell mussel Amblema plicata (Say, 1817), threeridge Anodonta imbecillis Say, 1829, paper pondshell Cyclonaias tuberculata (Rafinesque, 1820), purple wartyback Dromus dromas (Lea, 1834), dromedary pearlymussel Elliptio crassidens (Lamarck, 1819), elephant-ear Elliptio dilatata (Rafinesque, 1820), spike Epioblasma cf. capsaeformis (Lea, 1834), oyster mussel Epioblasma florentina walkeri (Wilson and Clark, 1914), tan riffleshell Fusconaia barnesiana (Lea, 1838), Tennessee pigtoe Fusconaia subrotunda (Lea, 1831), long-solid Lampsilis fasciola Rafinesque, 1820, wavy-rayed lampmussel Lampsilis ovata (Say, 1817), pocketbook Lasmigona costata (Rafinesque, 1820), fluted-shell Lasmigona holstonia (Lea, 1838), Tennessee heelsplitter Lexingtonia dolabelloides (Lea, 1840), slabside pearlymussel Medionidus conradicus (Lea, 1834), Cumberland moccasinshell Pleurobema oviforme (Conrad, 1834), Tennessee clubshell Ptychobranchus subtentum (Say, 1825), fluted kidneyshell Villosa iris (Lea, 1829), rainbow Villosa trabalis (Conrad, 1834), Cumberland bean Villosa vanuxemensis (Lea, 1838), mountain creekshell ied in abundance from rare to moderately common (Table 2). Shells of Pleurobema oviforme comprised 3% of all specimens collected at HR km 85.6, but was extremely rare or absent at all other localities. Only two individuals of L. ovata and three of Lexingtonia dolabelloides, both at HR km 86.5, were encountered during the July-October 1992 collecting period. L. ovata is moderately common through- out the upper Tennessee River system while L. dolabel- loides is becoming increasingly rare throughout its former range in east and middle Tennessee. These two taxa, like A. marginata and F. subrotunda, had not been recorded histori- cally from the Hiwassee River system. E. crassidens and Tritogonia verrucosa (Rafinesque, 1820) were reported from the Hiwassee River (Ortmann, 1918), but from the lower stretches in Meigs County which is now impounded by Chickamauga Reservoir. Neither species was found dur- ing this survey. Starnes and Bogan (1988) listed Alasmidonta viridis, Lasmigona holstonia and Villosa vanuxemensis, all typical of headwaters, small streams and creeks, as part of the Hiwassee River mussel fauna. It should be noted, how- ever, that the record of V. vanuxemensis in Starnes and Bogan (1988: 28, table 4) for the Hiwassee River is erro- neous. Inadvertently the designation for this taxon was placed in the column for the Hiwassee River when it should have been assigned to the Little Tennessee River. Nevertheless, none of these three species was located in the Hiwassee Ortmann This study Old Town (1918) X xX xX xX xX X X xX xX xX X X X xX xX xX xX xX xX X xX xX xX X xX xX xX xX xX X xX X xX X xX xX xX xX xX main channel collections of the Hiwassee River in the 1992 survey. In an effort to determine the presence of mussel populations in tributaries of the Hiwassee River, several small branches and creeks were collected at one or more localities in each. It became apparent that small, spring-fed branches with rocky substrata provided unsuitable habitat for unionids. Although none of these tributaries was col- lected throughout its entire length, mussels were found only in Conasauga and Spring creeks. A muskrat midden in Spring Creek ca. 400 m upstream from its confluence with the Hiwassee River (HR km 69.6) contained 16 shells of V. iris and 186 shells of V. vanuxemensis. Fourteen relic spec- imens of the latter species were found opposite the junc- tions of Forest Roads (Cherokee National Forest) Nos. 27 and 2005, a straight line distance of ca. 4.0 km northeast of the creek’s confluence with the Hiwassee River. Mussels were found in only one of four localities searched in Conasauga Creek (adjacent to TN Hwy. 310, ca. 4.0 km east of Etowah, McMinn County): seven relic specimens of V. vanuxemensis. Of the tributaries collected, Spring and Conasauga creeks appeared to provide habitat most suitable for mussels, but species diversity was low and populations localized. Two species comprising the mussel assemblages of the Hiwassee River, HR km 105.6 to km 85.6, are included in the federal list of Endangered and Threatened Wildlife and Plants. One of these, Villosa trabalis, was 24 AMER. MALAC. BULL. 11(1) (1994) Table 2. Freshwater mussel species and percent of each recorded from two collection localities, Hiwassee River, Polk County, Tennessee, 1992. Species HR km 98.8 HR km 85.6 Specimens Specimens N %o N % Alasmidonta marginata Say, 1818, elktoe 33 2.35 23 1.91 Elliptio dilatata (Rafinesque, 1820), spike 49 3.50 115 9.57 Epioblasma florentina walkeri (Wilson and Clark, 1914), tan riffleshell - - 16 1.33 Fusconaia barnesiana (Lea, 1838), Tennessee pigtoe 99 7.07 649 54.04 Fusconaia subrotunda (Lea, 1831), long-solid 3 0.21 11 0.91 Lampsilis fasciola Rafinesque, 1820, wavy-rayed lampmussel 38 2.71 31 2.58 Lampsilis ovata (Say, 1817), pocketbook - - 2 0.17 Lexingtonia dolabelloides (Lea, 1840), slabside pearlymussel - - 3 0.25 Pleurobema oviforme (Conrad, 1834), Tennessee clubshell 1 0.07 37 3.08 Villosa iris (Lea, 1829), rainbow 1,153 82.30 15 1.25 Villosa trabalis (Conrad, 1834), Cumberland bean 25 1.78 299 24.89 TOTAL 1,401 99.99 1,201 99.98 first reported from the Hiwassee River by Ortmann (1918), who considered the species to be extremely rare throughout its range. In a technical draft of a Recovery Plan for V. tra- balis, Ahlstedt (1983) listed 11 rivers and several creeks from which populations are, or were, known. The species appears to have been restricted to tributary streams of the Tennessee River and upper Cumberland River. We collect- ed specimens of V. trabalis at both the upper (HR km 98.8) and lower (HR km 85.6) locations, as well as below Forest Road 23 (HR km 91.7), suggesting its presence throughout the section of river between the Apalachia Dam and Powerhouse. At HR km 85.6, V. trabalis comprised nearly 24% (N = 299) of the shells collected, an indication of a stable and viable population. Of special significance was the discovery of a population of Epioblasma florentina walkeri at HR km 85.6, the area immediately above the Apalachia Powerhouse (Fig. 2). Sixteen specimens (five females and 11 males) were recovered, mostly in muskrat midden piles. Due to nacre and shell margin deterioration, seven of the specimens may be considered relics, while the other nine appear fresh enough to have been taken by muskrats during the winter/spring of 1992 (Fig. 3). Based on collection records since 1970, Neves (1991) concluded that the only known populations of E. f. walkeri are in the Middle Fork Holston River, Smyth and Washington counties, Virginia. These populations were based on two collection localities. However, Ahlstedt (in Neves, 1991) found five live speci- mens at a third locality in Smyth County, September 1985. One freshly dead specimen of E. f. walkeri was also recov- ered in the Duck River at Interstate Highway 65 bridge (Duck River km 243.0), Maury County, Tennessee, April 1988 by Ahlstedt (Neves, 1991), but its population status there is unknown. The Hiwassee River population of E. f. walkeri, based on recovery of 16 specimens (< 2% of 1201 shells collected in 1992 at this location), appears small and its viability questionable. Having discovered Epioblasma florentina walkeri in 1992 at HR km 85.6, three subsequent collecting trips (February 5, April 24, May 12, 1993) were taken in an effort to determine the population status of this mussel. Because of high water in February and April, only a few specimens of unionids were retrieved; however, one of those was a male E. f. walkeri with dried tissue adhering to the shell. From late March to early May the Powerhouse was shut down for repairs; this resulted in low water levels that permitted us to search beneath large boulders and rock ledges that had previously been inaccessible. On May 12, 1993, we collected 230 specimens, but with the exception of six recently eaten individuals, all were relic shells and no E. f. walkeri were found. Possibly spring high-water levels restricted muskrat predation. Shells of Fusconaia barne- siana (N = 137) and Villosa trabalis (N = 50) comprised 81% of this sample. Two relic specimens of Lexingtonia dolabelloides and Lampsilis ovata were also found, bring- ing the totals for those species from this collection locality to six and four, respectively. Of special interest was the recovery of a relic shell of a mature male Villosa vanuxemensis. As noted earlier, the section of Hiwassee River between Apalachia Dam and Powerhouse appears to be suitable habitat for this species, but for whatever reason(s) it has failed to establish viable populations. In addition, a badly fragmented and eroded shell of Anodonta imbecillis was found beneath a rock ledge in one of the deeper pools. Habitat for this species is marginal and its occurrence in the Hiwassee River at this PARMALEE AND HUGHES: MUSSELS OF THE HIWASSEE RIVER 22 cm Fig. 3. Examples of male (top row) and female (bottom row) Epioblasma florentina walkeri from the Hiwassee River (HR km 85.6-86.0), Polk County, Tennessee. locality is considered fortuitous. DISCUSSION The historic Hiwassee River might be thought of as a unit, a body of contained waterways flowing from the Blue Ridge physiographic province into the Ridge and Valley physiographic province and then into the Tennessee River. Yet several modifications have altered that continu- ity. First, Apalachia Dam impounds the river at HR km 105.6 forming Apalachia Lake upstream. At that point most of the water is diverted downstream to HR km 85.6, alter- ing the river downstream from Apalachia Dam from a mediumsized river to a first-order stream. At HR km 85.6 the diverted water from Apalachia Lake again joins the watershed as a result of water passing through the turbines at the Apalachia Powerhouse. Extremely fast currents and turbulent rapids prevail, abrupt- ly transforming a warm-water Hiwassee River into a trout stream for 44.0 km. At ca. HR km 41.6 the river is im- pounded by the backwaters from Chickamauga Reservoir for the remainder of its course. Basically, there are two mussel assemblages in this river, one occurring in the unal- tered stretch and the other in the lower impounded reaches. The paucity of specimens encountered thus far by malacologists in this 44.0 km stretch of river below the Powerhouse suggests that only rarely do individuals now become established, and viable populations are apparently nonexistent. On several collecting trips to the Hiwassee River (March 16, May 10-11, 1975; August 23, 1978; August 16, 1980) between HR km 58.5 and 72.8, S. A. Ahlstedt (pers. comm., 1992) obtained only 11 specimens representing eight taxa. These species included Fusconaia barnesiana (N = 2); Elliptio crassidens (N = 1); E. dilatata (N = 2); Lampsilis ovata (N = 1); Pleurobema oviforme (N = 2); Plethobasus cyphyus (Rafinesque, 1820) (sheepsnose) (N = 1); Potamilus alatus (Say, 1817) (pink heelsplitter) (N = 1); Villosa vanuxemensis (N = 1). Except for one live P. oviforme, all others were badly eroded relic shells. It should be noted, however, that the lower section of the Hiwassee River, near and into the impounded portion (Chickamauga Reservoir), does support populations of several species, e.g. those belonging to the genera Anodonta Lamarck, 1799, and Potamilus Rafinesque, 1818. Based on pre-impoundment and pre-powerhouse species distribution records provided by Ortmann (1918), 26 AMER. MALAC. BULL. 11(1) (1994) Starnes and Bogan (1988) listed 12 freshwater mussel taxa for the entire Hiwassee River system. Their number included three forms each of Pleurobema oviforme and Fusconaia barnesiana. Ortmann (1918) also noted the pres- ence of Lampsilis fasciola in the Hiwassee River, but it was omitted in the Starnes and Bogan (1988) list. In the case of Elliptio dilatata, Ortmann (1918: 556) commented that it was “...possibly the most widely distributed species in the upper Tennessee region, so that it is hardly required to name special localities.” By including FE. dilatata, 11 nom- inal species by our count comprised the mussel assemblage known to Ortmann (1918) in the Hiwassee River. Table 1 lists the eight species that Ortmann (1918) specifically col- lected in the Hiwassee River, Polk County, Tennessee. Shells of 18 freshwater mussel taxa recovered dur- ing the 1986-1987 archaeological excavations at Hiwassee Old Town, a former Cherokee village (but including some prehistoric Late Woodland features) comprising an area of approximately 140 ha along the Hiwassee River (HR km 65.6-68.8), Polk County, provided noteworthy comparative data with known historic assemblages (Table 1). All of the species represented at this aboriginal site are characteristic of small- to medium-sized tributaries of the upper Ten- nessee River system; eight of the 17 taxa identified from the Hiwassee Old Town samples have not been reported from modern collections. Of the 94 identifiable valves recovered, those of four species, Fusconaia barnesiana, F: subrotunda, Epioblasma cf. capsaeformis, and Ptycho- branchus subtentum comprised 65% of the samples. Indication of prehistoric diversity is also reflected in shell samples recovered from two late prehistoric (Mississipian: A.D. 1400-1500) aboriginal sites (Table 3). During archaeological salvage operations along the Ten- nessee River and lower Hiwassee River from 1936-1939 prior to completion of the Chickamauga Dam (1940), field crews supplied by the federal Works Progress Administration (WPA) carried out the excavations. Although not extensive, shell samples recovered at two late prehistoric sites [Ledford Island, 40BY13 (Bradley County), HR km 20.2; Mouse Creek, 40MN3 (McMinn County), HR km 24.0] provided additional records of species inhabiting the Hiwassee River prior to impound- ment. Of the 20 taxa identified from 167 valves (Table 3), 11 species are no longer a component of the known extant fauna or were not found in archaeological sites upstream. Combining the taxa reported in the literature (Ortmann, 1918; Starnes and Bogan, 1988) with those iden- tified from Ledford Island, Mouse Creeks and Hiwassee Old Town, and those presently inhabiting the unimpounded stretches, a minimum of 35 species are now known from the Hiwassee River. Pre-impoundment collections of the mussel fauna in the Hiwassee River were few in number, and published species accounts appeared in regional narrations (e.g. Ortmann, 1918) rather than in a publication dealing only with the assemblage of the Hiwassee River. Numerous Table 3. Freshwater mussels identified from two late prehistoric aboriginal sites along the lower Hiwassee River, Tennessee. Surface collections/ no provenience. Species Actinonaias ligamentina (Lamark, 1819), mucket Amblema plicata (Say, 1817), threeridge Cyclonaias tuberculata (Rafinesque, 1820), purple wartyback Dromus dromas (Lea, 1834), dromedary pearlymussel Elliptio crassidens (Lamarck, 1819), elephant-ear Elliptio dilatata (Rafinesque, 1820), spike Epioblasma torulosa (Rafinesque, 1820), tuberculed blossum Fusconaia subrotunda (Lea, 1831), long-solid Lampsilis ovata (Say, 1817), pocketbook Lexingtonia dolabelloides (Lea, 1840), slabside pearlymussel Ligumia recta (Lamarck, 1819), black sandshell Plethobasus cooperianus (Lea, 1834), orange-foot pimpleback Plethobasus cyphyus (Rafinesque, 1820), sheepnose Pleurobema clava (Lamarck, 1819), clubshell Pleurobema cordatum (Rafinesque, 1820), Ohio pigtoe Pleurobema plenum (Lea, 1840), rough pigtoe Potamilus alatus (Say, 1817), pink heelsplitter Ptychobranchus fasciolaris (Rafinesque, 1820), kidneyshell Ptychobranchus subtentum (Say, 1825), fluted kidneyshell Quadrula sparsa (Lea, 1841), Appalachian monkeyface TOTAL Ledford Island Ledford Island Mouse Creek (40BY 13) (40BY 13) (40MN3) WPA 1936-1939 PWP/MHH 1993 WPA 1936-1939 8 2 1 we) - — CcO— 1 3 3 28 11 13 2 2 é 10 13 5 1] - = 3 4 1 2 = . - - 1 3 2 . 3 2 - 3 : : : 1 : 5 1 2 - - 1 2 - 2 PARMALEE AND HUGHES: MUSSELS OF THE HIWASSEE RIVER 20 archaeological excavations of aboriginal sites along the river were undertaken prior to dam construction in the late 1930s; however, when samples of shell were retained, they were meager. Had greater quantities of shell been saved, they would have provided a more comprehensive list of the former Hiwassee River mussel taxa than now exists. The diversity of naiad species that probably occurred in the Hiwassee River may never be known. SUMMARY AND CONCLUSIONS There are extremely few stretches of medium- sized and large rivers in Tennessee that contain diverse mussel assemblages reminiscent of pre-impoundment con- ditions. Dam construction has brought about major adverse changes in molluscan faunas with the formation of reser- voirs and resulting changes in water depth and temperature, current, substratum, and amount of dissolved oxygen, as well as possible loss of fish hosts for glochidia. Pre- impoundment anthropogenic changes have also dealt a blow to mussel diversity as reflected in the differences in prehistoric mussel richness as opposed to a historic pre- impoundment decrease in species richness. The Hiwassee River typifies these changes throughout most of its course in North Carolina and Ten- nessee. One short stretch, however, between the Apalachia Dam and Apalachia Powerhouse (HR km 85.6-105.6), because it is relatively unaffected by the dam, serves as a refugium for numerous mussel species known to have inhabited much of the river prior to impoundment. Of the 13 taxa represented in our 1992-1993 collections, two (Epioblasma florentina walkeri and Villosa trabalis) are included on the federal list of Endangered and Threatened Wildlife and Plants. A third species, Lexingtonia dolabel- loides, rare in the Hiwassee River, has been proposed by the U.S. Fish and Wildlife Service as a candidate to receive protection under the Endangered Species Act of 1973 (Biggins, 1992). Freshwater mussel species identified from shells recovered from aboriginal sites, including those reported in the literature and known presently (1993) to inhabit free- flowing sections of the Hiwassee River, total 35 taxa. Parmalee et al. (1982) noted the occurrence of Anodonta imbecillis, A. suborbiculata Say, 1831, and Potamilus ohiensis (Rafinesque, 1820) at the impounded mouth of the Hiwassee River at its confluence with the Tennessee River (Chickamauga Reservoir). Although one or a few addition- al species may eventually become established or discovered in the lower impounded section of the Hiwassee River, lit- tle if any change is anticipated in mussel assemblages extant in Apalachia Lake upstream from Apalachia Dam and in the cold fast-flowing stretch of river downstream from the Apalachia Powerhouse. The section of Hiwassee River between the Apalachia Dam and Powerhouse is unique, providing stable and basically pristine mussel habi- tat. It should receive both state and federal protection to preserve extant plant and animal communities, and be con- sidered as a possible location for transplanting other threat- ened or endangered mollusks when techniques for such transplants are perfected. ACKNOWLEDGMENTS We extend our appreciation to Pat B. Ezzel, T.V.A., Norris, and Rupert Westfield, T.V.A., Athens, for providing specifications data on the Apalachia Dam and Powerhouse. Gary Rosenberg, Department of Malacology, Academy of Natural Sciences of Philadelphia, PA, kindly supplied collections data for mussels obtained by Steven A. Ahlstedt, T.V.A., Norris. We thank W. Miles Wright, Frank’ H. McClung Museum, for preparation of figures used in this paper. Acknowledged for their assis- tance in the collection of specimens are Patricia P. Adams; Richard G. Biggins and John Fridell, U.S. Fish and Wildlife Service, Ashville, NC; Kelly D. Helms; Lisa G., Rebecca F., Laura R., and Samuel H. Hughes. Arthur E. Bogan, Steven A. Ahlstedt, and Lynne P. Sullivan offered help- ful suggestions relative to this project. Thanks are extended to James Herrig, U.S. Forest Service, Cleveland, TN, for his cooperation in this study. We are grateful to Betty W. Creech for typing drafts of the manu- script. LITERATURE CITED Ahlstedt, S. A. 1983. Technical/agency draft recovery plan for the Cumberland bean pearly mussel Villosa trabalis (Conrad, 1834). U. S. Fish and Wildlife Service, Endangered Species Field Office, Atlanta, GA. USFWS Contract Number TV-60706A, 53 pp. Biggins, R. G. 1992. Notification of status review for six Tennessee and Cumberland River Basin freshwater mussels. Fish and Wildlife Service, Ashville, N.C. 7 pp. Hitch, R. K. and D. A. Etnier. 1974. Fishes of the Hiwassee River system - ecological and taxonomic considerations. Journal of the Tennnessee Academy of Science 49:81-87. Neves, R. J. 1991. Mollusks. In: Virginia’s Endangered Species, K. Terwilliger, coordinator, pp. 251-320. The McDonald and Woodward Publishing Company, Blacksburg, Virginia. Ortmann, A. E. 1918. The nayades (freshwater mussels) of the upper Tennessee drainage, with notes on synonymy and distribution. Proceedings of the American Philosophical Society 57(6):521-626. Parmalee, P. W., W. E. Klippel, and A. E. Bogan. 1982. Aboriginal and modern freshwater mussel assemblages (Pelecypoda: Unionidae) from the Chickamauga Reservoir, Tennessee. Brimleyana 8:75-90. Starnes, L. B. and A. E. Bogan. 1988. The mussels (Mollusca: Bivalvia: Unionidae) of Tennessee. American Malacological Bulletin 6(1):19-37. 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, and J. D. Williams. 1988. Common and scientific names of aquatic inver- tebrates from the United States and Canada: mollusks. American Fisheries Society Special Publication 16, Bethesda, MD, 277 pp. Date of manuscript acceptance: 18 January 1994 Life history of the endangered James spinymussel Pleurobema collina (Conrad, 1837) (Mollusca: Unionidae) Mark C. Hove* and Richard J. Neves Virginia Cooperative Fish and Wildlife Research Unit**, National Biological Survey, Department of Fisheries and Wildlife Sciences, Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061-0321, U.S.A. Abstract. The reproductive period, host fish requirements, and population characteristics of the James spinymussel [Pleurobema collina (Conrad, 1837)] were studied from 1987 to 1989 in the James River drainage, West Virginia and Virginia. This summer brooder was gravid from late May through early August and released the majority of glochidia in early June through late July. The mean fecundity was roughly 13,000 brooded eggs/female. Observations in the field and laboratory implicated fishes of the Cyprinidae as hosts for glochidia of the James spinymussel. Induced infestations of glochidia on fishes in the laboratory confirmed seven host species: the bluehead chub [Nocomis leptocephalus (Girard)], rosyside dace (Clinostomus fundu- loides Girard), satinfin shiner [Cyprinella analostana (Girard)], rosefin shiner [Lythrurus ardens (Cope)], central stoneroller [Campostoma anomalum (Rafinesque)], blacknose dace [Rhinichthys atratulus (Hermann)], and mountain redbelly dace [Phoxinus oreas (Cope)]. The age class structure of two pop- ulations, obtained by thin-sectioning valves collected in muskrat middens, ranged from 3 to 19 yrs with evidence of strong and weak year classes. Incidence of spines differed significantly among populations. The mean annual mortality rate of adults was 15.6 + 1.4%. The James spinymussel [Pleurobema collina (Con- rad, 1837)] is one of three spined species of freshwater mussels in the United States (Fig. 1). The historic range of this species, endemic to the James River watershed, is thought to have occurred upstream of Richmond, Virginia, throughout the larger tributaries of the river basin. However, a survey on the status of this species in 1984 revealed its occurrence in only two small tributaries to the James River, an approximately 90% reduction in range (Clarke and Neves, 1984). As a result of this reduction in range and its rarity, the James spinymussel was listed by the federal government as endangered in July 1988. Additional populations have been reported since its federal listing (Hove, 1990). As with most endemic mussel species, little is known of the life history and ecology of the James spiny- mussel. Boss and Clench (1967) described various anatom- ical characteristics and sexual dimorphism of Pleurobema collina. They suggested that it was probably a short-term brooder, releasing glochidia in the summer. Habitat of P. collina included streams ranging in size from 3 to 23 m wide and 15 to 100 cm deep (Clarke and Neves, 1984). *Present address: University of Minnesota, 200 Hodson Hall, 1980 Folwell Avenue, St. Paul, Minnesota 55108, U.S.A. **The Unit is jointly supported by the National Biological Survey, Virginia Department of Game and Inland Fisheries, Wildlife Management Institute, and Virginia Polytechnic Institute and State University. Fig. 1. Pleurobema collina from Craig Creek, Virginia. The species occupied sediments of cobble and sand in reaches with slow to moderate currents. The spinymussel occurred historically in larger streams such as the James River opposite the city of Maidens where the river is about 150 m wide (museum records, Ohio State University). Other aspects of its biology were unknown. The purpose of this study, therefore, was to determine the reproductive period, required fish hosts, and population characteristics of the James spinymussel in the headwaters of the James River, West Virginia and Virginia. MATERIALS AND METHODS Three populations of Pleurobema collina were monitored weekly during spring 1988 and 1989 to deter- mine the period of gravidity. These populations were in American Malacological Bulletin, Vol. 11(1) (1994):29-40 29 30 AMER. MALAC. BULL. 11(1) (1994) South Fork Potts Creek, Monroe County, West Virginia; Craig Creek, Craig County, Virginia; and Johns Creek, Craig County, Virginia. Gravidity was checked by separat- ing the valves approximately 1 cm to observe whether the outer gills were swollen with conglutinates. Observations were made on gill coloration, degree of gill inflation, and conglutinates released by gravid females. In spring 1988, eight gravid females at the Craig Creek study site were marked by scoring the right valve. Seven of these females were collected again in spring 1989 and placed in a 0.5 x 0.5-m area to facilitate their relocation. These mussels were monitored weekly from 26 May through 8 August 1989, and the numbers of gravid and nongravid females were recorded weekly. During each sampling, five to seven of the marked females were relocated and checked for gravidi- ty. To determine the period of glochidial release, stream drift was sampled in 1988 at the study site in South Fork Potts Creek. Drifting glochidia were collected with square-framed drift nets (0.45 m2) constructed of 130- micron nylon mesh fitted with removable cod-ends. Three drift nets were placed equidistantly, perpendicular to the current downstream of a 20 m long pool that held approxi- mately 30 adult Pleurobema collina. Drift samples were collected for 2 hrs between 1300 and 1700 hrs from 26 May through 12 August 1988 and preserved in 10% forma- lin, buffered with sodium borate to prevent dissolution of glochidial valves. Drift material was examined in a gridded petri dish under a dissecting microscope (25-40x), and glochidia were counted. Because the mussel fauna of South Fork Potts Creek consisted only of the squawfoot [Strophitus undulatus (Say, 1817)] and P. collina, glochidia of P. collina were readily identified by their smaller size and lack of hooks. Cross-sectional area of the stream, and the velocity of surface water 5 m downstream of the sam- pling site were used to estimate discharge. Densities of glochidia were computed from measurements of water depth and velocity at the net opening and from counts of glochidia/sample. A 90-day Ryan thermograph (Model J) recorded temperature continuously from 30 May to 14 July 1988. The determination of probable fish hosts was initiat- ed by examination of fishes collected by electrofishing in South Fork Potts Creek in spring and summer 1988. Weekly collections of fishes began on 23 May 1988 when gravid Pleurobema collina were first observed. Fishes were collected from a 50-m reach of stream below the drift sam- pling site. The collection and examination of fishes contin- ued through 4 August 1988. Captured fishes were anes- thetized, identified, visually inspected on the gills for glochidial infestations, and returned to the stream. Inci- dence of glochidial attachment was recorded. Laboratory experiments were conducted to confirm the identification of fish hosts of Pleurobema collina. Fishes were collected from streams outside of the James River drainage to eliminate the possibility of acquired immunity from prior exposure to glochidia (Neves et al., 1985). Test species were collected with backpack elec- troshocker and by seining. Collected fishes were held at 16- 23° C in test aquaria (40 | and 96 | capacity), at least 5 d prior to infestation. Mature glochidia were obtained from collected gravid females that were subsequently returned to their natal stream. When gravid females were brought into the laboratory, they were held at room temperature (approxi- mately 25° C) or in a recirculating circular stream (approx- imately 15° C). Mussels transported to the laboratory were held in beakers to ensure containment of glochidia. Females either released their conglutinates naturally within three to nine days or aborted them after handling. Glochidia were separated from aborted conglutinates by gently drawing the conglutinates in and out of a pipette. The procedures for infestation, subsequent examination of fishes, and recovery of juveniles were performed according to Zale and Neves (1982) with one exception. American eels [Anguilla rostrata (Lesueur)], because of difficulty in handling, were infested by placing them in 100 ml of water with 300 to 600 glochidia under vigorous aeration for seven to nine min. The effectiveness of this technique was con- firmed on four swallowtail shiners [Notropis procne (Cope)] prior to testing the eels. Juvenile mussels were identified by their opaque, dimpled valves and by their foot movements. A fish was considered a host if encystment and metamorphosis to the juvenile stage occurred. POPULATION CHARACTERISTICS Twenty-three Pleurobema collina were marked on 20 June 1988 upstream of the study site on South Fork Potts Creek to verify that growth lines formed annually. The specimens were given a unique mark, and the edges of their valves were double-notched to identify the 1988 annu- lus. The number of annuli, total length, and mark for each mussel were recorded. In late summer and fall 1989, these mussels were retrieved and inspected for recent annual growth rings. The ages of the specimens were determined by thin sectioning of valves with the technique described by Neves and Moyer (1988). Valves of Pleurobema collina were col- lected from muskrat middens at Johns Creek and its tribu- tary Dicks Creek in Craig County, Virginia. One hundred paired valves were sectioned, then aged using a dissecting microscope. Each valve was aged at least three times to provide a precise estimate. A von Bertalanffy growth equa- tion was computed from length-at-age data and was fit by nonlinear procedures to derive the parameters of the equa- tion (SAS Institute, 1982). The annual mortality rate (M) HOVE AND NEVES: JAMES SPINYMUSSEL LIFE HISTORY for 97 of the adult mussels (ages 4-19) collected in muskrat middens was computed using the estimator of Robson and Chapman (1961): M=1-T/ (XN,+T-1)) Variance = (T/DN,+T-1)) x ((T/ZN,+T-1))-((T-1)/EIN,+T-2))) where N, is the number in each successive age class (x) and T = N, + 2Ny + 3N3 +... XN x. This estimator assumes constant cohort recruitment and survival rate, and equal vulnerability of adults to muskrat predation. Because the incidence of spines was seemingly vari- able among streams sampled in the upper James River watershed, we tested these differences using multiple com- parisons for population proportions (Zar, 1984). Presence of spines on valves collected from Dicks, Johns, South Fork Potts, Craig, and Catawba creeks was compared. Common and scientific names for mussels and fish- es are according to Turgeon et al. (1988) and Robbins et al. (1991). RESULTS REPRODUCTIVE PERIOD The period of gravidity of Pleurobema collina was nearly the same in 1988 and 1989 (Table 1). Gravid females 31 of P. collina in South Fork Potts Creek were collected between 23 May and 9 August 1988, and between 23 May and 8 August 1989. Limited surveys in Johns and Craig Creeks in 1988 located gravid P. collina on 30 May in Craig Creek and between 27 June and 4 July in Johns Creek. In 1989, gravid females were observed in Craig Creek between 26 May and 11 July. The period of gravidity of the James spinymussel was longer in South Fork Potts Creek in 1989 than in either Johns Creek or Craig Creek. Efforts to determine the peak of the gravidity period were hampered by two weeks of high flows in early June 1989. On 26 May 1989, five of the eight females marked in 1988 were examined at the Craig Creek study site, and one of them was gravid. On 3 June, two more marked females were found and placed with the others. On this day, five of the seven females were gravid. During the following two weeks (9-23 June), the water was so murky or the cur- rent so swift that the marked females could not be collected for examination. The females were checked again on 27 June, and only two of the six examined were gravid. The marked females were examined again on 4 and 11 July, but none was gravid. Gravidity apparently peaked in mid-June during the high flow period. RELEASE OF GLOCHIDIA Table 1. Gravidity of female Pleurobema collina at study sites in Craig, Johns and South Fork Potts Creeks in 1988 and 1989. Collection Number Number Percentage of date Study site examined gravid females gravid! 1988 12 May Johns Creek l 0 23 May S. Fk. Potts Creek 36 15 30 May Craig Creek 18 10 27 June Johns Creek 8 6 04 July Johns Creek 6 4 11 July S. Fk. Potts Creek 15 4 14 July S. Fk. Potts Creek 5 2 25 July S. Fk. Potts Creek 16 7 09 August S. Fk. Potts Creek 11 11 12 August S. Fk. Potts Creek 8 0 1989 23 May S. Fk. Potts Creek 8 5 26 May Craig Creek 6 1 20% 10 2 03 June Craig Creek 6 5 83% S. Fk. Potts Creek 12 1 06 June S. Fk. Potts Creek 7 2 27 June Craig Creek 6 2 33% 04 July Craig Creek 7 l 14% 11 July Craig Creek 7 0 0% 16 July S. Fk. Potts Creek 8 5 28 July S. Fk. Potts Creek 17 5 08 August S. Fk. Potts Creek 22 1 ‘Females marked in 1988. 32 AMER. MALAC. BULL. 11(1) (1994) Drift sampling in 1988 provided a record of the changes in glochidial density with stream discharge (Table 2). Only individual glochidia (no conglutinates) were col- lected in drift nets. Glochidial densities peaked twice, once in late June (9.6 glochidia/m3) and again during mid-July (15.9 glochidia/m3). The mean summer water temperature stabilized around 23° C (Fig. 2). The densities of glochidia increased three days after the flow approached the mean summer level of 0.05 m3/sec on 23 June. Valves of the glochidia of Pleurobema collina are subovate and symmetrical. The exterior surface is smooth with very few pits. Glochidia are transparent and colorless, and the dorsal hinge is slightly convex. No significant dif- ference was evident in length, height, and hinge length of glochidia from South Fork Potts and Craig Creeks (p-val- ues: length = 0.39, height = 0.28, hinge length = 0.29). The length of valves ranged from 0.18 to 0.22 mm (mean + 1 SD = 0.19 + 0.01 mm), and the height of valves ranged from 0.14 to 0.20 mm (0.17 + 0.01 mm). The hinge length of the valves ranged from 0.10 to 0.16 mm (0.13 + 0.01 mm). The adductor muscle is attached to approximately 15% of the inside surface area of the valve and is just dor- sal and anterior to the valve’s center. Early in the gravidity period, the outer gills of females became significantly enlarged. As the glochidia matured, conglutinates changed color, and the hue of the gills changed from creamy white to pale gray or tan. Released conglutinates were subcylindrical, thin, and com- Table 2. Water temperature, stream discharge, and density of Pleurobema collina glochidia in South Fork Potts Creek, 1988. Sample Daily median Discharge Glochidial date temperature (°C) (m°/sec) density (N/ m’) 26 May 18 - 0 30 May 14 ~ 0 03 June 19 0.446 <1 06 June 16 0.209 <1 10 June 21 0.148 0 13 June 15 0.106 0 20 June 21 0.031 0 23 June 23 0.050 6.7 27 June 24 0.013 9.6 30 June = 0.025 5.6 05 July bg 0.010 0 08 July 25 0.010 1.6 11 July 25 0.012 15.9 14 July 24 0.024 0 18 July - 0.020 0 21 July sf 0.038 0 25 July - 0.025 (rosefin shiner) 17 5 Campostoma anomalum (Rafinesque) 16 (central stoneroller) Rhinichthys atratulus (Hermann) 15 (blacknose dace) 14 Phoxinus oreas (Cope) 9 (mountain redbelly dace) 17 "Includes pre-metamorphosed juveniles. Number of Juvenile! mussels Days to Temperature survivors collected metamorphosis range (°C) 0 — 0 16-21 4 41 48 15-18 11 — 13 15-18 0 — 7 15-18 14 14 30 15-18 10 44 34 15-18 0 — 0 15-18 5 2 40 15-18 5 8 36 15-18 1 13 26 16-23 12 3 23 16-23 9 6 50 15-18 1 1 29 16-23 13 8 44 15-18 HOVE AND NEVES: JAMES SPINYMUSSEL LIFE HISTORY Table 6. Fish species that did not serve as hosts for Pleurobema collina glochidia in laboratory experiments. Period of Number of Fish Number of Number of attachment premetamorphosed Temperature species fish infested survivors (days) juveniles range (°C) 1988 Catostomidae Catostomus commersoni (Lacepéde) 15 15 | — 16 (white sucker) Hypentelium nigricans (Lesueur) 11 11 2 — 20 (northern hog sucker) Ictaluridae Noturus insignis (Richardson) 18 18 8 — 22-23 (margined madtom) Percidae Etheostoma flabellare Rafinesque 8 6 10 — 15-18 (fantail darter) Centrarchidae Lepomis auritus (Linné) 13 11 12 — 15-18 (redbreast sunfish) Cyprinidae Notropis procne (Cope) 10 6 31 — 13 (swallowtail shiner) 1989 Percidae Etheostoma olmstedi Storer 1 1 6 — 13-15 (tessellated darter) Percina notogramma (Raney and Hubbs) 8 8 8 — 13-15 (stripeback darter) Centrarchidae Ambloplites rupestris (Rafinesque) 6 6 6 — 13-15 (rock bass) Cyprinidae Notropis procne (Cope) 4 4 55 = 13 (swallowtail shiner) Catostomidae Erimyzon oblongus (Mitchill) 8 8 2 — 13-15 (creek chubsucker) Cyprinidae Pimephales notatus (Rafinesque) 2) 9 12 — 13-15 (bluntnose minnow) Notemigonus crysoleucas (Mitchill) 6 5 41 4 13-15 (golden shiner) Umbridae Umbra pygmaea (DeKay) 6 6 12 — 13-15 (eastern mudminnow) Aphredoderidae Aphredoderus sayanus (Gilliams) 7 7 37 — 13-15 (pirate perch) Anguillidae Anguilla rostrata (Lesueur) 9 8 44 1 13-15 (American eel) Poeciliidae Gambusia affinis (Baird and Girard) 9 9 45 119 13-15 (western mosquitofish) Esocidae Esox niger Lesueur 9 8 25 22 13-15 (chain pickerel) Salmonidae Salvelinus fontinalis (Mitchill) 6 6 15 == 13-15 (brook trout) 'Not enumerated. 36 AMER. MALAC. BULL. 11(1) (1994) OO POF LE ERS BX SEASON AINE 5 SERN a ©. PRQQRRQIR a ~ eas \? OS = ents SOOO SOS OO" SR KAA AZ OX e KAAAAAA cx OOo od > m8 Os LO RR Sf CON, mee SERIO 2 00; SS EES o, me mm SESnsns Ss is xX? 05525050606 aS S x RN SOON SSS ~ x > 7 SERN ONO eateianee KR SSK Meee Le ee 2X SSS eaten i? OO 3 OO SIG c > 1M oP SS cS 22S SSSI Number of Mussels OX RLELE = ox 2, S88 IAN a4 . LX 0505 2% RG RR a, Me e%e% RO \ ene zy x ‘e OO © FD 3 53 <> ERS BR OOS ER <> OX C4 x o i“ > = Coe PSN QRS PRY RRO RRR x o, 2% x2 XS SZ O55 Me <> 0.05). In contrast, frequently sampled mussels had significantly lower whole wet weight at the end of the experiment (fre- quent = 1.377 g, infrequent = 1.455 g; Z = 2.163, P < 0.05). Disturbance and trauma resulting from sampling appear to have a measurable impact on growth in weight. Mortality rates were also affected by sampling. Animals measured infrequently during the study had signif- icantly lower mortality than those animals measured fre- quently indicating that removal and handling were detri- mental to survival (t-Test, t(9.05,11) = 4.202, P = 0.002). Infrequently measured Clinton River site animals had a mortality rate of 16.7% in 1990-1991, whereas frequently measured animals at the same site had a rate of 30.0%. The Thames River site in 1990-1991 showed a mortality rate of 17.1% for infrequently measured animals and a rate of 24.7% for those measured frequently. Similar results were again observed in the subsequent year. The Clinton River site had a mortality of 18.0% for frequently measured ani- mals and only 9.8% for those infrequently measured. The Thames River site also had an 18.0% mortality rate for fre- quently measured animals, but only 7.1% for those mea- sured infrequently. DISCUSSION These studies demonstrate that growth of individual zebra mussels can be studied in situ under relatively uni- form conditions using cages built of non-toxic materials. Individuals can be housed in cylindrical compartments with screening over each end. Cages of this design are relatively inexpensive and allow easy access to the animals for mea- surement or manipulation over the course of the experi- ment. It is clear that the compartment dimensions must be of adequate size to not restrict growth in any dimension. In our Lake St. Clair field studies, animals started in June 1991, at a mean length of 4.2 mm (17 mg wet weight) attained a length of 18-19 mm by the end of October and a length of 20-21 mm (110-130 mg) by May of the next year (Bitterman, 1992). Field data showed that not all growing seasons produced “good” growth, an observation that has been made by several other workers (see Dorgelo, 1993, and references therein). Annual and between-site variations in the growth of cage-reared zebra mussels will be more fully examined in a future report. Growth of caged zebra mussels in isolation obvi- ously differs somewhat from growth under more natural (usually crowded) conditions, where each mussel in a dense cluster is subject to strong intraspecific competition in the form of physical pressures from direct contact and collec- tive growth of adjacent mussels. However, the advantage to cage studies is that growth can be evaluated without con- founding effects of intraspecific competition, predation, or the more subtle effects of crowding. There is also the added advantage of being able to unambiguously monitor the growth of single individuals without needing to mark them. Previous studies have used cages to monitor growth, how- ever, these have involved groups of mussels in each cage (Smit et al., 1992; Dorgelo, 1993). The effect of cage-rear- ing on ratios of shell measurements observed in this study is one example of a difference arising from rearing in isola- tion. Width tended to be isometric with length among caged animals in contrast to uncaged animals which were not iso- metric. Allometric growth in length and width was observed in uncaged mussels regardless of density, suggest- ing that normal crowding has little impact on shell morpho- metrics. Based on field observations, it appears that year- to-year variations in lake conditions also exert a measurable influence on relative growth with consequences for allome- try. It has been asserted that morphometric features of zebra mussel shells vary only slightly among different lakes with no significant difference between lakes (Stanczy- kowska, 1964). In a survey of 11 Polish lakes, zebra mus- sels had L:W ratios with a mean of 2.24 in first-year mus- sels to 1.85 in five-year mussels. L:W ratios decreased with age at all sites. In contrast, L:H ratios increased with age from 1.93 in the first year to 2.06 in year five (Stanczy- kowska, 1977). Our L:W ratios are very similar to those reported by Stanczykowska, with mussels of 4 mm length having an initial L:W ratio of 2.4, decreasing to 1.8 for mussels of 20 mm length. L:H ratios reported here for Lake St. Clair ranged from 1.7 to 1.9. The main difference is that our data are for a single calendar year whereas Stanczy- kowska’s data are for growth over five years. Such differ- ences illustrate the influence of slower growth rates and longer time to maturity on allometry of growth in stable European populations in contrast to faster growth rates of recently invasive North American populations. Our data suggest that field density differences of the magnitude reported here have little or no consequences for shell growth morphometrics. However, growth occurring under caged conditions altered shell dimensions enough to change allometry to isometry and vice versa. Based on our cage study in Lake St. Clair, compartments of 30 mm diam- eter and 15 mm height are recommended as adequate for growth studies up to one year. Longer studies, especially under optimal growth conditions, may require larger com- partments. In Lake St. Clair, the occurrence of zebra mus- sels over 30 mm is not unusual, however such animals nor- mally constitute a relatively small percentage of the popula- 48 AMER. MALAC. BULL. 11(1) (1994) tion. In contrast, compartments of only 13 mm diameter markedly reduced the length and weight attained by test mussels compared to those in larger compartments even after only 114 days. Additionally, such growth-restricted mussels had different proportions of shell length, width, and height compared to uncaged mussels. Zebra mussels reared in intermediate size compartments showed only slight reduction in their length compared to animals in extra-large compartments. Although the differences were not significant, they suggest that cage-reared mussels may be responding to decreasing compartment space before they reach the absolute physical cage limits. Observed growth differences probably would have been greater had our study extended beyond 114 days. Our long-term cage studies used compartments of 26 mm diam- eter and 13 mm height, which yielded a mean maximum length of 20.8 mm over 336 days in 1991-1992 (a “good” growth year) and a mean maximum length of 16.6 mm over 343 days in 1990-1991 (a relatively poor growth year) (Bitterman, 1992, and unpublished data). Mean annual length and weight increments were 14.3 mm and 882 mg for 1990-1991, and 18.87 mm and 1540 mg for 1991-1992. Relatively large annual variation in growth may explain why rates reported in the North American literature are so divergent. For example, Hebert et al. (1989) reported Lake St. Clair zebra mussels grew at a rate of 10 mm/yr whereas Mackie (1991) found that most individuals in the same lake grew 15-20 mm/yr. Our overall growth rates for unrestricted cage- reared mussels (18.6 mm/yr, 0.051 mm/day) are compara- ble to those reported for European lakes. Dorgelo (1993) reported zebra mussel growth with a mean of 0.54-0.59 mm/wk (= 0.077-0.084 mm/day) for a Dutch oligotrophic lake, and ca. 0.050 mm/day for a mesooligotrophic lake. A Rhine River population with initial sizes of 4-7 mm grew at 14 mm/yr with virtually no net growth in winter (Neumann et al., 1993). Smit et al. (1992) reported growth rates for several Dutch lakes, including Lake IJsselmeer, where zebra mussels initially 4.2 mm added 12.35 mm in one growing season. Zebra mussel growth rates are similar to those of the Asiatic clam, Corbicula fluminea (Miller, 1774), but greater than those of the sphaeroidean clam, Pisidium casertanum (Poli, 1791). C. fluminea had a mean growth rate of 0.045-0.080 mm/day or 16-30 mm/yr (Aldridge and McMahon, 1978) and P. casertanum had a mean growth rate of 0.011 mm/day or 4 mm/yr (Mackie and Rooke, 1983). C. fluminea is an exotic clam from Asia and P. casertanum is one of many sphaeroidean clams native to American freshwaters that are of comparable size to zebra mussels. These growth rate differences are consistent with maximum size differences among these three species. In the restricted growth experiment, the small com- partments had a pronounced effect on zebra mussel growth, especially length, because that dimension was physically the most directly limited. Growth in width and height were not as constrained. Length of mussels in small compart- ments was reduced by 17.5% compared to that in extra- large compartments. Similar reductions occurred in width (15.7%) and height (11.4%). This flexibility in growth form is likely to be an adaptation to colonial life. Colonies can consist of tens to thousands of individuals of all age classes (Lewandowski, 1976). Densities in Mazurian Lakes have been recorded from 100-1000 individuals/m2. Lake St. Clair had densities as high as 200,000 individuals/m2 (Mackie, 1991). Older mussels tend to be directly attached to the substratum, often toward the center of the cluster, and covered by individuals of decreasing age toward the outer edge of the cluster. Other biofouling macroinverte- brates, such as barnacles, are known to change the dimen- sions of their growth due to the physical interference of crowding rather than differential exploitation of food or other resources (Connell, 1961). However, we saw no evi- dence in our study for lateral compression of the magnitude commonly observed in dense aggregations of barnacles or the marine mussel Mytilus edulis. Seed (1968) suggested that shell morphology in M. edulis was influenced by densi- ty and growth rate. Physical compression, which is greatest in areas of fast growth and high density, leads to elongated growth form; whereas low density populations with low compression have shells that tend to be higher (more trian- gular) in relation to length. Invading zebra mussels did not have a negative effect on test mussels in cages. This may be due to the aggregating life-habit in which zebra mussels naturally occur, making them relatively tolerant of crowding. Lewandowski (1976) observed that larvae settled more fre- quently on live zebra mussels than on empty shells or peb- bles. He stated that adults exert an attractive force on the settling mussels. The interactions between zebra mussels in aggrega- tions may be beneficial. For example, the irregular surfaces of mussel clusters create eddies increasing the amount of particulate food available (Frechette et al., 1989). Con- versely, life as part of a dense aggregation of mussels caus- es damage to some individuals. In Polish populations growth rates and maximum age decreased from overcrowd- ing (Stanczykowska, 1964). Both Wiktor (1963) and Stanczykowska (1977) found that individuals in the bottom layers of clusters had more deformations, attributed to the constant squeeze of neighboring mussels. Not only were the morphological features changed, but the caloric value of the dry body weight of the mollusks decreased as the density of the colony increased. As the population density increased, the body condition deteriorated. It was also noted that zebra mussels in dense populations consumed BITTERMAN, ET AL.: SHELL GROWTH AND MORPHOMETRICS OF ZEBRA MUSSELS 49 less food than those living in areas of lower mussel density (Stanczykowska, 1977). Hunter and Bailey (1992) observed elevated shell:tissue ratios and reduced zebra mussel condi- tion when crowded. In large individuals (20-30 mm length) from sparse populations, tissue mass was high in relation to shell mass yielding a low shell:tissue ratio of 9.5:1. In con- trast, similarly sized individuals from crowded populations had shell:tissue ratios averaging 15.5:1. ACKNOWLEDGMENTS The authors would like to thank the following peo- ple for their assistance during this study: Larry Laforet and Don MacLennan of the Ontario Ministry of Natural Resources for the use of their boat and assistance in field sampling; S. Jerrine Nichols of the U. S. Fish and Wildlife Service for supplying original cages, materials, and advice; and the following persons of the Michigan Department of Natural Resources: John Clevenger and Ken Koster for field sampling, Larry Shubel for equipment design and operation, and Al Sutton for drafting some of the figures. This study was partially funded by a Michigan Department of Natural Resources Youth Environmental Services Grant (Y903-170) and a Graduate Student Research Grant from Oakland University. LITERATURE CITED Aldridge, D. W. and R. F. McMahon. 1978. Growth, fecundity, and bioenergetics in a natural population of the Asiatic clam, Corbicula manilensis Philippi, from North Central Texas. Journal of Molluscan Siudies 44(1):49-70. Bitterman, A. M. 1992. Comparative growth and mortality rates of the zebra mussel, Dreissena polymorpha, from two sites in Lake St. Clair. Master’s Thesis, Oakland University, Rochester, Michigan. 58 pp. Connell, J. H. 1961. The effects of competition, predation by Thais lapil- lus, and other factors on natural populations of the barnacle Balanus balanoides. Ecological Monographs 31(1):61-103. Dorgelo, J. 1993. Growth and population structure of the zebra mussels (Dreissena polymorpha) in Dutch lakes differing in trophic state. In: Zebra Mussels: Biology, Impacts, and Control, T. F. Nalepa and D. W. Schloesser, eds. pp. 79-94. Lewis Publishers, Boca Raton, Florida. Frechette, M., C. A. Butman, and W. R. Geyer. 1989. The importance of boundary-layer flows in supplying phytoplankton to the benthic suspension feeder, Mytilus edulis L. Limnology and Oceanography 34(1):19-36. French, J. R. P. 1993. How well can fishes prey on zebra mussels in east- em North America? Fisheries 18(6):13-19. Haag, W. R., D. J. Berg, D. W. Garton, and J. L. Farris. 1993. Reduced survival and fitness in native bivalves in response to fouling by the introduced zebra mussel (Dreissena polymorpha) in western Lake Erie. Canadian Journal of Fisheries and Aquatic Science 50:13-19. Hebert, P. D. N., B. W. Muncaster, and G. L. Mackie. 1989. Ecological and genetic studies on Dreissena polymorpha (Pallas): a new mol- lusc in the Great Lakes. Canadian Journal of Fisheries and Aquatic Science 46:1587-1591. Hebert, P. D. N., C. C. Wilson, M. H. Murdoch, and R. Lazar. 1991. Demography and ecological impacts of the invading mollusc Dreissena polymorpha. Canadian Journal of Zoology 69:405-409. Hunter, R. D. and J. F. Bailey. 1992. Dreissena polymorpha (zebra mus- sel): colonization of soft substrata and some effects on unionid bivalves. The Nautilus 106(2):60-67. Leach, J. H. 1991. Biota of Lake St. Clair: habitat evaluation and environ- mental assessment. Hydrobiologia 219: 187-202. Leach, J. H. 1993. Impacts of the zebra mussel (Dreissena polymorpha) on water quality and fish spawning reefs in western Lake Erie. In: Zebra Mussels: Biology, Impacts, and Control, T. F. Nalepa and D. W. Schloesser, eds. pp. 381-397. Lewis Publishers, Boca Raton, Florida. Lewandowski, K. 1976. Unionidae as a substratum for Dreissena poly- morpha Pali. Polskie Archiwum Hydrobiologii 23(3):409-420. Mackie, G. L. and J. B. Rooke. 1983. Growth and production of three species of molluscs in six low alkalinity lakes in Ontario, Canada. Canadian Journal of Zoology 62:1474-1478. Mackie, G. L. 1991. Biology of the exotic zebra mussel, Dreissena poly- morpha, in relation to native bivalves and its potential impact in Lake St. Clair. Hydrobiologia 219:251-268. Nalepa, T. F. and D. W. Schloesser. 1993. Zebra Mussels: Biology, Impacts, and Control. Lewis Publishers, Boca Raton, Florida. 810 PPp- Neumann, D., J. Borcherding, and B. Jantz. 1993. Growth and seasonal reproduction of Dreissena polymorpha in the Rhine River and adja- cent waters. In: Zebra Mussels: Biology, lmpacts, and Control, T. F. Nalepa and D. W. Schloesser, eds. pp. 95-109. Lewis Publishers, Boca Raton, Florida. Roberts, L. 1990. Zebra mussel invasion threatens U. S. waters. Science 249: 1370-1372. Seed, R. 1968. Factors influencing shell shape in the mussel Mytilus edulis. Journal of the Marine Biological Association of the United Kingdom 48:561-584. Smit, H., A. Bij de Vaate, and A. Fioole. 1992. Shell growth of the zebra mussel (Dreissena polymorpha (Pallas)) in relation to selected physico-chemical parameters in the Lower Rhine and some associ- ated lakes. Archiv fiir Hydrobiologie 124(3):257-280. Stanczykowska, A. 1964. On the relationship between abundance, aggre- gations and “condition” of Dreissena polymorpha Pall. in 36 Mazurian lakes. Ekologia Polska (Seria A) 12:653-690. Stanczykowska, A. 1977. Ecology of Dreissena polymorpha (Pall.) (Bivalvia) in lakes. Polskie Archiwum Hydrobiologii 24(4):461- 530. Wiktor, L. 1963. Research on the ecology of Dreissena polymorpha Pall. in the Szczecin Lagoon. Ekologia Polska (Seria A) 11:275-280. Zar, J. H. 1984. Biostatistical Analysis, 2nd ed. Prentice Hall, Englewood Cliffs, New Jersey. 718 pp. Date of manuscript acceptance: 14 December 1993 Early development of the estuarine mollusk Polymesoda solida (Philippi, 1846) (Bivalvia: Corbiculidae) in Lake Maracaibo, Venezuela Yajaira G. de Severeyn, Héctor J. Severeyn, and Joseph J. Ewald Laboratory of Aquatic Invertebrate Culture, Biology Department, Universidad del Zulia, P. O. Box 1198, Maracaibo 4001A, Zulia, Venezuela Abstract. The estuarine clam, Polymesoda solida (Philippi, 1846), from Lake Maracaibo, Venezuela, was spawned and reared under laboratory condi- tions to monitor its early development. Spawning was induced via salinity changes and serotonin injections. Serotonin injections were more effective than salinity changes yielding viable gametes in 64% of the specimens used. Eggs and sperm were released into the water and fertilization was achieved by mix- ing 5 ml of eggs with | ml of sperm. After fertilization, the first two embryonic divisions occurred at | and 1.5 hr, trochophore larvae appeared between 24 and 30 hr, and free swimming straight-hinge veligers developed two days after the trochophores. Umbo stages with limited movement were seen after the fifth day. Trochophore and straight-hinged veliger stages were observed surviving inside the gelatinous envelope that surrounds the eggs. This envelope dis- solved in most cases when the trochophore stage was reached, but persisted in those experiments where the culture environment deteriorated. We postulate that this envelope plays an important role in protecting the early developmental stages under adverse conditions. For more than a century the existence of the estuar- ine bivalve Polymesoda solida in Lake Maracaibo, Vene- zuela (Fig. 1), has been known. This mollusk was believed, for more than 120 years, to be an endemic of this lake (Deshayes, 1854; Prime, 1865; Ten Broek, 1950; Rodri- guez, 1973), but Cosel (1977) and Garcia (1984) showed that populations of this species live in the Orinoco River delta in northeastern Venezuela and in coastal lagoons on the northeast Atlantic coast of Colombia. None of the pub- lished literature has described the eggs and larval development (Garcia, 1984; Severeyn et al., 1986, Severeyn, 1988, 1993). We describe here the development of P. solida. TAXONOMY Severeyn (1993) reviewed the status of Polymesoda species and considered P. arctata (Deshayes, 1854) from Lake Maracaibo a synonym of P. solida from the Atlantic coast of Nicaragua. The geographic range of P. solida therefore extends from the Orinoco River, Venezuela, to Gales Point, Belize (Severeyn, 1993). Within this range, four populations of P. solida have been described as differ- ent species. Species described as Polymesoda arctata from Venezuela, P. regalis (Prime, 1865) from Colombia, P. acuta (Prime, 1861), P. ordinaria (Prime, 1865), and P. boliviana (Clessin, 1879) from unknown localities on the Atlantic side of Central America are all P. solida. Samples of P. germana (Prime, 1870) from Tampico, Veracruz state, 72° Als Golfo de Venezuela x ~ LAGUNA DE SINAMAICA SAN RAFAEL DEL MOJAN LOS COQuITOS Lago de Maracaibo SUR AMERICA STUDIED AREA VENEZUELA Fig. 1. Area of study indicating collection stations at Los Coquitos and San Rafael del Mojan. American Malacological Bulletin, Vol. 11(1) (1994):51-56 51 a2 AMER. MALAC. BULL. 11(1) (1994) Mexico, were P. solida, but Severeyn (1993) considered the locality recorded doubtful because recent collections in the same locality have not found specimens. We follow Severeyn (1993) in adopting P. solida as the valid name for the species found along Venezuelan coasts. OBJECTIVE AND SIGNIFICANCE The density of Polymesoda solida populations in Lake Maracaibo (up to 200 specimens/m2) (Rodriguez, 1973) has made it an important source of human food. Its exploitation grew quickly in the late 1970’s, and dense pop- ulations in Laguna Sinamaica (Fig. 1) were depleted (Severeyn et al., 1986). Presently, at least 25,000 kilograms (whole weight including shells) are consumed monthly in the capital of Venezuela. The use of P. solida as food from underexploited areas will increase because populations of other marine species (e.g. Tivela and Donax) are now depleted after more than 30 years of commercial harvest. Despite a geographic range covering nine countries (Severeyn, 1993), Polymesoda solida has not been studied except for the population in the Lake Maracaibo, Vene- zuela, and the isolated report of Cosel (1977) in Colombian coastal lagoons. METHODS COLLECTION OF SPECIMENS The specimens of Polymesoda solida used in this study were collected between May and July 1985, from two stations, Los Coquitos and San Rafael del Mojan, on the northwestern coast of the Maracaibo Straits (Fig. 1). Salinities at collection sites varied from three to nine ppt. All specimens were collected in each of ten, 40 x 40 cm quadrats, five meters apart, along a transect parallel to the low tide water line. Each collection consisted of remov- ing the sediment to five cm deep, and sieving it through a two mm mesh. Sediment was mostly coarse to medium sand with 1-3% clay. Specimens larger than 25 mm were brought to the laboratory in plastic containers containing aerated water from the place of collection. This was done because there is evidence that Polymesoda solida, which reaches sexual maturity at 20 mm (Severeyn et al., 1986; Severeyn, 1988) may initiate spawning with a change in temperature, salini- ty, or other conditions. LABORATORY PROCEDURES In the laboratory, the clams were placed in 30 | glass aquaria containing sediment and filtered water from the collecting locality. These closed systems were aerated, with the salinity maintained at 5 + 1 ppt and temperature at 27 + 1° C. After one week of acclimation, animals were selected for spawning. SPAWNING AND LARVAL DEVELOPMENT Different techniques to provoke spawning under laboratory conditions have been tested in Polymesoda soli- da (see Garcia, 1984). These included sudden changes in salinity and temperature, mechanical stress (strikes), desic- cation, and chemical stress (pH changes, KCI and serotonin injections). Only salinity changes and serotonin injections have been successful. The use of these techniques in P. soli- da was described by Garcia (1984), Severeyn et al. (1986), and Ewald et al. (1986). Gibbons and Castagna (1984) originally described in detail the serotonin technique used for spawning induction in Mercenaria. Our experiments consisted of dropping or raising the salinity in the aquaria from that at the collection site to 0 or 35 ppt, respectively. These changes were always sudden. The rearing of embryos and larval stages followed the standard techniques of Loosanoff and Davis (1963). Embryos and larval stages were kept in 100 ml glass aquar- ia placed into environmentally controlled cabinets (Percival Co., Model I-35LL). Temperature was maintained at 26 + 1° C, and salinity at 5 + 1 ppt. A cycle of 12 hr darkness and 12 hr light simulated a natural daily cycle. After the trochophore stage was reached, bacterial contamination was controlled by adding five drops of Rifampicin (0.1 g/1) every two days to the 100 ml aquarium. FEEDING OF LARVAL STAGES Axenic algal cultures were used as food sources for veliger stages. The algae included two unidentified flagel- late species and Skeletonema sp. isolated from Lake Maracaibo waters, and /sochrysis galbana Parke obtained from Rosenstiel School of Marine and Atmospheric Sciences in Coral Gables, Florida. Local algal clones are part of the permanent collection of our laboratory and were isolated by the dilution technique in agar-agar Petri plates enriched with f/2 nutrient medium (Guillard and Ryther, 1962). Ten ml of algal concentrate, from a 30 | mass cul- ture, was added every two days to the aquaria containing 100-150 trochophore larvae. Every two days, five ml of the same algal concentrate was added to less densely populated aquaria containing 15-20 veliger larvae. Before each addi- tion of algae concentrate, water and detritus were removed and replaced with sterilized water of the same salinity, pre- pared with Instant Ocean (Aquarium Systems, Eastlake, Ohio). SEVEREYN ET AL.: EARLY DEVELOPMENT OF POLYMESODA SOLIDA 53 RESULTS SPAWNING AND FERTILIZATION Table 1 shows the results of those experiments that produced successful spawning of Polymesoda solida. Under our conditions, salinity change was not as successful as the use of serotonin for spawning induction. The four experi- ments that employed this chemical were successful in pro- ducing viable eggs and sperm. All other methods caused the production of large, hose-like packages of eggs from females (Fig. 2A) and balls of radially arranged spermato- zoa from males, oriented with their heads toward the center (Fig. 2B). The numbers of males and females that spawned were similar, slightly favoring females (55%). Almost 65% of the animals treated with serotonin spawned. Only speci- mens from Los Coquitos responded to the serotonin. Spermatozoa of Polymesoda solida have a long, cuneiform head with a mean length of 5 pm and a maxi- mum width of 2 pm. They possess a flagellum about 50 pm long. Eggs are completely spherical at spawning with a mean diameter of 60 pm. They are enclosed in a gelatinous envelope (King et al., 1986), called a hyaline capsule by Severeyn (1993), that increases their effective diameter to 120 pm (Fig. 3A, hc). Obtaining fertilized eggs was no problem. Nearly 100% of the eggs were fertilized using the techniques sug- gested by Loosanoff and Davis (1963). Unfertilized eggs in each experiment were removed to decrease the chance of bacterial contamination. EARLY CELL DIVISION AND LARVAL DEVELOPMENT Table 2 shows the developmental times (with maxi- ma and minima) for all stages. Fertilization was recognized by the appearance of a thick fertilization membrane (Fig. 3A, fm). Ten to 30 min later, a polar body appeared. One Table 1. Results of the experiments of spawning induction in Polymesoda solida (F, female; FT, free trochophore; M, male; MSR, maximum stage reached; SHV, straight-hinged veliger). Exp. Treatment N Spawned M F MSR 1 Sal. changes 25 3 2 l SHV 2 Sal. changes 25 2 1 l VEL 3 Serotonin 40 31 11 20 FT 4 Serotonin 16 12 7 5 SHV 5 Serotonin 20 6 4 2 Umbo 6 Serotonin 9 6 2. 4 SHV Total 135 60 27 33 Percent 44% 45% 55% Fig. 2. Abortive stages in Polymesoda solida. A) Female hose-like pack- ages (scale bar = 120 pm). B) Male ball of sperm (scale bar = 25 pm). and a half hr after fertilization, the first cleavage occurred (Fig. 3B). The second and third cleavages occurred 30 min and less than two hr after fertilization, respectively. Five to 8 hr after fertilization, most embryos were in blastula or gastrula stages. Trochophore larvae (Fig. 4A) were observed 24-36 hr after fertilization and straight-hinged veligers (Fig. 4B) were present after the second day. The few veligers that developed to the umbo stage appeared about five days after fertilization. The size of the two-day straight-hinged veliger varied from 137.5-176.0 pm and that of the few umbo stages observed between 270-290 pm. Despite the fact that both temperature and salinity were maintained, some embryos developed quickly and others slowly. The mortality of embryos prior to the tro- chophore stage (Fig. 4A) was low (5%), but then increased rapidly. A 70% mortality rate was observed at the tro- chophore stage, and only 5-7% of the embryos reached the straight-hinged veliger stage (Fig. 4B). Only 0.04% of the embryos reached the umbo stage. 54 AMER. MALAC. BULL. 11(1) (1994) Table 2. Development times after fertilization of embrionic and larval stages of Polymesoda solida under laboratory conditions. (*, capsule-free. Encapsulated trochophores appeared as early as 12 hr after fertilization). Stage Mean Time Range Polar body 20 min 10-30 min Ist cleavage 60 min 45-85 min 2nd cleavage 90 min 70-110 min Gastrula 5 hr 3-7 hr Blastula 7hr 4-9 hr Trochophore* 30 hr 14-42 hr Straight-hinged veliger 48 hr 36-68 hr Veliger (umbo) 5d 47d DISCUSSION SPAWNING The percentage of animals that spawned after treat- ment with serotonin (64.7%) was higher than those ob- tained by Gibbons and Castagna [1984; 45% in Geukensia demissa (Dillwyn, 1817) and 42% in Mercenaria merce- naria (Linné, 1758)]. However, the small number of spawning specimens (44.4%) suggests that most specimens were not ready for spawning despite the fact that they were collected during the spawning season. Severeyn (1988) demonstrated that Polymesoda solida has two spawning periods during the year, one between April and June and the other between November and January. We agree with Chanley (1975) who indicated that gonadal stage is a key factor controlling the possibility of spawning and the pro- duction of viable gametes under laboratory conditions, but these results suggest that the effectiveness of serotonin in inducing spawning is limited. The amount of sperm released by Polymesoda soli- da, 6.68 x 106 cells/ml/spawning event, was similar to the sperm density observed in Corbicula fluminea (Miller, 1774) (7.7 x 106 cells/ml/spawning; King et al., 1986). However, the number of eggs released by P. solida was larger than that observed in Corbicula. The fecundity of P. solida females was nearly 5000 eggs/100 ml/spawning while Corbicula only spawned seven eggs/500 ml. This dif- ference is clearly in response to different reproductive strategies. THE GELATINOUS ENVELOPE A feature of Polymesoda solida eggs that needs spe- cial comment is the presence of a hyaline capsule (Fig. 3A, hc). This capsule is present from the time eggs are liberated and normally disappears after the trochophore stage is reached. Occasionally, we observed that it remained as the trochophore developed, and sometimes through more advanced larval stages. Straight-hinged stages, actively moving within the capsule up to a month after fertilization, were observed. We considered these encapsulated larval stages anomalous. The role of the capsule is controversial. Severeyn (1993) argued that it could play an important role in pro- tecting the eggs and in their ability to disperse within and between estuarine areas. Similar capsules are quite com- mon in the Corbiculoidea (Corbiculidae and Sphaertidae) and Unionoidea (Mackie, 1984). Severeyn (1993) verified its presence in four species of Polymesoda. Olsen (1976) reported it in P. caroliniana (Bosc, 1801). King et al. (1986) also found it in North American Corbicula fluminea and although it was not mentioned by Morton (1989) in P. erosa (Solander, 1786) (= P. coaxans Rumph, 1741) it is evident in some of his figures. Considering our rearing con- ditions, we hypothesize that this capsule may protect advanced larval stages as long as unfavorable environmen- Fig. 3. Early embrionic stages of Polymesoda solida. A) Recently fertil- ized egg showing the fertilization membrane (fm) and the hyaline capsule (hc) (scale bar = 25 pm). B) First cleavage (two-celled embryo) (scale bar = 25 pm). SEVEREYN ET AL.: EARLY DEVELOPMENT OF POLYMESODA SOLIDA 55 Fig. 4. Larval stages of Polymesoda solida. A) Trochophore larva (scale bar = 25 pm). B) Straight-hinged veliger (scale bar = 35 pm). tal conditions persist. This coincides with observations made by Olsen (1976) who remarked that in P. caroliniana the hyaline capsules disappeared after fertilization, leaving embryonic and larval stages capsule-free. However in some cases, the capsule remained until the trochophore stage was reached. Further research is needed to clarify the nature and function of the hyaline capsule. DEVELOPMENT The results suggest that Polymesoda solida has a larval planktonic cycle that could be as short as five days, likely reaching the umbo stage in one week. Meta- morphosis of P. solida was not successful under our labora- tory conditions but we believe that bacterial contamination and subsequent mortalities were important factors con- straining this final phase of the larval cycle. In a few cases we observed umbo-stage individuals displaying crawling behavior, therefore assumed to be very close to reaching metamorphosis. We suspect that unfavorable conditions (e.g. unsuitable substratum, bacterial contamination, inade- quate salinity, etc.) may extend this seven-day cycle at least as long as one month. Apparently, the presence of the hya- line capsule allows this cycle extension. These results are based upon laboratory experiments and may not necessarily reflect the natural situation. The broad variation of developmental times for the encapsulat- ed trochophore, the capsule-free trochophore and the straight-hinged veliger stages suggests that these stages may be particularly sensitive to environmental changes. The developmental mean times observed in Polymesoda solida (Table 2) differ from those reported for P. caroliniana and Corbicula fluminea. The dissimilarities may be due to: (a) taxonomic differences, (b) ecological differences, and/or, (c) experimental artifacts. C. fluminea belongs to another genus and P. caroliniana and P. solida, although congeneric, are placed in remotely related subgen- era (i.e. Polymesoda and Neocyrena; Severeyn, 1993). The three species live in distinct habitats. C. fluminea inhabits freshwater, while both Polymesoda species are estuarine. P. caroliniana lives mostly in temperate conditions, while P. solida is exclusively tropical (Severeyn, 1993). CONCLUSIONS The populations of Polymesoda solida studied from Lake Maracaibo in western Venezuela developed from fer- tilized eggs to late-stage veligers (umbo) in five to seven days, but may prolong this up to one month. Observations also indicated that development is highly variable. After fertilization, the gastrula stage was reached in three h and the first larval stage, an encapsulated trochophore, in 12 hr. Under normal conditions, the capsule surrounding the tro- chophore disappeared 30 hr after fertilization and the straight-hinged veliger stage was reached in 48 hr. The advanced veliger or umbo stage appeared two days later. Sometimes, the capsule persisted and the trochophore was able to continue its development within it. Larvae that remained encapsulated reached the straight-hinged veliger stage and can persist as such up to one month. After this, the larvae died. The prolonged encapsulated phase suggests that between the trochophore and the straight-hinged veliger stages, the larvae are able to use stored food reserves, probably originating in the egg reserves. It is also possible that some nutrients can diffuse through the capsule to be absorbed by the larva. Eventually the larva dies, if not liberated from the capsule. Additional research must be done to answer a series of questions arising from this investigation. Are develop- mental times in nature similar to those reported here? Are the prolonged encapsulated stages present in nature? What conditions stimulate the maintenance of the capsule or 56 AMER. MALAC allow it to dissolve? Does the capsule allow diffusion of nutrients into it? If not, how can the larvae survive such a long period (up to a month) without an external food sup- ply? Does the capsule increase bouyancy and the likelihood of long-term dispersal? How can salinity changes affect lar- val stages within the capsule? Why was it not possible to achieve metamorphosis of Polymesoda solida under our laboratory conditions? ACKNOWLEDGMENTS The authors wish to acknowledge Dr. Edward Iversen of the Rosenstiel School of Marine and Atmospheric Sciences of the University of Miami, Coral Gables, for donation of algal cultures, and Jackeline Vilchez and Roberta Mora for caring for the algal mass cultures. Our thanks to Dr. Eric van den Berghe for advice and revision of part of this manuscript. This research was partially sponsored by a grant from Consejo de Desarrollo Cientifico y Humanistico (CONDES) of the University of Zulia, Venezuela. Additional facilities were provided by the Biology Department, School of Sciences, University of Zulia, Maracaibo, Venezuela. LITERATURE CITED Chanley, P. 1975. Laboratory culture of assorted bivalve mollusks. /n: Culture of Invertebrate Animals, W. L. Smith and M. H. Chanley, eds. pp. 297-318. Plenum Press, New York. Cosel, R. 1977. The genus Polymesoda on the north coast of South America (Bivalvia: Corbiculidae). Archiv fiir Molluskenkunde 108 (4-6):202-214. Deshayes, G. 1854. Description of new species of shells, from the collec- tion of H. Cuming. Proceedings of the Zoological Society of London 12:13-23. Ewald, J., H. Severeyn, and Y. Garcia de Severeyn. 1986. Induccién del desove de la almeja Polymesoda arctata (Bivalvia: Corbiculidae) mediante el uso de Serotonina. Acta Cientifica Venezolana XXXVII, suppl. 1:4. Garcia, Y. 1984. Biologia y ecologia de la almeja Polymesoda arctata Deshayes (Bivalvia: Corbiculidae) en el Lago de Maracaibo. . BULL. 11(1) (1994) Trabajo Especial de Grado, Universidad del Zulia, Facultad Experimental de Ciencias, Departamento de Biologia, Maracaibo. 128 pp. Gibbons, M. and M. Castagna. 1984. Serotonin as an inducer of spawning in six bivalve species. Aquaculture 40:189-191. Guillard, R. and J. Ryther. 1962. Studies of marine planktonic diatoms. I. Cyclotella nana Husted, and Dentonula confervacae (Cleve) Gran. Canadian Journal of Microbiology 8:229-239. King, C., C. Langdon and C. Counts III. 1986. Spawning and early devel- opment of Corbicula fluminea (Bivalvia: Corbiculidae) in laborato- ry culture. American Malacological Bulletin 4:81-88. Loosanoff, V. and H. Davis. 1963. Rearing of bivalve mollusks. Advances in Marine Biology 1:1-136. Mackie, G. L. 1984. Bivalves. In: The Mollusca. Vol. 7. Reproduction, A. Tompa, N. Verdonk and A. Biggelar, eds. pp. 351-418, Academic Press, New York. Morton, B. 1989. The functional morphology of the organs of the mantle cavity of Batissa violacea (Lamarck, 1797) (Bivalvia: Corbiculacea). American Malacological Bulletin 7(1):73-79. Olsen, L. 1976. Reproductive cycles of Polymesoda caroliniana (Bosc) and Rangia cuneata (Gray) with aspects of dessication in the adults, and fertilization and early larval stages in P. caroliniana. Doctoral dissertation, Florida State University, Tallahassee. 117 pp. Prime, T. 1865. Monograph of American Corbiculadae. Smithsonian Miscellaneous Collections 145, 80 pp. Rodriguez, G. 1973. El sistema de Maracaibo. IVIC, Caracas, Venezuela. 395 pp. Severeyn, H. 1988. El ciclo de vida de la almeja Polymesoda arctata (Bivalvia: Corbiculidae) en el Lago de Maracaibo. Trabajo de Ascenso, Universidad del Zulia, Facultad de Ciencias, Departa- mento de Biologia, Maracaibo. 33 pp., 10 tab., 2 figs. Severeyn, H. 1993. Taxonomic revision and phylogeny of the genus Polymesoda Rafinesque, 1820 (Bivalvia: Corbiculidae). Doctoral dissertation, University of Maryland, Eastern Shore Campus, MEES Program. 424 pp., 50 figs., 7 tabs., 9 appen. Severeyn, H., J. Ewald, Y. Garcia de Severeyn, R. Rodriguez and F. Morales. 1986. Estudio de las estrategias reproductivas y adaptati- vas de la almeja Polymesoda arctata en el Lago de Maracaibo. Informe Final, Proyecto de Investigacion CONDES-FEC, Universidad del Zulia, Maracaibo. 21 pp., 3 tab., 7 figs. Ten Broek, A. 1950. On some brackish water Mollusca from the Lake of Maracaibo. Zoologische Mededeligen 30(8):79-87. Date of manuscript acceptance: 18 February 1994 Gametogenic cycle of the Carolina marshclam, Polymesoda caroliniana (Bosc, 1801), from coastal Georgia Randal L. Walker! and Peter B. Heffernan!. 2 1Shellfish Research Laboratory, Marine Extension Service, 20 Ocean Science Circle, Savannah, Georgia 31411-1011, U.S.A. 2Marine Institute, 80 Harcourt Street, Dublin 2, Ireland Abstract. The gametogenic cycle of a brackish water bivalve, Polymesoda caroliniana (Bosc, 1801), was studied from December 1983 to February 1986 on Skidaway Island, Georgia, U.S.A. The population studied occurred in an irregularly flooded mosquito-control ditch which drains into the Skidaway River. Staging criteria were used to describe gametogenic development from histological preparations. Several features were quantitatively analyzed for males (% gonad area and % spermatozoa area) and females (% gonad area, % oocyte area, and egg number) using photo-planimetry (image analysis). A uni- modal gametogenic cycle was evident in 1984 and 1985. In 1984, late development occurred from December 1983 to July 1984 with ripe individuals occur- ring from June to September. Spawning occurred in August and September. In 1985, ripe individuals occurred in July and August with spawning occurring in August and September. Gonadal area levels were consistent from year to year. Males and females produced similar amounts of gametes during 1985, while males had significantly higher levels and a more protracted spawning period during 1984. Sex ratios were 1:1 with no hermaphrodites detected. Although the Carolina marshclam, Polymesoda caroliniana (Bosc, 1801) (Bivalvia: Corbiculidae), is a dominant macro-invertebrate within the brackish water of southeastern U.S. and Gulf of Mexico marshes, little is known of its biology. P. caroliniana ranges from Virginia to Mexico (Schalie, 1933) and occurs intertidally among the root mats of various marsh plant species (Harry, 1942; Andrews and Cook, 1951) or on muddy creek banks and bottoms (Schalie, 1933; Tabb and Moore, 1971; Hoese, 1973). Schalie (1933) described the distribution, shell mor- phology, and anatomy of P. caroliniana, while Olsen (1972, 1975) characterized it as a non-selective filter feeder. Fairbanks (1963) observed that the spawning period of P. caroliniana overlaps with that of another brackish water bivalve, Rangia cuneata (Sowerby, 1831), in Louisiana, while Duobinis and Hackney (1978) briefly described P. caroliniana recruitment periods (February and July) in Mississippi. Polymesoda caroliniana spawns from July to September in a population on the northwest Florida Gulf coast (Olsen, 1976). In an irregularly flooded population in the Mississippi marsh, ripe P. caroliniana individuals were found in May, July, August, and October (Hackney, 1983). Hackney (1983) was uncertain whether this indicated a pro- longed single breeding period or three discrete spawning periods. No studies on the gametogenic cycle of P. car- oliniana from Atlantic coastal populations have been undertaken. Due to the dearth of life history information on this species, its importance as a macroconsumer, and its abundance within the salt-marsh ecosystem, the reproduc- tive cycle of P. caroliniana for a population from the estu- arine waters of Georgia is described. MATERIALS AND METHODS Sampling and Tissue Processing A mean of 19.6 (range 15-20) Polymesoda car- oliniana individuals was collected monthly from December 1983 to February 1986 from a mosquito-control ditch locat- ed on the northwestern tip of Skidaway Island, Georgia. These ditches occur within a maritime forest and drain into the Skidaway River. The ditch-water salinity ranged from 0 ppt during rain runoff at low tide to 28 ppt on high spring tides. Clams occurred at the elevated end of the ditch and were irregularly covered by the tides. They were obtained from a mud substrate within the root mat of salt marsh grasses, Spartina alternata Loisel and Juncus roamerianus Scheele. Shell length measurements and tissue processing (histology) were performed as described previously (Heffernan et al., 1989a, b). Individuals ranged in shell length (longest possible measurement, i.e. anterior-posteri- or) from 14.6 to 33.6 mm, with a mean of 25.9 mm. Qualitative Reproductive Analysis The staging criteria described by Morton (1985) American Malacological Bulletin, Vol. 11(1) (1994):57-66 57 58 AMER. MALAC. BULL. 11(1) (1994) were employed for comparative purposes. Individuals were thus ascribed, on the basis of morphological observations, to one of the following stages: Inactive; Male or Female Developing; Male or Female Ripe; Male or Female Partially Spawned; and Male or Female Spent. Monthly gonad index (GI) values were computed (as described pre- viously, Heffernan et al., 1989a, b) using the following developmental stage scoring system: Inactive = 1; Developing = 4; Ripe = 5; Partially Spawned = 3; and Spent = 2. Quantitative Reproductive Analysis Two fields per specimen with a minimum separa- tion of 80-100 jm (more often ca. 300 pm) within the tis- sue block were printed on a Javelin! video printer via a TV camera system mounted on a standard compound micro- scope (10x objective) (Heffernan and Walker, 1989a). These prints contained elements of epithelial, gonadal, con- nective, and digestive tissues and had a standard field size of 7,566 mm2. Given a mean magnification factor of 244.9 x (+ 2.1, SE) this represents an actual field area of 0.125 mm2.2 Using total field area as the standard, several male and female area measurements were calculated from prints using a Sigma-Scan! digitizing tablet, and expressed as per- centage values. We have termed this method of image analysis “photo-planimetry.” Thus, mean values calculated from data measured on both specimen prints were evaluat- ed for each individual, depending on sex, for the following: Males - % Gonad and % Spermatozoa; Females - % Gonad, % Eggs, and Egg Number (manually counted). The per- centage values indicate how much of the field area was occupied by the gametogenic feature being measured, e.g. egg area. Sex ratios were tested against a 1:1 ratio with Chi-square tests (Steel and Torrie, 1960). RESULTS The gametogenic cycle of Polymesoda caroliniana from Skidaway Island, was ascertained from the combina- tion of qualitative and quantitative data collected from December 1983 to February 1986. Monthly qualitative and quantitative data assessments of reproductive conditions are \Mention of a trade name does not signify endorsement by the University of Georgia. 2Enforced camera (TV) replacement during the course of the study required replacement of prints taken with the original system. Consequently, the presently reported field dimensions differ unavoidably, due to the differing magnification factors of the two TV cameras, from those reported in an earlier Research Note (Heffernan and Walker, 1989a). Those dimensions were based solely on data generated by the original camera system. illustrated in Figs. 1 and 2 and Figs. 3 and 4, respectively. In Georgia, P. caroliniana displayed an unimodal gameto- genic cycle, with spawning in the late summer in both years studied. 1984-1985 SPERMATOGENESIS In 1984, late developmental stages of maturation were present from December 1983 to July 1984 with the percentage of ripe stages dominant in July and August (Fig. 1). Partial spawning commenced in August with the major- ity of animals having spawned by September. Presumably (no data gathered in October), spawning continued through December, although at a reduced level. During November and December early maturation stages dominated. Mean monthly gonad index values for males and females were tested by analysis of variance. Overall, no significant dif- ferences between males and females were detected (P = 0.3382) (Fig. 2). In 1985, early stages of maturation domi- nated until March (25%) and July (20%) when late and ripe stages (20%) occurred (Fig. 1). Mean gonad index did not change significantly for males from December 1983 to June 1984 before a significant increase to 4.2 occurred in July. The male gonad index decreased significantly in August and September, indicating spawning. By August, late stages dominated with some spawning occurring. Spawning animals dominated in September. Early matura- tion stages were present for the remainder of the year. Analysis of gonad indices (GI) (Fig. 2) revealed no signifi- cant change in GI from January to April, but a significant drop from April to May (2.9 to 2.3). This drop in GI values was followed by a rapid and significant increase from May to August (3.6). A significantly rapid drop in values from 3.6 to 2.6 occurred from August to September, indicating spawning. A significant increase in value occurred from September (2.6) to October (3.4), with no significant change in values thereafter until March 1986. The drop in GI in April-May is indicative of spawning, but neither the staging data (Fig. 1) nor the quantitative data support this suggestion. The decline in August-September does reflect spawning and quantitative data support this. The quantitative data for males revealed that per- cent total field area of gonadal tissue remained at approxi- mately 57%, with no significant changes from December 1983 to June 1984 (Fig. 3A). No significant differences in percent total field area for gonadal tissue (ca. 62%) occurred from November 1984 to February 1986, with the exception of the significant drop in May 1985. Although no significant differences in percent lumen (i.e. field area occupied by follicle lumena) in males were detected between the months from February 1984 to September 1984, an obvious decline was present from May to July. The large standard error present in the June sample resulted a9 GAMETOGENIC CYCLE OF POLYMESODA CAROLINIANA WALKER AND HEFFERNAN: 40 indifferent N = resorbing N BANIEU] % ] Spawning [_] early ff] ripe 78.8 8 8 8.8 8. “~ @eesaeasn VLLLLLLL LLL LLL 43714; 000 00,0000, $7723°@ @@eeaesean 3/PIN % * * * * ee ee ee eae TE —— ieee DJFMAMJJASONDJEMAMJJASONDJFE gjewsa4 % 1985 Fig. 1. Qualitative data illustrating the sex and gonad developmental stages of Polymesoda caroliniana from a Skidaway Island, Georgia population. The length of each area represents the percentage frequency of clams in each developmental stage. (*, no sampling). 1984 60 AMER. MALAC. BULL. 11(1) (1994) in no significant difference being determined in the data between May and July for pair-wise comparisons; however, it was obvious that significant differences occurred between the May (7%) and July (ca. 0%) data. The lumen percent- age remained at approximately 0% until October before increasing to 5% in November. The lumen percentage remained at approximately 4% until March 1985 before increasing to 7% in April. In month-wise comparisons, no significant increase was revealed due to large standard errors in the May sample; however, April and May values were significantly higher than the July and August values. After August, values increased and remained at approxi- mately 5% during the remainder of the study. The area occupied by sperm tissue remained at approximately 8% to 9% (of field) from December 1983 to July 1984 before significantly increasing and peaking in August and September 1984. The sperm percentage signif- icantly dropped to approximately 4% by November and remained at this level until April. Thus, males matured in both years by August and spawned by October in 1985 and by at least November in 1984 (October 1984 data absent). These quantitative data support the conclusion that a uni- modal male gametogenic cycle occurs in this Georgia pop- ulation of P. caroliniana. 1984-1985 OOGENESIS In 1984, early maturation stages were dominant in animals from December 1983 to May 1984, although late maturation stages were present (Fig. 1). In June, late matu- ration stages were dominant with some animals ripe and spawning. By July, all animals were in the late maturation Gonad Index (GI) DJF MAM J J * x 1984 A S$ 0 N D J or ripe stage. Ripe individuals (41%) were dominant in August with 36% being in the spawning stage. Spawning individuals were dominant (36%) in September. Individuals were in the early maturation stages from November to May 1985. In June 1985, late maturation Stages were observed and became the dominant stage until August. Ripe stages were present in July and August, with spawning occurring in August and September. Animals in the early stage of maturation were present throughout the remainder of the study. As was the case in males, female GI data supported the staging data. Female GI remained constant from December 1983 to June 1984 before signifi- cantly increasing to a maximum (4.2) in July. Significant declines occurred from July to September. GI levels reached a plateau again from November 1984 until May 1985 before a rapid increase occurred. Female GI levels peaked in July (4.1) before significantly dropping in August and September. GI levels reached a plateau again by October 1985. Quantitative data for the female showed that gonadal tissue remained consistent, at approximately 55% (of field), from January until June 1984. In 1985 this value was slightly higher (60%) from November 1984 to June 1985. Rapid reductions in gonadal tissue area occurred between June, July, and August 1984 (Fig. 4A). No signifi- cant differences occurred between August and September 1984. In 1985, similar significant reductions occurred between June and July with no significant differences being detected between July and September. These reductions in percent total field area occupied by gonadal tissue occurred during the period corresponding to late development and —=— Males --@-- Females FMAM J 1985 JAS ON D J F * * Fig. 2. Gonad index values for male and female Polymesoda caroliniana in a population from Skidaway Island, Georgia. (*, no sampling). WALKER AND HEFFERNAN: GAMETOGENIC CYCLE OF POLYMESODA CAROLINIANA 61 > “% of Field Occupied by Gonad % of Field Occupied by Spermatazoa DJF MAMJJASONDJFMAMJSJSASON DJ F * * * ® 1984 * 1985 Fig. 3. Composite of quantitative data (obtained using image analysis of print microscopic fields) representing the state of gonad conditions for Polymesoda caroliniana males from Skidaway Island, Georgia population. (A) Mean percentage of print field area occupied by gonad. (B) Mean percentage of print field area occupied by spermatozoa. Vertical bars represent two standard errors above the mean. Horizontal bars indicate periods of statistically significant (t-tests) changes in the features being measured. These have been included to highlight likely spawning events. (*, no sampling). maturity (ripe stages). Minimum values were obtained dur- ing spawning periods (August-September 1984 and July- August 1985). Area occupied by oocytes (Fig. 4B), dis- played no significant changes from February 1984 to June 1984 and from November 1984 to June 1985. Significant decreases in gonadal area occupied by oocytes occurred from June to September in both years. Mean number of eggs (Fig. 4C) remained at zero from December 1983 to May 1984 and from November 1984 to June 1985. Egg production (mean number of eggs) peaked in July of both years (Table 1). In 1984, egg production remained high through September (i.e. no significant differences) and dropped to zero by November. In 1985, a significant reduc- tion in egg production occurred in August and reached zero production by September. The two sets of quantitative data thus reveal a unimodal spawning peak from August to September confirming qualitative results. Table 1. Gametogenic peak production values for Polymesoda carolini- ana during 1984 and 1985. Data given in % area or number of eggs + one standard error. (NS, not significant; Sign., significance). 1984 1985 t-test df Sign. Female % Gonad 55.9 +2.9 (June) % Egg 14.3 + 1.5 (Aug) # Egg 30.4 + 2.7 (July) 57.3 45.2 (June) 0.230 16 NS 18.742.0 (July) 1.730 19 NS 34.344.1 July) 0.792 17 NS Male % Gonad 83.4+3.3 (July) % Sperm 24.3 +5.5 (Aug) 71.6+5.9(Aug) 1.760 8 NS 19.9+3.8(Aug) 0.663 10 NS COMPARISON OF YEARLY DATA Comparison of spawning events for Polymesoda caroliniana between years reveal similar patterns (major 62 AMER. MALAC. BULL. 11(1) (1994) spawning in August and September) in both years. Yet, during 1984, a protracted spawning period occurred for both males and females, as compared to 1985, as indicated by the qualitative (Fig. 1) and quantitative data (Figs. 3, 4). In 1984, spawning in the males and females appeared to be extended, as evidenced by females spawning in June, August and September and males spawning in August, September, October and November (Fig. 1) and by the plateauing effect observed in the quantitative data for 1984 (Figs. 3, 4). For the percentage of sperm tissue, a peak was obtained in August 1984 with no significant differences detected between August and September (thus a plateau) before the significant drop by November (Fig. 3C). Mean egg count peaked in July and did not decrease significantly until September, before dropping to zero by November. In 1985, values rapidly increased from June to a July peak and then significantly dropped by August. Thus, no plateauing effect occurred and a more rapid spawning event resulted in 1985, as compared to 1984. Although a more protracted spawning event occurred in 1984 as compared to 1985, gametogenic pro- duction levels each year were equal (Table 1). There were no significant differences in male gonadal area or sperma- tozoa levels at maturity during 1984 and 1985. Similarly, female gonadal area, oocyte area and mean number of eggs were equal each year (Table 2). Polymesoda caroliniana is dioecious. Individuals (N = 425) ranging from 14.6 to 33.6 mm in shell length were sexed during this study. Females (N = 224) slightly outnumbered males (N = 201), but the ratio did not signifi- cantly differ (Chi-square = 1.24, p > 0.05) from an equal (1:1) sex ratio. No hermaphrodites were detected. Mean ambient sea water temperature and salinity data for the Skidaway River are given in Fig. 5. Ambient sea water temperatures peaked in August 1984 and early September 1985. DISCUSSION The Polymesoda caroliniana population from a high marsh site in Georgia has a single spawning season occurring in August and September. The unimodal gameto- genic cycle of P. caroliniana described herein for a coastal Georgia population is similar to the patterns described for populations in the northwest Florida Gulf coast and a Mississippi marsh. Olsen (1976) observed a unimodal spawning pattern from July to September in a Florida popu- lation, which is similar to our observed pattern of August- September spawning during 1985 (Fig. 1). Hackney (1983) observed ripe P. caroliniana individuals in Mississippi in May, July, August and October, but was uncertain whether this indicated a single prolonged breeding period or three distinct spawning periods. The spawning pattern exhibited in Mississippi is similar to the Georgia P. caroliniana spawning pattern of 1984, where spawning was observed in June and again in August-September for females and from August-September and then November-December for males (Fig. 1). Unfortunately, an October sample was not taken in 1984, so we do not know if the male pattern is Table 2. Comparison of gonadal production levels among ripe males and females of Polymesoda caroliniana at the Skidaway Island site during 1984 and 1985 and three other bivalve species from populations at House Creek, Little Tybee Island, Georgia. (NS, not significant; S, significant; Sign., significance). Gonadal Area (%) Male Female A. Polymesoda caroliniana 1984 83.4 55.9 (June) 1985 S75 71.6 (Aug./May) B. Geukensia demissa (from Heffernan and Walker, 1989b) 1984 54.1 (5.2)-Jul. 40.2 (5.3)-Jul. 1985 59.7 (2.4)-Jul. 54.9 (6.6)-Jul. C. Mercenaria mercenaria (from Heffernan et al., 1989a) 1984 Spring 94.5 (1.4)-Feb. 94.3 (1.4)-Mar. 1984 Fall 91.4 (2.1)-Sept. 92.5 (1.8)-Sept. 1985 Spring 90.0 (2.2)-Jan. 96.5 (1.1)-Jan. 1985 Fall 92.7 (1.0)-Sept. 95.3 (1.6)-Sept. 1985 Winter 93.4 (1.1)-Dec. 92.0 (1.9)-Nov. D. Crassostrea virginica (from Heffernan et al., 1989b) 1984 85.1 (7.4)-Apr. 64.4 (9.8)-Jul. 96.9 (0.7)-Jun. 99.6 (0.8)-Jun. Gamete Area (%) Sign. Sperm Ova Sign. S (p< 0.001) 24.3 14.3 (Aug.) NS NS 19.9 18.7 (Aug//July) NS NS 24.0 (2.3)-Aug. 33.3 (2.2)-Aug. S (p < 0.02) NS 27.0 (1.4)-Jul. 29.5 (3.9)-Jul. NS NS 38.9 (4.5)-Mar. 33.6 (2.2)-Mar. NS NS 32.1 (1.6)-Aug. 21.5 (1.0)-Sept. S (p< 0.001) NS 67.1 (4.4)-Apr. 26.5 (0.7)-Mar. S (p< 0.001) NS 16.0 (1.5)-Sept. 19.0 (1.9)-Sept. NS NS 28.4 (3.0)-Dec. 23.9 (0.6)-Dec. NS NS 28.3 (2.3)-Apr. 22.3 (4.2)-Jul. NS NS 52.7 (4.2)-Jun. 44.3 (4.1)-Jun. NS WALKER AND HEFFERNAN: GAMETOGENIC CYCLE OF POLYMESODA CAROLINIANA 63 A 100 Ss e 80 ec re) 60 So as = 40 (=) > ] 20 B 80 60 40 % of Field Occupied by Oocytes st =s—s a2 —58 pe os as. oe DJFMAMJJASONDJFMAMJJ ASON DJF * * * * * 1984 1985 Fig. 4. Composite of quantitative data (obtained using image analysis of print fields) representing the state of gonadal condition for Polymesoda caroliniana females from Skidaway Island, Georgia population. (A) Mean percentage of print field area occupied by gonad. (B) Mean percentage of print field area occupied by oocytes. (C) Mean number of eggs per print field area. Vertical bars represent two standard errors above the mean. Horizontal bars indicate peri- ods of statistically significant (t-tests) changes in the features being measured. These have been included to highlight likely spawning events. (*, no sam- pling). 40 20 Mean Egg Number per Field 20 truly unimodal or bimodal. Hackney (1983) observed no that in Georgia one spawning event occurs. The quantita- ripe individuals in either his June or September samples, tive data also supports a unimodal spawning event for P. i.e. periods between spawnings, leading him to consider caroliniana in this Georgia population. that three spawning periods occurred. In our data, the per- Our results are also similar to the spawning pattern centage of ripe individuals increased from June to August, observed for Polymesoda erosa (Solander, 1786) from a with ripe individuals occurring in July, leading us to believe Hong Kong mangrove population. P. erosa commences 64 AMER. MALAC. BULL. 11(1) (1994) 32 28 RO > RO © Temperature (°C) 1983 1984 (ydd) Ayiuies —— Temperature —e— Salinity 1985 Fig. 5. Mean ambient water temperatures and salinities for the Skidaway River that occasionally inundates the Polymesoda caroliniana population located in an irregularly flooded mosquito-control ditch on Skidaway Island, Georgia. active gametogenesis in April and May and by June mature gametes occur with spawning taking place during summer (Morton, 1985). This type of reproductive pattern is simi- lar to our 1985 results, where late active stages were first observed in June and spawning finished by September. However, it differs from the 1984 results where late active stages were present from December 1983 through July 1984. In 1984 females spawned from June to possibly October, while males spawned from August until December. Obviously, temporal effects between years can alter the gametogenic cycle of a species. Temporal differences in gametogenic cycles have been recorded for other Georgia bivalves. Three bivalve species collected from the same site in Wassaw Sound showed marked differences in their reproductive patterns between 1984 and 1985. Mercenaria mercenaria (Linné, 1758) showed evidence of trimodal spawning in 1984, but bimodal spawning in 1985 (Heffernan et al., 1989a). Crassostrea virginica (Gmelin, 1791) showed evidence of spawning from June to November in 1984, but only from August to November in 1985 (Heffernan et al., 1989b). Geukensia demissa (Dillwyn, 1817) spawned from August to either October (females) or December (males) in 1984, but only from July to September in 1985 (both sexes) (Heffernan and Walker, 1989b). In a Spisula solidissima similis (Say, 1822) population in coastal Georgia, clams spawned from March to May in 1990, but from April to June in 1991 (Kanti et al., 1993). Equal gametogenic production levels between years, even with different timing patterns (protracted or rapid), have been recorded in other studies (Table 2). Gonadal area levels of Geukensia demissa were consistent from year to year (Heffernan and Walker, 1989b). G. demissa males and females produced similar amounts of gametes in 1985, but females had a significantly higher out- put of gametes in 1984. In Mercenaria mercenaria and Crassostrea virginica (fide Heffernan et al., 1989a, b), greater levels of reproductive output were detected in 1985 than in 1984. Thus, temporal variations may or may not occur within populations. In the two cases where temporal variations did not occur (Geukensia and Polymesoda), there is a relatively short spawning season. When temporal vari- ations were detected, either a polymodal (Mercenaria) or a prolonged spawning season occurs (Crassostrea) from May WALKER AND HEFFERNAN: GAMETOGENIC CYCLE OF POLYMESODA CAROLINIANA 65 to October. For Polymesoda caroliniana, clams in 1984 had a much more protracted spawning period than in 1985, although gametogenic production levels each year were equal (Table 1). The factor(s) responsible for the observed temporal variations in gametogenesis between years is unknown. Temperature is a major factor in controlling gametogenesis in marine bivalves (Giese, 1959). For Spisula solidissima (Dillwyn, 1817) populations, Ropes (1968) noted that clams generally spawned once per year when lower water temperatures prevailed, whereas, in warm-water years, clams spawned twice. For Mercenaria mercenaria in Georgia, Heffernan et al. (1989a) observed that clams spawned for a third time during a warmer than usual winter, whereas, clams previously spawned only in spring and fall. For P. caroliniana in this study, water tem- perature patterns for 1984 and 1985 were similar (Fig. 5). Salinity can also regulate gametogenesis in marine bivalves (Giese, 1959). For Polymesoda caroliniana in this study, differences in salinity were observed between years, with salinities dropping lower during winter 1984 than in winter 1985. Although differences in salinity patterns were observed between years in this study (Fig. 5), salinity read- ings from the Skidaway River, into which these drainage ditches empty, are not necessarily reflective of salinities within the ditches at any one point in time. These ditches are flooded irregularly by the tide. During low tide, rain runoff can drastically alter the salinity within the ditches, as compared to the relatively vast volume of water in the river. These periodic changes in salinity due to rain runoff can play a role in regulating the gametogenic cycle of P. car- oliniana, but the dynamics of the interaction between changes in salinity and spawning of P. caroliniana are unknown. An examination of gametogenic production levels (% gonad area) from each sex during peak maturity peri- ods, reveal that males had higher production levels than females in 1984, while production levels were equal in 1985 (Table 2A). In similar studies, no significant differences in gonadal area between sexes were recorded for Geukensia demissa (Table 2B), Mercenaria mercenaria (Table 2C), and Crassostrea virginica (Table 2D); however, significant differences in percent gamete area between sexes per year were detected in one year for G. demissa (Table 2B) and during the spring spawning of 1985 for M. mercenaria (Table 2C). In other reproductive studies, females of the Iceland scallop, Chlamys islandica (Miller, 1776) showed higher ova production values than males for sperm produc- tion (Sundet and Lee, 1984; Vahl and Sundet, 1985). Obviously, much variation within reproductive physiology can occur between years as well as between sexes within a year. The dynamics controlling such events are not well defined. Based on the results of this study, Polymesoda caroliniana is dioecious. Sex ratio did not deviate signifi- cantly from 1:1 during this study and no hermaphrodites were recorded. Unfortunately, neither Olsen (1976) nor Hackney (1983) gave information pertaining to sex ratio in their studies; however, Olsen (1976) did note that no her- maphrodites were recorded among the 236 individuals examined. Polymesoda erosa is also dioecious and of the 167 specimens examined in Hong Kong, 51.5% were females, 38.5% males, 0.5% hermaphrodites and 9.5% indeterminate sex (Morton, 1985). The results of this study should be interpreted with care when expanding the data to reproduction of the P. car- oliniana population of coastal Georgia. P. caroliniana occurs in a variety of habitats within the coastal waters of Georgia. In higher saline areas, P. caroliniana occurs high in the intertidal zone usually either in small creeks or salt marsh areas that receive rain runoff; whereas, in brackish water areas farther inshore, clams can occur lower in the intertidal zone and onto subtidal creek bottoms (Walker, pers. obs.). The reproductive cycle of clams from more brackish and less hostile (in terms of length of aerial expo- sure time) environments can be different in timing and length of spawning. ACKNOWLEDGMENTS This work was supported by Georgia Sea Grant Project number NA84AA-D-00072. Dr. H. Langston helped greatly at the histological and analytical stages of this work. The loan of an Autotechnicon and microtome from Georgia Institute of Technology greatly facilitated this research as well as three earlier reproductive studies. Dr. R. Hanson allowed us to use his digitizing system and Mr. C. Robertson provided much appreciated assistance during this phase of the study. Mr. G. Paulk, Mr. D. Head and Ms. P. Adams provided much appreciated technical assistance. The statistical advice of Dr. J. W. Crenshaw, Jr. is duly acknowledged. Mrs. D. Thompson typed the manuscript. Ms. A. Boyette and Ms. S. McIntosh provided all the graphic materials. The editorial comments of two anonymous reviewers are appreciated. LITERATURE CITED Andrews, J. D. and C. Cook. 1951. Range and habitat of the clam Polymesoda caroliniana (Bosc) in Virginia (Family Cycladidae). Ecology 32(4):758-760. Duobinis, E. M. and C. T. Hackney. 1978. Seasonal and spatial distribu- tion of the Carolina marsh clam, Polymesoda caroliniana (Pelecypoda: Corbiculidae), in a Mississippi tidal marsh. Association of Southeastern Biologists Bulletin 25:44. Fairbanks, L. D. 1963. Biodemographic studies of the clam, Rangia cuneata Gray. Tulane Studies in Zoology 10(1):3-47. Giese, A. C. 1959. Comparative physiology. Annual reproductive cycles of marine invertebrates. Annual Review of Physiology 21:547-576. 66 AMER. MALAC. BULL. 11(1) (1994) Hackney, C. T. 1983. A note on the reproductive season of the Carolina marsh clam, Polymesoda caroliniana (Bosc), in an irregularly flooded Mississippi marsh. Gulf Research Reports 7:281-284. Harry, H. W. 1942. List of Mollusca of Grand Isle, Louisiana, recorded from the Louisiana State University Marine Laboratory 1929-1941. Occasional Papers of the Marine Laboratory, Louisiana State University (1):1-13. Heffernan, P. B. and R. L. Walker. 1989a. Quantitative image analysis methods for use in histological studies of bivalve reproduction. Journal of Molluscan Studies 55:135-137. Heffernan, P. B. and R. L. Walker. 1989b. Gametogenic cycles of three bivalves in Wassaw Sound, Georgia III: Geukensia demissa (Dillwyn, 1817). Journal of Shellfish Research 8:327-334. Heffernan, P. B., R. L. Walker, and J. L. Carr. 1989a. Gametogenic cycles of three bivalves in Wassaw Sound, Georgia I: Mercenaria merce- naria (L.). Journal of Shellfish Research 8:51-60. Heffernan, P. B., R. L. Walker, and J. L. Carr. 1989b. Gametogenic cycles of three bivalves in Wassaw Sound, Georgia II: Crassostrea virginica (Gmelin). Journal of Shellfish Research 8:61-70. Hoese, H. D. 1973. Abundance of low salinity clam, Rangia cuneata, in southwestern Louisiana. Proceedings of the National Shellfisheries Association 63:99-106. Kanti, A., P. B. Heffernan, and R. L. Walker. 1993. Gametogenesis cycle of the southern surf clam Spisula solidissima similis from St. Catherines Sound, Georgia. Journal of Shellfish Research 12:255- 261. Morton, B. 1985. The reproductive strategy of the mangrove bivalve Polymesoda (Geloina) erosa (Bivalvia: Corbiculoidea) in Hong Kong. Malacological Review 18:83-89. Olsen, L. A. 1972. Comparative functional morphology of feeding mech- anisms in Rangia cuneata (Gray) and Polymedosa caroliniana (Bosc). Proceedings of the National Shellfisheries Association 63:4. (Abstract) Olsen, L. A. 1975. Ingested material in two species of estuarine bivalves: Rangia cuneata (Gray) and Polymesoda caroliniana (Bosc). Proceedings of the National Shellfisheries Association 66: 103-104. Olsen, L. A. 1976. Reproductive cycles of Polymesoda caroliniana (Bosc) and Rangia cuneata (Gray), with aspects of desiccation in the adults, and fertilization and early larval stages in P. caroliniana. Doctoral dissertation, Florida State University, Tallahassee, Florida. 117 pp. Ropes, J. W. 1968. Reproductive cycle of the surf clam, Spisula solidissi- ma, in offshore New Jersey. Biological Bulletin 135:349-365. Schalie, H. V. 1933. Notes on the brackish water bivalve, Polymesoda caroliniana (Bosc). Occasional Papers of the Museum of Zoology, University of Michigan 11:1-9. Steel, R. and J. Torrie. 1960. Principles and Procedures of Statistics. McGraw-Hill Book Co., New York. 481 pp. Sundet, J. H. and J. B. Lee. 1984. Seasonal variations in gamete develop- ment in the Iceland scallop, Chlamys islandica. Journal of the Marine Biological Association of the United Kingdom 64:411-416. Tabb, D. C. and D. R. Moore. 1971. Discovery of the Carolina marsh- clam, Polymesoda caroliniana (Bosc), a supposed Florida disjunct species in Everglades National Park, Florida. Gulf Research Reports 3:265-277. Vahl, O. and J. H. Sundet. 1985. Is sperm really so cheap? Jn: Marine Biology of Polar Regions and Effects of Stress on Marine Organisms. J. S. Gray and M. E. Christiansen, eds. John Wiley & Sons Ltd., Chichester. pp. 281-285. Date of manuscript acceptance: 31 May 1994 Morphometric and biochemical changes in two age classes of the tropical scallop, Argopecten ventricosus, under laboratory conditions Janzel R. Villalaz G. Departamento de Biologia Acuatica, Universidad de Panama, Panama, Republica de Panama Abstract. A laboratory study was carried out in Lewes, Delaware, U.S.A., to (a) compare, through controlled laboratory experiments, morphometric and biochemical changes in the digestive gland, adductor muscle, mantle-gills and gonad of eight- and 16-month-old tropical scallops Argopecten ventricosus (Sowerby, 1842), and (b) determine the effect of age on the content of carbohydrates and protein in the gonad. The gonadal index of eight- and 16-month-old scallops declined through the first 40 days of the experiment. Also, carbohydrate content declined significantly in the digestive gland and adductor muscle. This suggests energy was used either for somatic growth and/or reproduction. Some carbohydrates could have been used for reproduction, but were not transferred directly to the gonad before day 40. Adductor muscle dry weight and protein content of eight- and 16-month-old scallops increased for the first 40 days of the experiment, then declined around day 55. Simultaneously, around day 55, both groups of scallops had noticeable increases in gonadal dry weight and protein content. This suggests that the adductor muscle stores protein during the first 40 days, then protein catabolism in the adductor muscle contributes to reproductive activity of the gonad. Protein content in the mantle was higher in eight- than in 16-month-old scallops. This suggests mantle-gill protein provides better support for somatic growth in eight- than in 16-month old tropical scallops. The laboratory study in Delaware suggests that one-year-old scallops will spawn in the field in March. The reproductive cycle of bivalves includes the Mytilus edulis Linné, 1758. following stages: (a) vegetative (juveniles); (b) activation; Reproductive strategies used by scallops at different (c) growth and gametogenesis; (d) maturation; (e) rest- ages have been studied by several authors: Argopecten irra- ing stage. According to Sastry (1975) the timing of the dians by Bricelj et al. (1987) using physiological condi- reproductive cycle is a genetic response to the environment, tions, and Epp et al. (1988) using condition indices. Blake and involves interaction of exogenous and endogenous fac- (1972) studied the interactions among age, temperature, tors. Giese (1959) identified the exogenous factors as tem- neurosections and reproductive stage in A. irradians. perature, salinity, day length and food abundance; endo- Allocation of chemical compounds in different size classes genous factors included age, metabolism and neuro- of Chlamys islandica (Miller, 1776) was studied by Sundet endocrines. and Vahl (1981) and in A. purpuratus (Lamarck, 1819) by Age is an endogenous factor affecting gametogene- Martinez (1991). sis in scallops (Barber and Blake, 1991). With a reduction The literature indicates that bivalves can obtain in the rate of somatic growth as scallops increase in size, energy either directly from food, or from storage substrates increase in the energy allocated for gamete production in organs and tissues, such as the digestive gland (Taylor occurs. According to Bayne and Newell (1983) this process and Venn, 1979), the adductor muscle (Ansell, 1974; Epp et of redistribution from somatic to reproductive energy sets al., 1988), or the mantle (Barber and Blake, 1981; Lowe et the limit for species size. Thompson and MacDonald al., 1982). Also, a fluctuation in utilization of storage com- (1991) suggested there is a reduction in energy available for pounds (lipids, carbohydrates and proteins) has been ob- growth as the organism increases in size, depending on the served in these bivalves relative to the reproductive cycle. combination of food available and efficiency to convert this The objectives of this study were: (a) to compare, food into tissue. However, according to Blake (1972) a through controlled laboratory experiments, morphometric minimum age (or size) is required for gametogenesis to and biochemical changes in the digestive gland, adductor begin in Argopecten irradians (Lamarck, 1819). muscle, mantle-gills and gonad between eight- and 16- Scallops of different ages have distinct metabo- month-old tropical scallops, Argopecten ventricosus lisms. Small individuals invest more energy in growth than (Sowerby, 1842), and (b) to determine the effect of age on in reproduction. Bayne and Newell (1983) found a higher the content of carbohydrates and protein in the gonad of metabolic rate in smaller than in larger individuals of these scallops. American Malacological Bulletin, Vol. 11(1) (1994):67-72 67 68 AMER. MALAC. BULL. 11(1) (1994) MATERIALS AND METHODS Two groups of Argopecten ventricosus were com- pared. One group (eight-month-old) was observed in November 1989, and a second (16-month-old) in May 1990. REPRODUCTIVE CONDITION Morphometric data were obtained from wet weights of the digestive gland, adductor muscle, mantle-gill and gonad. Indices for each organ were defined by the follow- ing formula: dry weight of organ Index = x 100 dry weight of soft parts Biochemical determinations included carbohydrate and protein levels from the digestive gland, adductor mus- cle, mantle-gills and gonad. Determination of carbohy- drates followed the technique presented by Barber and Blake (1981), using phenol in water and sulfuric acid, and reading the optical density at 490 nm (Lowry ef al., 1951; Mann and Gallager, 1985). Proteins were measured using the Bio-Rad Protein Assay (BIO RAD, Richmond, California), a dye-binding assay based on the differential color change of a dye in response to various concentrations of protein (Bradford, 1976). CULTURE Scallops were collected off Farallon Beach, Panama Bay, Panama, with a 1 m dredge, in 10 m depth. Animals were maintained in fiberglass tanks with running sea water in the marine laboratory of the University of Panama on Naos Island. Scallops (wrapped in wet towels, and held at 19°C) were transported by air to the College of Marine Studies in Lewes, Delaware, U.S.A., in September 1988. In Lewes, 40 scallops were conditioned and main- tained in a 200 | recirculating seawater tank at 19°C and 30 ppt salinity. A combined algal diet (50:50) of Jsochrysis galbana Parke (C-ISO) and Chaetoceros calcitrans Paulsen was provided daily. In February 1989, 20 scallops were induced to spawn by thermal stimulation (26°C) and frequent water changes. Larvae were reared in a 200 | tank at 24°C, and a combined algal diet was provided daily. Eight-Month-Old Scallops In October 1989, 90 eight-month-old scallops from the group spawned in February 1989 were separated into three groups (30 scallops each), and their sizes and weights were determined. Initial condition was analyzed for ten individuals by determining wet and dry weights of the digestive gland, adductor muscle, mantle-gills and gonad, and a biochemical analysis was performed for each organ. Thirty of the 90 scallops were placed in each of three aquaria (40 | each) containing aerated filtered sea- water. A combination of monocultures (50:50) of Isochrysis galbana (T-ISO) and Chaetoceros gracilis at high concen- trations was added daily to the three aquaria (this phyto- plankton concentration was observed in the field in Panama during the dry season). In addition, as a control, several scallops were placed in normal densities of phytoplankton observed during the dry season, and the amount of phyto- plankton cleared by the scallops was determined. This experiment was performed every other week, to determine if sufficient food was being provided to the experimental scallops. Quantities of phytoplankton consumed by experi- mental scallops were determined daily by direct count with a hemocytometer and weekly with a Coulter Counter. The correction factor for a plateau in normal rate of growth of phytoplankton was determined by holding an aquarium with phytoplankton for only 24 h. Three aquaria were maintained at 20°C, the temper- ature observed in the field in Panama during the dry season. Water temperature was maintained in a circulating bath (Forma Scientific). Salinity was measured daily using a refractometer; pH of the water was recorded daily. In each of the three aquaria, ten scallops were sacri- ficed at 40 and again at 58 days. The condition of each scallop was determined by the wet and dry weights of the digestive diverticulum, adductor muscle, mantle-gills and gonad, and a biochemical analysis of each organ was per- formed. Morphometric measurements of scallops and mean levels of biochemical compounds were compared using ANOVA. 16-Month-Old Scallops In May 1990, sizes and weights were determined for 150 Argopecten ventricosus, from the group spawned in the laboratory in February 1989. Initial condition was ana- lyzed for ten individuals by determining dry weights of digestive gland, adductor muscle, mantle-gills and gonad. A biochemical analysis was also completed on ten scallops. At this time scallops ranged from 18 to 24 mm in size. Sixty-nine of these scallops were placed in an aquarium containing filtered aerated seawater. These scal- lops were treated in a similar manner to those in the previ- ous experiment. Fifteen scallops from the experiment were sacri- ficed at 36 and again at 52 days. Wet and dry weights of the digestive gland, adductor muscle, mantle-gills and gonad of control and experimental scallops were determined. A bio- chemical analysis of these organs was also performed. Morphometric measurements of scallops and mean levels of biochemical compounds were statistically analyzed and compared with the group of eight-month-old scallops utiliz- ing an ANOVA. VILLALAZ: MORPHOMETRIC AND BIOCHEMICAL CHANGES IN ARGOPECTEN 69 RESULTS MORPHOMETRIC ANALYSIS Eight- and 16-month-old scallops increased continu- ously and significantly in shell height during the experi- mental period (Fig. 1). Total weight increased significantly in both age classes to day 52, but declined in 16-month-old scallops after day 52 (Fig. 2). Gonadal dry weight increased significantly in eight- and 16-month-old scallops during the experiment (Fig. 3). Gonadal index declined during the first 40 days of the experiment, then a significant increase occurred in both age classes (Fig. 4). Digestive gland dry weight was stable in eight- month-old scallops until day 40 of the experiment, then declined by day 58 (Fig. 5). In 16-month-old scallops, digestive gland dry weight declined until day 36, then increased significantly by day 52 (Fig. 5). Adductor muscle and mantle-gill dry weights of eight-month-old scallops increased until day 40 (Figs. 6 and 8). However, by day 58, weight of these organs either decreased or remained stable. Adductor muscle and mantle-gill dry weights of 16- month-old scallops increased significantly during the exper- iment (Figs. 6 and 8). However, by day 58 the adductor Table 1. Percent glucose of the digestive gland, adductor muscle, mantle- gill and gonad of Argopecten ventricosus. Age Day Gland Muscle Mantle Gonad 8 8 34.62 22.93 12.79 -- 8 40 8.84 2.99 4.32 4.60 8 58 12.22 -- 2.55 12.91 16 1 27.57 24.04 7.73 7.81 16 36 15.08 8.24 8.01 7.78 16 52 26.12 21.31 8.31 2.39 Table 2. Mean glucose content (standard error in parentheses) of the digestive gland, adductor muscle, mantle-gill and gonad of Argopecten ventricosus. Age Day Gland Muscle Mantle Gonad 8 8 5.44 4.16 1.38 -- (1.15) (2.21) (0.88) 8 40 1.54 1.72 0.94 0.03 (0.85) (1.94) (0.84) 8 58 1.51 -- 0.34 0.04 (0.30) (0.12) 16 1 5.76 10.33 1.75 0.24 (2.32) (2.77) (0.59) 16 36 2.36 3.42 1.57 0.48 (2.57) (3.40) (1.75) 16 52 8.04 14.64 3.62 0.16 (5.72) (8.77) (2.66) muscle index had declined (Fig. 7). BIOCHEMICAL ANALYSIS Percent and content of glucose in the gonad were similar in the two age classes of scallops (Tables | and 2). Percent of gonadal protein was not statistically different (Table 3), although protein content was significantly higher in 16-month than in eight-month-old scallops (Table 4). Glucose percent and content in the digestive gland and glucose percent in the mantle-gills were not significant- ly different between both age groups of scallops (Tables 1 and 2). Content of glucose in the mantle-gills was signifi- cantly higher in 16-month than in eight-month-old scallops. Percent and content of protein in the digestive gland were not significantly different between both groups of scallops. However, percent and content of protein in the mantle-gills were significantly higher in eight-month than in 16-month-old scallops (Tables 3 and 4). Percent and content of glucose in muscle were not significantly different between groups (Tables | and 2), but percent and content of protein in this organ were higher in eight-month than in 16-month-old scallops (Tables 3 and 4). Table 3. Percent protein of the digestive gland, adductor muscle, mantle- gills and gonad of Argopecten ventricosus. Age Day Gland Muscle Mantle Gonad 8 8 7.91 22.31 19.87 9.48 8 40 12.99 45.52 32.82 7.58 8 58 9.28 -- 9.69 15.38 16 1 10.87 15.87 11.49 14.76 16 36 9.78 24.50 8.97 7.58 16 52 14.9] 22.34 12.41 7.04 Table 4. Mean protein content (standard error in parentheses) of the digestive gland, adductor muscle, mantle-gills and gonad of Argopecten ventricosus. Age Day Gland Muscle Mantle Gonad 8 8 1.36 4.22 2.87 0.05 (1.11) (2.30) (1.80) 8 40 2.11 20.42 7.72 0.13 (0.82) (15.06) (7.45) 8 58 1.35 —- 1.67 0.10 (0.77) (0.71) 16 1 2.21 7.07 2.88 0.46 (0.59) (4.30) (1.68) 16 36 1.49 10.45 1.65 0.47 (0.60) (6.44) (0.40) 16 52 4.17 17.87 4.93 0.47 (2.29) (12.55) (1.82) DAYS 7 XdCNI TVGVNOD (3) LHDIGM LOM TVLOL dO NI AMER. MALAC. BULL. 11(1) (1994) 16-month : a ae ee 1 || 1 —— 0 10 20 30 40 50° 60 DAYS a v ak Es seen eee ees eee ee ee | 30 25 rf fo) oO ro) oO fo) re) SS a N N - (ww) LHSIGH TISHS (301) GVNOD JO LHDIGM AUG AO NI 70 DAYS DAYS 4.5 3.50 (91) GNVID JO LHDIGM AUC JO NI ee! DH — | To) am 1 fo) Te) ro) Te) + a) De) N (3u1) AIOSAW dO LHDISM AUC dO NT *—_@—4 a Ww) A wo oO wo fo) oS o p= oi) Dey ”m N N 10 20 30 40 50 60 70 0) 70 60 DAYS DAYS VILLALAZ: MORPHOMETRIC AND BIOCHEMICAL CHANGES IN ARGOPECTEN oh 60 T T =e T T T 40 - 5 a ADDUCTOR MUSCLE INDEX x f 30 L 1 1 a n ! ! 0) 10 20 30 40 50 60 70 DAYS 4.0 T T T iT T T T WG Oo T FH ! t-<$H1 KH eae LN OF DRY WEIGHT OF MANTLE-—GILLS (mg) 0) 10 20 30 40 50 60 70 DAYS Figs. 1-8. Comparison of morphometric parameters over time in eight- and 16-month-old Argopecten ventricosus in the laboratory. Fig. 1. Shell height. Fig. 2. Total weight (natural logarithm). Fig. 3. Gonadal dry weight (natural logarithm). Fig. 4. Gonadal index. Fig. 5. Digestive gland dry weight (natural loga- rithm). Fig. 6. Adductor muscle dry weight (natural logarithm). Fig. 7. Adductor muscle index. Fig. 8. Mantle-gill dry weight (natural logarithm). All sym- bols as in Fig. 1. DISCUSSION According to Félix-Pico (1991), the size of Argopecten ventricosus at reproductive maturity is uncer- tain. However, Tripp-Quezada (1985) reported initial spawning of A. ventricosus at an age of six months. Blake (1972) noted that a minmum age had to be reached before cytoplasmatic growth of oocytes could be initiated. This implies that eight-month-old scallops in my experiment had reached a sufficient age to start gametogenesis. The gonadal index of eight- and 16-month-old scal- lops declined during the first 40 days of the experiment. Also, a significant decline in carbohydrate content was observed in the digestive gland and adductor muscle. This could suggest energy was used either for somatic tissue and/or reproduction. I assume some of these carbohydrates were used for reproduction, but were not transferred direct- ly to the gonad before day 40. After day 40, these stored carbohydrates were transformed into gonadal lipids. Several authors have suggested a conversion of pre-stored glycogen into lipids (Gabbott, 1975; Barber and Blake, 1985b). There were increases in dry weight and protein con- tent of adductor muscle during the first 40 days of the experiment, then declines were observed around day 55 in dry weight and protein content of the adductor muscle. Simultaneously, around day 55, eight- and 16-month-old scallops showed noticeable increases in gonadal dry weight and protein content. Barber and Blake (1985a) considered the adductor muscle in Argopecten irradians a long-term storage organ to support vitellogenesis and spawning. I suggest the adductor muscle of A. ventricosus stores protein during the first 40 days, then protein catabolism in the adductor muscle supports reproductive activity of the gonad. I observed a similar pattern of storing proteins in the adductor muscle in both eight- and 16-month-old A. ventri- cosus. This similarity in reproductive strategy in both age classes differs from the condition observed in A. purpura- tus by Martinez (1991), in which the seasonal cycle of adductor muscle protein was significantly different among different size classes. I conclude that the energy metabo- lism in A. ventricosus is similar in eight- and 16-month-old individuals with respect to gametogenesis. The protein content in the mantle was higher in eight- than in 16-month-old scallops, an indentical situation reported by Martinez (1991) for different age classes of Argopecten purpuratus. She suggested this organ stored energy to be used for growth rather than reproduction. This suggests that mantle-gill protein can provide better support for somatic growth in eight- than in 16-month-old tropical scallops. ACNOWLEDGMENTS This work was partially supported by the International Foundation for Science (IFS), and the Smithsonian Tropical Research Institute. I am grateful to my advisors, Dr. Melbourne Carriker, Dr. Kent Price, Dr. Charles Epifanio, Mr. Michael Castgana, and Dr. Bruce Barber. (p: AMER. MALAC. BULL. 11(1) (1994) LITERATURE CITED Ansell, A. 1974. Seasonal changes in biochemical composition of the bivalve Chlamys septemradiata from the Clyde Sea area. Marine Biology 25:85-99. Barber, B. and N. Blake. 1981. Energy storage and utilization in relation to gametogenesis in Argopecten irradians concentricus (Say). Journal of Experimental Marine Biology and Ecology 52:121-134. Barber, B. and N. Blake. 1985a. Intra-organ biochemical transformations associated with oogenesis in the bay scallop, Argopecten irradians concentricus (Say), as indicated by 14C incorporation. Biological Bulletin 168:39-49. Barber, B. and N. Blake. 1985b. Substrate catabolism related to reproduc- tion in the bay scallop, Argopecten irradians concentricus, as determined by O/N and RQ physiological indexes. Marine Biology 87:13-18. Barber, B. and N. Blake. 1991. Reproductive physiology. Jn: Scallops: Biology, Ecology and Aquaculture, Vol. 21. S. Shumway, ed. 1095 pp. Elsevier, Amsterdam. Bayne, B. L. and R. C. Newell. 1983. Physiological energetics of marine molluscs. In: The Mollusca, Vol. 4, Part 1. A. S.M. Saleuddin and K. M. Wilbur, eds. pp. 407-515. Academic Press, New York. Blake, N. 1972. Environmental regulation of neurosecretion and repro- ductive activity in the bay scallop, Aequipecten irradians (Lamarck). Doctoral Thesis, University of Rhode Island, Kingston. 161 pp. Bricelj, V. M., J. Epp, and R. E. Malouf. 1987. Intraspecific variation in reproductive and somatic growth of bay scallops Argopecten irra- dians. Marine Ecology - Progress Series 36:123-137. Bradford, M. M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein dye binding. Analytical Biochemistry 72(1/2):248-254. Epp, J., V. M. Bricelj, and R. E. Malouf. 1988. Seasonal partitioning and utilization of energy reserves in two age classes of the bay scallop Argopecten irradians irradians Lamarck. Journal of Experimental Marine Biology and Ecology 121:113-136. Félix-Pico, E. 1991. México. In: Scallops: Biology, Ecology and Aqua- culture, Vol. 21. S. Shumway, ed. 1095 pp. Elsevier, Amsterdam. Gabbott, P. A. 1975. Storage cycles in marine bivalve molluscs: a hypoth- esis concerning the relationship between glycogen metabolism and gametogenesis. Jn: Proceedings of the Ninth European Marine Biology Symposium. H. Barnes, ed. pp. 191-211. Aberdeen University Press. Giese, A. C. 1959. Comparative physiology: annual reproductive cycles of marine invertebrates. Annual Reviews in Physiology 21:547-576. Lowe, D. M., M. N. Moore, and B. L. Bayne. 1982. Aspects of gametoge- nesis in the marine mussel Mytilus edulis L. Journal of the Marine Biological Association of the United Kingdom 62:133-145. Lowry, O. H., N. J. Rosenbrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent, Journal of Biological Chemistry 193:265-275. Mann, R. and S. Gallager. 1985. Physiological and biochemical energetics of larvae of Teredo navalis L. and Bankia gouldi (Bartsch) (Bivalvia: Teredinidae). Journal of Experimental Marine Biology and Ecology 85(3):211-228. Martinez, G. 1991. Seasonal variation in biochemical composition of three size classes of the Chilean scallop Argopecten purpurpatus Lamarck, 1819. The Veliger 34:335-343. Sastry, A. N. 1975. Physiology and ecology of reproduction in marine invertebrates. In: Physiological Ecology of Estuarine Organisms. F. J. Vernberg, ed. pp. 279-299. University of South Carolina Press, Columbia. Sundet, J. H. and O. Vahl. 1981. Seasonal changes in dry weight and bio- chemical composition of the tissues of sexually immature Iceland scallops, Chlamys islandica. Journal of the Marine Biological Association of the United Kingdom 64:411-416. Taylor, A. C.and T. J. Venn. 1979. Seasonal variation in weight and bio- chemical composition of the tissues of the queen scallop, Chlamys opercularis, from the Clyde Sea area. Journal of the Marine Biological Association of the United Kingdom 59:605-621. Thompson, R. J. and B. A. MacDonald. 1991. Physiological integrations and energy partitioning. Jn: Scallops: Biology, Ecology and Aquaculture, Vol. 21. S. Shumway, ed. 1095 pp. Elsevier, Amsterdam. Tripp-Quezada, A. 1985. Explotacion y cultivo de la almeja catarina Argopecten circularis en Baja California Sur. Tesis de Maestria, CICIMAR-IPN, La Paz, B. C. S., México. Date of manuscript acceptance: 13 June 1994 An apparent hybrid zone between freshwater gastropod species Elimia livescens and E. virginica (Gastropoda: Pleuroceridae) Thomas S. Bianchi!*, George M. Davis and David Strayer! \Institute of Ecosystem Studies, Millbrook, New York, 12545, U.S.A. 2Department of Malacology, Academy of Natural Sciences of Philadelphia, Philadelphia, Pennsylvania 19103, U.S.A. Abstract. The purpose of this study was to determine evidence for genetic hybridization between sympatric populations of the pleurocerids Elimia livescens (Menke, 1830) and E. virginica (Say, 1817) in New York state, U.S.A. E. livescens, an Interior Basin species, was once totally isolated from E. virginica, an Atlantic Slope species, by the Alleghenian Divide during glaciation. Populations of Elimia livescens and E. virginica from three different regions across New York were compared on the basis of allozyme genetics involving 22 loci and 39 alleles. E. livescens and E. virginica collected from regions of allopatry, had alternatively fixed alleles at eight of the 22 loci ana- lyzed. Genetic differentiation was found at 11 of the 22 loci. Nei’s genetic distance (D) was 0.322 between the two species in the allopatric regions. There were no significant genetic differences between the populations of each species within a particular region. However, there were significantly higher levels of heterozygosity for both species in a narrow region of sympatry than in the allopatric regions. Despite substantial genetic differences between these two species, seven of the 22 loci indicate evidence for hybridization and introgression occurring in the zone of contact. Because these species existed in allopatry at the peak of glaciation (and prior to opening of the Erie Canal) we feel these differences cannot be explained by differentiation along a continuous gradient of populations. Freshwater gastropods of the genus Elimia [= (Ortmann, 1913; Strayer, 1987). Contact between these Goniobasis, according to Burch (1989)] have been of con- species has now been established as a consequence of a dis- siderable interest in recent years to evolutionary biologists persal route created by the opening of the Erie Canal in because of their extensive endemic radiation in the south- 1825; it is possible but not likely that such contact was ini- eastern United States (Chambers, 1978, 1980, 1982; Dillon tiated ca. 13,000 years B.P. with the Mohawk River outlet and Davis, 1980; Dillon, 1984). The low levels of heterozy- of the Great Lakes (Strayer, 1987). Immigration routes such gosity and gene flow among these populations are believed as this can act to reduce the amount of differentiation to be primarily caused by rare events of dispersal by Elimia among conspecific populations because of enhanced gene between drainage basins (Chambers, 1982; Dillon, 1984). flow or even allow congeners to hybridize. As examples, it The genetic divergence among these isolated populations has been demonstrated that the opening of the Erie Canal also appears to be related to gene flow restriction created has resulted in hybridization within both bivalve genera by physical barriers (e.g. rivers, dams) (Dillon, 1984). In Anodonta and Lampsilis (Bivalvia: Unionidae) (Kat, 1986). fact, the amount of divergence among these populations of The purpose of this study was to determine the Elimia is greater than that reported for Drosophila by Ayala genetic variability among northeastern populations of et al. (1974). Despite the extensive amount of work on the Elimia livescens and E. virginica in regions of sympatry radiation of Elimia spp. in the southeastern United States, and allopatry. We will also present evidence for genetic virtually nothing is known about the genetics of Elimia hybridization between these two species in the zone of con- populations in other regions. tact in central New York state. In the northeastern United States, populations of Elimia livescens (Menke, 1830), an Interior Basin species, were totally isolated from E. virginica (Say, 1817), an MATERIALS AND METHODS Atlantic Slope species, by the Alleghenian Divide during : : Animals and Localities. Seven populations of Elimia Wisconsinan and perhaps earlier episodes of glaciation livescens and E. virginica were sampled in three different ePrescnn Address: Dept of Ecology, bvolution/and Oreaniemal regions of New York state from Buffalo southeast to Ulster Biology, Tulane University, New Orleans, Louisiana 70118, County (Fig. 1). Shell morphology is very distinct between U.S.A. E. livescens and E. virginica (Fig. 2). Identifications at the American Malacological Bulletin, Vol. 11(1) (1994):73-78 73 74 AMER. MALAC. BULL. 11(1) (1994) LAKE ONTARIO SALINA’ HAITI ISLAND Fig. 1. Map of New York state, showing drainage systems and locations (solid circles) where populations of Elimia virginica and E. livescens were sampled. Dashed line indicates the drainage divide between St. Lawrence and Atlantic Basins. species level were made on the basis of shell phenotype. Region | contains allopatric populations of E. livescens, region 3 contains allopatric populations of E. virginica, and region 2 contains both species in sympatry (Fig. 1). In the region of sympatry there was a blurring of shell phenotypes between the species and the distinction between the mor- photypes was less clear. Snails with EF. virginica-type shells dominated in Haiti Island, and E. livescens-type shells dominated in Salina. In the Mudlock site there was a greater phenotypic mixture and we sorted the snails into two groups: 1) those most resembling the EF. livescens-type with some shells appearing to be pure E. livescens; 2) those most resembling the FE. virginica-type with some shells appearing to be pure E. virginica. As an outgroup for com- parative purposes an additional FE. virginica population was sampled at an outer geographic region on the Potomac River in Harpers Ferry, West Virginia (approximately 500 km from the Ulster location). Electrophoresis. Snails were shipped live to the Academy of Natural Sciences of Philadelphia (ANSP), where they were frozen at -70°C in tissue buffer. Specimens to be ana- lyzed using electrophoresis were hand-ground in a small amount (5-10 1) of deionized water. The shells were not removed, however, they were scrubbed clean to remove any additional organic material on the surface of the shell before grinding. Eleven loci served to demonstrate genetic differenti- ation among populations: AAT-1, APH, EST, GP1, ISDH, LAP, NADD, OCT, 6PGD, PGM, and XDH. Invariant loci were ACPH-1, ACPH-2, AO, FUM, G6PD, MDH-1, MPI, SDH, SOD-1, and SOD-2. There were two loci with popu- lations fixed for alternative alleles (LAP, XDH) and nine loci exhibiting polymorphism (Table 1). Horizontal starch-gel electrophoresis was carried out using the general methods of Ayala et al. (1973) as modified by Dillon and Davis (1980) and Davis et al. (1981). Six different buffer systems were used: 1) tris-cit- rate (TC), pH 6; 2) TC, pH 8; 3) tris-edta-borate (TEB), pH 8; 4) TEB, pH 9.1; 5) TEB, pH 9.1 gel buffer and TEB, pH 8 tray buffer; and 6) TC, pH 8.75 gel buffer and sodi- um-borate (Poulik), pH 7.6 tray buffer. Starch gels were prepared using 34 g of Electrostarch and 250 ml of gel buffer. Run times for each of the gels and buffers are as reported in Davis et al. (1988). Wicks of No. 3 Whatman filter paper were saturated with supernatant from the homogenized tissue from each individual. They were blotted and applied, one wick from each individual, to be run concurrently. The gels were sliced into three layers for staining. Agar overlays were employed for all enzyme assays except AAT and LAP for which aqueous solutions were used. Standard recipes for all Fig. 2. Representative shells of Elimia virginica (A) and E. livescens (B) collected from allopatric localities (Ulster and Buffalo Creek, respective- ly). BIANCHI ET AL.: APPARENT HYBRID ZONE IN ELIMIA ® Table 1. Allele frequencies from 11 loci (where polymorphism is exhibit- ed at least once) of seven populations of Elimia collected along the Erie Canal. Species names are based on shell phenotype. H* = presumed hybridization and introgression. Hybrid group (1) represents E. livescens shell phenotype with a trend toward E. virginica; hybrid ( 2) represents E. virginica shell phenotype with a trend towards E. livescens. (*, diagnostic). Locus E. virginica E.livescens_ E. virginica E. livescens Ulster H*(1) H*(2) Buffalo Mudlock Mudlock _ ATT-1: 100 1.000 0.097 0.711 - - 97 - 0.903 0.289 1.000 APH: 100 1.000 1.000 1.000 0.983 98 - - - 0.017 EST: 100 0.979 1.000 1.000 1.000 98 0.021 - = = GPI 100 1.000 0.016 0.974 - * 97 - 0.984 0.025 1.000 ISDH: 100 1.000 - 0.211 0.586 * 98 - 0.042 0.789 0.362 98 - 0.958 - 0.002 LAP: 100 1.000 - 1.000 - x 97 - 1.000 ~ 1.000 NADD: 100 1.000 0.210 0.763 ~ * 105 - 0.790 0.237 1.000 OCT 106 - - 0.263 - = 103 - - 0.079 ~ 100 1.000 - 0.632 ~ 96 - 0.548 - 0.817 91 - 0.452 0.026 0.183 6PGD: 100 1.000 - 0.250 - * 103 - 1.000 0.750 1.000 PGM: 103 - 0.974 0.026 1.000 * 100 1.000 - 0.974 - 97 - 0.026 - - 94 - - - - XDH: 100 1.000 - 1.000 - ig 98 - 1.000 - 1.000 systems are in Poulik (1957), Brewer (1970), and Shaw and Prasad (1970). The recipe for NADD was developed in the ANSP laboratory in 1982 by Robin Hadlock. Computations on genetic data were made using a version of NY-SYS (Rohlf et al., 1972) and BIOSYS-1 (Swofford and Selander, 1981). The standard Nei’s distance (1978) and the Cavalli-Sforza and Edwards (1967) Arc dis- tance coefficients were calculated and phenograms prepared using the numerical taxonomy program NT-SYS and UPGMA option. Voucher specimens (two shells of each) from some populations were cataloged into the ANSP collections: Elimia virginica from Harpers Ferry (ANSP 373457); E. virginica from Ulster County (ANSP 373458); E. livescens combined from Cazenovia and Buffalo Creeks (however, only snails from Buffalo Creek were used for electrophore- sis). As small samples were taken from Ulster, Salina, and Harpers Ferry, West Virginia, frequency data are not listed here. Nei’s and Arc Genetic Distances. We include data on Nei’s genetic distance (D) because it is a standard of long use and thus allows comparisons with the literature. The relative value of Nei’s, Rogers’, and the Cavalli-Sforza and Edwards’ Arc distances have been discussed (Davis et al., 1988). In short, given closely related populations and species, Nei’s D will compact values at the lower end of the distance scale. Nei’s D is a squared distance value that will rise linearly with time, as more genetic divergence accumu- lates. There is no upper limit because Nei’s D is not metric. The Arc distance is metric and provides a better under- standing of relationships from zero similarity (no shared alleles) to 100% identity. RESULTS We obtained 22 loci with 39 alleles (Tables 1, 2). There were 11 polymorphic loci of which nine were diag- nostic (Table 1). The outgroup for this study was Elimia virginica from West Virginia (N = 4). It was monomorphic at all loci and matched the gene frequencies we observed in the Ulster population of this species from the Hudson River. The Ulster population was polymorphic at only one locus. In considering both populations, a mean of 24 individuals was examined from each locus. E. livescens from Buffalo Creek was monomorphic at seven (78%) of the nine diag- nostic loci. Table 2. Genetic variability at 22 loci in all populations (standard error in parentheses). (A, allopatric; S, sympatric). Population Mean sample Mean no. of Poly- Mean size per alleles morphic heterozygosity locus per locus loci (%) (direct count) Elimia virginica, 4.0 1.0 ) 0 West Virginia (A) (0.0) (0.0) E. virginica, 20.4 1.0 0.5 0.002 Ulster (A) (1.0) (0.0) E. livescens, 25.8 1.2 13.6 0.014 Buffalo Creek (1.4) (0.1) (A) E. virginica, 6.0 1.0 4.5 0.015 Haiti Isl. (S) (0.0) (0.0) E. virginica, 18.9 1.4 31.8 0.044 Mudlock (S) (0.0) (0.2) E. livescens, 5.0 1.2 18.2 0.035 Salina (S) (0.0) (0.1) E. livescens, 29.0 1.3 213 0.050 Mudlock (S) (1.0) (0.1) 76 AMER. MALAC. BULL. 11(1) (1994) Nei’s D was 0.332 between the outgroup Elimia vir- ginica and Buffalo Creek E. livescens; Arc distance was 0.577. The two distance values between the two popula- tions of pure E. virginica were 0 and 0.020, respectively. The numbers are adequate for determining genetic distance. As discussed in Davis (1983), genetic distances are not much affected by sample size. To gain a more refined esti- mate of genetic distance between two species than that based on only two individuals necessitates the use of 30 or more individuals (Sarich, 1977). A sample of two individu- als will yield a heterozygosity (H) within 2.5% of that cal- culated from a much larger sample size; there is less than 0.1 difference in D (Gormon and Renzi, 1979). Emphasis in this report is placed on examining what is occurring at Mudlock; it is clear that hybridization and introgression are evident. There are no F1 generation snails involved as there is fixation for alternative alleles at the LAP locus. However, evidence for introgression is found at seven (78%) of the diagnostic loci. Most striking are data for the individuals considered to be hybrids that graded to pure Elimia virginica. At the 6PGD locus there was a fre- quency of 0.750 for the FE. livescens alleles; at the NADD locus this was 0.237, and 0.289 at the ATT locus. A signifi- cant amount of E. virginica alleles was found in individuals that were suspected of being hybrids with shell characters grading to pure E. livescens only at the AAT-1 and NADD loci. As shown in Table 2, the lowest heterozygosity was in the West Virginia population of Elimia virginica (H = 0) followed by the Ulster populations of EF. virginica (H = 0.002); the control EF. livescens population had an H of 0.014. Hybridization at the Mudlock groupings was indicat- ed by H values of over three times that found for control E. livescens (i.e 0.044 or 0.050) (Table 2). Deviation from Hardy-Weinberg expectations was encountered at one locus (ISDH) of the Elimia livescens population from Buffalo Creek. There was a pronounced deficiency of the 100/98 heterozygotes. While the number of snails used from Mudlock classified as E. virginica or hybrids had a mean of 18.9 per locus, there was a clear trend for heterozygote deficiency at AAT-1, NADD, OCT and 6PGD. Increasing sample size would probably not eliminate these deficiencies. DISCUSSION We acknowledge the two greatest weaknesses of the study: 1) on the basis of shells there was no exact method to differentiate populations of hybrids from “pure” parental types; and 2) there were insufficient individuals to preserve phenotypic classes (i.e. it was impossible to remove ani- mals for extraction of enzymes without destroying the shells) to voucher the choices we made in selecting the hybrid groupings at Mudlock. GENETIC DISTANCE AND HETEROZYGOSITY This study brings to 11 the number of Elimia species studied in terms of allozyme population genetics. Chambers (1980) studied E. vanhyningiana (Goodrich, 1921), E. floridensis (Reeve, 1860), E. athearni (Clench and Turner, 1956), E. albanyensis (Lea, 1864), E. curvi- costata (Reeve, 1861), and E. dickinsoni (Clench and Turner, 1956) from Florida. Dillon and Davis (1980) stud- ied E. proxima (Say, 1825), E. simplex (Say, 1825), and E. semicarinata (Say, 1829) from Virginia and North Carolina. There are some interesting patterns that surface when analyzing mean genetic distances and mean heterozy- gosities between Elimia species in data provided by three studies (Chambers, 1980; Dillon and Davis, 1980; Dillon, 1988): 1) species pairs have Nei’s distances averaging from 0.33-0.59 (standard deviations range from 0.33-0.99); 2) conspecific populations can have relatively large mean dif- ferences (0.12-0.15; Chambers, 1980; Dillon and Davis, 1980), however, those observed in this study were quite low; and 3) mean heterozygosity is very low (0.01-0.08). As Dillon and Davis (1980) pointed out, there is consider- able fixation of alternative alleles both between and among conspecifics. The implication for conspecifics is that gene flow between populations of the same species or species complex is probably very low, even within the same drainage system. Because of these general observations and the very low levels of heterozygosity among populations of all northern species thus studied, the situation along the Erie Canal stands out prominently. Although this study is preliminary, the collective evidence indicates the first case of hybridization and introgression within the Pleuroceridae. HYBRID ZONES Very few examples of hybrid zones between mol- luscan species have been reported. The greatest in-depth analyses involve marine bivalves of the genus Mytilus [see extensive review and study, Vainola and Huilsom (1991)]. Other examples involve land snail genera such as Cepaea and Cerion [reviewed by Barton and Hewitt (1985)], union- id clams of the genera Anodonta and Lampsilis (Kat, 1986), and Mercenaria (Dillon and Manzi, 1989; Dillon, 1992). The zone of Elimia species overlap in New York is, as thus far known, relatively narrow, < 40 km. The width of many hybrid zones is typically considerably larger than the dispersal for the individual organism, implying that neutral BIANCHI ET AL.: APPARENT HYBRID ZONE IN ELIMIA aT diffusion is occurring for genes that control the traits being observed (Hewitt, 1988). The dispersal rate for these Elimia species in this region is not known. As is characteristic of hybrid zones, polymorphism increases (increased mean het- erozygosity), hybrizymes increase, and heterozygote defi- ciency is frequently observed. In this study we feel there is evidence for introgression of E. virginica genes into the western-most population investigated, i.e. Buffalo Creek E. livescens, considering the seven diagnostic loci that show no overlap between allopatric populations of E. livescens and E. virginica. Snails from Salina (N = 5) matched gene frequen- cies for Elimia livescens at most loci but showed some allelic overlap with E. virginica at AAT-1:100 and NADD:100 (E. livescens was fixed at allele 97 at those two loci) indicating some introgression. Results at the ISDH locus also suggest introgression as ISDH:96 had a high fre- quency in the Salina and Mudlock E. livescens; this allele was not found in E. virginica. This allele was at low fre- quency in Buffalo E. livescens, a population that also had ISDH:98, not seen in West Virginia or Ulster E. virginica but seen in the populations at Mudlock and Salina. Several interpretations are possible. The ISDH:100 allele could be a shared primitive allele; the ISDH:96 allele could have recently evolved in mid-state E. livescens or as a hybrizyme. Overall, given the very low heterozygosity found in general in species of Elimia, and the mixture of alleles at the ISDH locus for Salina, Mudlock, and Buffalo snails, the data suggest that introgression is occurring. Vainola and Huilsom (1991) very effectively showed the occurrence of a hybrid zone in Mytilus popula- tions, however, their diagnostic loci were not nearly as dis- tinct as the allopatric populations when compared to this study. Although we were not able to show statistical signifi- cance of heterozygote deficiency of snails in the contact zone (Mudlock station), the magnitude of the difference is considerably more than that seen among some hybrid popu- lations of Mytilus. The appearance of a novel polymorphic locus (PGM) at Haiti Island, not found in either of the allopatric populations of E. livescens or E. virginica may seem peculiar. However, this pattern, although not well understood, has been observed in other hybrid zones; in particular Woodruff and Gould (1980) observed this in Cerion populations. The age of this zone is probably very recent, with contact of Interior Basin Elimia livescens with northeastern slope E. virginica made possible by the opening of the Erie Canal in 1825. The detection of this hybrid zone should be of considerable interest to students of speciation, especially given the genetic structures of Elimia populations. Local populations within a drainage system are often highly dif- ferentiated genetically; gene flow is evidently slow. In a transplantation experiment mixing conspecific populations of E. proxima with fixed different alleles at one or two loci, introduced genes spread at 15-20 m upstream and 5-10 downstream per year (Dillon, 1988). In some cases the progress of hybridization, which can result in fusion, extinction, or speciation, can actually occur at a rate of one gene at a time (Bert and Harrison, 1988). ACKNOWLEDGMENTS We thank Jo Ann Bianchi and Joseph O’Brien for their invalu- able assistance in the field. This is a contribution from the Institute of Ecosystem Studies, and the Molecular Genetics Laboratory of the Department of Malacology, The Academy of Natural Sciences of Philadelphia. We acknowledge Caryl Hesterman who did electrophoretic runs and scoring of gels. This work was supported in part by the Theodore Roosevelt Fund, Museum of Natural History, New York, to T. S. Bianchi, and by an NSF grant BSR-85002796 to G. M. Davis. We very much appreciate Rob Dillon’s patience in reviewing the manuscript. LITERATURE CITED Ayala, F., D. Hedgecock, G. Zumwalt and J. Valentine. 1973. Genetic variation in Tridacna maxima, an ecological analog of some unsuc- cessful evolutionary lineages. Evolution 27:177-191. Ayala, F. J.. M. L. Tracey, D. Hedgecock, and R. C. Richmond. 1974. Genetic differentiation during the speciation process in Drosophila. Evolution 28:576-592. Barton, N. H. and G. M. Hewitt. 1985. Analysis of hybrid zones. Annual Review of Ecology and Systematics 16:113-148. Bert, T. M. and R. G. Harrison. 1988. Hybridization in western Atlantic stone crabs (genus Menippe): evolutionary history and ecological context influence species interactions. Evolution 42:528-544. Brewer, G. J. 1970. An Introduction to Isozyme Techniques. Academic Press, New York. 186 pp. Burch, J. B. 1989. North American Freshwater Snails. Malacological Publications, Hamburg, Michigan. 365 pp. Cavalli-Sforza, L. L. and A. W. F. Edwards. 1967. Phylogenetic analysis: models and estimation procedures. Evolution 21:550-570. Chambers, S. M. 1978. An electrophoretically detected sibling species of “Goniobasis floridensis.” Malacologia \7:157-162. Chambers, S. M. 1980. Genetic divergence between populations of Goniobasis (Pleuroceridae) occupying different drainage systems. Malacologia 20:63-81. Chambers, S. M. 1982. Chromosomal evidence for parallel evolution of shell sculpture pattern in Goniobasis. Evolution 36:1 13-120. Davis, G. M. 1983. Relative roles of molecular genetics, anatomy, mor- phometrics and ecology in assessing relationships among North American Unionidae (Bivalvia). In: Protein Polymorphism: Adaptive and Taxonomic Significance, G. S. Oxford and D. Rollinson, eds. pp. 193-222. Systematics Association Special Volume 24. Academic Press, London. Davis, G. M., V. Forbes and G. Lopez. 1988. Species status of northeast- ern American Hydrobia (Gastropoda: Prosobranchia): ecology, morphology and molecular genetics. Proceedings of the Academy of Natural Sciences of Philadelphia 140:191-246. Davis, G. M., W. H. Heard, S. L. H. Fuller and C. Hesterman. 1981. Molecular genetics and speciation in Elliptio and its relationship to other taxa in North American Unionidae (Bivalvia). Biological 78 AMER. MALAC. BULL. 11(1) (1994) Journal of the Linnean Society 15:131-150. Dillon, R. T. 1984. Geographic distance, environmental difference, and divergence between isolated populations. Systematic Zoology 33:69-82. Dillon, R. T. 1988. Evolution from transplants between genetically distinct populations of freshwater snails. Genetica 76:111-119. Dillon, R. T. 1992. Minimal hybridization between populations of the hard clams, Mercenaria mercenaria and Mercenaria campechiensis, co- occurring in South Carolina. Bulletin of Marine Science 50:411- 416. Dillon, R. T. and G. M. Davis. 1980. The Goniobasis of southern Virginia and northwestern North Carolina: genetic and shell morphometric relationships. Malacologia 20:83-98. Dillon, R. T. and J. J. Manzi. 1989. Genetics and shell morphology in a hybrid zone between the hard clams Mercenaria mercenaria and M. campechiensis. Marine Biology 100:217-222. Gorman, G. C. and J. Renzi. 1979. Genetic distance and heterozygosity estimates in electrophoretic studies: effects of sample size. Copeia 79:242-249. Hewitt, G. M. 1988. Hybrid zones - natural laboratories for evolutionary studies. Trends in Ecology and Evolution 3:158-167. Kat, P. W. 1986. Hybridization in a unionid faunal suture zone. Malacologia 27:107-125. Nei, M. 1978. Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics 89:583-590. Ortmann, A. E. 1913. The Alleghenian Divide, and its influence upon the freshwater fauna. Proceedings of the American Philosophical Society 52:287-390. Poulik, M. D. 1957. Starch gel electrophoresis in a discontinuous buffer system. Nature 180:1477-1479. Rohlf, F. J., J. Kishpaugh, and D. Kirk. 1972. NT-SYS: Numerical Taxonomy System of Multivariate Statistical Programs. Stony Brook, New York. Sarich, V. M. 1977. Rates, sample sizes and the neutrality hypothesis for electrophoresis in evolutionary studies. Nature 265:24-28. Shaw, C. and D. Prasad. 1970. Starch gel electrophoresis of enzymes - a compilation of recipes. Biochemical Genetics 4:297-320. Strayer, D. 1987. Ecology and zoogeography of the freshwater mollusks of the Hudson River basin. Malacological Review 20:1-68. Swofford, D. L. and R. Selander. 1981. B/OSYS-1. A computer program for the analysis of allelic variation in genetics. University of Illinois, Urbana. 60 pp. Vainola, R.and M. M. Huilsom. 1991. Genetic divergence and a hybrid zone between Baltic and North Sea Mytilus populations (Mytilidae; Mollusca). Biological Journal of the Linnean Society 43:127-148. Woodruff, D. S. and S. J. Gould. 1980. Geographic differentiation and speciation in Cerion - a preliminary discussion of patterns and processes. Biological Journal of the Linnean Society 14:389-416. Date of manuscript acceptance: 7 July 1993 Research Note Morphological differences between zebra and quagga mussel spermatozoa Dana R. Denson and Shiao Y. Wang* Department of Biological Sciences, University of Southern Mississippi, Hattiesburg, Mississippi 39406-5018 U.S.A. Abstract: Sperm morphology of the zebra mussel, Dreissena polymorpha (Pallas, 1771), and of the currently-unnamed quagga mussel are described from light and scanning electron microscopy. Differences in shape of the sperm head and acrosome are described as a potentially useful tool in distinguish- ing the two species. There are currently two species of dreissenid mus- sels that have invaded North America. The appearance of the first species, Dreissena polymorpha (Pallas, 1771), was described by Hebert et al. in 1989. A description of a sec- ond species, whose identity is not known but which is cur- rently called the quagga mussel, was made by May and Marsden in 1992**. In addition to genetic differences, the two species differ in shell morphology (Pathy and Mackie, 1993). The quagga mussel lacks the acute angle or carina between the ventral and dorsal surfaces of the zebra mussel shell. This difference in shell morphology can be difficult to discern in some mussels, especially to those who exam- ine mussels only occasionally. The difference in sperm morphology described in this note could be helpful in the positive distinction between male zebra and quagga mus- sels. As a part of an ongoing study of the reproductive cycle of zebra mussels, samples of quagga mussels were also examined to determine if there were differences in the reproductive patterns of the two species. Upon examination of fixed and stained soft tissues of mussels under light microscopy, it became apparent that there were clear and consistent differences in gross morphology of the sperma- tozoa. Examination of simple squash preparations of small amounts of viscera from live specimens revealed that dif- *Correspondence to: Shiao Y. Wang, Department of Biological Sciences, University of Southern Mississippi, Hattiesburg, MS 39406-5018, U.S.A. **The quagga mussel has since been identified as Dreissena bugonsis Andrusov, 1897, by G. Rosenberg and M. L. Ludyanskiy, and A. P. Spidle, J. E. Marsden and B. May in two publications to appear in vol- ume 51 of the Canadian Journal of Fisheries and Aquatic Sciences. ferentiation between males of the two species was possible without extensive sample preparation. To examine the mor- phological differences more closely, zebra and quagga mus- sel spermatozoa were examined by scanning electron microscopy. The mussels examined were collected on 22 July 1993 from the Black Rock Lock adjacent to the Niagara River in Buffalo, New York. Both zebra and quagga mus- sels were found at the same site. For electron microscopy, squashes of a small portion of the visceral mass from sever- al individuals of each species were made for quick determi- nation of sex. The remaining tissues were fixed for two hours at 7°C in 2.5% glutaraldehyde and 0.1 M cacodylic acid, pH 7.2. Tissues were then washed three times in 0.1 M cacodylate buffer followed by 2% osmium tetroxide solution for ten minutes each. The samples were dehydrat- > e.) Fig. 1. Comparison of the spermatozoan morphology of zebra mussel (A) and quagga mussel (B). Scale = | pm. American Malacological Bulletin, Vol. 11(1) (1994):79-81 79 80 AMER. MALAC. BULL. 11(1) (1994) ed stepwise in a series of solutions containing 50, 60, 75, 90 and 100% ethanol. Tissues were dried in a critical point dryer before being coated with gold under an argon atmos- phere. The prepared samples were examined using an Electroscan environmental scanning electron microscope. The sperm heads of zebra mussels are straight, short, and blunt, whereas those of the quagga mussel are curved, longer, and more pointed, somewhat reminiscent of a scaphopod shell (Fig. 1). The difference in morphology is apparent even in an unstained squash preparation under rel- atively low magnification (400X). Electron photomicro- graphs show the described differences clearly (Figs. 2 -3). Measurements using electron photomicrographs revealed that the mean length of Dreissena polymorpha sperm heads is 4.0 + 0.1 pm (N = 5), whereas that of quag- ga mussels is 4.7 + 0.2 pm (N = 7). Mean width of zebra mussel sperm at the widest point is 1.6 + 0.1 pm (N = 4), as compared to 1.3 + 0.1 pm (N = 4) in quagga mussels. The zebra mussel sperm head is more bulbous in appearance Fig. 2. Scanning electron photomicrographs of the spermatozoa of zebra mussels. Scale = 5 pm (A), | pm (B). Fig. 3. Scanning electron photomicrographs of the spermatozoa of quagga mussels. Scale = 5 pm (A), | pm (B). than that of the quagga mussel sperm, which has a more sharply pointed outline. Again, the most distinguishing fea- ture is the straight, short appearance of zebra mussel sperm in contrast to the curved, longer appearance of quagga mus- sel sperm. Differences are also apparent between the acro- somes of the two sperm types. In the zebra mussel, the acrosome is more bulbous and oval in shape than that of the quagga mussel. In the latter species, the acrosome is wide at the point of attachment to the rest of the sperm head, but tapers more sharply toward the tip. In both types, the pri- mary acrosome vesicle is visible as a small nib at the apex of the acrosome. This structure is functionally important, as its membrane covering disarticulates at fertilization to fuse with the plasma membrane of the ovum (Dan, 1970). Spermatozoa of both zebra and quagga mussels pos- sess elongated flagella, which are at least several times the length of the head portion. Flagella of the sperm of both species project from the broad base of the head, and this DENSON AND WANG: ZEBRA AND QUAGGA MUSSEL SPERMATOZOA 81 junction is surrounded by four rounded structures which contain mitochondria (Mackie, 1984). Sperm morphology has been used as a tool in the study of systematic zoology by a number of workers. Such investigations have included oligochaetes (Jamieson, 1984), ascidians (Franzen, 1992), and rodents (Roldan et al., 1992). In addition, similar studies of gastropod spermatozoa have been employed in the examination of congeneric limpets (Hodgson and Bernard, 1988; Jamieson ef al., 1991). With regard to mussels, if positive distinction between the two species of Dreissena based on shell mor- phology is not possible, the described differences in the appearance of the sperm could be helpful. Of course, the usefulness of this technique is limited to male mussels, and only to that part of the reproductive cycle during which sperm are present, generally between March and September. ACKNOWLEDGMENTS We are grateful to Gary L. Dye for collecting the mussels and Raymond W. Scheetz for expertise and guidance in sample preparation and electron microscopy. Funding for this study was provided by the U. S. Army Engineer Waterways Experiment Station, Vicksburg, Mississippi. LITERATURE CITED Dan, J. C. 1970. The acrosome process membrane. /n: Comparative Spermatology, B. Bacetti, ed. pp. 487-498. Academic Press, New York, New York. Franzen, A. 1992. Spermatozoan ultrastructure and spermatogenesis in aplousobranch ascidians, with some phylogenetic considerations. Marine Biology 113:77-87. Hebert, P. D. N., B. W. Muncaster, and G. L. Mackie. 1989. Ecological and genetic studies on Dreissena polymorpha (Pallas): a new mollusc in the Great Lakes. Canadian Journal of Fisheries and Aquatic Sciences 46: 1587-1591. Hodgson, A. N. and R. T. F. Bernard. 1988. A comparison of the structure of the spermatozoa and spermatogenesis of 16 species of patelid limpet (Mollusca: Gastropoda: Archaeogastropoda). Journal of Morphology 195:205-233. Jamieson, B. G. M. 1984. A phenetic and cladistic study of spermatozoal ultrastructure in Oligochaeta (Annelida). Hydrobiologia 115:3-13. Jamieson, B. G. M., A. N. Hodgson, and R. T. F. Bernard. 1991. Phylogenetic trends and variation in the ultrastructures of the spermatozoa of sympatric species of South African patellid limpets (Archaeogastropoda, Mollusca). Invertebrate Reproduction and Development 20: 137-146. Mackie, G. L. 1984. Bivalves. In: The Mollusca: Vol. 7, Reproduction. A. S. Tompa, N. H. Verdonk, and J. A. M. van den Biggelaar, eds. pp. 378-379. Academic Press, Orlando, Florida. May, B. and J. E. Marsden. 1992. Genetic identification and implications of another invasive species of dreissenid mussel in the Great Lakes. Canadian Journal of Fisheries and Aquatic Sciences 49:1501-1506. Pathy, D. A. and G. L. Mackie. 1993. Comparative shell morphology of Dreissena polymorpha, Mytilopsis leucophyta, and the “quagga” mussel (Bivalvia: Dreissenidae) in North America. Canadian Journal of Zoology 71:1012-1023. Roldan, E. R. S., M. Gomendio, and A. D. Vitullo. 1992. The evolution of eutherian spermatozoa and underlying selective forces: female selection and sperm competition. Biological Reviews 67:55 1-593. Date of manuscript acceptance: 31 January 1994 AMERICAN MALACOLOGICAL UNION FINANCIAL REPORT General Accounts INFLOWS Membership Dues 11004.00 1992 65.00 1993 9410.00 1994 1529.00 Life Membership 450.00 Bulletin Sales 630.50 Subscriptions (Vol. 10) 291.00 Subscriptions (Vol. 11) 222.00 Back Issues & Suppliments 117.50 Sales of Other Publications 809.92 Recovered Publishing Costs 4576.60 Page Charges 3125.00 Reprints 1451.60 Donations 607.00 Symposium Fund 357.00 Student Award 250.00 Interest (Bank Accounts) 1796.25 Miscellaneous Income 78.24 FROM Symposium Endowment Bond Fund (Interest) 37.62 FROM Symposium Endowment Stock Fund (Interest) __ 106,90 TOTAL INFLOWS 20097.03 OUTFLOWS Office Expenses 897.96 Telephone/Postage 545.75 Printing 201.09 Miscellaneous 151.12 Bank Charges 301.28 Dues in Other Organizations 230.00 Incorporation Fees 20.00 Insurance/Bond 391.00 Accounting Fees 250.00 Publication Costs 12819.58 Bulletin 10(1) 9056.37 AMU News 23(3), 24(1) & 24(2) 3763.2) Travel Expenses 1500.00 Symposium Speakers 1500.00 Student Paper Awards 750.00 TO Symposium Endowment Bond Fund 14150.00 TO Symposium Endowment Stock Fund 21222.41 TO Life Membership Endowment Stock Fund 4950.00 TOTAL OUTFLOWS 58982.23 NET CASH FLOW (38885.20) CASH BALANCE (1/1/93) 95132.91 CASH BALANCE (12/31/93) 56247.71* * Note an additional $40,322.41 in endowment funds invested in Vanguard bond and index stock accounts. 83 SPECIAL PUBLICATIONS OF THE AMERICAN MALACOLOGICAL BULLETIN The Special Publication Series of the American Malacological Bulletin was begun to disseminate collected sets of papers with similar or related themes in a single volume. To date, three such issues have been published, each the result of a special convened symposium. The three Special Editions are PERSPECTIVES IN MALACOLOGY, PRO- CEEDINGS OF THE SECOND INTERNATIONAL CORBICULA SYMPOSIUM, and PROCEEDINGS OF THE SYMPOSIUM ON ENTRAINMENT OF LARVAL OYSTERS. Additional Special Editions are planned for the future. PERSPECTIVES IN MALACOLOGY (Sp. Ed. #1, July 1985) offers a wide range of papers dealing with molluscan biology of interest to professionals and amateurs alike. These papers were presented as part of a symposium held in honor of Professor M. R. Carriker at the time of his retirement and highlight a variety of recent advances in numerous facets of the study of molluscs. PERSPECTIVES IN MALACOLOGY offers insight into some of the frontiers of molluscan biology ranging from deep-sea hydrothermal vent malacofauna to chemical ecology of oyster drills. The PROCEEDINGS OF THE SECOND INTERNATIONAL CORBICULA SYMPOSIUM (Sp. Ed. #2, June 1986) con- tains numerous papers on this exotic bivalve that has become a significant ‘‘pest’’ organism of power plants and other industries using cooling waters. The proliferation, spread, functional biology, attempts at industrial control, taxonomy, and many other topics of interest to the malacologist and industrial biologist are addressed in this important special publication. The third special edition of the American Malacological Bulletin, PROCEEDINGS OF THE SYMPOSIUM ON ENTRAIN- MENT OF LARVAL OYSTERS (Sp. Ed. #3, October 1986) contains important review papers on the larval biology of the American oyster Crassostrea virginica, as well as intriguing papers on factors that limit productivity of these bivalves and limitations that exist on their dispersal and survival. The impact of cutter-head dredges is addressed in this special edition with special emphasis on the Chesapeake Bay system. To order your copies of PERSPECTIVES IN MALACOLOGY, PROCEEDINGS OF THE SECOND INTERNATIONAL CORBICULA SYMPOSIUM and PROCEEDINGS OF THE SYMPOSIUM ON ENTRAINMENT OF LARVAL OYSTERS, simply fill out the form below. Enclose check or money order made out to the AMERICAN MALACOLOGICAL BULLETIN or include VISA or MasterCard number and expiration date. PERSPECTIVES IN 2ND INTERNATIONAL ENTRAINMENT OF MALACOLOGY CORBICULA SYMPOSIUM LARVAL OYSTERS Special Edition No. 1 Special Edition No. 2 Special Edition No. 3 AMERICAN MALACOLOGICAL BULLETIN AMU Members $10.00 $20.00 $14.00 Non-AMU Members $15.00 $28.00 _ $20.00 Unitas Members $12.00 $22.00 $16.00 Institutions $25.00 $37.00 $28.00 Foreign Airmail $10.00 $10.00 $10.00 Foreign Seamail $ 5.00 $5.00 $ 5.00 SUBTOTALS $ i a nee $ TOTAL ENCLOSED $ (check or money order made out to: AMERICAN MALACOLOGICAL BULLETIN) Name: MasterCard/VISA No. Mailing Address: Expiration Date: Send Orders To: Dr. Paula Mikkelsen Department of Malacology Delaware Museum of Natural History Card Holder Box 3937 Wilmington, Delaware 19807-0937, U.S.A. Signature: 84 61st ANNUAL MEETING THE AMERICAN MALACOLOGICAL UNION HILO, HAWAII JUNE 8-12, 1995 (Arrive Hilo June 7 - leave June 13) Aloha komo mai kau ma hale — welcome to my house. Make your reservations now for Hilo, Hawai’i! The 1995 meeting of the American Malacological Union will be held at the University of Hawaii at Hilo, in Hilo, Hawaii. The meeting is a celebration of islands, beginning with a keynote address by Hawaii’s well known naturalist and photographer Bill Mull who will describe with tales and slides some of the wonders of Hawaiian animals and plants, and especially land shells. Three symposia are scheduled: biogeography convened by Gustav Paulay; cephalopods convened by Richard Young; and conservation convened by Elaine Hoagland. Plan to contribute a paper and/or poster for other sessions. The annual auction of books and shells will be exciting. Social events include the President’s Welcome, a sun- set reception at Lyman House Museum, a barbecue, and final banquet. Field trips on June 11 include tidepools and snorkeling, lava tubes, rainforest, kipukas (oases with snails on Mauna Kea), and the volcano. Accommodations will be in University of Hawaii at Hilo dormitories on campus with a meal plan. Accommodations will run $25 per night per bed in a four room suite, and the meal plan, covering most meals and coffee breaks from arrival June 7 to leaving June 13 will be about $150.00. If you prefer a hotel, you may make your own reservations. A list of hotels along with rates will be included with registration forms in January. Van service to and from the airport for accompanying guests and family are scheduled. For further information please contact: E. Alison Kay President, AMU Department of Zoology University of Hawaii 2538 The Mall Honolulu, Hawaii 96822 (808) 956-8620 (Phone) (808) 956-9812 FAX eakay@zoogate.zoo.hawaii.edu (E-Mail) 85 CONTRIBUTOR INFORMATION The American Malacological Bulletin serves as an outlet for reporting notable contributions in malacological research. Manuscripts concerning any aspect of original, unpublished research, important short reports, and detailed reviews dealing with molluscs will be considered for publi- cation. Each original manuscript and accompanying illustra- tions must be submitted with two additional copies for review purposes. Text must be typed on one side of 8-1/2 x 11 inch bond paper, double-spaced, and all pages num- bered consecutively with numbers appearing in the upper right hand corner of each page. Leave ample margins on all sides. Form of the manuscript should follow that outlined in the Council of Biology Editors Style Manual (sixth edition, 1994). This can be purchased from the CBE, 11 S. LaSalle Street, Suite 1400, Chicago, IL 60603, U.S.A. Text, when appropriate, should be arranged in sections as follows: 1. Cover page with title, author(s) and address(es), and suggested running title of no more than 50 characters and spaces. Authors should also supply five key words, placed at the base of this page, for indexing purposes. Abstract (less than 5% of manuscript length) 3. Text of manuscript starting with a brief introduction followed by methodology, results, and discussion. Separate sections of text with centered subtitles in cap- ital letters. Acknowledgments 5. Literature cited Figure captions All binomens must include the author attributed to that taxon the first time the name appears in the manuscript [e.g. Crassostrea virginica (Gmelin)]. This includes non- molluscan taxa. The full generic name along with specific epithet should be written out the first time that taxon is referred to in each paragragh. The generic name can be abbreviated in the remainder of the paragraph as follows: C. virginica. References should be cited within text as follows: Hillis (1989) or (Hillis, 1989). Dual authorship should be cited as follows: Yonge and Thompson (1976) or (Yonge and Thompson, 1976). Multiple authors of a single article should be cited as follows: Beattie et al. (1980) or (Beattie et al., 1980). In the literature cited section of the manuscript refer- ences must also be typed double spaced. All authors must be fully identified, listed alphabetically and journal titles must be unabbreviated. Citations should appear as follows: Beattie, J. H, K. K. Chew, and W. K. Hershberger. 1980. Differential survival of selected strains of Pacific oys- ters (Crassostrea gigas) during summer mortality. Proceedings of the National Shellfisheries Association 70(2):184-189. Hillis, D. M. 1989. Genetic consequences of partial self fertilization on population of Liguus fasciatus (Mollusca: Pulmonata: Bulimulidae). American Malacological Bulletin 7(1):7-12. Seed, R. 1980. Shell growth and form in the Bivalvia. In: Skeletal Growth of Aquatic Organisms, D. C. Rhoads and R. A. Lutz, eds. pp. 23-67. Plenum Press, New York. Yonge, C. M. and T. E. Thompson. 1976. Living Marine Molluscs. William Collins Son & Co., Ltd., London. 288 pp. Illustrations should be clearly detailed and readily reproducible. All line drawings should be in black, high quality ink. Photographs must be on glossy, high contrast paper. All diagrams must be numbered in the lower right hand corners and adequately labeled with sufficiently large labels to prevent obscurance with reduction by one half. Magnification bars must appear on the figure, or the cap- tion must read Horizontal field width = xmm or xpm. All measurements must be in metric units. All illustrations submitted for publication must be fully cropped, mounted on a firm white backing ready for reproduction, and have author’s name, paper title, and figure number on the back. All figures in plates must be nearly contiguous. Additional figures submitted for review purposes must be of high quality reproduction. Xerographic reproduction of pho- tomicrographs or detailed photographs will not be accept- ablefor review. Abbreviations used in figures should occur in the figure caption. Indicate in text margins the appropri- ate location in which figures should appear. Color illustra- tions can be included at extra cost to the author. Original illustrations will be returned to author if requested. Any manuscript not conforming to AMB format will be returned to the author for revision. Final submission of accepted, revised manuscripts should include two typed copies of the text, tables, etc. and an additional copy in electronic form on 3.5” diskette. The electronic version should be readable as non-formatted ASCII files. New Taxa. The Bulletin welcomes complete descriptions of new molluscan taxa. Establishment of new taxa must conform with the International Code of Zoological Nomenclature (1985). Descriptions of new species-level taxa must include the following information in the order as given: higher taxon designation as needed for clarity; fami- ly name with author and date; generic name with author and date; Genus species author sp. nov. followed by numeration of all figures and tables; complete synonymy (if any); listing of type material with holotype and any other type material clearly designated along with complete museum catalogue or accession information; listing of all additional non-type material also with full museum deposi- tion information; type locality; diagnosis and full descrip- tion of material done in telegraphic style including measurements and zoogeographic distribution as neces- sary; discussion; etymology. Descriptions of new supraspe- cific taxa should include type species (for new genus) or type genus (for new family), diagnosis and full description done in telegraphic style, and list of included taxa. Proofs. Page proofs will be sent to the author and must be checked for printer’s errors and returned to the printer within a three day period. Significant changes in text, other than printer errors, will produce publishing charges that will be billed to the author. Mailing. All overseas mailing must be done via airmail. The American Malacological Union will not be responsible for deferred publication of manuscripts delayed in surface mail. Charges. There are no mandatory page costs to authors lacking financial support. Authors with institutional, grant or other research support will be billed for page charges. The current rate is $35.00 per printed page. Acceptance and ultimate publication is in no way based on ability to pay page costs. Reprints. Order forms and reprint cost information will be sent with page proofs. The author receiving the order form is responsible for insuring that orders for any coau- thors are also placed at that time. Submission. Submit all manuscripts to Dr. Ronald B. Toll, Editor-in-Chief, American Malacological Bulletin, Depart- ment of Biology, Wesleyan College, 4760 Forsyth Road, Macon, Georgia 31297-4299 U.S.A. Subscription Costs. Institutional subscriptions are avail- able at a cost of $32.00 per volume. [Volumes | and 2 are available for $18.00 per volume]. Membership in the American Malacological Union, which includes personal subscriptions to the Bulletin, is available for $25.00 ($15.00 for students). All prices quoted are in U.S. funds. Outside the U.S. postal zones, add $5.00 seamail and $10.00 airmail per volume or membership. For subscrip- tion information contact Dr. Paula Mikkelsen, Managing Editor, American Malacological Bulletin, Department of Malacology, Delaware Museum of Natural History, Box 3937, Wilmington, Delaware, 19807-0937, U.S.A. AMERICAN MALACOLOGICAL BULLETIN VOLUME 11 ) NUMBER 2 Biannual Journal of the American Malacological Union CONTENTS Larval and juvenile shells of four North Atlantic scaphopod species. ASTORIA IRD STEINER sp scgucsccgsnpuvatesscptn devs ccvatinnsnsiesenpesteaveisersecavsbato steadeasicaibensvidetaQemctelaedadnasiiWvave: 87 The relevance of passive dispersal for the biogeography of Caribbean mollusks. RUDOLF S. SCHELTEMA ..0.00....ececcecccceccessssessesceesescescesecsesecseseecseceuseecssssesesseussscecesseaseese 99 Reproductivity seasonality, periodicity, and associated behavior in a colony of Strombus pugilis (Mollusca: Gastropoda) in Puerto Rico. CEL ALVY NA TORRE DD scientist cp tcspevscsada laaaidituanaspeiacneSenoddeneunlandivans dah oignvsbtentenbideusvedaukendd atyecersels 117 The stygobiont genus Bythiospeum in Austria: a basic revision and anatomical description of B. cf. geyeri from Vienna (Caenogastropoda: Hydrobiidae). IVEAR TEIN HAASE pss scnsicastucostansaptyacydoesanisis adovacctassacaeasachcechctaadecestdenindssassbunnsiberentasdenesuseoaiveutevstiats 123 HPLC analysis of chloroplast pigments from the marine ascoglossan Tridachia crispata (Morch, 1863) (Mollusca: Opisthobranchia). . RICHARD A. ROLLER and THOMAS S. BIANCHI ..0........ccccccccccsssssssssscecscssesssecsceeceecsesasseeaeees 139 Population ecology of the endangered Ouachita rock-pocketbook mussel, Arkansia wheeleri (Bivalvia: Unionidae), in the Kiamichi River, Oklahoma. CARYN C. VAUGHN and MARK PYRON ..........c.cccccccccsescsccsessessescessesecacsecsevevseescsacsctacecaceecaeeaeees 145 Infestations of glochidia on fishes in the Barren River, Kentucky. JEFFREY L. WEISS and JAMES B. LAYZER ...0.....ccccccsscscsesescsesssseescecscscscsesscecsvassesessseseecseaeeees 153 1994 SYMPOSIUM ON GULF OF MEXICO MOLLUSCA Ecology of infaunal Mollusca in south Texas estuaries. PAUL A. MONTAGNA and. RICHARD D, KALK Bo o...22-.csisssssssaisssesossesseescasecesesorasosesesinsendossosatenssensteetdsanadaiasneashasacensessivocets 163 The hypobranchial gland of the estuarine snail Stramonita haemastoma canaliculata (Gray) (Prosobranchia: Muricidae): a light and electron microscopical study. RICHARD A. ROLLER, JOHN D. RICKETT, and WILLIAM B. STICKLE 000... cccccscesescscsscsesscssssessseceesesscsevseessceacseuassesecaeseaecasessssescassacaceaasaeeees 177 The estuarine clam Rangia cuneata (Gray) as a biomonitor of heavy metals under _ laboratory and field conditions. MARC A. McCONNELL and RICHARD C. HARRED .c.ccsssescsccccseccssecssnasavaeebtosoinccsnuiiiniavucapd¥usesoeoacesssbensndsobeaveneeseduroesesccadetuelvenaele 191 (continued on back cover) AMERICAN MALACOLOGICAL BULLETIN RONALD B. TOLL, Editor-in-Chief Department of Biology Wesleyan College Macon, Georgia 31297-4299 MELBOURNE R. CARRIKER BOARD OF EDITORS PAULA M. MIKKELSEN, Managing Editor Department of Malacology Delaware Museum of Natural History ASSOCIATE EDITORS College of Marine Studies University of Delaware Lewes, Delaware 19958 GEORGE M. DAVIS Department of Malacology The Academy of Natural Sciences Philadelphia, Pennsylvania 19103 R. TUCKER ABBOTT Melbourne, Florida, U.S.A. JOHN A. ALLEN Millport, United Kingdom JOHN M. ARNOLD Honolulu, Hawaii, U.S.A. JOSEPH C. BRITTON Fort Worth, Texas, U.S.A. JOHN B. BURCH Ann Arbor, Michigan, U.S.A. EDWIN W. CAKE, JR. Ocean Springs, Mississippi, U.S.A. PETER CALOW Sheffield, United Kingdom JOSEPH G. CARTER Chapel Hill, North Carolina, U.S.A. ARTHUR L. CLARKE Portland, Texas, U.S.A. CLEMENT L. COUNTS, III Wallops Island, Virginia, U.S.A. THOMAS DIETZ Baton Rouge, Louisiana, U.S.A. WILLIAM K. EMERSON New York, New York, U.S.A. DOROTHEA FRANZEN Bloomington, Illinois, U.S.A. ROGER HANLON Galveston, Texas, U.S.A. Wilmington, Delaware 19807-0937 W. D. RUSSELL-HUNTER Department of Biology Syracuse University Syracuse, New York 13210 RUDIGER BIELER, Ex Officio Department of Zoology Field Museum of Natural History Chicago, IIlinois 60605 BOARD OF REVIEWERS JOSEPH HELLER Jerusalem, Israel ROBERT E. HILLMAN Duxbury, Massachusetts, U.S.A. K. ELAINE HOAGLAND Washington, D.C., U.S.A. VICTOR S. KENNEDY Cambridge, Maryland, U.S.A. ALAN J. KOHN Seattle, Washington, U.S.A. LOUISE RUSSERT KRAEMER Fayetteville, Arkansas, U.S.A. JOHN N. KRAEUTER Baltimore, Maryland, U.S.A. ALAN M. KUZIRIAN Woods Hole, Massachusetts, U.S.A. RICHARD A. LUTZ Piscataway, New Jersey, U.S.A. GERALD L. MACKIE Guelph, Ontario, Canada EMILE A. MALEK New Orleans, Louisiana, U.S.A. MICHAEL MAZURKIEWICZ Portland, Maine, U.S.A. JAMES H. McLEAN Los Angeles, California, U.S.A. ROBERT F. MCMAHON Arlington, Texas, U.S.A. THOMAS R. WALLER Department of Paleobiology Smithsonian Institution Washington, D. C. 29560 ANDREW C. MILLER Vicksburg, Mississippi, U.S.A. BRIAN MORTON Hong Kong JAMES J. MURRAY, JR. Charlottesville, Virginia, U.S.A. RICHARD NEVES Blacksburg, Virginia, U.S.A. JAMES W. NYBAKKEN Moss Landing, California, U.S.A. A. RICHARD PALMER Edmonton, Canada WINSTON F. PONDER Sydney, Australia CLYDE F. E. ROPER Washington, D.C., U.S.A. NORMAN W. RUNHAM Bangor, United Kingdom AMELIE SCHELTEMA Woods Hole, Massachusetts, U.S.A. DAVID H. STANSBERY Columbus, Ohio, U.S.A. FRED G. THOMPSON Gainesville, Florida, U.S.A. Cover: Io fluvialis (Say, 1825) is the logo of the American Malacological Union. THE AMERICAN MALACOLOGICAL BULLETIN is the official journal publication of the American Malacological Union. AMER. MALAC. BULL. 11(2) ISSN 0740-2783 AMERICAN MALACOLOGICAL BULLETIN VOLUME 11 NUMBER 2 SMITHSON TAR SEP 11.1995 LIBRARIES Larval and juvenile shells of four North Atlantic scaphopod species. GERHARD STUN iyi. 35005555 Sco eysccscassagovsvnvivsayignaessnyeavaburie Goecetuevasduutsdeiusdiseoucsuarstonteheueeinansneaeiesuael 87 Biannual Journal of the American Malacological Unio. CONTENTS The relevance of passive dispersal for the biogeography of Caribbean mollusks (RU DOLE Sz SCRE TRINA cots ceccyseves nssstnaseseastaascvenvennesdecssinesorsiin aster eee eee 99 Reproductivity seasonality, periodicity, and associated behavior in a colony of Strombus pugilis (Mollusca: Gastropoda) in Puerto Rico. SEPA WIN Ay BUD psoas a re cagvy Gas sao cSt css Ses cess fac dd oee shew ea av me cadena eas beep eaedecenes 117 The stygobiont genus Bythiospeum in Austria: a basic revision and anatomical description of B. cf. geyeri from Vienna (Caenogastropoda: Hydrobiidae). IVER TEIN EAB so oo sasingsvusnaea ven dren pss Sa cvstiaipep bea eed eee teeedeed seed ican eeiseosen atte teeon ets 123 HPLC analysis of chloroplast pigments from the marine ascoglossan Tridachia crispata (Morch, 1863) (Mollusca: Opisthobranchia). RICHARD A. ROLLER and THOMAS S. BIANCHI |... eeeecceecesceeneeeceecceaceeesacensescencenss 139 Population ecology of the endangered Ouachita rock-pocketbook mussel, Arkansia wheeleri (Bivalvia: Unionidae), in the Kiamichi River, Oklahoma. CARYN'C.. VAUGHN and MARK PY RON soc yecsccccheasedeczoceosds2caitee seinen aevinasdedacseeaanciapieidovaepsoetons 145 Infestations of glochidia on fishes in the Barren River, Kentucky. JEFFREY L. WEISS and JAMES B. LAYZER ..0000....eeecccsccsesscssssceseeceeaesseenssacenssecessececesesneeeeaees 153 1994 SYMPOSIUM ON GULF OF MEXICO MOLLUSCA Ecology of infaunal Mollusca in south Texas estuaries. PAUL A. MONTAGNA ane RIC BEAR DD ANC oe suscencTisw esse ucts sds scaseeaeads ote seb delobawtdevet ves avd las catcassteaeseovitdisuaaniidiomsialees 163 The hypobranchial gland of the estuarine snail Stramonita haemastoma canaliculata (Gray) (Prosobranchia: Muricidae): a light and electron microscopical study. RICHARD A. ROLLER, JOHN D. RICKETT, and VVUIGICUN NE Be SC Tse cara cerestecsesiis os ta pcesaitiva a eve h va tues cee veces hae evehcs exio ussuntuerzvaetev sexta e eens 177 The estuarine clam Rangia cuneata (Gray) as a biomonitor of heavy metals under laboratory and field conditions. MARC A. McCONNELL and RE A DG EEA BR aoa cas aace da vs uea rae dU Tote tafe casa tat a aeeD ea aac bec cudbges Peeeae dhoa ces at ava bchawnsbben ase: 191 Ecological notes and patterns of dispersal in the recently introduced mussel, Perna perna (Linné, 1758), in the Gulf of Mexico. DAVID W. HICKS and DOTEN W 2 TUNNEL gs FIR ssa sessscnsstcsicncctntaneccivssacasnsencetanieassenesnutatetadsassuullcuscsessavseseeesaqtasiecteuscaceanerrasens 203 PIN@hicial REPOF vicsascasnssveasionasasswescavousesiane paces pesasasossbanasunnanusdandalcaccanehunsunn dt ceseznuncesstestegucuautvorsssuevsssvecanssayeseare 207 PRAIA ATU IR sass cabana cat sete ony eusaisateas vo RRs vane Pa Renee paemn oS ane coe aean ate eee ee 208 Mid TVS TINORA IN cs cas casvoicaaxcnsidacasxnyecesienbeeehensuiaan caro cineo'deusssnsivguwesesistanehensveuiee sasve vannaeneens busmaunnevusneneneescasveuasssuansaneeee 209 Tape EO. VIN 1: nesses acs inssnaircs cialeus diners nspnciin a tac aing deiio hs ania tuts sa osesnca webge Canaan name cetase ois cideied tao taananetanen ee testas 210 il Larval and juvenile shells of four North Atlantic scaphopod species Gerhard Steiner Institute of Zoology, University of Vienna, Althanstr. 14, A-1090 Vienna, Austria Abstract. Ontogenetically early shells of four North Atlantic scaphopod species, Antalis occidentalis Stimpson, 1851, Entalina quinquangularis (Forbes, 1844), Pulsellum lofotensis (M. Sars, 1865), and Cadulus subfusiformis (M. Sars, 1865) were examined by SEM. The larval shell of previously described scaphopods can be divided into the bulbous protoconch A and the more or less elongated protoconch B. In A. occidentalis and E. quinquangularis, a smooth teleoconch A is distinguished from the ribbed teleoconch B. The protoconch of A. occidentalis matches previous reports on A. tarentinum (Lamarck, 1818) (type 1). E. quinquangularis and P. lofotensis have a protoconch with a marked constriction, but without annulations (type 2). In C. sub- fusifomis, only teleoconchs A and B were found, but no distinct protoconchs. Teleoconch A of this species resembles a juvenile Pulsellum, before it forms a constriction separating it from the rapidly expanding teleoconch B. In addition to shell morphometrics, the microstructure is described for each species. The results are compared to earlier descriptions of Recent and fossil scaphopod protoconchs: type | is restricted to the Dentaliida, type 2 typical for Gadilida except for the presumably derived condition in C. subfusifomis. The first to describe and figure larval shells of the dentaliid scaphopods Antalis dentalis (Linné, 1766) and A. entalis (Linné, 1758) were Lacaze-Duthiers (1856-1857) and Kowalevsky (1883). Sars (1865) gave the first report of a gadilid larval shell, Entalina quinquangularis (Forbes, 1844). Since then scaphopod larval shells have been occa- sionally reported (Table 1), but mostly as by-products of other investigations. Scarabino (1979) noted differences between the orders regarding larval shells: Dentaliida have several annulations, Gadilida only one. The minute larval structures are also preserved in fossils; MacNeil and Dockery (1984), Ruggieri (1987) and Engeser et al. (1993) described lower Jurassic to Pliocene specimens. The major difficulty in the study of early scaphopod shell development is the dynamics of the mantle-shell mor- phology. The posterior mantle margin starts to dissolve or decollate the tip of the shell soon after metamorphosis so as to maintain an apical shell aperture large enough for the increasing respiratory needs (Lacaze-Duthiers, 1856-1857; Steiner, 1991; Reynolds, 1992). Although the onset of this shell-destructing process seems to vary between species (Engeser et al., 1993), the larval shell is lost in all but very young specimens, and various stages of erosion may be seen. Engeser et al. (1993) sketched a general outline of a scaphopod larval shell. Based upon this concept and the reports in the literature (Table 1), the outline of a dentaliid larval shell in Fig. 1 serves as morphological reference for the species examined in this paper. The larval shell or pro- toconch shows two structurally different regions. The earli- est part of the shell is a bilaterally symmetrical, saddle- or clasp-shaped secretion that is not fused ventrally (Bandel, 1982). Engeser et al. (1993) coined the name “genae” for this bulging part of the shell; here it is termed protoconch A. The younger part of the larval shell, protoconch B, extends both posterior and anterior to protoconch A. The posterior part is the chimney-like fumarium (Engeser et al., 1993); the anterior one shows the characteristic annula- tions, followed by a short cylindrical region at the transition Fig. 1. General outline of dentaliid larval shell, ventral view; modified after Engeser et al. (1993). (f, fumarium; g, genae; i, increment lines; pc A, protoconch A; pc B, protoconch B,; r, ribs; s, suture; tc A, teleoconch A; tc B, teleoconch B). American Malacological Bulletin, Vol. 11(2) (1995):87-98 87 88 AMER. MALAC. BULL. 11(2) (1995) Table 1. Larval shells of Recent (r) and fossil (f) scaphopods previously described and/or illustrated (*). DENTALIIDA Antalis tarentinum (Lamarck, 1818) Antalis cf. dentalis (Linné, 1766) Antalis sp. “A” Antalis cf. pseudofissura Janssen, 1978 Fissidentalium phaneum (Dall, 1895) Fissidentalium jaffaensis Cotton and Ludbrook, 1938 Fissidentalium laqueatum (Verrill, 1885) Dentalium sp. “Dentalium” zephyrinum Casey, 1903 “Dentalium” polygonuum Casey, 1903 “Dentalium rectum” Gmelin, 1790 Episiphon hyperhemileuron (Verco, 1911) GADILIDA Entalina quinquangularis (Forbes, 1844) Entalinopsis callithrix (Dall, 1889) Entalinopsis sp. Heteroschismoides subterfissum (Jeffreys, 1877) “Laevidentalium” sp. Suevidontus jaegeri Engeser, Riedel and Bandel, 1993 Baltodentalium weitschati Engeser and Riedel, 1992 Gadila turgida (Meyer, 1886) to the teleoconch. There is a suture marking the line of ven- tral mantle and shell fusion extending from between the bulging sides of protoconch A onto protoconch B. In many scaphopods with longitudinal sculpture, the teleoconch has two or three sections. The early adult shell, teleoconch A, lacks ribs or striae and shows transverse growth lines only. The ribbed part is teleoconch B. In the genus Antalis, for example, a third section, teleoconch C, is differentiated, where the longitudinal ribs have faded and the shell is smooth again. The letters “A” and “B” are preferred to numbers to avoid confusion or inadequate homologization with prodissoconch sequences of gastropods and bivalves. Only one (see Discussion) of the dentaliid species studied has a constriction at the anterior end of protoconch B as fig- ured by Engeser et al. (1993). Therefore, such a constric- tion is not part of the common dentaliid pattern. The only report on early shell development in the family Gadilidae is that on the fossil Gadila turgida by Engeser et al. (1993). The larval shell, being somewhat bell-shaped, lacking a fumarium, and having “The suture and the genae ... only poorly developed” (Engeser et al., 1993:90), has only little resemblance with the pattern out- lined above. The larval and juvenile shells of four North Atlantic species of Scaphopoda are described and morpho- metrically characterized in this paper. Three of them have not been described previously. Sars’ (1865) description of the larval shell of Entalina quinquangularis is updated. Lacaze-Duthiers, 1856-1857 Bandel, 1982 Engeser et al., 1993 Kowalevsky, 1883 Scarabino, 1979 Engeser et al., 1993 Burch and Burch, 1989 Cotton and Godfrey, 1940 Henderson, 1920 Engeser et al., 1993 MacNeil and Dockery, 1984 MacNeil and Dockery, 1984 Ruggieri, 1987 Cotton and Godfrey, 1940 Sars, 1965 Scarabino, 1979 Engeser et al., 1993 Davies, 1987 Engeser et al., 1993 Engeser et al., 1993 Engeser et al., 1993 Engeser et al., 1993 fos Min Willens Wiens ens Sites iin in Bn in Ss Os Be Bs} bea rita 1 tn cr etic a cor Sees | * %* * &* & * * * * &* * * x * © &* &* ¥ & * MATERIAL AND METHODS Antalis occidentalis Stimpson, 1851, Entalina quin- quangularis, Pulsellum lofotensis (M. Sars, 1865), and Cadulus subfusiformis (M. Sars, 1865) were collected off Saksvik, Trondheimfjord, Norway (63° 27’N, 10° 32’E) from 100-238 m depth, using a benthic sled with 1 mm mesh size. The sediment was carefully washed through a 250 pm nylon net, the animals sorted under a binocular microscope and preserved in 70% ethanol. All specimens used in this study were collected at this site or at other loca- tions in Trondheimfjord. Juvenile specimens of C. sub- fusiformis as described in this paper were also found in Raunefjord near Bergen. For scanning electron microscopy (SEM), the samples were transferred to absolute ethanol prior to air-drying, mounted, sputter-coated, and examined with a Jeol-JMS 09 scanning electron microscope. Morphometric measurements were taken as indicat- ed in Fig. 2 from SEM photomicrographs. Lengths of the total shell, protoconch (and its components protoconch A and protoconch B), teleoconch, and, if present, teleoconch A and teleoconch B, were measured. Shell diameters were taken at the apical aperture, maximum width of protoconch A (which is in all cases the maximum diameter of the entire protoconch), initial width of the teleoconch (identical with the terminal diameter of the protoconch), basal aperture, and, if present, width at the transition from teleoconch A to STEINER: SCAPHOPOD LARVAL AND JUVENILE SHELLS 89 iw) \ oa 18 me) oO | (a) 1 D-te Bi i-pe 8 — L-pc Fig. 2. Diagram of shell measurements. (D-ap, apical diameter, D-pc, maximum protoconch diameter; D-tc, maximum teleoconch diameter, D- tc Bi, initial teleoconch B diameter; D-tc i, initial teleoconch diameter; L- pe, length of protoconch; L-pe A, length of protoconch A; L-pc B, length of protoconch B; L-tc, total length of teleoconch; L-tc A, length teleo- conch A; L-tc B, length of teleoconch B; L-tot, total length). teleoconch B. An additional measurement in Cadulus sub- fusiformis was the maximum diameter of teleoconch A, because it is not identical with the initial diameter of teleo- conch B. A common measurement for curvature is the arc, the maximum normal distance between the shell and the line between anterior and posterior dorsal shell margins (Ludbrook, 1960). The length: width ratio of the protoconch was calculated from its total length and the maximum diam- eter. The coefficients of variation (CV; %) were calculated (standard deviation/mean x 100) to provide a unit- and dimension-free indicator of the variability of each parame- ter (Zar, 1984). Their comparison beyond that in variability profiles seemed not appropriate, because there is no test of significance for the CV of more than two characters (Sokal and Braumann, 1980) and not all measures are of the same dimensionality (Lande, 1977). Principal Component Analysis (PCA) based on the variance/convariance matrix of log-transformed data was accomplished using Statgraphics Plus ver. 5.2 (Graphic Software Systems, Inc.). Of the program’s output, only Eigenvalues and the weights for each measurement on each axis were used. The princi- pal components were calculated using the equation Yj = (xq 7 X’)) X Wi + (x2 7 X’9) XWj2 +... + (Xj = X’}) X Wij where Yj is the principal component of the i-th axis, x the measurement, X’ the global mean, and w the weight of the measurement on the i-th axis (Johnson and Wichern, 1988). I preferred not to use other standardizations of the log- transformed data to maintain the mutual positions of the data points. This ensures that, using the global means and weights given in Table 3, morphometric data of other speci- mens can be plotted onto the resulting diagrams, and thus compared with those of the present study. RESULTS SHELL MORPHOLOGY Antalis occidentalis The five juvenile specimens found with the larval shell still attached ranged from 1.49-3.43 mm in length (Table 2) and showed various stages of apical erosion (Fig. 3). The larval shell fits the general dentaliid pattern. Its mean length is 465 + 61 pm, the length: width ratio (L: W) is 2.62. The protoconch diameter provided the least variable measure. In contrast to all other protoconch parameters (variability 14.78-19.319%), it varied by only 4.68% (Table 2). Several growth lines separated the smooth and bulging sides of protoconch A from the suture (Fig. 3G). The fumarium was more or less eroded in all specimens; therefore, both the orientation and the width of the apical shell orifice varied. Protoconch B had five to seven annula- tions. The transition from protoconch B to teleoconch A is straight or almost so (Figs. 3A-E). Although the diameter of protoconch A is slightly larger than that of protoconch B, there is no constriction. Teleoconch A is smooth with the exception of growth lines and about 1.8 mm long. Anteriorly, the longitudinal ribs of teleoconch B appear. Fig. 3. Larval and juvenile shells of Antalis occidentalis. A. Smallest specimen in lateral view; teleoconch B is not developed yet. B-C. Dorsal (B) and lateral (C) view of specimens with teleoconch B, the latter being the only one with protoconch and teleoconch forming an angle. D. Larval shell with fumarium almost eroded. E-F. Detail of B showing the widened apical shell opening. G. Ventral view of A showing the well developed protoconch A (genae) and the suture extending onto protoconch B. Scale bars = 100 pm. 90 AMER. MALAC. BULL. 11(2) (1995) Table 2. Means (+ S.D.) of shell measurements in pm. [Abbreviations as in Fig. 2; also, CV = coefficient of variation expressed as percent (S.D. x 100/ mean), D-tc A = maximum teleoconch A diameter (Cadulus subfusiformis only), L:W = protoconch length: width ratio, ]. N = 5 (Antalis occidentalis), 15 (Entalina quinquangularis), 11 (Pulsellum lofotensis), and 52 (Cadulus subfusiformis). Antalis occidentalis Mean CV Mean Species A. Total Shell Measurements L-tot 2583 + 709 27.46 1612 + 347 D-ap 67 + 11 16.78 53 + 26 D-tc 379 + 60 15.91 310 + 63 arc Si 22 42.63 177 + 52 B. Protoconch Measurements L-pe 465 + 69 14.78 264 + 31 L-pe A 186 + 29 15.86 183 + 12 L-pc B 279 + 54 19.31 81 + 26 D-pc A 178 + 8 4.68 108 + 4 L:W 2.62 + 0.39 14.80 2.43 + 0.27 C. Teleoconch Measurements L-tc 2118 + 650 30.70 1345 + 357 Teleoconch A L-tc A 1387 + 306 22.07 545 + 124 D-tc i 175 + 10 5.75 109 + 6 D-tc A — — — Teleoconch B L-tc B 732 + 377 =51.48 532 + 298 D-tc Bi 320 + 35 10.78 211 + 12 Entalina quinquanqularis Pulsellum lofotensis Cadulus subfusiformis Mean CV Mean CV 21.52 951 + 192 20.18 1182 + 99 8.39 49.47 77 +17 21.85 163 + 31 18.88 20.30 229 + 63 27.51 328 + 152 46.21 29.18 59 + 20 33.76 64 + 26 40.63 11.59 205 + 29 13.99 — — 6.71 122 + 28 23.23 — — 31.99 83 + 17 20.86 — — 3.38 109 + 4 3.53 — — 10.92 1.89 + 0.28 14.77 — — 26.56 748 + 205 27.44 — — 22.82 — — 802 + 226 28.12 5.31 109 + 5 4.80 — — — — 236 + 12 5.21 5.71 —- — 208 18 8.56 d 4 56.02 — — 376 + 266 70.59 + Among teleoconch parameters, the initial diameters of teleoconch A and B were most invariable (CV 5.14% and 9.64, respectively; Table 2), but the variability of teleo- conch A length (19.74%) was about average. Entalina quinquangularis The 15 specimens with larval shells ranged from 1.07-2.34 mm long; the larval shells measured about 264 + 30 ppm in length, with L:W 2.43. The larval shell is smooth, Fig. 4. Larval and juvenile shells of Entalina quinquangularis. A-C. Lateral views of entire specimens. D. Detail of B showing the widened apical opening and the suture. E. Specimen with the most clearly distin- guishable protoconch A and fumarium. Scale bars = 100 pm. sometimes with growth lines near the border to teleoconch A (Fig. 4). Protoconch A was less set off from protoconch B than in Antalis occidentalis (Figs. 4A, E) and, therefore, the border between them less obvious. It lies in the area where the larval shell starts to decrease in diameter. The ventral midline is marked by a distinct suture. The fumari- um is short, smooth, and eroded in most specimens. Protoconch B forms a constriction, beyond which the shell increases abruptly in diameter forming a conspicuous annu- lus preceeding the youngest part of protoconch B. A slight angle is formed either by the protoconch constriction or by the axes of the proto- and teleoconch. Like in A. occidental- is, there is a typical teleoconch A with growth lines only (Figs. 3A-C). The five characteristic ribs of the species’ teleoconch B_ become visible about 0.8 mm from the larval shell. The circular anterior shell opening then becomes more and more pentagonal. Pulsellum lofotensis The 11 specimens of Pulsellum lofotensis with almost complete larval shells ranged from 0.64-1.21 mm in length; the mean protoconch length was 205 + 27 pm, with L:W 1.89. Generally, the protoconch resembles that of Entalina quinquangularis. It is, however, significantly shorter and stouter in shape (Fig. 5), as the lower L:W expresses (see below). The larval shell has a more rounded STEINER: SCAPHOPOD LARVAL AND JUVENILE SHELLS 91 Table 3. Principal Component Analysis: Eigenvalues, global means of measurements [log(gl-x)] and their weights (w) for each Principal Component (PC I- X). Abbreviations as in Fig. 2. PC I PC II PC Ill PCIV PC V PC VI PC VII PC VIII PC IX PCX Eigenvalue 0.1288 0.1147 0.0301 0.0099 0.0065 0.0031 0.0018 0.0006 0.0002 3.28E-S log(gl-x) w w w w Ww w w w w w L-tot 3.151 0.4801 -0.0576 0.0747 -0.2083 -0.1744 -0.2212 0.0352 -0.1698 0.0427 0.7784 L-pc 2.409 0.2174 -0.1754 -0.2967 -0.0080 0.4955 0.0926 0.7610 0.0070 -0.0180 -0.0154 L-tc 3.056 0.5323 -0.0228 0.1497 -0.2606 -0.2977 -0.2526 0.1153 -0.2544 0.1093 -0.6193 D-pc 2.069 0.1142 -0.1921 -0.0163 -0.1953 -0.0452 -0.1528 -0.0621 0.3890 -0.8533 -0.0544 D-ap 1.776 -0.0621 -0.1658 0.7914 0.0212 0.5297 -0.2400 -0.0270 -0.0536 0.0017 -0.0043 D-tc i 2.070 0.1149 -0.1823 -0.0103 -0.1473 -0.0155 -0.1573 -0.0474 0.8123 0.4943 -0.0187 D-tc 2.441 0.3007 -0.0179 0.2978 -0.2024 -0.0054 0.8736 -0.0802 0.0978 -0.0210 0.0037 L-pc A 2.194 0.2013 0.0392 -0.3931 -0.3414 0.5718 -0.0436 -0.5710 -0.1538 0.0572 -0.0700 L-pc B 1.984 0.3254 -0.5876 -0.1028 0.6802 -0.0157 0.0525 -0.2577 -0.0653 0.0054 -0.0438 arc 1.972 0.4147 0.7216 0.0585 0.4562 0.1590 -0.0719 -0.0241 0.2342 -0.0987 -0.0131 % total variance 43.54 38.78 10.19 3.36 2.20 1.05 0.60 0.19 0.07 0.01 appearance: protoconch A is not clearly demarcated from protoconch B and a suture is not evident. All specimens show at least slight signs of apical erosion or truncation (Fig. 5), and so the presence of a fumarium cannot be con- firmed. In most shells, the transition from the constriction of protoconch B to the annulus is not as abrupt as in E. quinquangularis. The cylindrical section of protoconch B bordering the teleoconch is similar to that of other scaphopods. Because there are only growth lines and no longitudinal sculpture, teleoconchs A and B can not be dis- tinguished. The anterior opening of the teleoconch is circu- lar. Cadulus subfusiformis Bulbous larval shells like those described above have not been found in this species. Instead, aberrant looking scaphopods with somewhat trumpet-like shells (Fig. 6) Fig. 5. Larval and juvenile shells of Pulsellum lofotensis. A. Complete protoconch. B. Onset of apical erosion. C. Half of protoconch eroded. D. Lateroventral view of protoconch with widened apical opening. E. Detail of B, showing the different surface textures of protoconch and teleoconch. Scale bars = 100 pm. Fig. 6. Juvenile shells of Cadulus subfusiformis. A-B. Teleoconch A. C- D. Eroding teleoconch A and commencing teleoconch B. E-F. Incomplete, stub-like teleoconchs B; teleoconch A eroded. Scale bar = 100 pm. turned out to be juvenile Cadulus subfusiformis, showing the typical radula morphology of the species. The apical part of the shell, teleoconch A, is a slowly expanding tube, like the teleoconch of Pulsellum lofotensis but almost straight. Then, after a short constricted section, a rapidly expanding teleoconch B is formed. The length of these trumpet-like shells varied little (0.89-1.39 mm; mean 1.18 + 0.1 mm; N = 52; see also Table 2), but the relative propor- tion of the two sections does, indicating well synchronized anterior secretion and posterior truncation processes. Animals which have not yet started to secrete teleoconch B are very similar to juvenile P. lofotensis, but can be identi- fied by their slightly depressed anterior shell opening. The shell surface is always smooth, without any longitudinal sculpture. Faint growth lines are present on teleoconch A; teleoconch B lacks even those. Surface texture of teleo- conch B is amorphous and completely smooth under SEM 2 AMER. MALAC. BULL. 11(2) (1995) yy Fig. 7. Shell growth series of juvenile Cadulus subfusiformis (diagram- matic). (Fig. 6C), but that of teleoconch A is crystalline. About 44% of the specimens (23 of 52) have an area where the early crystalline texture alternates with the amorphous teleoconch texture on teleoconch A. Thus, a series of annu- lations are seen under SEM (Figs. 6A, B, D). Occasionally, I found stub-like, incomplete teleo- conchs of Cadulus subfusiformis without the typical anteri- or constriction (Figs. 6E, F). From the various shapes and stages of the trumpet-like and stub-like shells, a growth series can be inferred (Fig. 7). SHELL MICROSTRUCTURE The shells of all species examined have a layer of prismatic crystals, the only component of the most recently secreted shell parts (Fig. 8A). In the older portions, the prismatic layer is lined by a crossed lamellar layer. Close to its outer border, the lamellae are slightly irregular. The first- order lamellae are perpendicular to the long axis of the shell and, in juvenile specimens, form an angle of about 45° with the shell diameter (Fig. 8B). In adult specimens, the lamellae become almost tangential to the internal shell lumen. A distinct layer of organic material outside the cal- careous ones (periostracum) was evident only in three spec- imens of Cadulus subfusiformis, but was definitely lacking in the other species examined. The prismatic shell layer in Antalis occidentalis was 2.0-2.3 jm thick near the anterior shell edge, but up to 3 pm in the fumarium (Fig. 8A). In Entalina quin- quangularis, the prisms were 1.7-2.3 pm high (Fig. 8C), those of Pulsellum lofotensis were between 1.3 and 1.8 pm (Figs. 8B, D, E). Compared to the other species, the crossed lamellar layer in E. quinquangularis rapidly increases in thickness behind the anterior shell edge. The shell thus quickly gains strength and stability, but only slowly increases in internal diameter. Cadulus subfusiformis has an outer prismatic layer 1.3-2.0 pm thick. In the apical por- tions of teleoconch A, the prisms have rounded tops with pits between them, but those of teleoconch B fit tightly together, each giving rise to the respective surface texture (see above). Being formed by the outer prismatic layer only, the incomplete definitive teleoconch is extremely thin and fragile, so that manipulation and transport often dam- aged the shell edge. When the teleoconch is completed, Fig. 8. Shell microstructure: A. Antalis occidentalis; prismatic layer at the fumarium margin (apical part of protoconch B). B. Pulsellum lofotensis; prismatic layer at the apical margin of protoconch A. C. Entalina quin- quangularis; transversal fracture of early teleoconch B, prismatic and thick cross-lamellate layers. D. Pulsellum lofotensis; transversal fracture of early teleoconch, prismatic and cross-lamellate layers. E. Pulsellum lofotensis; change in shell surface texture at the transition from protoconch (left) to teleoconch (right). F. Cadulus subfusiformis; transversal fracture of early teleoconch B; outer prismatic and cross-lamellate layers lined by flat inner prismatic layer. Scale bars = 1 pm (A-B); 5 pm (D-F); 10 pm (C). STEINER: SCAPHOPOD LARVAL AND JUVENILE SHELLS ee t of Variation icien Coeff L—-tot D-tc L-—pce D-ap orc L-pcA L-pcB L/w D-tcl L-tcB D-pcA L-te L-tcA O-tcBi Fig. 9. Variability profiles of 14 characters for Antalis occidentalis (Ao) Entalina quinquangularis (Eq), Pulsellum lofotensis (P|) and Cadulus subfusiformis (Cs). The vertical bars indicate the range of CV for each character for all species, taken from Table 2. Abbreviations of characters as in Fig. 2. however, it is multi-layered, very tough and resistant to fracture. The mature shell is comprised of the outer pris- matic layer, the crossed lamellar layer, each about 2 pm thick, and a 0.5-1 pm thin inner prismatic layer (Fig. 8F). MORPHOMETRIC ANALYSIS The coefficients of variation (CV) for 14 shell mea- surements of each species (as listed in Table 2) are plotted as variability profiles in Fig. 9. For Cadulus subfusiformis, only teleoconch measurements were available. The three most invariable characters for all species were the maxi- mum diameter of the protoconch (D-pc A) and the diameter at the transitions from proto- to teleoconch (D-tc i) and from teleoconch A to B (D-tc Bi, missing in Pulsellum lofotensis). The length of teleoconch B, when present, was the most variable character. Congruency between profiles was high on the right side of L-pcB, where some proto- conch and the teleoconch characters were plotted. It was lowest for the total shell measurements. Ten length and diameter measurements of total shell, protoconch and teleoconch, of Antalis occidentalis, Entalina quinquangularis and Pulsellum lofotensis were used for Principal Component Analysis (PCA). The mea- surements of teleoconchs A and B (missing in P. lofotensis) and the data of Cadulus subfusiformis (protoconch not known) were excluded. Eigenvalues, weights and global means of each measurement are listed in Table 3. The first two axes (PC I and II) of the PCA represent 82.33%, the first three, 92.52% of total variance. The plotted PC I and II (Fig. 10A) clearly separated A. occidentalis from the two gadilid species that were close to each other but did not overlap. In the plot of PC I and III (Fig. 10B), however, E. quinquangularis and P. lofotensis separated more clearly, while the former overlapped considerably with A. occiden- talis. PC I was mainly a length axis: all characters except for D-ap were loaded positively, L-tc and L-tot having the highest loads (= weights). PC II reflected the inverse rela- tive correlation of arc and some protoconch measurements: A. occidentalis having the longest protoconch B was almost straight, but the gadilid species with a low L-pc B value were more curved. The other negatively loaded characters on PC II were D-pc, D-tc i and L-pc. Positive loads on the diameters of the shell apertures (D-ap, D-tc) and negative loads on protoconch length (L-pe A, L-pc) determined PC III, indicating that with increasing teleoconch diameter, protoconch A became shorter, in relative and absolute terms, and its aperture larger (see Discussion). In addition, linear regressions were drawn for each species (Fig. 10). The principal components of log-transformed mea- surements of the juvenile shells described and listed by Engeser et al. (1993) were similarly calculated. The results are plotted in Fig. 10C. Significant differences between the L:W ratio of Entalina quinquangularis and Pulsellum lofotensis were 94 AMER. MALAC demonstrated by a STUDENT t-test for equal variances. The t-statistic of 4.95 with 23 D.F. was significant at a = 0.001 to reject HO (no differences in means). DISCUSSION The previously reported scaphopod larval and juve- nile shells (Table 1) and those described above present three different types (Fig. 11). The first and best-known type has a bulging, very distinct protoconch A, a long pro- toconch B with three to nine annulations, and no constric- tion at the border to the teleoconch (Fig. 11A). It is report- PC 0.8 °o . BULL. 11(2) (1995) Antolis occidentalis ed from four species of the genus Antalis, three species of Fissidentalium, Episiphon hyperhemileuron, and four pre- sumably dentaliid species of questionable generic position. In the second type, protoconch A is less clearly distinguish- able from protoconch B than is often the case in the type 1 larval shell. Protoconch B is comparatively short and forms a constriction and a subsequent marked annulus bordering the teleoconch (Fig. 11B, C). This type of protoconch is known from four species of the genera Entalina, Entalinopsis and Heteroschismoides belonging to the gadilid suborder Entalimorpha, the gadilimorph genus Pulsellum, and from three species of uncertain systematic position (“Laevidentalium” sp., Baltodentalium weitschati, * Entolino quinquongularis y=0.162x+0.245 C Pulsellum lofotensis -0.8 y=0.577x+0.309 0.5 y=0.744x-0.165 . ve a “ / vad Pas y=0.263x-0.132 Fig. 10. Plots of Principal Component Analysis of ten characters of Antalis occidentalis, Entalina quinquangularis and Pulsellum lofotensis (see text). A. Plot of PC I and II. B. Plot of PC I and III. Linear regression equations are given as y = kx + d. C. Plot of PC I and II showing the species areas as in A (crosses indicating means) and the fossil specimens from Engeser et al. (1993). (Ao, Antalis occidentalis; D, Dentalium sp.; Dp, Dentalium pseudofissura; E, Entalinopsis sp.; Eq, Entalina quinquangularis, L, “Laevidentalium” sp.; Pl, Pulsellum lofotensis, S, Suevidontus jaegeri). STEINER: SCAPHOPOD LARVAL AND JUVENILE SHELLS 95 Suevidontus jaegeri). Cadulus subfusiformis and the confamiliar, pale- ocenic Gadila turgida differ from both these types: there is no trace of a clasp-like or bulbous protoconch and, in C. subfusiformis, there is a constriction between teleoconch A and B (Fig. 11D). Instead, G. turgida has a bell-shaped pro- toconch (Engeser et al., 1993) that is not similar to those of other scaphopods. The shell development of C. sub- fusiformis resembles that of G. turgida in having transverse annulations in the surface texture of early shell parts and in having an anteriorly constricted teleoconch. However, G. turgida lacks the constriction between teleoconch A and B. It seems likely that these closely related species have a sim- ilar mode of shell development and perhaps the same type of protoconch. Only ontogenetic studies will reveal whether such a bell-shaped larval sheil also exists in C. sub- fusiformis. Besides complete lack in ontogeny, several rea- sons may account for the absence of a protoconch in this material of C. subfusiformis: loss in the course of washing and sorting, very early truncation, or lack of calcification. The juvenile shell described for C. subfusiformis definitely is a teleoconch, because the animals found within these shells were unequivocally adult in their organization with- out even remnants of larval features (e.g. a ciliary locomo- tory organ). The shells were perfectly uniform with the ear- liest parts already tubular and cylindrical without a trace of a suture. It is by no means certain, however, that teleoconch A of this species is homologous with that of the others described here. C. subfusiformis and G. turgida obviously represent a third type of early ontogenetic shell morpholo- gy, although in the former, no protoconch has yet been found. The Principle Component Analysis (PCA) is a pow- erful tool of data reduction to demonstrate and test similari- ties and differences of complex morphologies. The ten measurements of identical units allowed the use of their variance/covariance matrix. To test the robustness of the dataset, PCA was also performed with the correlation matrix, both with and without adding the L:W ratio to the data. Their results (not shown) were qualitatively identical, indicating that these ten parameters represent a stable dataset. PCA of the protoconch measurements only showed an overlap of Entalina quinquangularis and Pulsellum lofotensis. Because protoconchs alone are unlikely to be found in sediment samples or collections, I prefer to include the measurements of the early teleoconch in the analysis. As shown in Fig. 10, the juvenile shells of dentali- ids and gadilids were clearly separated. This is not only a question of size, although P. lofotensis and Antalis occiden- talis were distinct on principal component (PC) I, mainly a size axis. On the same axis, E. quinquangularis, however, overlapped widely with A. occidentalis. The latter was sep- arated from the gadilid species on PC II, demonstrating the inverse correlation of arc and length of protoconch B. The distinction between E. quinquangularis and P. lofotensis was mainly on PC I. In the plot of PC I/II (Fig. 10B), the separation of these two species became clearer: the smaller P. lofotensis has a wider apical aperture (D-ap) than the longer E. quinquangularis, but a smaller anterior aperture and a shorter protoconch. The slope of the linear regressions for each species can be used as an indicator of isometric growth of measure- ments loaded on the PCs when it is approximately 1, as in Fig. 10B for Pulsellum lofotensis and Entalina quinquangu- laris. In this case, apical and anterior aperture increased (log-) proportionally with length, but protoconch length, especially that of protoconch A, decreased. Table 3 provides all parameters to reproduce PCA calculations for data of additional specimens of these and of other species using the given equation. The positions of the two “Dentalium” specimens cf Engeser et al. (1993) in the PCI-PCII-plot (Fig. 10C) grouped with the area occu- pied by Antalis occidentalis. All other specimens showed type-2 protoconch morphology and grouped with Entalina quinquangularis and Pulsellum lofotensis, supporting the above ordinal assignment. For another fossil larval shell described by Engeser et al. (1993), “Laevidentalium” sp., this is of particular importance. Generic assignment to Laevidentalium refers to the order Dentaliida, but the proto- conch with constriction and annulus shows the typical type- 2 features of the Gadilida. Although it does not clearly group with P. lofotensis, the juvenile shell closely resem- bles that of Pulsellum species I have seen, both in shape and surface sculpture. The protoconch is larger than in P. lofotensis but the mean L:W (1.89 + 0.13, CV = 7.11) cal- culated from seven specimens listed by Engeser et al. (1993) was exactly the same. STUDENT t-tests showed “Laevidentalium” to be as different from FE. quinquangu- laris as was P. lofotensis, with an even higher significance, but not different from P. lofotensis. If this species is indeed closely related to the genus Pulsellum and can accordingly be grouped within the Pulsellidae, the paleontological record of the family would not begin in the Paleocene but earlier in the middle Jurassic. Origin for the Pulsellidae in this period is in much better accordance with the earliest record of its proposed sister-group, the family Gadilidae (Steiner, 1992) dating from the early Cretaceous (Emerson, 1962). Comparing the coefficients of variation (CV) (Table 2, Fig. 9), it becomes clear that three measurements are almost constant (CV 3.38-10.78) in all four species. These are all measurements of width: maximum protoconch diam- eter, and the initial diameters of teleoconch (terminal width of protoconch) and teleoconch B. Although lecithotrophic larvae are definitely known from the genus Antalis only (Lacaze-Duthiers, 1856-1857; Kowalevsky, 1883), there is 96 AMER. MALAC. BULL. 11(2) (1995) tcB tcA pcB pcA | | | | | D Fig. 11. Comparison of the types of larval and juvenile scaphopod shells. A. Dentaliida (Antalis, Fissidentalium, Episiphon). B-C. Gadilida: Entalimorpha and Gadilimorpha: Pulsellidae (B. Entalina, Heteroschismoides, C. Entalinopsis, Baltodentalium, Suevidontus ; Pulsellum not figured). D. Cadulus subfusiformis (Gadilida: Gadili- morpha: Gadilidae). (Pa, protoconch A; Pb, protoconch B; Ta, teleoconch A; Tb, teleoconch B). no indication that Pusellum lofotensis and Cadulus sub- fusiformis differ in their development (Steiner, unpubl. obs). The small variability of the maximum and terminal protoconch diameters is probably a consequence of lecithotrophic development with its limited resources of matter and energy. The maximum protoconch diameter is directly correlated with egg diameter, the terminal diameter of the protoconch perhaps only in regard to the limited nutrients (see below). The uniformity of initial teleoconch B diameter, however, is unlikely to be due to the available amount of yolk, unless teleoconch A is secreted prior to the onset of feeding. According to Lacaze-Duthiers (1856- 1857), definite mouth and anal openings are present after nine days of development, but he distinguished the adult gut topology only after 30 days because of the opaque yolk remnants. Thus, it is not clear when feeding actually com- mences. Therefore, it is not impossible that in Antalis occi- dentalis and Entalina quinquangularis, teleoconch A 1s secreted under non-feeding conditions. Although also secreted in the presumed lecithotrophic phase, protoconch length measurements are generally more variable having CVs between 6.71 and 32.99, L-pc B being the least con- stant. The exception of Pulsellum lofotensis, where the CV of L-pc A exceeds that of L-pc B, is explained by the high- er number of apically truncated specimens rather than by the variability of the complete shell. If limitation of yolk is the only determining factor for protoconch measurements, why are lengths more variable than diameters? In addition to resource limitation, one can assume genetically fixed, species-specific thresholds in diameter, at which the next growth phase is initiated. Such thresholds may also exist for teleoconch features and appear to be a more plausible explanation of why D-tc Bi is one of the most constant measurements, but L-tc A is not. Another example would be the constriction and subsequent formation of teleoconch B in C. subfusiformis after the determined D-tc A of 236 yim is reached. Apart from already discussed features, Cadulus sub- fusiformis differs from the other species also by its hardly variable total shell length (CV 8.39), while its teleoconch B length has the highest CV (70.59) of all measurements taken in this study. There appears to be a high synchroniza- tion of shell growth at the anterior end and shell truncation at the apex. Consequently, teleoconch A of up to 1 mm length is lost before teleoconch B of about 2 mm length is completed. The available data on larval shell morphology permit a systematic correlation of protoconch types at the ordinal level. Type 1 protoconch is characteristic of the Dentaliida and type 2 of Gadilida except for the genus Cadulus that shows type 3. The mode of shell development of G. turgida with a weakly differentiated protoconch A may be interme- diate between the Gadilida-type and Cadulus. This correlation brings new light to some problems regarding classification of conchological material, particu- larly fossil scaphopods. Engeser et al. (1993) described two fossil species with protoconchs, Suevidontus jaegeri and Baltodentalium weitschati, but left their family assign- ment open. According to the morphological and morpho- metric results presented here, they belong to the order Gadilida. Their type 2 larval shell places them in the order Gadilda, and their conspicuous longitudinal sculpture of teleoconch B is characteristic for some genera of the subor- der Entalimorpha like Entalina or Entalinopsis. Both S. jaegeri and B. weitschati have a very short (or absent?) teleoconch A resembling Entalinopsis callithrix and Entalinopsis sp. (Fig. 11C). Although polarities of proto- conch character states are far from clear, it is plausible that the short (or absent) teleoconch A is a synapomorphy of these four species and suggests close phylogenetic relation- ships. There is little albeit varying information on the teleoconch microstructure of Scaphopoda. Only a brief review of the literature is given here; future investigations will need to cover shell structure comparatively on a broad STEINER: SCAPHOPOD LARVAL AND JUVENILE SHELLS 97 taxonomic base. One organic and two to four aragonitic layers may be present. A distinct periostracum is reported in recent papers for Dentalium rectum, Pictodentalium vernedei (Sowerby, 1860) (fide Haas, 1972), and Antalis vulgaris (DaCosta, 1778) (= tarentinum) (fide Alzuria, 1985). This purely organic layer seems to be lacking in Rhabdus rectius (Reynolds, 1992) and in all Gadilida examined so far (Shimek and Steiner, in press). This is most surprising in the suborder Gadilimorpha where most of the shells are smooth and lusterous. However, three of the 60+ specimens of Cadulus subfusiformis examined in this study had a thin, homogenous, probably organic outer shell layer. Reasons for the lack of the periostracum could be the ethanol preservation or, more likely, early and rapid erosion, because it is also lacking in air-dried shells. The outermost calcareous component is prismatic, with prisms oriented perpendicular to the shell axis (lacking in R. rec- tius). It may rest on a thin amorphous (Shimek and Steiner, in press) or complex crossed lamellar layer (Haas, 1972), followed by an ordinary crossed lamellar layer. Towards the internal lumen, the lamellae may become narrower and more aciculate-like (Haas, 1972; Alzuria, 1985). Especially in the apical teleoconch region, there may be a layer of short prisms perpendicular to the shell axis, a myostracum- like structure. The ontogenetically early shells described above have only one or two layers. The outer prismatic layer is always present. It becomes coated from the inside when the mantle epithelium starts to secrete the crossed lamellar structure. The calcification of the protoconch is believed to take place before (Lacaze-Duthiers, 1856-1857) or at metamorphosis (Engeser et al., 1993). Meta- morphosis, however, does not correlate with ventral man- tle/shell fusion, which occurs during the secretion of proto- conch B. The obliteration of the velum-like larval locomo- tory organ is a good indicator of the transition to adult orga- nization. It loses its locomotory function as it becomes enclosed by the elongating tubular shell (Lacaze-Duthiers, 1856-1857). The absence of growth lines on the protoconch indicates that it is formed by the shell gland and/or mantle as a purely organic structure and that it calcifies at once (Bandel, 1982; Engeser et al., 1993), as in many Archaeogastropoda and pteriomorph Bivalvia (Bandel, 1982; Waller, 1981). Another similarity to these groups is the in toto calcification of the organic protoconch, where the crystals grow from surface to surface, leaving no unmineralized periostracum (Bandel, pers. comm.). This result, in particular, underlines the primitive position of Scaphopoda among the Conchifera in regard to larval development. On the other hand, there are increment lines on the fumarium and the suture indicating their origin as mineralized shell components. Again, we need develop- mental analyses to decide whether these two structures are added to the shell during or after protoconch secretion and mineralization. Which type of larval shell represents (or is closer to) the ancestral state is hitherto uncertain. The Cadulus-type is probably not the plesiomorphic state, because larval shells with a bulging protoconch A are present in dentaliids and both suborders of gadilids. In addition, form and microstructure of the Cadulus teleoconch itself seems to be highly derived. Consequently, the bulging larval shell with a ventral gap found in Dentaliida and both clades of Gadilida is likely to be plesiomorphic. Engeser et al. (1993) assumed a longer larval and swimming phase for dentaliids with their longer type-1 protoconch, although there are no signs, whatsoever, that Scaphopoda have planktotrophic lar- val stages (Engeser et al., 1993; Steiner, unpubl. obs.). ACKNOWLEDGMENTS I am indebted to Jon-Ame Sneli for his invitation to Trondheim Biological Station and his help there. Ronald L. Shimek (Wilsall, Montana), besides providing other helpful comments on the manuscript, encouraged me to attempt the quantitative analysis, and Hans L. Nemeschkal (University of Vienna) guided me through it. I thank Luitfried Salvini-Plawen (University of Vienna), Klaus Bandel (University of Hamburg), Theo Engeser (University of Hamburg) and Thomas R. Waller (Smithsonian Institution, Washington, DC) for their most valuable critical comments. Mariti Rauscher (University of Vienna) was a great help with the English. This study was part of project no. P8522-BIO financed by the Austrian Fonds zur Forderung der wissenschaftlichen Forschung. LITERATURE CITED Alzuria, M. 1985. Ultrastructura de la concha en Dentalium vulgare (DaCosta, 1778) (Mollusca; Scaphopoda). /berus 5:1 1-19. Bandel, K. 1982. Morphologie und Bildung der friihontogenetischen Gehause bei conchiferen Mollusken. Facies 7:1-198. Burch, B. L. and T. A. Burch. 1989. On the retention of the embryonic shell of Fissidentalium phaneum Dall from Hawaii (abstract). Program and Abstracts, 55th Annual Meeting, American Malacological Union:29. Cotton, B. C. and F. K. Godfrey. 1940. The molluscs of South Australia: Scaphopoda, Cephalopoda, Aplacophora and Crepipoda. Jn: Handbook of the Flora and Fauna of South Australia Part IT. H. M. Hale, ed. pp. 317-341. Government Printer, Adelaide. Davies, G. J. 1987. Aspects of the Biology and Ecology of Deep-Sea Scaphopoda (Mollusca). Doctoral Dissertation, Heriot-Watt University, Edinburgh, Scotland. 194 pp. Emerson, W. K. 1962. A classification of the scaphopod molluscs. Journal of Paleontology 36:461-482. Engeser, T. S., F. Riedel, and K. Bandel. 1993. Early ontogenetic shells of Recent and fossil Scaphopoda. Scripta Geologica, Special Issue 2:83-100. Haas, W. 1972. Micro- and ultrastructure of Recent and fossil Scaphopoda. /nternational Geological Congress 24, sect.7: 15-19. Henderson, J. B. 1920. A monography of the East-American scaphopod Mollusca. United States National Museum Bulletin 111:1-177. 98 AMER. MALAC Johnson, R. A. and D. W. Wichern. 1988. Applied Multivariate Statistical Analysis, 2nd ed. Prentice Hall, Englewood Cliffs, New Jersey. 607 pp. Kowalevsky, A. 1883. Etude sur l’embryogénie du Dentale. Annales du Musée d'Histoire Naturelle de Marseille 1(7):54 pp. Lacaze-Duthiers, H. 1856-1857. Histoire de l’organisation et du développement du Dentale. Annales des Sciences Naturelles 4, Zool.:6:225-281; 7:5-51, 171-255; 8:18-44. Lande, R. 1977. On comparing coefficients of variation. Systematic Zoology 26:214-217. Ludbrook, N. H. 1960. Scaphopoda. Jn: Treatise on Invertebrate Paleontology, Part I, Mollusca 1, R. C. Moore, ed. pp. 137-141. Geological Society of America and University of Kansas Press, Lawrence, Kansas. MacNeil, F. A. and D. T. Dockery III. 1984. Lower Oligocene Gastropoda, Scaphopoda, and Cephalopoda of the Vicksburg Group in Mississippi. Bulletin of the Mississippi Department of Natural Research 124:1-415. Reynolds, P. D. 1992. Mantle-mediated shell decollation increases posteri- or aperture size in Dentalium rectius. The Veliger 35:26-35. Ruggieri, G. 1987. La protoconca del Dentalium rectum. Bolletino Malacologico 23:413-416. Sars, M. 1865. Malakozoologiske Jagtagelser. II. Nye Arter af Slaegten . BULL. 11(2) (1995) Siphonodentalium. Forhandlinger Videnskabs-Selskabet i Christiania 1864:296-315. Scarabino, V. 1979. Les Scaphopodes Bathyaux et Abyssaux de l’Atlantique Occidental: Nouvelle Classification pour I’ Ensemble de la Classe. Doctoral Dissertation, Université d’ Aix-Marseille. 154 pp. Shimek, R. L. and G. Steiner. In press. Scaphopoda. In: Microscopic Anatomy of Invertebrates, F. W. Harrison and A. Kohn, eds. Liss Inc., New York. Sokal, R. R. and C. A. Braumann. 1980. Significance tests for coefficients of variation and variability profiles. Systematic Zoology 29:50-66 Steiner, G. 1991. Observations on the anatomy of the scaphopod mantle, and the description of a new family, the Fustiariidae. American Malacological Bulletin 9:1-20. Steiner, G. 1992. Phylogeny and classification of Scaphopoda. Journal of Molluscan Studies 58:385-400. Waller, T. 1981. Functional morphology and development of veliger lar- vae of the European oyster, Ostrea edulis Linné. Smithsonian Contributions to Zoology 328: 1-70. Zar, J. H. 1984. Biostatistical Analysis, 2nd ed. Prentice Hall, Englewood Cliffs, New Jersey. 718 pp. Date of manuscript acceptance: 23 October 1994 The relevance of passive dispersal for the biogeography of Caribbean mollusks’ Rudolf S. Scheltema Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543, U.S. A. Abstract. Occurrence of teleplanic veliger larvae of sublittoral species in major currents of the tropical Atlantic Ocean supports the hypothesis that pas- sive dispersal of larvae contributes importantly to the biogeography of shoal-water molluscan species. Rafting may play a secondary role. Criticisms of pas- sive larval dispersal as a factor in molluscan biogeography are: (1) that larvae have a fixed life span too short to account for long-distance dispersal: both laboratory and field data show this assertation to be mistaken, (2) that larvae after a time lose their competence to metamorphose: whenever tested, teleplanic larvae have been shown to retain their ability to metamorphose; (3) that dispersal is random and cannot explain congruent nonrandom associations among widely differing taxa: dispersal is largely accounted for by advection along major ocean currents which provide quasi-permanent or seasonally reoccurring corridors for the transport of planktonic larvae; therefore various taxa necessarily are dispersed over similar routes, leading to congruent geographic distribu- tions; and (4) that some species lacking long planktonic larval stages nevertheless have wide geographic ranges and consequently there is little relationship between mode of reproduction and geographic range of species: those species lacking extended planktonic larval stages yet having wide geographic ranges are mostly sessile epibenthic forms that can attach to hard substrata, and consequently are preadapted for passive dispersal by rafting. Dispersal by human agencies also has contributed within historic time to the geographic distribution of marine mollusks. Such alternative modes of dispersal do not negate the importance of larval transport. Despite demonstrable evidence for widespread passive larval dispersal and in some instances also for the transport by rafting of molluscan species, ecological constraints place restrictions upon where and when new colonists can survive and reproduce. Geotectonics and sea-floor spreading have been significant in controlling the pattern and extent of passive dispersal over geologic time. The clos- ing of the Tethys Seaway during the Oligocene and Early Miocene resulted in the isolation of the southwestern tropical Pacific Ocean from the Mediterranean Sea and tropical eastern Atlantic Ocean. Similarly, the closing of the corridor between North and South America divided the tropical eastern Pacific Ocean from the Caribbean Sea. Continental drift and sea-floor spreading resulted in the initial formation and subsequent enlargement of the Atlantic Basin and has led subsequent- ly to the “mid-Atlantic barrier” which today acts as a filter between the tropical eastern Atlantic Ocean and Caribbean Sea. Both tectonic events and sea- floor spreading have placed important constraints on the dispersal between tropical faunas. Faunal studies of the amphi-Atlantic distributions of marine benthic mollusks further support the hypothesis of passive dispersal. Low endemism on oceanic islands suggests that initial colonization must be largely accomplished by teleplanic larvae. Available evidence shows that although no single process can completely explain the present composition of the Caribbean molluscan fauna, passive dispersal of planktonic veliger larvae must have played and continues to play an important role both in colonization and in maintaining genetic continuity between widely disjunct regions of the tropical Atlantic Ocean. Recent new techniques now available which reveal enzyme variation and mitochondrial DNA polymorphism can revolutionize the study of bio- geography. They can make possible, for example, the identification of closely similar larvae in instances where morphological characteristics are inadequate and can allow measurements of genetic exchange or gene flow and the genetic relationships between widely separated populations. Cladistic analysis, although revealing little about the processes leading to present geographic distributions, can help reconstruct large-scale geographic relationships as related to the evolution of taxa. “It requires a very unusual mind to undertake analy- mollusks. It is not enough simply to make inferences about sis of the obvious.” A. N. Whitehead. the origin of existing patterns without explicit evidence for To understand the origin of the Caribbean sublittoral the processes that have been responsible for these observed molluscan fauna requires not only a comparison among _ Patterns. spatially disjunct tropical regions throughout the world but In this paper I will: (1) summarize evidence for pas- also a consideration of the processes, both biologic and sive dispersal of mollusks in Atlantic tropical waters, in geologic, which may have brought about and continue to particular data on the transport of planktonic larvae and on influence the observed geographic distributions of tropical the rafting of sessile epibenthic species; (2) consider objec- tions that have been raised against such passive dispersal to I This review is based on a contribution to the 1992 AMU Symposium on explain the observed geographi emismibubons 2) ead Biology of Caribbean Mollusks. See American Malacological Bulletin rize the relevance of geotectonic events such as the closing 10(2):179 ff. of seaways and sea-floor spreading and make inferences American Malacological Bulletin, Vol. 11(2) (1995):99-115 99 100 AMER. MALAC. BULL. 11(2) (1995) about how these geologic processes could have affected passive dispersal and influenced present-day distributional patterns of tropical mollusks, in particular the Caribbean molluscan fauna, and (4) discuss other evidence that sug- gests that passive long-distance dispersal has and continues to play an important role in the initial colonization and sub- sequent genetic exchange between widely disjunct popula- tions of Atlantic tropical molluscan species (Scheltema, 1968, 1971b, 1992). THE ROLE OF PASSIVE DISPERSAL IN ESTABLISHING GEOGRAPHIC DISTRIBUTION OF TROPICAL WESTERN ATLANTIC MOLLUSKS There are three principal means by which sublittoral mollusks can be passively dispersed, namely: (1) by the transport of their planktonic larvae, (2) by “rafting” on drifting objects, and (3) by the activities of humans within historic time, i.e. approximately the past 500 years. PASSIVE DISPERSAL OF LARVAE Passive long-distance dispersal of planktonic larvae in tropical waters of the Atlantic Ocean has been the object of continuing studies for more than two decades (Schel- tema, 1964, 1968, 1971a, b, 1986; Mileikovsky, 1966; Laursen, 1981). Four principal currents potentially can serve to transport larval forms of tropical sublittoral species over long distances (Fig. 1). The North and South Equatorial Currents (Fig. 1, NEC and SEC) flow from east to west, i.e. from West Africa to South America; both are surface currents usually extending no deeper than 5O m. Velocities of the North Equatorial Current have been esti- mated at 25 to 35 cm/sec, while that of the South Equatorial Current ranges from 25 to 90 cm/sec in some regions. An Equatorial Undercurrent (Fig. 1, EUC) passes from west to east directly beneath the westwardly flowing South Equatorial Current at depths greater than 50 m (Metcalf er al., 1962); its maximum velocity occurs between 50 and 100 m depth and sometimes exceeds 100 cm/sec (Fig. 2). Within the undercurrent, water temperature drops from ca. 26°C at 50 m to 20°C at 100 m. Such temperatures are suf- ficiently warm to sustain larvae of most tropical mollusks. A fourth current, the Equatorial Countercurrent (Fig. 1, ECC) is found in the region between 5° and 10° N latitude. It is seasonal in its occurrence, flows from west to east toward the coast of West Africa, and is fully developed between June and August when it can attain a velocity greater than 35 cm/sec. (Richardson, 1984). In the fall it diminishes, then disappears completely in January to reap- pear subsequently in early summer (Fig. 3). Fig. 1. Schematic illustration of major ocean currents of the tropical Atlantic Ocean. ECC, seasonally occurring Equatorial Countercurrent flowing from the western to the eastern equatorial Atlantic (see Fig. 3); EUC (large open arrows), Equatorial Undercurrent flowing from Brazil to West Africa at a depth below 50 m (see Fig. 2); NEC, North Equatorial Current flowing from northwestern West Africa to the Caribbean; SEC, South Equatorial Current flowing from West Africa to Brazil. SCHELTEMA: PASSIVE DISPERSAL IN BIOGEOGRAPHY 101 100 Depth (m) 200 Fig. 2. North-south section across the Equator at 13°30, W illustrating the extent of the Equatorial Undercurrent. Continuous, unbroken isopleths denote current shear in cm/sec and illustrate relative differences in current velocity at various depths. Direction of undercurrent is toward the east (into the Fig.). Vertical lines show position of stations. Dashed lines indi- cate 26° and 20° C isotherms. (Based on Metcalf et al., 1962:2503, fig. 6). Consider now the distribution of teleplanic or “long- distance” veliger larvae of sublittoral species in relation to this tropical North Atlantic circulation. Among the most common molluscan larvae that have been encountered are those of prosobranch gastropods. At least 20 families are represented in samples within the upper 100 m. Two espe- cially prominent are the Architectonicidae and the Ranellidae (=Cymatiidae). For example, among the Architectonicidae larvae of approximately 30 recognizable species have been encountered in the tropical and warm temperate North Atlantic Ocean. Study of the adult forms has shown that many Architectonicidae have very wide geo- graphic distributions (Merrill, 1970; Garcia-Talavera, 1982; Bieler, 1993); indeed Bieler et al. (1986:286) ventured that “most species are ... amphi-Atlantic.” However, only six species of architectonicid larvae have actually been described and related to their adult forms by a comparison of their larval shell to the protoconch of identified post-lar- vae; two of these are Atlantic species, viz. Philippia krebsii (Morch, 1875) and Architectonica nobilis Réding, 1798. Both are amphi-Atlantic in their geographic distribution (Scheltema, 1971b; Laursen, 1981). Thirty-four species of Ranellidae are known from the tropical Atlantic Ocean and among these over half have an amphi-Atlantic distribution (Garcia-Talavera, 1987). Larvae of 12 Atlantic species have been described (Scheltema, 1971b; Laursen, 1981). The distributions of veliger larvae belonging to the Architectonicidae and Ranellidae obtained from plankton tows taken in the tropical Atlantic are shown in Fig. 4. Among bivalve mollusks, the veliger larvae of the members of Pinnidae are among the largest known, some attaining a length of 700 pm before metamorphosis (Yoshida, 1956; Ota, 1961; Booth, 1979) and are readily recognized by their characteristic triangular-shaped valves. Eight species assigned to two genera, viz. Pinna and Atrina, are known from tropical and temperate Atlantic waters, and their geographic distributions were briefly described by Scheltema and Scheltema (1984). Two tropical species belonging to the genus Pinna are amphi-Atlantic in distrib- ution; in contrast, all species of the genus Atrina are broad- ly distributed either along the coastlines of South America or West Africa, but none occur on both sides of the tropical Atlantic Ocean. Veligers of the wood-boring bivalve family Teredinidae also occur in epipelagic waters of the tropical and warm temperate Atlantic Ocean (Scheltema, 1971c). Identification is possible from the configuration of the hinge teeth and shape of the valves, i.e. greater dorsoven- trally than anteroposteriorly. Other bivalve veligers known from the Atlantic epipelagic zone include the Erycinidae, Limidae, Lucinidae, Mytilidae, Ostreidae, Pholadidae, and Velocity (cm/sec) -10 JIFMAMJIJ AS ON D Month Fig. 3. Seasonal variation in mean monthly velocity of the Equatorial Countercurrent determined from ship’s drift in the region bounded by 5° and 8° N latitude and 25° and 30° W longitude. Positive values indicate eastward velocity in cm/sec; negative values indicate westward flow. The countercurrent is at its maximum between June and November when its velocity ranges between 10 and 30 cm/sec. Evidence from moored current meters and from drifting buoys during 1983 showed peak velocities up to 40 cm/sec in June and July, i.e. earlier and with higher maximum veloci- ties than those obtained from the set of ships. (Modified after Richardson, 1984:748, fig. 4). 102 AMER. MALAC. BULL. 11(2) (1995) Pteriidae. The distribution of bivalve veligers encountered in the tropical Atlantic Ocean is shown in Fig. 5. The direction of mean net transport of larvae is along the axis of the major ocean currents. It cannot be assumed however that planktonic larvae will always behave as com- pletely passive particles. The possibility remains that some larvae may control their vertical distributions and if this is so, they may utilize the Equatorial Undercurrent (Fig. 2) which moves in the opposite direction to that near the sur- face. There is at present, however, only little known about the behavior of teleplanic larvae (Scheltema, 1972a). Evidence from the observed distributions of teleplan- ic larvae and from advection by ocean currents provides a compelling argument for long-distance larval dispersal of tropical molluscan species. Minimum and maximum aver- age current velocities of 25-100 cm/sec can result in pas- sive dispersal of larvae between 22 and 90 km/day and crossing of the tropical Atlantic Ocean (ca. 5000 km) in 55- 195 days or a mean of about four months. What is the likelihood that a larva will be transported passively across the Atlantic Basin? The drift coefficient for the tropical and warm temperate Atlantic Ocean (i.e. the likelihood that a specific larva will be successfully dis- persed across the Atlantic Basin) falls between 0.01 and 0.001. These values were derived from drift bottle data (Scheltema, 1971b). Although the odds seem very small, they must be related to fecundity, which in most gastropods with planktonic development is very high and, between 10*-10° eggs/female/yr. Among bivalves, this value can be even higher, > 107 eggs/female/yr. Although the likelihood that any particular larva will be transported across the tropi- cal Atlantic Ocean is very small indeed, there is a reason- able chance that at least some will succeed. For example, for a species that produces 0.5 million eggs/female/yr, at a drift coefficient of 0.01, between 5 and 50 larvae/female/yr will be successful, even when allowing 99% mortality from predation (Scheltema, 1971b). Plankton samples confirm that larvae actually are advected over great distances by ocean currents (Figs. 4 and 5) and laboratory observations show that teleplanic larvae taken from the plankton in fact can delay settlement and metamorphosis for periods up to five months after collection (Scheltema, 1971b). RAFTING An alternative to passive dispersal by means of planktonic larvae is the transport of juvenile and adult organisms by rafting on floating objects. Drifting materials that serve as rafts include plant remains such as logs, parts of fruits [e.g. coconut husks, seed pods and pits (Guppy, 1917; Gunn, 1968), mangrove leaves (Wehrtmann and Ditell, 1990)] or drifting algae (Vallentin, 1895; Arnaud et al., 1976; Highsmith, 1985). Algae could be less effective Fig. 4. Distribution of sampling locations where gastropod veliger larvae were encountered. Large filled circles, locations where larvae of the family Architectonicidae were found. Large open circles, locations where larvae of Ranellidae were encountered. Large divided circles, locations where both Architectonicidae and Ranellidae were found in the same sample. Triangles, locations where veliger larvae of 18 other gastropod families have been taken in plankton nets including Naticidae, Cypraeidae, Thaididae, Triphoridae, Coralliophilidae, Bursidae, and Tonnidae. Small open circles, locations where gas- tropod veligers were absent. (Redrawn after Scheltema, 1986:300, fig. 4.) SCHELTEMA: PASSIVE DISPERSAL IN BIOGEOGRAPHY 103 60° 0° Fig. 5. Distribution of sampling locations where the bivalve veliger larvae were encountered. Large filled circles, locations where larvae of the family Pinnidae were encountered. Large open circles, locations where Teredinidae larvae were found. Large divided circles, locations where both pinnid and terri- dinid larvae co-occurred. Triangles, locations where larvae of other bivalve families occurred. Small open circles, sampling locations lacking bivalve larvae. in tropical waters than at high latitudes where gastropods and bivalves can be transported in the holdfasts of enor- mous kelps. Carquist (1974) gave a useful summary of plants which potentially serve as rafts. Bivalves that are cemented to hard surfaces (e.g. oysters; Stenzel, 1971), that are attached by a byssus (e.g. mussels, Pinnidae), or indeed most newly settled bivalve species, can be rafted on almost any floating object. The Teredinidae, or shipworms, and other wood-boring mollusks are readily dispersed on float- ing timber (Turner, 1966). Lyrodus pedicellatus (Quatrefages, 1849), which has a planktonic larval life of about one day, has nonetheless a circumtemperate and trop- ical distribution. It has been suggested by Marche-Marchad (1968) that the wide geographic distributions of large volu- tid gastropods of the genera Cymba (= Cymbium) and Adelomelon which lack planktonic larval stages, can be explained by the dispersal of their egg capsules attached to drifting objects. Another form of rafting reported by Frazier et al. (1985) is the transport of bivalves and gas- tropods on the carapaces of sea turtles. Pumice that originates from volcanic eruptions can serve as floats for a variety of invertebrate organisms (Scheltema, 1977). Floating pieces varying in size from a few millimeters to two meters in diameter were reported by Jokiel (1989), and Fushimi et al. (1991) found concentra- tions up to 500 per 1850 m2 in the Kiroshio, North Equatorial, and Subtropical Counter Currents. Through knowledge of its relative composition, (viz. Na2O and Kr20O; FeO, TiO2 and MgO; or ZnO, CuO and SrO), it is possible to infer the origins of drifting pumice by compari- son with material arising from known volcanic distur- bances. Jokiel (1984) made observations on reproductively mature colonies of the coral Poecillopora which he judged to be two to three years of age and also the calcareous encrusting remains of oysters, serpulid worms, and calcare- ous algae attached to drifting pumice (Jokiel 1989, 1990a). From all of this anecdotal evidence, it seems that rafting must play a roll in the passive dispersal of at least some molluscan species, but such transport necessarily must be restricted to forms that spend most of their adult life on hard substrata. It is highly unlikely that infaunal “soft-bottom” molluscan species will ever be rafted over long distances. This is because of their inability to survive for long periods outside their normal habitat and the improbability that they can remain affixed to drifting objects for significant periods of time. Unlike planktonic larvae which may be dispersed by sub-surface currents (e.g. the Equatorial Undercurrent), rafting is restricted entirely to the surface. It seems probable that rafting plays only a sec- ondary role in the dispersal of mollusks. HUMAN ACTIVITIES Finally, there is the possibility of dispersal of mol- luscan species by the activities of humans. Both attach- 104 AMER. MALAC. BULL. 11(2) (1995) ment of mollusks to the hulls of ships and the practices used in shellfish culture have led to the dispersal of numer- ous gastropod and bivalve species. Recently, ballast water of ships has also been implicated in the transport of plank- tonic larvae over great distances. Carleton (1985) and Williams et al. (1988) have reviewed and given evidence for the importance of human intervention to the geographic distribution of marine invertebrate species. However, such introductions of species are probably only of marginal importance in the historical biogeography of the Caribbean as they include only a very restricted period of time (ca. 500 yrs). OBJECTIONS TO THE HYPOTHESIS OF PASSIVE LONG-DISTANCE DISPERSAL Four major criticisms have been raised against pas- sive dispersal of planktonic larvae as a significant factor in determining the patterns of geographic distribution among benthic species. The first of these was most recently expressed by Jokiel (1990b:60) who maintained that “lar- vae generally have a fixed life span” and “the majority of invertebrates ... have larval competence periods of only a few weeks.” Jokiel concluded that “even the most rapid ocean currents can transport drifting objects only a few hundred km during the life span of a typical larva.” This view has its origin in a paper by Thorson (1961:472) who concluded that “eighty percent of all bottom invertebrates with pelagic larvae have a planktonic life of less than six weeks ... [and therefore] even if ... larvae are transported by especially rapid currents, they never seem to have even the slightest chance to cross the larger ocean basins.” Thorson (1961) believed that only 5.5% of invertebrates had a potential planktonic life exceeding twelve weeks. However, this conclusion was based upon cold-temperate, North Sea and Baltic Sea species, and Thorson himself (1961:473) allowed that tropical species can differ from such cold-temperate forms. Thorson’s data on mollusks were quite limited (Thorson, 1961:460, fig. 1). Only 18 species of prosobranch gastropods were listed, and among these none had a development exceeding nine weeks (most less than five weeks). Now 30 years later, at least 20 fami- lies of gastropods are known to have long-distance larvae; indeed within the family Architectonicidae alone, more than 30 species of teleplanic larvae have been collected in the open waters of the North and South Atlantic Oceans. Thorson (1961:466) was actually aware of at least six fami- lies of “pronounced long-distance larvae among proso- branchs” and guessed that these “might last for six months,” but somehow he did not include these in his tabu- lations. Subsequent observations made in the laboratory have shown that at least some tropical gastropod veligers taken far out at sea and probably already several months old when captured can delay settlement for an additional period exceeding four months duration (Scheltema, 1971b). Bivalves also were considered by Thorson. Data for 37 species led him to conclude that most bivalves have planktonic larvae for less than five weeks. However, this estimate, as that for gastropods, was based upon Baltic Sea and North Sea species and ought not to be extrapolated to tropical forms. Bivalve veligers belonging to sublittoral species and taken in the Gulf Stream, Sargasso Sea, North and South Equatorial Currents, and the Canary, Brazil, and Antilles Currents, suggest from their position at the time of capture that they also must have been adrift very much longer than the five weeks inferred by Thorson. Contrary to Jokiel’s (1990b:60) assertion, most invertebrate larvae do not “have a fixed life span.” In fact, the duration of planktonic development is usually very flex- ible. The length of the “precompetent” stage, which is the time of larval growth and development, is thought to be determined largely by seawater temperature and by the amount and “quality” of available food, i.e. by both the species and “condition” as well as by the number of such food organisms (e.g. for bivalves see Davis and Guillard, 1958; Bayne, 1965; Loosanoff et al., 1951; Stickney, 1964; Pechenik et al., 1990; Sprung, 1984; for gastropods see Pilkington and Fretter, 1970; Scheltema, 1967; Pechenik and Fisher, 1979; Lima and Pechenik, 1985). When the “competence” to settle has been achieved, often there may be, in the absence of an appropriate cue, a further delay in settlement and metamorphosis. Such a delay was demon- strated for prosobranch gastropods in Nassarius obsoletus (Say, 1822) by Scheltema (1956, 1961) and in Phestilla sibogae Bergh, 1905, by Hadfield and Karlson (1969); delay in metamorphosis has also been established for bivalves, viz. Ostrea edulis Linné, 1758, by Cole and Knight-Jones (1949), and Knight-Jones (1951), and for Crassostrea virginica (Gmelin, 1791) by Crisp (1967). The generality of such settlement responses has been amply confirmed experimentally for numerous other molluscan species. A particularly striking example of the prolonga- tion of planktonic larval life in the absence of an appropri- ate cue is that of the gastropod Aplysia juliana (Quoy and Gaimard, 1832) which in the laboratory becomes compe- tent to settle after only 30 days but in the absence of a cue (in this instance a green alga), can delay settlement for more than 200 days (Kempf, 1981)! It should be obvious that the length of the delay in settlement and metamorpho- sis could vary widely if a cue is not encountered although there is some evidence that with increasing lengths of time in the plankton the larvae of some invertebrates may become less discriminating and respond to alternative and presumably less favorable cues. The statement by Jokiel (1990b:66) that “even the SCHELTEMA: PASSIVE DISPERSAL IN BIOGEOGRAPHY 105 most rapid currents can transport objects only few hundred kilometers during the life span of a typical larva” ignores a large body of existing evidence. This question was addressed almost 30 years ago (Scheltema, 1966:87, table 1) when it was demonstrated that the surface current veloci- ty was well within the time required for the larva of the gastropod Cymatium parthenopeum (von Salis, 1793) to cross the North Atlantic to the Azores. The current veloci- ties in the North and South Equatorial Currents are quite sufficient to transport larvae from West Africa to South America in four to six months; likewise both the Equatorial Undercurrent and Equatorial Countercurrent (during sum- mer months) are sufficient to passively disperse larvae in the opposite direction from South America to West Africa. A second criticism has been expressed by Mitton et al. (1989:360) who noted that it is not known “whether a significant proportion of larvae found in the middle of the ocean are [still] capable of metamorphosis and survival,” and hence whether teleplanic larvae can contribute to downstream populations. Laboratory experimenis have shown that at least some molluscan larvae taken alive from the open waters of the Atlantic Ocean retain the capacity to metamorphose. For example, veligers of gastropod species in the families Architectonicidae, Ranellidae, Coral- liophilidae, Cypraeidae, Naticidae, Neritidae, Ovulidae, Strombidae, Tonnidae, Triphoridae, among others, as well as in the bivalve families Pinnidae, Erycinidae, Limidae, Mytilidae, Ostreidae, Pholadidae, Pteriidae, and Thyasiridae have been shown to metamorphose in the labo- ratory when removed from samples obtained hundreds of kilometers out at sea (Scheltema, unpub. data). Clearly additional observations on a wider spectrum of molluscan species and families would be useful. Nevertheless, obser- vations up to now show that in most known instances teleplanic larvae do in fact retain their ability to metamor- phose. A third objection is the claim by vicariance biogeog- raphers that long-distance or “jump” dispersal cannot explain patterns of distribution or the relationships between widely separated faunas because such a process is random and therefore cannot account for the congruent, nonrandom geographic associations found among numerous, widely differing taxa. For example, Croizat et al. (1974:267) maintained that the “means of dispersal ... vary from species to species and do not explain patterns of biotic dis- tribution.” This assertion is arguably true for terrestrial forms but is irrelevant for a wide spectrum of sublittoral marine mollusks. This is because the principal means of dispersal among terrestrial species is active migration by adults (Scheltema, 1986, 1989a), whereas most tropical sublittoral or benthic marine molluscan species (ca. 70- 80%) have planktonic larval stages which are passively dis- persed (Thorson, 1950). It is true that passive dispersal can result from both eddy diffusion and advection. However, eddy diffusion is the random dispersal that results from turbulent flow. Its horizontal component along a two-dimensional surface can be measured by the mean change in distance between freely drifting objects. Dye experiments in a number of locations in the North Atlantic Ocean have shown that over the peri- od of one month, eddy diffusion alone accounted on aver- age for maximum dispersal to within a radius of about 100 km (Okubo, 1971). But the larva dispersed passively by eddy diffusion theoretically can be found anywhere within such a 100 km radius, i.e. both the direction and rate of dis- persal will be unpredictable. Thus while 100 km defines the upper limit of dispersal, most larvae will be transported over much shorter distances (see Okubo, 1994, for a more comprehensive and mathematical treatment of eddy diffu- sion and its role in larval dispersal). Advection, in contrast, is the non-random, directed, horizontal transport within a defined current regime and is the horizontal displacement with respect to a set of fixed coordinates. The direction of the mean net transport will be along the axis of the current. The rate of flow of the major tropical Atlantic circulation, as already related above, ranges from about 25 to more than 100 cm/sec; by compari- son the maximum mean rate of horizontal dispersal from eddy diffusion will scarcely ever exceed 4 cm/sec and its net direction will be random. Because eddy diffusion and advection occur simultaneously, larvae will be both ran- domly dispersed with respect to one another while at the same time advected along the axis of the current. At a bio- geographically relevant scale, it is reasonable to conclude that dispersal results largely from advection along ocean currents; eddy diffusion probably plays only a minor role in the geographic distribution of sublittoral marine molluscan species (Scheltema, 1986:fig. 1) although possibly it has other significant consequences (Strathmann, 1974). Planktonic larvae of widely different species can and do co-occur within the same major ocean currents. The major current systems of the tropical Atlantic Ocean pro- vide quasi-permanent or in some instances seasonably reoc- curring corridors for the transport of molluscan larvae quite irrespective of their taxonomic affinity. All passively advected larvae are necessarily dispersed over similar routes and such transport can be expected, on average, to result in congruent geographic distributions among widely different molluscan taxa. A fundamental difference exists between active dispersal directly related to the behavior of an individual, as commonly encountered among terrestrial organisms, and passive dispersal which depends solely upon the direction and velocity of ocean currents. A fourth criticism due to Turner (1966), Bhaud (1982), Jackson (1986), and most recently Jokiel (1990b), is that many benthic invertebrate species, even though they 106 AMER. MALAC. BULL. 11(2) (1995) lack a long planktonic larval development nevertheless have very large geographic ranges. It already has been shown above that rafting can provide an alternative means of pas- sive dispersal, but also that such a mode of transport neces- sarily is restricted to organisms that can remain attached to hard substrata over long intervals of time (e.g. bryozoans, hydrozoans, barnacles, teredinids, some bivalves, etc.). Excluded, as may be verified by examining floating objects, are infaunal species normally found living within benthic sediments. Further, because the proportion of tropical sub- littoral species that lack planktonic larval stages is relative- ly small, rafting although sometimes effective is limited to only a small proportion of benthic mollusks. Another explanation for large disjunct geographic distributions of mollusks results from human activity, and also has already been considered. The role of human intervention evidently is increasing and is sometimes of great economic impor- tance. Nonetheless, it must be relatively minor when con- sidering the biogeography of the entire Caribbean mollus- can fauna. EVIDENCE FOR ECOLOGICAL FACTORS RESTRICTING THE DISTRIBUTION OF TROPICAL WESTERN ATLANTIC MOLLUSKS Bhaud (1993) argued and concluded correctly that the capacity for dispersal is not the sole condition which contributes to the geographic distribution of benthic species. Despite the readily demonstrable, widespread pas- sive dispersal of larvae, and also in some instances the raft- ing of molluscan species over very long distances, there are also ecological constraints which place restrictions upon where and when newly settled colonizers can survive and reproduce. It is obvious that species well-adapted for the tropics cannot exist at high latitudes; likewise most polar and cold-temperate species will not be able to survive in the tropics. Indeed, temperature sometimes plays a significant role in limiting the latitudinal distribution of benthic species. Such a relationship was already well-established more than a century ago. Darwin (1859:311) remarked that “Change of climate must have had a powerful influence on migration: a region when its climate was different may have been a high road for migration, but now impassible.” Bousfield and Thomas (1975:figs. 1-9) have considered such an instance of climatic change on the post glacial dis- tribution of littoral marine invertebrates along the Canadian Atlantic Region over the past 15,000 years. Hutchins (1947) has usefully discussed the contemporary role of temperature in a monograph entitled “The bases for temper- ature zonation in geographical distribution.” As an illustration, consider the blue mussel Mytilus edulis Linné, 1758, a boreal species which in the western North Atlantic Ocean is found southward along the North American coast as far as Beaufort, North Carolina (ca. 34°30’N) where it occurs only irregularly and then only during the winter and early spring. Here adult size or reproductive maturity are never achieved. In contrast, pro- ceeding from Cape Hatteras northward (> 35° N) M. edulis can be found in any suitable habitat throughout the year. Beaufort populations clearly must represent the extreme southern limit at which larvae will survive and to which they are successfully dispersed, but newly attached juve- niles never survive Summer seawater temperatures which range between 25° and 30° C. Although expatriate or ““pseudo-populations” can be established early during most years, no permanent or reproducing populations have ever been established (Wells and Gray, 1960). The opposite situation has been demonstrated for the tropical amphi-Atlantic gastropod, Cymatium parthen- opeum. Larvae are known to occur throughout the Gulf Stream and North Atlantic Drift (Scheltema 1966, 1971a, b), and adult specimens of this tropical form have been found alive by dingle fishermen off the western Irish coast (O’Riordan, 1977). It is inferred that the origin of such individuals is most likely to have been the Caribbean and that their veligers have been passively transported along the Gulf Stream and North Atlantic Drift. Because there is at least a 10°C difference in summer surface temperature between the Caribbean and the western coast of Ireland, it seems improbable that C. parthenopeum will ever repro- duce off the western Irish coast and that therefore this pop- ulation probably is an expatriate, non-reproducing one. On a smaller scale, other characteristics of the envi- ronment will limit spatial distribution or sometimes the geographic range of species. In most instances, species adapted to rocky intertidal shores cannot exist upon inter- tidal marshes [although exceptions exist, e.g. Littorina lit- torea (Linné, 1758)] nor can infaunal species of intertidal flats exist in the surf of sandy coastal beaches. Physical heterogeneity as well as various biological interactions at differing spatial scales (from meters to kilometers) can result in disjunct distribution within a species’ geographic range. Although ecological factors can set distinct limits for the distribution of species, e.g. temperature may deter- mine the northward and southward latitudinal range, much of the small scale physical and biological heterogeneity of the environment can play only a secondary or minor role in determining species distribution on a biogeographically rel- evant scale. Because this review deals primarily with pas- sive dispersal, a detailed discussion of the ecological effects on biogeographic distribution is not undertaken here. SCHELTEMA: PASSIVE DISPERSAL IN BIOGEOGRAPHY 107 GEOTECTONICS, SEA-FLOOR SPREADING, AND GEOGRAPHIC DISTRIBUTION OF TROPICAL MOLLUSCAN FAUNAS Any consideration of the contemporary distribution of marine sublittoral molluscan faunas must also account for tectonic events in the geologic past and how such activi- ty could have affected the extent of passive dispersal (Darwin, 1859:311). Significant among these are such vic- ariant events as: (1) the opening and closing of seaways or corridors which may either facilitate or, alternatively, com- pletely prevent any passive dispersal (Hallam, 1973), and (2) continental drift and sea-floor spreading leading to the initial formation and subsequent enlargement of ocean basins (Berggren and Hollister, 1974). TETHYS, THE CENTRAL AMERICAN CORRIDOR, AND THE PASSIVE DISPERSAL OF TROPICAL MOLLUSKS During the Middle Cretaceous and Early Tertiary, the Tethys Sea provided a route through which species could be passively dispersed between the tropical waters of the Indo-Pacific and eastern proto-Atlantic Oceans (Fig. 6). Similarly the corridor between North and South America allowed passive dispersal between the tropical western Atlantic and eastern Pacific Oceans. Thus during the Middle Cretaceous, a circumglobal seaway existed, provid- ing continuity among tropical oceans throughout the world. Luyendyk et al. (1972) inferred from tank studies using a planetary vorticity model that during the Middle Cretaceous (ca. 100 million years ago) a circumglobal cir- culation existed from east to west with an average velocity of 2-4 knots (i.e. ca. 100-200 cm/sec). More recently, Barron and Peterson (1989:685) devised a three-dimension- al numerical model of Middle Cretaceous circulation which also included sea-level and climatic changes and (in con- trast to Luyendyk et al., 1972) concluded that although cir- culation in the tropical Pacific Ocean did indeed appear to flow from east to west, that of the Tethys Sea showed “a clockwise flow, which produced a general eastwardly flow along [its] ... northern margin. The clockwise, gyre-type circulations that develop between North and South America and between Eurasia and Africa yield components of [both] westward [and] ... eastward flow.” Van Andel (1979:18) Fig. 6. Disposition of continents during the Middle Cretaceous (slashed areas) and their present positions (stippled outlines) showing the displacement of continents over the past 100 million years. Arrows indicate the circumtropical connections by way of the Tethys Sea (between Asia and Africa) and through the corridor between North and South America. During the next 35 million years, the North Atlantic enlarged and the South Atlantic began to open. (Modified from Briden et al., 1974:figs. 1, 5; see also Fig. 7). 108 AMER. MALAC. BULL. 11(2) (1995) believed that “for some time during the Miocene, Pacific water may have flowed eastward into the Caribbean through the isthmus area.” Observations on contemporary long-distance trans- port of planktonic larvae together with a knowledge of major seaways, corridors, and ocean basins during the Tertiary allows inferences to be made about passive disper- sal of molluscan veligers during the geologic past. However, to deal with specific instances, one must know something about the life history of fossil species (Jablonski and Lutz, 1983). Such information can be obtained in well- preserved fossil gastropods from the protoconch, i.e. the larval shell at the apex of the adult or juvenile shell (Thorson, 1950; Shuto, 1974; Jung, 1975; Scheltema, 1978) and from the prodissoconch or larval shell at the umbo of postlarval bivalves (Ockelmann, 1965; Lutz and Jablonski, 1978). Inspection of the protoconch or prodissoconch makes it possible not only to determine the mode of devel- opment, i.e. whether or not a species had planktonic devel- opment, but sometimes also to roughly approximate the length of time adrift, and hence to discriminate between species with actaeplanic larvae that have relatively short planktonic lives of only a few weeks (Scheltema, 1989b:186) and those with teleplanic veligers that may remain planktonic for many months. For example, compar- ison of the protoconchs of fossil specimens of Architec- tonica nobilis from the Miocene, Pliocene, and Pleistocene (Woodring, 1959:165) with those of contemporary speci- mens of the same species shows that the mode of develop- ment remained unchanged and allows the inference that long-distance dispersal of A. nobilis larvae must have been much the same during the Miocene as it is today (Scheltema, 1979). Similar inferences can be made with extinct congeners, especially when the life histories of pre- sent-day species are known. When the life histories of Tertiary mollusks are con- sidered in relation to the known tectonic events since the late Oligocene (a period of about 40 million years), it seems highly probable that species with planktonic larvae would have had the opportunity for passive two-way disper- sal between both the tropical western Indo-Pacific and east- erm Atlantic Oceans (Kauffman, 1975) through the Tethys seaway and between the Caribbean and eastern tropical Pacific Ocean by way of the Central American Corridor (Woodring, 1966). Hence, tropical Tertiary species with teleplanic larvae, such as Architectonicidae and Ranellidae, could have been dispersed during the early Tertiary by the ocean circulation that existed at that time. However, the Tethys was a shallow epeiric sea, so it is likely that even a species with a relatively short planktonic larval develop- ment could have passed through to the eastern proto- Atlantic by stepwise dispersal over many generations (Scheltema, 1989a). Kauffman (1975:183, Fig. 2) made substantially similar arguments for the dispersal of bivalve larvae and wrote, “The larvae of Cretaceous bivalves should have been able to disperse through the Euramerican Region and across the Atlantic extension of the Tethys in short ‘geologically instantaneous’ periods of time.” The open circumglobal seaway first became ob- structed in the southeast Asian region by the Early Oligocene (ca. 35 million years ago) and in the Mediter- ranean and Near East in the Late Oligocene or Early Miocene. The corridor between North and South America closed much later, in the Early Pliocene (Van Andel, 1979:17), and resulted from the eastward movement of the Caribbean plate (Williams, 1986:11, fig. 6). As a result of such tectonic activity, three major marine tropical biogeo- graphical regions appear to have arisen. These were: (a) the Atlantic (including the Mediterranean); (b) the Indo- West Pacific including the Indian Ocean, the western Pacific, and the Central Pacific Islands; and (c) the East Pacific, including the tropical western coastline of Central and South America. Bieler recognized these three regions in a study of the Architectonicidae (Bieler et al., 1986; Bieler, 1993) and proposed that a major radiation leading to Recent species took place before the separation of the major oceans in the Miocene and Pliocene. He concluded, “The differences between the three modern architectonicid faunas can be explained by Post-Pliocene extinction of dif- ferent parts of the Neogene stock in the Eastern Pacific and in the Atlantic” (Bieler et al., 1986:236). Among the Ranellidae, the relationship between the Indo-Pacific, Atlantic, and East Pacific is not as easily demonstrated because of taxonomic difficulties within the family (Beu and Kay, 1988), but it is apparent nonetheless that much of the biogeography of species in this family is explicable only by reference to long-distance dispersal of teleplanic larvae and to the tectonic events that occurred during the Middle and Late Tertiary. SEA FLOOR SPREADING, PALEOCIRCULATION, AND THE PASSIVE DISPERSAL OF TROPICAL MOLLUSKS IN THE ATLANTIC OCEAN The paleogeography and paleocirculation of the Alantic Ocean, from its origin in the Triassic ca. 200 mil- lion years ago until the present, were summarized by Berggren and Hollister (1974). Sea-floor spreading throughout the Tertiary resulted in the ever-increasing size of the Atlantic Basin and a concomitant increase in the dis- tance between the tropical eastern Atlantic Ocean and the Caribbean Sea (Fallow, 1979; Fallow and Dromgoole, 1980). As the Atlantic Ocean increased in size (Fig. 7), the possibility of passive trans-Atlantic dispersal on ocean cur- rents became increasingly more difficult for those species whose larvae had only relatively short planktonic develop- SCHELTEMA: PASSIVE DISPERSAL IN BIOGEOGRAPHY 109 ment or whose capacity to delay settlement was limited to only a few weeks. Such time constraints did not apply to rafted organisms so long as they were able to survive at sea, but rafting as already noted is restricted mostly to epibenth- ic species that live their adult life permanently attached to hard substrata. Differences in the amount of time required for dis- persal between the tropical eastern and western Atlantic at various epochs throughout the Cenozoic can be estimated from the mean width of the Atlantic Basin and from the inferred surface currents based on experimental tank studies (Luyendyk et al., 1972). Values in Table 1 show that dur- ing the Paleogene, passive dispersal of planktonic larvae would have required two to five weeks, well within the capacity of most contemporary molluscan species. How- ever, after the Oligocene and throughout the Neogene the time needed for passive transport across the Atlantic Ocean progressively increased owing to the greater distance between continents (Berggren and Hollister, 1974) and also to a decrease in current velocity (Luyendyk et al., 1972), so that in the Holocene two to six months are required. Indeed, the Cenozoic fossil record appears to show an inverse relationship in the similarity coefficient of inverte- 15° brate genera occurring on either side of the Atlantic Ocean and the width of the Atlantic Basin (Fallow, 1979; Fallow and Dromgoole, 1980). An increase in the size of the Atlantic Basin seems to have acted as a “filter” for exchange between the fauna of the eastern and western tropical Atlantic. Only species having larvae with a long planktonic development and capacity to delay settlement are today able to successfully make a trans-Atlantic cross- ing. Yet, a substantial number of species appear to have such a capacity (Garcia-Talavera, 1982). Obviously, successful trans-Atlantic crossing of lar- vae must have been much more likely in the Early Tertiary than it is today; nonetheless the veligers of many contem- porary molluscan species continue to have the potential to cross the “mid-Atlantic barrier.” CONSIDERATIONS ARISING FROM GEOGRAPHIC DISTRIBUTION The account given in the previous pages shows that the principal source of the Caribbean molluscan fauna most likely was the eastern tropical Atlantic, which in turn was Fig. 7. Atlantic paleogeography and inferred paleocirculation at the end of the Cretaceous, 65 million years ago. Small arrows show surface circulation. Question marks show regions where some uncertainty still exists regarding the direction of currents. Dark dashed line indicates approximate position of the Mid-Atlantic Ridge from which sea-floor spreading is initiated and which has resulted in the ever-increasing size of the Atlantic Basin. Shallow sills occurred at the mouth of the Tethys Sea (“Mediterranean”) and the corridor between North and South America. (Modified after Berggren and Hollister, 1974, with additional data from a three-dimensional numerical model by Barron and Peterson, 1989). 110 AMER. MALAC. BULL. 11(2) (1995) derived largely during the Late Cretaceous and Early Tertiary by way of the Tethys Sea. The question arises, what can a comparison of the contemporary faunas indicate about such relationships? What does geographic distribu- tion reveal about the possible role of passive dispersal? EVIDENCE FOR THE DISTRIBUTION OF GENERA Among the gastropod genera occuring on the West African coast, approximately two-thirds are found also in the western tropical Atlantic. Specifically, of the 287 con- temporary gastropod genera recorded by Nicklés (1950), Gofas et al. (1989), and Nordsieck and Garcia-Talavera (1979) along the west coast of Africa and closely associat- ed islands, 198 or about 65% occur also in the Caribbean Sea. Similarly among the 94 West African bivalve genera, 70 or about 75% are also found in the tropical and warm- temperate western Atlantic. Furthermore, a comparison between the tropical eastern Atlantic and eastern Pacific shows that 140 or 49% of West African gastropod genera also occur in the tropical eastern Pacific. The above values have been derived by the comparison of West African mol- luscan genera from sources cited above with those recorded by Abbott (1974) in the Caribbean and by Keen (1971) in the eastern Pacific. Whether or not these values are truly representative remains problematic; they however can be of only limited value for making inferences about the relationship of the Caribbean and eastern Pacific fauna to that of the eastern tropical Atlantic Ocean inasmuch as no fossil genera have been included. The values perhaps do allow the tentative conclusion that the Caribbean fauna is represented (along with some endemic taxa) by an attenuated representation of tropical eastern Atlantic genera, and that the eastern tropi- » cal Pacific fauna is derived largely from the tropical Atlantic. This is so because the wide expanse of the Pacific Ocean acts as a substantial barrier to dispersal and as Darwin (1859) observed is an almost “impossible barrier” for the migration of coastal marine animals from the west- ern to eastern tropical Pacific (but see Emerson, 1991; Scheltema, 1988). EVIDENCE FROM SPECIES DISTRIBUTIONS Recognition of the differences and similarities of species composition between the eastern and western tropi- cal Atlantic faunas may lead to further insights about processes that have led to the present Caribbean fauna. Such comparisons between faunas are most easily accom- plished by use of a database such as that devised by Rosenberg (1993) for the Caribbean. It is however doubtful that a meaningful or reasonably complete compilation of West African marine molluscan species could be made from the existing literature. Table 1. Time required for passively transported larvae to cross the tropi- cal Atlantic Ocean during the Late Cretaceous and Cenozoic (after Scheltema, 1986). Time required Average width Geological epoch to transverse of ocean (km)* or period Atlantic (wk)** 2,248 Upper Cretaceous to Tertiary 1.8-3.6 2,507 Paleogene to Eocene 2.1-4.2 3,215 Eocene to Oligocene 2.5-5.2 3,852 Early Miocene 6.4-19.1 4,194 Late Miocene 8.3-25.0 4,752 Holocene 9.2-28.3 *Values taken from Fallow and Dromgoole (1980); mean width of the pre- sent tropical Atlantic Ocean taken from contemporary charts. **Computed from current velocities derived experimentally by Luyendyk et al. (1972). Tank model studies suggest velocities of 3.7-7.4 wk during the Early Tertiary. At the close of the Tethys Sea (Oligocene), the current velocity was reduced to 1-3 km/hr and these values were used in computa- tions made between Early Miocene and Holocene. Notwithstanding, some useful information on amphi- Atlantic molluscan distribution does exist, which although incomplete allows at least some comparisons. Templado et al. (1990) have complied an annotated list of 182 species of opisthobranchs among which 22.5% have amphi-Atlantic distributions. Although some of these species may have rafted on algae, it was considered that most of the distribu- tions could be accounted for by larval dispersal. Edmunds (1977) determined that among 37 species of opisthobranchs from Ghana, 43% also occurred in the Caribbean or Brazil, and Marcus and Marcus (1966) estimated that 29% of the West African opisthobranchs from the Gulf of Guinea were amphi-Atlantic. Nordsieck and Garcia-Talavara (1979) list- ed 50 species (8.7%) of prosobranch gastropods from Macronesia (mostly from the Canary Islands) that also occur in the Caribbean Sea. Garcia-Talavera (1982) enu- merated 102 amphi-Atlantic species of prosobranch gas- tropods from the Canary Islands. In a study of the family Ranellidae from the Atlantic Ocean, Garcia-Talavera (1987) found 17 of 36 species were known to have amphi- Atlantic distributions. Most Architectonicidae are thought to have amphi-Atlantic ranges (Bieler et al., 1986). From the above data one is encouraged to believe that the Atlantic Basin is not at present a complete barrier to exchange between the tropical eastern and western Atlantic and that genetic continuity between widely separated popu- lations of some molluscan species is possible even today. On the other hand, the closure of the corridor between North and South America (i.e. the Panamanian Isthmus) has served as a complete barrier to dispersal by any natural means since the Early Pliocene thus precluding any exchange for 3 million years. Creditable inferences about SCHELTEMA: PASSIVE DISPERSAL IN BIOGEOGRAPHY 111 the role of dispersal for determining geographic distribu- tions of molluscan species requires a knowledge of natural history and mode of reproduction but such data are at pre- sent largely wanting for most species. EXTINCTION AND SPECIATION Two possible reasons have been suggested for the differences encountered between the faunas on opposite shores of the tropical Atlantic Ocean. These are: (1) extinc- tion, i.e. when a species becomes extinct in one of the two disjunct regions of its original range, and (2) speciation, in which the two disjunct regions wherein a species originally occurred become sufficiently isolated that gene flow is completely prevented and, owing to differential phyletic changes between the populations, allopatric speciation eventually occurs. Vermeij (1978) allowed that it is impos- sible from Recent fauna alone to determine whether differ- ential extinction or speciation have occurred; it is necessary to consult the fossil record in order to address such ques- tions. A number of authors have recently dealt with the question of extinction of Caribbean mollusks, particularly in comparison with the eastern Pacific (e.g. Allmon et al., 1993; Jackson et al., 1993). It is not within the scope of the present paper to deal with this question. ROLE OF DISPERSAL The question of isolation necessarily is concerned with the lack of dispersal between two widely disjunct regions within the range of a species. Whenever a widely distributed genus includes only few species, one may sus- pect that dispersal has played, and continues to play, a role in the distribution of such a genus. Genetic continuity is maintained between populations and consequently allopatric speciation is unlikely to occur. In contrast, a widely distributed genus with many species is likely to result when populations become isolated in a heterogenous environment thereby leading to species radiation. There is in fact some circumstantial evidence that a relationship exists between the capacity to disperse and the amphi-Atlantic distribution of species. For example, among the 102 amphi-Atlantic species considered by Garcia-Talavera (1982) 96% are known to have planktonic development. Scheltema (1989b) found that among 87 species of gastropods for which the mode of development was known, 22 had teleplanic larvae and among these, 19 species were known to have amphi-Atlantic distribution. Is there then any evidence that morphological differences between widely separated populations are related to the capacity of larvae to disperse? Scheltema (1972b:113) found “among five gastropod species ... a remarkable direct correspondence between the estimated frequency of larval dispersal and the degree of morphological similarity of eastern and western Atlantic populations. Adults of species estimated to have a high frequency of larval dispersal showed little or no differences between the eastern and western-Atlantic. Conversely, species that have a restricted larval dispersal ... [were] represented by different sub- species on either side of the Atlantic ... [This] suggested that the frequency of larval dispersal and the rate of gene- flow are related” in the five species examined. Since the time of these observations, it now has become possible through new techniques in molecular biology to make esti- mates of gene flow, yet no research seems to have revealed the relationship between any tropical amphi-Atlantic mol- luscan species. COLONIZATION OF ISLANDS The colonization of Atlantic oceanic islands pro- vides yet other compelling evidence for the effectiveness of passive long-distance dispersal in establishing the geo- graphic distribution of species and in affecting the compo- sition of tropical Atlantic faunas. Indeed, island coloniza- tion seems to provide a model of how long-distance disper- sal works. Ascension Island (probably of Late Pleistocene age) is located in the tropical South Atlantic (7.57° S; 14.22° W) where it arises from the Mid-Atlantic Ridge. Rosewater (1975) found that the endemism of marine mol- lusks was less than 8% and that the remaining 92% were continental species with about equal numbers originating from the eastern and western Atlantic. Rosewater consid- ered that the continental species must have originally colo- nized the island as teleplanic larvae. Analogous situations have been described by Pawson (1978) for echinoderms and for decapod Crustacea by Manning and Chace (1990). Older oceanic islands originating from the Mid-Atlantic Ridge at some earlier geologic epoch (e.g. the Canary and Cape Verde Archipelagos) show much higher endemisms of marine mollusks than that found on Ascension Island owing, it must be supposed, to the longer time over which species radiation can have occurred. The Canary and Cape Verde Islands as a result of sea-floor spreading have moved eastward from their point of origin on the Mid-Atlantic Ridge, and as a result are closer to the source of passively dispersing planktonic larvae from the west coast of Africa and more remote from any western Atlantic source. Consequently these island groups naturally have a larger proportion of West African marine mollusks and a much lower percentage of South American species. In a contrast- ing example, Leal (1991) found in the study of four tropical oceanic islands off the coast of Brazil (viz. Atol das Roccas, Fernando do Noronha, Trinidade, and Martin Vaz) that the total number of Brazilian species decreased with increasing distance from the continent and that there occurred only a very small proportion of eastern Atlantic forms. In the case of the Canary and Cape Verde Islands, one can speculate that most of the endemics will be related 112 AMER. MALAC. BULL. 11(2) (1995) to western Atlantic species owing to the ever-increasing distance from a western Atlantic source and the consequent restriction over time of genetic exchange with South American species, resulting eventually in complete isola- tion and allopatric speciation. CONCLUSIONS No single process can completely explain the con- temporary composition of the Caribbean tropical fauna. It has been argued from the evidence presented here that most of the Caribbean tropical molluscan fauna had its origin by passive dispersal of planktonic larvae originating from the eastern tropical Atlantic and ultimately from the Indo- Pacific by way of the Tethys Sea. With the close of the Tethys Sea during the Oligocene and Miocene, the eastern Atlantic Ocean became isolated from the Indo-Pacific region. Likewise, the close of the North and South American Corridor during the Early Pliocene resulted in complete isolation of the Caribbean from the eastern Pacific fauna. Disappearance of these two seaways resulted in almost complete isolation between faunas of the affected regions. Indeed, natural selection or genetic drift in some instances has led to allopatric speciation and to geminate pairs or sister species (e.g. between the tropical eastern Pacific and western Atlantic; Vermeij, 1978). Alternatively, some species have become extinct on either one or the other side of the interposed barriers. The initial formation of the Atlantic Ocean during the Triassic and its enlargement by sea-floor spreading throughout the Cretaceous and Tertiary resulted in the establishment of a “mid-Atlantic barrier” which increasing- ly over time has acted as a filter for passive dispersal. The result is a reduction in the number of species found in com- mon between the eastern tropical Atlantic and Caribbean Sea, particularly among forms with a restricted capacity for long-distance dispersal (i.e. those having a relatively short planktonic larval development or an inability to prolong planktonic life) or which are not adapted for long-distance rafting. However, the biogeographic evidence suggests that within the recent geologic past (i.e. Pleistocene) and even into the present, long-distance larval dispersal continues to play an important role, both for the initial colonization of oceanic islands (e.g. Ascension Island) and for genetic exchange or “gene flow” between the eastern and western tropical Atlantic Ocean. Biogeography is now at a turning point where new techniques, e.g. the measurement of enzyme variation and mitochondrial DNA polymorphism, will make it possible to address questions heretofore largely intractable. Some of these possibilities using such techniques were summarized by Ward (1989). Olson et al. (1991:357) wrote that “the inability to distinguish between marine invertebrate larvae of closely related taxa is a longstanding problem ... within groups such as bivalves ... The larvae of many species are so similar that they cannot be identified by morphology alone. Using the polymerase chain reaction [PCR] to amplify and sequence mitochondrial DNA, ... [it is possi- ble] to distinguish between morphologically identical co- occurring larvae of two species .... [The] technique should prove effective for the identification of larvae even when they are far from their adult population, such as open-ocean (teleplanic) larvae.” The measurement of geographic variation by the use of morphological characters can now be augmented by new biochemical techniques, and it also could become possible to make realistic measurements of genetic exchange between disjunct, widely-distributed populations. Finally Cladistics, although largely ignoring the processes that determine the distribution of species, can allow a better understanding of existing geographic relationships such as that between the eastern and western Atlantic by recogniz- ing large scale patterns of distribution in relation to the evo- lution of molluscan species. However, better knowledge of the systematics of contemporary and fossil species will be required. ACKNOWLEDGMENTS The superb library of the Marine Biological Laboratory and the Woods Hole Oceanographic Institution facilitated the writing of this review paper. My own research on various aspects of larval dispersal has been supported over a period of more than 20 years by a number of grants from the National Science Foundation, most recently OCE92-01499. Thanks go to Amélie H. Scheltema for her careful reading of an earlier version of the manuscript and to Ethel F. LeFave for patiently enduring numerous revisions and rendering the final text into a form suitable for publication. Contribution No. 8391 from the Woods Hole Oceanographic Institution. LITERATURE CITED Abbott, R. T. 1974. American Seashells. Van Nostrand Reinhold Co., New York. 663 pp. Allmon, W. D., G. Rosenberg, R. W. Portell, and K. S. Schindler. 1993. Diversity of Atlantic coastal plain mollusks since the Pliocene. Science 260: 1626-1629. Arnaud, F., P. M. Arnaud, A. Intés, and P. LeLoeuff. 1976. Transport d’invertébrés benthiquesentre |’ Afrique du Sud et Sainte Héléne par les laminaires (Phaeophycea). Bulletin du Muséum National d’ Histoire Naturelle, Paris, 3° Ser., No. 384, Ecologie Générale 30: 49-55. Barron, E. J. and W. H. Peterson. 1989. Model simulation of the Cretaceous ocean circulation. Science 244:684-686. Bayne, B. L. 1965. Growth and delay of metamorphosis of the larvae of SCHELTEMA: PASSIVE DISPERSAL IN BIOGEOGRAPHY 113 Mytilus edulis (L). Ophelia 2:1-47. Berggren, W. A. and C. D. Hollister. 1974. Paleogeography, paleobio- geography and the history of circulation in the Atlantic Ocean. Studies in Paleo-Oceanography, Society of Economic Paleontology and Mineralogy, Special Publication 20: 126-186. Beu, A. G. and E. A. Kay. 1988. Taxonomy of gastropods of the families Ranellidae (= Cymatiidae) and Bursidae. Part IV. The Cymatium pileare complex. Journal of the Royal Society of New Zealand 18:185-223. Bhaud, M. 1982. Relations entre stratégies de reproduction et aire de repartition chez les annélides polychétes. Oceanologicia Acta 5:465-472. Bhaud, M. R. 1993. Relationship between larval type and geographic range in marine species: complementary observations on gas- tropods. Oceanologica Acta 16:191-198. Bieler, R. 1993. Architectonicidae of the Indo-Pacific (Mollusca, Gastropoda). Abhandlungen des Naturwissenschaftlichen Vereins in Hamburg (NF) 30:377 pp. Bieler, R., A. S. Merrill, and K. J. Boss. 1986. Faunal relationships of the western Atlantic Architectonicidae [abstract]. American Malacological Bulletin 4:236. Booth, J. D. 1979. Common bivalve larvae from New Zealand: Pteriacea, Anomiacea, Ostreacea. New Zealand Journal of Marine and Freshwater Research 13:131-139. Bousfield, E. L. and M. L. H. Thomas. 1975. Postglacial changes in distri- bution of littoral marine invertebrates in the Canadian Atlantic Region. Proceedings of the Nova Scotia Institute of Science 27 (Suppl. 3):47-60. Carlton, J. 1985. Trans-oceanic and inter-oceanic dispersal and coastal marine organisms: the biology of ballast water. Oceanography and Marine Biology Annual Review 23:3 13-371. Carquist, S. 1974. Island Biology. Columbia University Press, New York. ix + 660 pp. Cole, H. A. and E. W. Knight-Jones. 1949. The setting behavior of larvae of the European flat oyster, Ostrea edulis L. and its influence on methods of cultivation and spot collection. Fishery Investigations, Ministry of Agriculture and Fisheries, Ser. I, 17(3):1-39. Crisp, D. J. 1967. Chemical factors inducing settlement in Crassostrea virginica (Gmelin). Journal of Animal Ecology 36:329-336. Croizat, L., G. Nelson, and D. E. Rosen. 1974. Centers of origin and relat- ed concepts. Systematic Zoology 23:265-287. Darwin, C. 1859. On the Origin of Species by Means of Natural Selection. John Murray, London. 490 pp. Davis, H. C. and R. R. Guillard. 1958. Relative value of ten genera of micro-organisms as food for oyster and clam larvae. Fishery Bulletin 58:293-304. Edmunds, M. 1977. Larval development, oceanic currents and the origin of the opisthobranch fauna of Ghana. Journal of Molluscan Studies 43:301-308. Emerson, W. K. 1991. First records for Cymatium mundum (Gould) in the eastern Pacific Ocean, with comments on the zoogeography of the tropical trans-Pacific tonnacean and non-tonnacean prosobranch gastropods with Indo-Pacific faunal affinities in west American waters. The Nautilus 105:62-80. Fallow, W. C. 1979. Trans-North Atlantic similarity among Mesozoic and Cenozoic invertebrates correlated with widening of the ocean basin. Journal of Geology 7:398-400. Fallow, W. C. and E. L. Dromgoole. 1980. Faunal similarity across the South Atlantic among Mesozoic and Cenozoic invertebrates cor- related with widening of ocean basins. Journal of Geology 88:723-727. Frazier, J., D. Margaritoules, K. Muldoon, C. W. Potter, J. Rosewater, C. Ruckeschel, and S. Sales. 1985. Epizoan communities on marine turtles. I. Bivalve and gastropod molluscs. Publicacione della Stazione: Zoologica di Napoli, I. Marine Ecology 6:127-140. Fushimi, K., M. Nishikawa, and H. Mitsuda. 1991. Floating pumices in the western North Pacific. Oceanographical Magazine [Tokyo] 41:59-74, Garcia-Talavera, F. 1982. Los Moluscos Gasteropodos Afiatlanticos (Estudio Paleo y Biogeografico de las Especies Bentonicas Litorales). Secretariado de Publicaciones de la Universidad de la Laguna, Tenerife, Canarias, Coleccion Monografias 10:352 pp. Garcia-Talavera, F. 1987. The family Ranellidae, Gray, 1854 (= Cymatiidae, Iredale, 1913) in the Atlantic; zoogeographical con- siderations. Bolletino Malacologico 23:243-258. Gofas, S., J. P. Alfonso, and M. Brandao. 1989. Conchas and Moluscos de Angola. Universidade Agostinho Neto/Elf Aquitaine Angola. 140 + 1V pp. Gunn, C. R. 1968. Stranded seeds and fruits from the southeastern shore of Florida. Garden Journal [New York Botanical Garden] Mar.- Apr. 1968:43-54. Guppy, H. B. 1917. Plants, Seeds, and Currents in the East Indies and Azores. William Norgate, London. 531 pp. Hadfield, M. G. and R. H. Karlson. 1969. Externally induced metamor- phosis in a marine gastropod [abstract]. American Zoologist 9:1122. Hallam, A. 1973. Distributional patterns in contemporary terrestrial and marine animals. /n: Organisms and Continents through Time, N. F. Hughes, ed. pp. 93-105. Systematics Association, London, Special Papers in Paleontology No. 12, Publication No. 9. Highsmith, R. C. 1985. Floating and algal rafting as potential dispersal mechanisms in brooding invertebrates. Marine Ecology Progress Series 24:169-179. Hutchins, L. W. 1947. The bases for temperature zonation in geographi- cal distribution. Ecological Monographs 17:325-335. Jablonski, D. and R. A. Lutz. 1983. Larval ecology of marine benthic invertebrates: paleobiological implications. Biological Reviews 58:21-89. Jackson, J. B. C. 1986. Modes of dispersal of clonal benthic invertebrates: consequences for species distribution and genetic structure of local populations. Bulletin of Marine Science 39:588-606. Jackson, J. B. C., P. Jung, A. G. Coates, and L. S. Collins. 1993. Diversity and extinction of tropical American mollusks and emer- gence of the Isthmus of Panama. Science 260: 1624-1626. Jokiel, P. L. 1984. Long-distance dispersal of reef corals by rafting. Coral Reefs 3:69-76. Jokiel, P. L. 1989. Rafting of reef corals and other organisms at Kwajalein Atoll. Marine Biology 101:483-493. Jokiel, P. L. 1990a. Transport of reef corals into the Great Barrier Reef. Nature 347:665-667. Jokiel, P. L. 1990b. Long-distance dispersal by rafting: re-emergence of an old hypothesis. Endeavour 14:66-73. Jung, P. 1975. Quaternary larval gastropods from Leg 15, Site 147, Deep Sea Drilling Project. Preliminary Report. The Veliger 18:109- 126. Kauffman, E. G. 1975. Dispersal and biostrategraphic potential of Cretaceous benthonic Bivalvia in the Western Interior. Geological Association of Canada, Special Paper 13:165-194. Keen, A. M. 1971. Sea Shells of Tropical West America, 2nd ed. Stanford University Press, Stanford, California. xiv + 1064 pp. Kempf, S. C. 1981. Long-lived larvae of the gastropod Aplysia juliana: do they disperse and metamorphose or just slowly fade away? Marine Ecology Progress Series 6:61-65. Knight-Jones, E. W. 1951. Aspects of the settling behaviour of larvae of Ostrea edulis on the Essex oyster beds. Rapport et Proces - Verbaux de Reunions, Conseil Internationale pour I’ Exploration 114 AMER. MALAC. BULL. 11(2) (1995) de la Mer 128(2):30-34. Laursen, D. 1981. Taxonomy and distribution of teleplanic prosobranch larvae in the North Atlantic. Dana Report 89:44 pp. Leal, J. H. 1991. Marine Prosbranch Gastropods from Oceanic Islands off Brazil. Universal Book Services/Dr. W. Backhuys, Oegstgeest, Netherlands. x + 419 pp. Lima, G. M. and J. A. Pechenik. 1985. The influence of temperature on growth rate and length of larval life in the gastropod Crepidula plana Say. Journal of Experimental Marine Biology and Ecology 90:55-71. Loosanoff, V. L., W. S. Miller, and P. B. Smith. 1951. Growth and settling of larvae of Venus mercenaria in relation to temperature. Journal of Marine Research 10:59-81. Lutz, R. A. and D. Jablonski. 1978. Cretaceous bivalve larvae. Science 199:439-440, Luyendyk, B. P., D. Forsyth, and J. D. Phillips. 1972. Experimental approach to the paleocirculation of the oceanic surface waters. Geological Society of America Bulletin 83:2649-2664. Manning, R. B. and F. A. Chace. 1990. Decapod and stomatopod Crustacea from Ascension Island, South Atlantic Ocean. Smithsonian Contributions to Zoology 503:91 pp. Marche-Marchad, I. 1968. Remarques sur de développement chez les Cymba, prosobranches volutides et l’hypothése de leur origine Sud Americaine. Bulletin de Institute Fondamental d’ Afrique Noire 30A: 1028-1037. Marcus, E. and E. Marcus. 1966. Opisthobranchs from tropical West Africa. Pilsbry Deep-Sea Biological Expedition to the Gulf of Guinea, 1964-65. Studies in Tropical Oceanography 4:152-208. Merrill, A. S. 1970. The Family Architectonicidae (Gastropoda: Mollusca) in the Western and Eastern Atlantic. Doctoral Dissertation, University of Delaware, Newark. 338 pp. Metcalf, W. G., A. D. Voorhis, and M. C. Stalcup. 1962. The Atlantic Equatorial Undercurrent. Journal of Geophysics 67:2499-2508. Mileikovsky, S. A. 1966. [Range of dispersion of pelagic larvae of ben- thic invertebrates by currents and the migratory role of this disper- sion taking Gastropoda and Lamellibranchia as examples (In Russian). Okeanologiya 6:482-492. Mitton, J. B., C. J. Berg, and K. S. Orr. 1989. Population structure, larval dispersal, and gene flow in the queen conch, Strombus gigas of the Caribbean. Biological Bulletin 177:356-362. Nicklés, M. 1950. Mollusques Testacés Marins de la Céte Occidentale d’ Afrique. Paul Lechevalier, Paris. viii + 269 pp. Nordsieck, F. and F. Garcia-Talavera. 1979. Molluscos Marinos de Canarias y Madera (Gastropoda). Aula de Cultura de Tenerife. 208 pp, 46 pls. Ockelmann, K. W. 1965. Developmental types in marine bivalves and their distribution along the Atlantic coast of Europe. In: Proceedings of the First European Malacological Congress [1962], L. R. Cox and J. F. Peake, eds. pp. 25-35. London. Okubo, A. 1971. Oceanic diffusion diagrams. Deep-Sea Research 18:789-802. Okubo, A. 1994. The role of diffusion and related physical processes in dispersal and recruitment of marine populations. In: The Bio- physics of Marine Larval Dispersal, P. W. Sammarco and M. L. Heron, eds. pp. 5-32. American Geophysical Union, Washington, D.C. Olson, R. R., J. A. Runstadler, and T. D. Kocher. 1991. Whose larvae? Nature 351:357-358. O’Riordan, C. E. 1977. Marine fauna notes from the National Museum of Ireland. /rish Naturalist Journal 19:6-7. Ota, S. 1961. Identification of the larva of Pinna atrina japonica (Reeve). Bulletin of the Japanese Society of Scientific Fisheries 27:107- 112. Pawson, D. L. 1978. The echinoderm fauna of Ascension Island, South Atlantic Ocean. Smithsonian Contributions to Marine Science 2:31 pp. Pechenik, J. A. and N. S. Fisher. 1979. Feeding assimilation and growth of the mud snail larvae, Nassarius obsoletus (Say) on three differ- ent algae diets. Journal of Experimental Marine Biology and Ecology 38:57-80. Pechenik, J. A., L. S. Eyster, J. Widdows, and B. L. Bayne. 1990. The influence of food concentration and temperature on growth and morphological differentiation of blue mussel Mytilus edulis L. lar- vae. Journal of Experimental Marine Biology and Ecology 136:47-64. Pilkington, M. C. and V. Fretter. 1970. Some factors affecting growth of prosobranch veligers. Helgolander Wissenschaftliche Meeresuntersuchungen 20:576-593. Richardson, P. L. 1984. Drifting buoy trajectories in the Atlantic North Equatorial Counter-current during 1983. Geophysical Research Letters 11:745-748. Rosenberg, G. 1993. A database approach to studies of molluscan taxon- omy, biogeography and diversity, with examples from western Atlantic marine gastropods. American Malacological Bulletin 10:257-266. Rosewater, J. 1975. An annotated list of the marine mollusks of Ascension Island, South Atlantic Ocean. Smithsonian Con- tributions to Zoology 189:41 pp. Scheltema, R. S. 1956. The effect of substrate on the length of planktonic existence in Nassarius obsoletus [abstract]. Biological Bulletin 111:312. Scheltema, R. S. 1961. Metamorphosis of the veliger larvae of Nassarius obsoletus (Gastropoda) in response to bottom sediment. Biological Bulletin 120:92-109. Scheltema, R. S. 1964. Origin and dispersal of invertebrate larvae in the North Atlantic [abstract]. American Zoologist 4:299. Scheltema, R. S. 1966. Evidence for trans-Atlantic transport of gastropod larvae belonging to the genus Cymatium. Deep Sea Research 13:81-95. Scheltema, R. S. 1967. The relationship of temperature to the larval development of Nassarius obsoletus (Gastropoda). Biological Bulletin 132:253-26S. Scheltema, R. S. 1968. Dispersal of larvae by equatorial currents and its importance to the zoogeography of shoal-water tropical species. Nature 217:1159-1162. Scheltema, R. S. 1971a. The dispersal of larvae of shoal-water benthic invertebrate species over long distances by ocean currents. Jn: Fourth European Marine Biology Symposium, D. Crisp, ed. pp. 7- 28. Cambridge University Press, Cambridge, U. K. Scheltema, R. S. 1971b. Larval dispersal as a means of genetic exchange between geographically separated populations of shoal-water ben- thic marine gastropods. Biological Bulletin 140:284-322. Scheltema, R. S. 1971c. Dispersal of phytoplanktotrophic shipworm lar- vae over long distances. Marine Biology 11:5-11. Scheltema, R. S. 1972a. Eastward and westward dispersal across the trop- ical Atlantic Ocean of larvae belonging to the genus Bursa (Prosobranchia, Mesogastropoda, Bursidae) Internationale Revue der gesamten Hydrobiologie 57:863-873. Scheltema, R. S. 1972b. Dispersal of larvae as a means of genetic exchange between widely separated populations of shoal water benthic invertebrate species. In: Fifth European Marine Biology Symposium, B. Battaglia, ed. pp. 101-114. Piccin Editore, Padova, Italy. Scheltema, R. S. 1977. Dispersal of marine invertebrate organisms: pale- obiogeographic and biostratigraphic implications. In: Concepts and Methods of Biostratigraphy, E. G. Kauffman and J. E. Hazel, SCHELTEMA: PASSIVE DISPERSAL IN BIOGEOGRAPHY 115 eds. pp. 73-108. Dowden, Hutchinson and Ross, Stroudsburg, Pennsylvania. xiii + 658 pp. Scheltema, R. S. 1978. On the relationship between dispersal of pelagic veliger larvae and the evolution of marine prosobranch gas- tropods. In: Marine Organisms - Genetics Ecology and Evolution, B. Battaglia and A. J. Beardmore, eds. pp. 303-322. Plenum Press, New York. Scheltema, R. S. 1979. Dispersal of pelagic larvae and the zoogeography of Tertiary marine benthic gastropods. Jn: Historical Biogeography, Plate Tectonics, and the Changing Environment, J. Gray and A. J. Boucot, eds. pp. 391-397. Oregon State University Press, Corvallis. Scheltema, R. S. 1986. On dispersal and planktonic larvae of benthic invertebrates: an eclectic overview and summary of problems. Bulletin of Marine Science 39:290-322. Scheltema, R. S. 1988. Initial evidence for the transport of teleplanic lar- vae of benthic invertebrates across the east Pacific barrier. Biological Bulletin 174:145-152. Scheltema, R. S. 1989a. On the children of benthic invertebrates: their ramblings and migration in time and space. In: Environmental Quality and Ecosystem Stability, Vol. 4, B. E. Spanier, Y. Steinberger, and M. Luria, eds. pp. 93-112. Israel Society for Ecology and Environmental Quality Publications, Jerusalem, Israel. Scheltema, R. S. 1989b. Planktonic and non-planktonic development among prosobranch gastropods and its relationship to the geo- graphic range of species. In: Reproduction Genetics and Distributions of Marine Organisms, J. S. Ryland and P. A. Tyler, eds. pp. 183-188. Olsen and Olsen, Fredensborg, Denmark. Scheltema, R. S. 1992. Passive dispersal of planktonic larval and the bio- geography of tropical sublittoral invertebrate species. In: Marine Eutrophication and Population Dynamics, G. Colombo, I. Ferrari, V. U. Ceccherelli, and R. Rossi, eds. pp. 195-202. Olsen and Olsen, Fredensborg, Denmark. Scheltema, R. S. and A. H. Scheltema. 1984. Larval dispersal and the geographic range among species of the bivalve family Pinnidae [abstract]. American Zoologist 24: 132. Shuto, T. 1974. Larval ecology of prosobranch gastropods and its bearing on biogeography and paleontology. Lethaia 7:239-256. Sprung, M. 1984. Physiological energetics of mussel larvae (Mytilus edulis). Il. Food uptake. Marine Ecology Progress Series 17:295- 305. Stenzel, H. B. 1971. Oysters - dispersal. /n: Treatise on Invertebrate Paleontology, Part N, Vol. 3, Mollusca 6, Bivalvia, R. C. Moore, ed. pp. 1035-1036. Geological Society of America and University of Kansas Press, Lawrence, Kansas. Stickney, A. P. 1964. Salinity, temperature and food requirements of soft-shell clam larvae in laboratory culture. Ecology 45:283-291. Strathmann, R. 1974. The spread of sibling larvae of sedentary marine invertebrates. American Naturalist 108:29-44. Templado, J., A. A. Luque, and J. Ortea. 1990. A commented check-list of the amphi-Atlantic Ascoglossa and Nudibranchia (Mollusca: Opisthobranchia). Atti Congreso Sorrento 1987, Lavori, Societd Italiano Malacologico 23:295-326. Thorson, G. 1950. Reproduction and larval ecology of marine bottom invertebrates. Biological Reviews 24:1-45. Thorson, G. 1961. Length of pelagic larval life in marine bottom inverte- brates as related to larval transport by ocean currents. In: Oceanography, M. Sears, ed. pp. 455-474. Publication 67, American Association for the Advancement of Science, Washington, D. C. Turner, R. D. 1966. A Survey and Illustrated Catalogue of the Teredinidae (Mollusca: Bivalvia). Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts. 265 pp. Vallentin, R. 1895. Some remarks on the dispersal of marine animals by means of seaweeds. Annals and Magazine of Natural History, Ser. 6, 16:418-423. Van Andel, T. H. 1979. An eclectic overview of plate tectonics, paleo- geography, and paleoceanography. In: Historical Biogeography, Plate Tectonics, and the Changing Environment, J. Gray and A. J. Boucot, eds. pp. 9-25. Oregon State University Press, Corvallis. Vermeij, G. T. 1978. Biogeography and Adaptation. Patterns of Marine Life. Harvard University Press, Cambridge, Massachusetts. xvi + 332 pp. Ward, R. D. 1989. Molecular population genetics of marine organisms. In: Reproduction, Genetics and Distributions of Marine Organisms, J.S. Ryland and P. A. Tyler, eds. pp. 235-249. Olsen and Olsen, Fredensborg, Denmark. Wehrtmann, I. S. and A. I. Ditell. 1990. Utilization of floating mangrove leaves as a transport mechanism of estuarine organisms with emphasis on decapod Crustacea. Marine Ecology Progress Series 60:67-73. Wells, H. W. and I. E. Gray. 1960. The seasonal occurrence of Mytilus edulis on the Carolina coast as a result of transport around Cape Hatteras. Biological Bulletin 119:550-559. Williams, C. A. 1986. An ocean-wide view of Palaeogene plate tectonic events. Palaeogeography, Palaeoclimatology, and Palaeoecology 57:3-25. Williams, R. J., F. B. Griffiths, E. J. Van der Wal, and J. Kelley. 1988. Cargo vessel ballast water as a vector for the transport of non- indigenous species. Estuarine Coastal and Shelf Science 26:409- 420. Woodring, W. P. 1959. Geology and paleontology of Canal Zone and adjoining parts of Panama. Descriptions of Tertiary mollusks (Gastropods: Vermetidae to Thaididae). United States Geological Survey, Professional Paper 306B: 147-239, 14 pls. Woodring, W. P. 1966. The Panama land bridge as a sea barrier. Proceedings of the American Philosophical Society 11:425-433. Yoshida, H. 1956. Early life-history of useful bivalves in the Ariake Sea. Journal of the Shimonoseki College of Fisheries 6:115-122. Date of manuscript acceptance: 28 January 1994 Reproductive seasonality, periodicity, and associated behavior in a colony of Strombus pugilis (Mollusca: Gastropoda) in Puerto Rico Shawna E. Reed* Department of Marine Sciences, University of Puerto Rico, P.O. Box 908, Lajas, Puerto Rico 00667 Abstract. This work focused on defining seasonal reproductive dynamics and individual reproductive behavior for the West Indian fighting conch, Strombus pugilis Linné, 1758, in Puerto Rico. A colony of S. pugilis was followed from September 1987 to September 1989. For both years, reproductive activity began in March, when water temperature rose, and ended in November, when temperature dropped. By December, the entire colony had buried and remained so until re-emergence in February. During the reproductive season, mating and spawning activity occurred on a monthly basis with some irregu- larity. The monthly periodicity was significantly correlated with the lunar cycle, with peak spawning occurring in the second week following a full moon. Observations of conch held in an outdoor tank at the laboratory showed that winter burial, spring emergence, and reproductive seasonality and periodicity coincided with the same events in the field, despite a different ambient temperature regime. The reproductive season in strombids is protracted with multiple spawnings (D’Asaro, 1965; Brownell, 1977; Kuwamura et al., 1983; Weil and Laughlin, 1984). Repro- ductive activity decreases or ceases altogether during win- ter months, possibly due to the drop in water temperature (Brownell, 1977; Kuwamura et al., 1983). Spawning has been observed year-round for Strombus pugilis Linné, 1758, in Barbados (Bradshaw-Hawkins, 1982) and S. costa- tus Gmelin, 1791, in Puerto Rico and the Turks and Caicos Islands (pers. obs.), but ceases in winter in the Yucatan (L. A. Rodriguez Gil, pers. comm.). Gametogenesis occurs year-round in S. gigas Linné, 1758, in Belize (Egan, 1985), although spawning has not been observed. Stoner et al. (1992) suggested that reproductive activity in S. gigas in the Bahamas is influenced by a combination of daylength and water temperature. The variability in reproductive sea- sonality among strombids at different locales does not allow extrapolation of such data to other areas. Reproductive periodicity within a season has not been noted in any strombid species with the exception of Strombus pugilis (fide Percharde, 1968, 1970). Percharde (1968, 1970) casually noted cyclic burying and reproduc- tive behavior, and suggested that lunar periodicity acted as a cue to induce spawning activity in S. pugilis. Repro- ductive periodicity was observed in a population of S. *Present Address: c/o 24265 - 60th Ave., Langley, B.C., Canada V3A 6H4 pugilis in Puerto Rico (Sanders, 1988), but not studied. This study was undertaken to follow the yearly cycle of reproduction in a colony of Strombus pugilis to document reproductive seasonality and periodicity, if any, in Puerto Rico. Secondary objectives were to observe behavior associated with the reproductive cycle and to com- pare observations made in the field to those made at the laboratory. METHODS The colony of Strombus pugilis chosen for study was located one km off the town of La Parguera on the southwestern coast of Puerto Rico, at a depth of 10 m, near Corona de Piedra reef. The substratum was muddy-sand with patches of interspersed macroalgae (primarily Halimeda, Ulva, Penicillus, and Caulerpa spp.). Beginning in October 1988, bottom water temperature was recorded to the nearest 0.1°C. Surveys of this colony were initiated as informal weekly observations beginning in September 1987 and con- tinuing until February 1988. A more rigorous procedure was initiated in March 1988. The study was concluded in September 1989 due to disappearance (presumably harvest) of the entire colony during the first week of September. All observations were carried out using SCUBA. Observation periods varied throughout the day. The initial weekly observation periods document- American Malacological Bulletin, Vol. 11(2) (1995):117-121 117 118 AMER. MALAC. BULL. 11(2) (1995) ed cyclic burying and reproductive behavior. One hr obser- vation periods were found to be sufficient to survey the approximately 2000 m2 area inhabited by the colony. Numbers of individuals buried or not buried, or engaged in spawning or copulation, were recorded and converted to percentages for graphical purposes. Beginning in June 1988, conch were tagged for individual identification. Tags consisted of embossed DYMO-brand vinyl tape, numbered consecutively, with a hole punched at one end through which a rubber band was threaded. The rubber band was then wrapped around the spire of the conch and held in place by the spines on the shell. Tag loss was negligible as the rubber bands main- tained their integrity for at least the 15-month duration of the tagging portion of the study. Each conch was sexed by observing the foot when extended from the shell: males have verges, and females have egg grooves. A total of 1805 conch were tagged in this colony, of which 1231 were sexed (also see Reed, 1993, for a discussion of sex ratios in this colony); only three juveniles were observed. Those animals used for experimental purposes, such as tank observations (see below), are not included in the 1231 used for field observations. Only 34 tagged conch were found dead during the course of study. Approximately 100 individuals per hr of observa- tion could be accounted for if tag numbers were recorded. When the entire colony was buried, except for a few, isolat- ed individuals (not engaged in reproductive activity), pro- portions buried were rounded off to 100% and proportions of reproductively active were set to zero. The colony was observed on a weekly basis except when spawning was intensive, when observation periods were increased to three per week or more. Activities recorded were: “buried” (conch immersed into the substra- tum), “unburied” (conch on surface of substratum, and actively moving, with a trail visible behind it, or feeding), “spawning” (egg mass visibly being extruded), and “copu- lating” (male with verge extended under the shell of the female). Completely buried conch were found by observ- ing the substratum for a protruding eye-stalk. The angle of burial was determined by measuring the depth of the spire of the shell below the surface of the substratum. Equal numbers of males and females (approxi- mately 25 of each sex at any given time) were collected from the study site as needed and maintained in an outdoor, concrete tank at the laboratory with sandy substratum of 10 cm depth and free-flowing seawater. These conch were also individually tagged and observed daily (including at night) to complement field observations and supplement behavioral data when weather and sea conditions prohibited field activities. Water temperature of the tank was recorded. Statistical analyses were performed using the SYSTAT package for IBM-PCs (SYSTAT Inc., 1992). Proportions were arc-sine transformed before analysis to ensure homogeneity of variances. RESULTS SEASONAL BEHAVIOR Variations in burial and reproductive activity are summarized in Fig. 1. For both years, the entire colony remained buried for at least two months, starting in mid- November until emerging in late January in 1988 (early February in 1989). Burial during this time period differed from the rest of the year in that conch dug in apex down, and remained at a 45° angle, with only a small hole through which an eye stalk could protrude. The siphonal notch was one cm below the surface of the substratum. The substra- tum was flat and appeared undisturbed where conch were buried. Occasionally, an individual would emerge and rebury less than a meter away. Because of the cohesive nature of the substratum, original holes and characteristic marks made by the operculum during movement remained in evidence for a week or more. Winter burial coincided with the annual drop in water temperature for this area, and emergence coincided with the spring equinox and annual rise in water tempera- ture. Bottom water temperature began dropping in the sec- ond week of November 1988 from approximately 28°C to 26°C by the third week of December 1988, and remained stable at that temperature until the first week of February 1989, when water temperature began rising. Although no temperature data were recorded at the study site in 1987 and early 1988, a similar drop in water temperature was first noted by divers in November 1987, as was a rise in temperature in February 1988 (later verified from water temperatures recorded at the Department of Marine Sciences, University of Puerto Rico, located approximately 1.5 km away). Burial during non-winter months was superficial with the conch dug into the substratum, apex down, at a shallower angle of 15° to 30°. The siphonal canal and eye notch(es) of the shell rested level with the substratum. REPRODUCTIVE PERIODICITY During the first emergence of the year, no repro- ductive activity was observed; however, the conch were actively feeding. The second, or “spring,” emergence sig- nalled the onset of spawning, which continued on a cyclic basis until the next winter burial. No lag was observed between copulation and onset of spawning in 1988 or 1989, although copulations during the spawning season were fre- quently observed; first copulations of the season could have occurred outside of the observation times. REED: SPAWNING PERIODICITY IN STROMBUS PUGILIS 119 = % 5 Daylength (hours) N 100 Buried % ° VAAN Val » $ 80 ; o Full moon ils + oO e : \ W\ a o [4 ct) =r: of | oO aS fe apa =o ee $2 g 2 $$ 22 ¢ € EEE EE BP 1988 1989 Fig 1. Annual cycle of burial and reproduction in a colony of Strombus pugilis in Puerto Rico, from January 1988 to August 1989, including daylength curves and dates of full moons. Conch maintained in the outdoor tank over winter emerged from the substratum and began reproductive activi- ty concurrently with those in the field; however, relative to the field, water temperature of the tank did not rise. Tank temperature remained constant (23.5°C throughout the win- ter, until the end of March), and was 3° lower than that in the field (26.5°C). From March to November, behavior of the colony followed a roughly monthly cycle with week-long spawning peaks. During the fourth week of the cycle, the conch buried superficially. These cycles of behavior were not upset by natural disturbances such as tropical storms, e.g. Hurricane Gilbert (12 Sept 1988). However, the conch missed one cycle each year: that which followed the longest day, June 21. No spawning appeared to have taken place anywhere in the study area because egg masses, or rem- nants thereof, were not found (egg mass remnants usually remained in evidence for up to two weeks after the eggs themselves had hatched). This cyclic reproduction appeared to be associated with the lunar cycle, as spawning peaks were most frequently observed after a full moon. To test the regularity and strength of these cycles, two analyses were made. To test the regularity of the cycles, a sine wave function was fit to the data using non-linear regression. The model was: y = a + b * sine (week - c), where y equals the arc-sine transformation of the proportion of individuals reproductively active, and week equals the angle equivalent to lunar week (e.g. week 1 = 0 radians, week 2 = p/2 radi- ans, etc.); a is the shift in the sine function along the y-axis (y-intercept), b is the amplitude, and c is the phase shift (a, b, and c are constants). The regression was significant (p < 0.05) with a = 6.653 + 1.306 SE, b = 5.645 + 1.846 SE, and c = 0.007 + 0.327 SE. Since c was small and not signifi- cantly different from zero, peak reproductive activity was found to be centered around the second week following the full moon. However, the correlation coefficient was low (r2 = 0.400; p < 0.05), indicating that there was variability in the timing and/or strength of the spawning peaks. Spawning peaked during the second week after a full moon in 10 out of 14 cases (Fig. 1). The strength of these peaks was tested using an analysis of variance (ANOVA), to test for significant differences in the propor- tion of conch reproductively active (arc-sine transformed) among weeks. The ANOVA results (Table 1) showed a sig- nificant difference in mean percentage of spawners on a weekly basis (p = 0.031). Reproductive activity was great- est during the second week after a full moon relative to the other three weeks (p < 0.007; using post-hoc orthogonal contrasts). 120 AMER. MALAC. BULL. 11(2) (1995) Table 1. Analysis of variance used to test for differences among weeks, with respect to strength of reproductive peaks. Source Sum of Degrees of Mean‘ F-ratio _— Probability Squares Freedom Square week 924.101 3 308.034 3.186 0.031 error 5026.929 52 96.672 DISCUSSION Reproductive seasonality was observed in Strombus pugilis, with all reproductive activity ceasing from mid-November until mid-March. A similar annual cycle of behavior was noted in other colonies of S. pugilis located in the general area (Sanders, 1988; pers. obs.), as well as in other strombid species, such as S. gigas (fide Weil and Laughlin, 1984; pers. obs.), S. Juhuanus Linné, 1758 (Catterall and Poiner, 1983; Kuwamura et al., 1983), and S. costatus (fide L. A. Rodriguez Gil, pers. comm.) although the last spawned year-round with variable intensi- ty in Puerto Rico (R. S. Appeldoorn, pers. comm.). Be- cause long-term observations have not been made on popu- lations of the last two species, this type of seasonal behav- ior is not well documented, and probably differs by locale. The reproductive season of Strombus pugilis coin- cided with changes in daylength and water temperature. Spring emergence coincided with both increasing daylength and water temperature. Consequently, increasing daylength may be a major factor influencing spring emergence. Stoner et al. (1988) also suggested that spring emergence in juvenile S. gigas was influenced by increasing daylength and water temperature. The cessation of reproductive activity and onset of winter burial coincided with both decreasing daylength and water temperature in both the field and the laboratory. In Strombus gigas, Stoner et al. (1992) found that reproduc- tion gradually decreased with decreasing daylength but ceased altogether when temperature began dropping. The drop in water temperature during the winter also affected metabolic rate in S. pugilis (Sanders, 1990), suggesting that cessation of reproduction is determined by energetic requirements, especially because the intensity of reproduc- tion was so high during the season. Low temperature not only reduces the parental metabolic rate but also extends hatching time (Rodriguez Gil et al., 1991) such that larger eggs would have to be produced in winter in order to ensure viability and larval survival (Hughes, 1986). Kuwamura et al. (1983) found that the gonadal index of S. luhuanus decreased when temperature dropped, and that reproductive activity ceased altogether during winter months in Japan. Given these observations, temperature appears to be the main factor influencing reproductive seasonality in strom- bids. However, because of the confounding of daylength and temperature in field observations, controlled experi- ments will be needed to examine the differential effects of daylength and temperature on reproduction. During the reproductive season, spawning occurred on a monthly basis for Strombus pugilis, with a peak in spawning generally falling in the second week after a full moon. Such cyclic behavior could be a response to either internal endogenous rhythms or to external environ- mental factors, or more likely, a combination of both (Sollberger, 1965). No such periodicity has been reported for other species of Strombus. One possible explanation for the correlation of spawning with lunar periodicity is that larval release (hatching occurs 3-5 days after the egg mass is laid) would occur during the period of neap tides, which could act to restrict the dispersal of larvae to offshore waters where they may perish. From their work in the Grenadines, Posada and Appeldoorn (1994) felt there exists some mechanism limiting dispersal of Strombus larvae. However, tidal cur- rents along the south coast of Puerto Rico are exceedingly small (Kjerfve, 1986), so differences between spring and neap tides may not be biologically significant in this partic- ular area, but spawning at certain point(s) during the tidal cycle could have evolved as a mechanism for the retention and transport of larvae to inshore nursery grounds (Posada and Appeldoorn, 1994; Stoner et al., 1992). Using the lunar cycle to cue the periodicity would also serve to ensure that all members of the population were ripe and reproduc- tively active at the same time (Fretter and Graham, 1964). In the present study, while the reproductive cycle correlated positively with lunar phase, the correlation was imprecise. In four out of 14 cases, spawning peaks did not occur during the second week after a full moon. Furthermore, for both years one spawning cycle was missed, that coincident with the longest day and full moon thereafter, with conch remaining superficially buried for almost a month. Thus, lunar periodicity cannot be seen as the sole factor regulating reproductive activity, but may work in conjunction with, or be influenced by, other as yet undetermined factors. ACKNOWLEDGMENTS I would like to thank Dr. R. S. Appeldoorn for his help in the preparation of the original chapter of my thesis upon which this manu- script is based. LITERATURE CITED Bradshaw-Hawkins, V. I. 1982. Contributions to the Natural History of the West Indian Fighting Conch, Strombus pugilis Linnaeus 1758, REED: SPAWNING PERIODICITY IN STROMBUS PUGILIS (21 with Emphasis on Reproduction. Master’s Thesis, McGill University, Montreal, Canada. 131 pp. Brownell, W. N. 1977. Reproduction, laboratory culture, and growth of Strombus gigas, S. costatus and S. pugilus [sic] in Los Roques, Venezuela. Bulletin of Marine Science 27(5):668-680. Catterall, C. P. and I. R. Poiner. 1983. Age- and sex-dependent patterns of aggregation in the tropical gastropod Strombus luhuanus. Marine Biology 77:171-182. D’Asaro, C. N. 1965. Organogenesis, development, and metamorphosis in the queen conch, Strombus gigas, with notes on breeding habits. Bulletin of Marine Science 15(2):359-416. Egan, B. D. 1985. Aspects of the Reproductive Biology of Strombus gigas. Master’s Thesis, University of British Columbia, Vancouver, Canada. 147 pp. Fretter, V. and A. Graham. 1964. Reproduction. In: Physiology of Mollusca, vol. 1, K. M. Wilbur and C. M. Yonge, eds. pp. 127- 164. Academic Press, New York. Hughes, R. N. 1986. A Functional Biology of Marine Gastropods. John Hopkins University Press, Baltimore, Maryland. 245 pp. Kjerfve, B. 1986. Tides of the Caribbean Sea. Journal of Geophysical Research 86(CS):4243-4247. Kuwamura, T., R. Fukao, M. Nishida, K. Wada, and Y. Yanagisawa. 1983. Reproductive biology of the gastropod Strombus luhuanus (Strombidae). Publications of the Seto Marine Biological Laboratory 28(5-6):433-443. Percharde, P. L. 1968. Notes on distribution and underwater observations on the molluscan genus Strombus as found in the waters of Trinidad and Tobago. Caribbean Journal of Science 8(1-2):47-53. Percharde, P. L. 1970. Further underwater observations on the molluscan genus Strombus Linné as found in the waters of Trinidad and Tobago. Caribbean Journal of Science 10(1-2):73-77. Posada, J. and R. S. Appeldoorn. 1994. Preliminary observations on the distribution of Strombus larvae in the eastern Caribbean. In: Queen Conch Biology, Fisheries and Mariculture, R. S. Appeldoom and B. Rodriguez Q., eds. pp. 191-200. Fundacién Cientifica Los Roques, Caracas, Venezuela. Reed, S. E. 1993. Size differences between sexes (including masculin- ized females) in Strombus pugilis (Mesogastropoda: Strombidae). Journal of Shellfish Research 12: 77-79. Rodriguez Gil, L. A., J. Ogawa, and C. A. Martinez-Palacios. 1991. Hatching of the queen conch, Strombus gigas L., based on early life studies. Aquaculture and Fisheries Management 22:7-13. Sanders, I. M. 1988. Energy Relations in a Population of Strombus pugilis. Doctoral Dissertation, University of Puerto Rico, Mayaguez, Puerto Rico. 130 pp. Sanders, I. M. 1990. Seasonal changes in oxygen consumption of the West Indian fighting conch, Strombus pugilis Linnaeus, 1758. Journal of Shellfish Research 9(1):63-65. Sollberger, A. 1965. Biological Rhythm Research. Elsevier Publishing Co., Amsterdam, Netherlands. 461 pp. Stoner, A. W., R. N. Lipcius, L. S. Marshall, Jr., and A. T. Bardales. 1988. Synchronous emergence and mass migration in juvenile queen conch. Marine Ecology - Progress Series 49(1):51-55. Stoner, A. W., V. J. Sandt, and I. F. Boidron-Metairon. 1992. Seasonality in reproductive activity and larval abundance of the queen conch, Strombus gigas. Fisheries Bulletin 90(1):161-170. SYSTAT, Inc. 1992. SYSTAT for Windows: Statistics, Version 5. Evanston, Illinois. 750 pp. Weil, E. M. and R. G. Laughlin. 1984. Biology, population dynamics, and reproduction of the queen conch Strombus gigas Linné in the Archipiélago de Los Roques National Park. Journal of Shellfish Research 4(1):45-62. Date of manuscript acceptance: 29 June 1994 ee OS The stygobiont genus Bythiospeum in Austria: a basic revision and anatomical description of B. cf. geyeri from Vienna (Caenogastropoda: Hydrobiidae) Martin Haase Institut fiir Zoologie der Universitat Wien, Althanstr. 14, A-1090 Wien, Austria. Abstract. The type series of the eight Austrian nominal species of the stygobiont genus Bythiospeum, all originally described only from their shells, are compared morphometrically. Two series attributed to B. geyeri (Fuchs, 1925) and one sample of an hitherto undescribed species from the type locality of B. tschapecki (Clessin, 1878), are also included in this investigation. Several of the taxa have been classified as subspecies or synonyms of other taxa in a vari- ety of combinations. These classifications were based on poorly defined, i.e. non-quantified, similarities. It is demonstrated that, in the absence of anatomical data, systematic/taxonomic speculations exclusively based on conchological similarities are pointless and misleading. In addition, Bythiospeum cf. geyeri from the groundwater of the river Danube in Vienna is described anatomically. It lacks eyes and epidermal pig- ment, and is characterized by a number of presumably plesiomorphic (with respect to other hydrobiid genera) features, such as a simple penis, well devel- oped hypobranchial gland, and a duct connecting the prostate and mantle cavity. The genus Bythiospeum Bourguignat, 1882 [syn- onyms: Vitrella Clessin, 1877, non Swainson, 1840; Lartetia Boettger, 1905, non Bourguignat, 1869; Paladilhiopsis Pavlovic, 1913 (cf. Zilch, 1970; Giusti and Pezzoli, 1982)] is exclusively stygobiont. It comprises more than 100 nominal species and subspecies and is reported from the Netherlands throughout western and central Europe, the Balkans, Asia minor, to the Caucasus and Usbekistan (Boettger, 1905; Bole and Velkovrh, 1986; Kuijper and Gittenberger, 1993). The great majority of these nominal species and subspecies has been only vague- ly described from shells and there is great confusion as to possible synonymies, because shell characters provide little information for hydrobiid systematics. A comprehensive revision of the genus would have to be based on exact mor- phometric descriptions and, primarily, on anatomical inves- tigations. However, stygobiont hydrobiids are rarely encountered alive, which is rather discouraging if one con- siders the great number of taxa. Thus, it is not surprising that the few revisionary papers concentrate on limited geo- graphic areas. Boeters (1984b) developed some ideas for the revision of the German species and Bernasconi (1990) focused on France, Switzerland, and also Germany. The recent discovery of a considerable number of living specimens of Bythiospeum cf. geyeri (Fuchs, 1925) in the groundwater of the river Danube in Vienna gave rise to a revision of the eight Austrian taxa: B. tschapecki (Clessin, 1878), B. pfeifferi (Clessin, 1890), B. geyeri, B. elseri (Fuchs, 1929), B. noricum (Fuchs, 1929), B. borman- ni (Stojaspal, 1978), B. cisterciensorum (Reischitz, 1983), and B. reisalpense (Reischiitz, 1983). Throughout a series of papers, Reischiitz (1981, 1983a, b, c) considered only B. tschapecki and B. pfeifferi as valid species. The other taxa were treated either as species, subspecies (in a variety of combinations), or synonyms. Reischiitz’s argument was based on conchological comparisons, but his argumentation is not convincing, because it lacks any quantification. The present paper provides a morphometric analysis including the type series of the nominal Austrian species of the genus Bythiospeum, presumptive topotypes of B. pfeif- feri, a series of two populations attributed to B. geyeri by previous authors (e.g. Klemm, 1960; Pospisil, 1989; Reischiitz, 1988b), and one lot of an undescribed species from the type locality of B. tschapecki, with which it has been confused by Reischiitz (1983b). The anatomy of B. cf. geyeri from the groundwater of the river Danube in Vienna is described in full detail and its evolutionary significance discussed. MATERIALS AND METHODS Part of Clessin’s collection including the type series of Bythiospeum pfeifferi deposited in the State Museum of Natural History in Stuttgart (SMNS) was destroyed during American Malacological Bulletin, Vol. 11(2) (1995): 123-137 123 124 AMER. MALAC. BULL. 11(2) (1995) World War II (Zilch, 1970; Niederhfer, pers. comm.). But four shells of Clessin’s original material of B. pfeifferi still exist in the Museum of Natural History in Vienna (NHMW). For the species described by Fuchs and Clessin, lecto- and paralectotypes are designated from the syntypic series. All series except that of B. cf. geyeri from well T3 in Vienna consist of empty shells collected in caves, wells, springs and riverine alluvia. Material examined (see Fig. | for localities): Bythiospeum bormanni: coll. Reischiitz: cave Barenloch near Mixnitz. B. cisterciensorum: NHMW 82577: spring in Sieben- brunn/Tirnitz. B. elseri: lectotype: NHMW 87187; paralectotypes: NHMW 33180: Gresten; [NHMW/K 48830: Traun near Wels, | shell, not used in this analysis]. B. geyeri: lectotype: NHMW 87186; paralectotypes: NHMW/K 6815, NHMW/E 33143, NHMW/E 33144, NHMW 53279: all from Eichhornquelle (Quelle = spring) in Sch6nbithel/Donau; coll. Rusnov, NHMW/E 1630, NHMW/K 44000: all allu- vium of the Danube in Schénbihel. B. cf. geyeri: NHMW 5276: Weidlingbach; well T3 at the Eberschiittwasser in Vienna. B. noricum: lectotype: NHMW 87188; paralectotypes: NHMW/K 48785: Gaflenz in Weyer/Enns. B. pfeifferi: lectotype: NHMW 87189; paralectotypes: NHMW 22382; presumptive topotypes: NHMW 29471, NHMW/E 1622, NHMW/K 48826: all Kremsmunster. B. reisalpense: NHMW 82576: spring in Kleinzell. Weyer: B. noricum Kremsmunster: B. pfeifferi B. tschapecki: lectotype: SMNS ZI 9429/1; paralectotypes: SMNS ZI 9429/2-4; Buchkogelhohle near St. Martin. B. sp.: coll. Reischiitz: Buchkogelhohle near St. Martin. Shells were measured with a dissecting microscope fitted with an ocular micrometer. Only shells with a fully developed peristome were used (cf. Fig. 8A). Parameters measured included shell height (sh), shell width (sw), aper- ture height (ah), and aperture width (aw). In addition, the ratios sh:sw, ah:aw, sh:ah and sw:aw were calculated. The means and the confidence intervals of the means (p = 0.05) of all eight parameters were plotted to sun ray diagrams, whose axes are limited by the respective minima and maxi- ma. Differences among series are considered statistically significant if the confidence intervals do not overlap. In addition, pairwise Euclidean distances between the series were calculated from the z-transformed means of each vari- able. Cluster analyses [Unweighted Pair Group Method using Arithmetic averaging; UPGMA] and minimum span- ning trees were computed based on all eight parameters (overall similarity) and on the four ratios (shape), respec- tively. UPGMA cluster analysis was carried out using the program NEIGHBOR of Felsenstein’s (1993) package PHYLIP, version 3.5c. NEIGHBOR generates UPGMA dendrograms with half the branch lengths produced by the usually used algorithm of Sneath and Sokal (1973). Therefore, the branch lengths computed by NEIGHBOR were multiplied by two. The program SLINK, employing a single-linkage algorithm, written by Dr. Nemeschkal (Vienna), was used to generate the minimum spanning trees. Discriminant analysis and analysis of variance could Gresten: B. e/seri Schénbuhel: B. geyer B. cf. geyeri Weidlingbach: B. cf. geyer Vienna: B. cf. geyer Kleinzell: B. reisalpense Siebenbrunn: B. cisterciensorum Mixnitz: B. bormanni St. Martin: B. tschapecki B. sp. Fig. 1. Map of Austria showing the localities of the series investigated in this study. HAASE: AUSTRIAN BYTHIOSPEUM 125 not be performed because the variances of the measure- ments are not homogeneous (Bartlett’s test). The living specimens of Bythiospeum cf. geyeri (11 females, 3 males) were collected with a double-packer sampler connected to a hand-pump (Danielopol and Niederreiter, 1987) in the groundwater monitoring well T3 (a pipe with a diameter of 4.8 cm) near the Eber- schiittwasser, a backwater of the river Danube, in Vienna, Austria, from a depth of 10-12 m by Dr. P. Pospisil in September 1992. The animals were fixed unrelaxed with Bouin’s fixative. Eight females and two males were embed- ded in paraffin and sectioned at 7 pm. The serial sections were Stained with Heidenhain’s Azan. The anatomy was reconstructed from the series of one female and one male using the computer program PC3D (Jandel Scientific). Two females and one male were dissected. The head-foot of the male was critical-point dried after removal of the mantle and investigated with a scanning electron microscope (SEM). One radula and representative shells were also examined with the SEM. All of these methods are described in more detail in Haase (1992). RESULTS MORPHOMETRICS The shell measurements of all species and popula- tions investigated are given in Table 1. The series from the Eichhornquelle, Schénbihel (both Bythiospeum geyeri) and Kremsmiinster (B. pfeifferi, collected by Pfeiffer), respec- tively, were pooled in order to increase the number of speci- mens for the analysis. This is justified because Fuchs’s and Pfeiffer’s original series were partly distributed among sev- eral collectors (indicated by the lables; Wawra, pers. comm.), whose collections are now deposited in NHMW. Shell parameters of the four populations of B. geyeri (including B. cf. geyeri) are compared in Fig. 2. The signifi- cant differences are summarized in Table 2. Shell parameters of the type series of the Austrian species of Bythiospeum (with B. geyeri represented by the series from the Eichhornquelle) and B. sp. from St. Martin are contrasted in Fig. 3. The significant differences of Fig. 3 are summarized in Table 3. Bythiospeum elseri, B. geyeri and B. noricum differ only in size parameters. B. pfeifferi can also be distinguished from B. geyeri and B. noricum only in size. The same holds for comparison of B. tschapec- ki with B. geyeri and B. noricum, and B. sp. with B. elseri. B. tschapecki and B. pfeifferi cannot be distinguished at all. However, one has to be careful drawing conclusions from Figs. 2 and 3, because several samples contain only a few specimens, which results in quite large confidence intervals and thus less clear separation between the series. Figs. 4 and 5 show UPGMA phenogram and a mini- mum spanning tree, respectively, based on the Euclidean distances in Table 4 considering size and shape parameters. Figs. 6 and 7 are based on the shape parameters only. These dendrograms and minimum spanning trees are dealt with in more detail in the Discussion. DESCRIPTION OF BYTHIOSPEUM CF. GEYERI FROM VIENNA SHELL The shell (Fig. 8) is unsculptured and comprises up to five whorls. The outer lip is curved (Fig. 8B). The proto- conch (Fig. 8C) is smooth and has 1.5 whorls. Measurements are given in Table 1. OPERCULUM The light orange operculum is paucispiral with an excentric nucleus. EXTERNAL FEATURES The epidermis is unpigmented. Yellow granules lie in the musculature and connective tissue of the foot. Black granules are dispersed in the connective tissue around the stomach and the distal lobes of the digestive gland (Fig. 9). These black granules lie close to the nuclei of the cells of the connective tissue, which have large vacuoles. The snails have no eyes, not even rudiments. The tentacles bear a median row of cilia (Fig. 10). MANTLE CAVITY EXCEPT DISTAL GENITAL ORGANS The ctenidium consists of 14-16 filaments. The hypobranchial gland is very massive and extends almost the full length of the pallial oviduct (Fig. 11A). In the roof of the mantle cavity the intestine forms a loop, which is about half as long as the albumen gland in most specimens. Some individuals, such as the one depicted in Fig. I11A, have a somewhat shorter loop. DIGESTIVE SYSTEM The radula (Fig. 12) includes more than 100 rows and is described by the formula R: y = ,L:414,Mi: 18 - 21, M2: 16 - 19. Esophagus and digestive gland open side- by-side into the proximal end of the stomach, which is bent toward the digestive gland. Stomach and digestive gland are not very close and are connected via a duct. The digestive gland is a large sac with four to five prominent, distal lobes (Fig. 13). NERVOUS SYSTEM The cerebral and pleural ganglia are separated by short, but distinct, connectives so that the ganglia are not 126 Table 1. Shell morphometry; measurements in mm, s*100/x in %. (ah, aperture height; aw, aperture width; max, maximum; min, minimum; N, number of specimens; s, standard deviation; s*100/x, coefficient of variation; sh, shell height; sw, shell width; w, maximum number of whorls; x, mean). B. bormanni Mixnitz N=17 w = 4.75 B. cisterciensorum Siebenbrunn N=4 w = 4.75 B. elseri Gresten N=7 w = 4.7 B. geyeri Eichhornquelle N=8 w=5.5 B. cf. geyeri Schonbihel N=15 w =5.25 B. cf. geyeri Vienna, T3 N=15 w =5.0 B. cf. geyeri Weidlingbach N=10 w =5.0 B. noricum Weyer N=5 w=5.5 B. pfeifferi Kremsminster N = 20 w =5.0 B. reisalpense Kleinzell N= 13 w =5.25 B. sp. St. Martin N=5 w=5.5 B. tschapecki St. Martin AMER. MALAC. BULL. 11(2) (1995) sh 1.73 2.03 1.87 0.09 4.58 2.03 2.58 2223 0.25 11.42 1.75 2.10 1.90 0.11 5.92 2:23 2.60 2.38 0.12 5.20 1.90 2.45 2.20 0.15 7.01 1.83 2.48 2.17 0.19 8.87 1.68 2.30 2.03 0.21 10.08 1.88 2.28 2.06 0.17 8.39 22s 3.08 2.59 0.23 8.91 1.70 2.60 2.04 0.23 11.42 2.08 2:93 2.31 0.17 7A5 2.59 3.41 3.00 0.34 11.23 SW 0.85 1.03 0.91 0.05 4.95 1.13 1.45 1.26 0.14 10.79 0.75 0.83 0.79 0.03 4.03 0.93 1.08 1.03 0.05 4.88 0.78 1.08 0.97 0.07 6.89 0.83 1.08 0.94 0.07 7.39 0.75 1.00 0.86 0.09 10.45 0.85 1.05 0.94 0.07 7.93 0.92 1.31 1.16 0.10 8.25 0.90 1.20 1.05 0.09 8.55 0.92 1.03 0.97 0.05 4.78 1.15 1.80 1.38 0.29 21.21 aw 0.58 0.68 0.62 0.03 4.36 0.88 1.08 0.96 0.09 9.60 0.50 0.55 0.52 0.02 3.31 0.65 0.80 0.70 0.05 6.61 0.50 0.70 0.65 0.05 8.12 0.53 0.75 0.63 0.06 9.51 0.48 0.70 0.58 0.07 12.82 0.53 0.65 0.59 0.05 8.33 0.64 0.97 0.80 0.09 11.15 0.63 0.85 0.73 0.07 9.15 0.56 0.64 0.61 0.03 5.35 0.82 1.02 0.91 0.09 9.92 sh:sw 95 2.32 2.07 0.10 4.84 1.67 1.80 1.76 0.06 3.42 2.25 2.55 2.40 0.11 4.54 2.19 2.60 2.33 0.13 5.46 2.03 2.45 221 0.11 4.97 2.14 2.46 2.30 0.09 4.03 2.16 2.58 2.37 0.15 6.30 2.03 2.42 2.21 0.14 6.38 2.08 2.49 2.23 0.11 5.04 1.74 2.17 1.93 0.11 5.92 2.19 2.68 2.37 0.18 7.65 1.89 2.37 2.21 0.22 9.86 ah:aw 0.96 1.13 1.06 0.05 4.36 1.03 1.06 1.05 0.01 1.20 1.14 1.24 1.17 0.04 3.30 1.11 1.19 1.16 0.03 2.35 1.14 1.26 122 0.04 3.30 1.11 1.28 1.21 0.05 3.71 1.11 1229 1.19 0.06 4.88 1.14 1.25 1.20 0.04 3.64 1.07 1.25 1.15 0.05 4.20 1.07 1.24 1.13 0.05 4.43 1.09 L222 1.16 0.05 3.99 1.14 1.23 1.17 0.04 3.44 sh:ah 2.74 2.93 2.83 0.07 2.28 2.14 2.29 2.21 0.06 2.77 3.00 3.36 3.11 0.13 4.14 2.68 3.15 2.94 0.13 4.56 2.61 3.04 2.80 0.12 4.39 2.69 3.11 2.84 0.11 3.93 2.75 3.15 2.96 0.12 3.92 2.77 3.13 2.95 0.16 5.56 2.48 3.20 2.82 0.16 5.75 2.27 2.67 2.48 0.12 4.75 3.13 3.38 3.28 0.11 3.40 212 2.91 2.81 0.10 3.54 SW:aw 1.36 1.58 1.46 0.08 5.14 1:25 1.40 1.31 0.08 5.70 1.43 1.57 1.51 0.05 3.01 1.34 1.54 1.47 0.07 4.51 1.39 1.58 1.50 0.06 3.88 1.39 1.61 1.49 0.07 4.52 1.43 1.58 1.49 0.06 3.94 1.50 1.76 1.60 0.10 6.22 1.29 1.63 1.46 0.10 6.91 1.39 1.56 1.45 0.05 3.60 1.48 1.70 1.61 0.08 5.01 1.31 1.77 1.51 0.20 13.42 HAASE: AUSTRIAN BYTHIOSPEUM sh\2.60 | 127 mate 68 fh | H 1.34 3.15 Ver >. {119 64 aa sh:ah Ae 0.93 111 : / . 129 /an:aw \aw | A 2.60) sh:sw E Ss Vv Ww Fig. 2. Sun ray diagrams comparing shell parameters in the four series of, or attributed to, Bythiospeum geyeri. A. Comparison of all four series. B. Each series on a separate diagram. The means of the variables of each series are connected; in A the confidence intervals are also plotted on the axes, which are limited by the respective maxima and minima. The series are identified by the initial letter of their respective locality. (ah, aperture height; aw, aperture width; E, Eichhornquelle; S, alluvium of the river Danube in Schénbiihel; sh, shell height; sw, shell width; V, Vienna; W, Weidlingbach.) fused. The pleurosupraesophageal connective is much longer than the pleurosubesophageal connective. The tenta- cle nerves arise from a ganglionic thickening at the distal end of the cerebral ganglia. FEMALE GENITAL SYSTEM (FIGS. 11, 13A) The ovary is a simple sac. The glandular, renal oviduct makes a wide loop. The gonopericardial duct branches off at the proximal end of the renal oviduct. It cannot be determined from sections whether this is still an open duct or a massive strand of tissue. Albumen and cap- sule glands are both divided into two histologically differ- ently staining sections. The genital chamber lies behind the albumen gland. The former is inclined towards the latter. The seminal receptacle is ventral to the bursa copulatrix. MALE GENITAL SYSTEM (FIGS. 10, 14) The testis is a simple sac. The seminal vesicle is weakly developed with only a few coils. The proximal vas deferens opens into the prostate near its posterior end. The distal vas deferens originates close to its tip. The penis is simple; it has no appendages, neither muscular nor glandular. The prostate overlies the mantle cavity only by its anterior third; it is connected to the mantle cavity through a duct, which arises somewhat behind the distal vas deferens. 128 AMER. MALAC. BULL. 11(2) (1995) sh|3.41 ——pbo Fig. 3. Sun ray diagrams contrasting shell parameters in the type series of the Austrian taxa of Bythiospeum and B. sp. from St. Martin, A. Comparison of all nine series. B. Each series on a separate diagram; the series are identified by their first two initials, see also legend of Fig. 2. (bo, B. bormanni; ci, B. cister- ciensorum, el, B. elseri;, ge, B. geyeri; no, B. noricum, pf, B. pfeifferi; re, B. reisalpense; sp, B. sp.; ts, B. tschapecki) HAASE: AUSTRIAN BYTHIOSPEUM 12? Table 2. Significantly different shell parameters extracted from Fig. 2. The parameters are abbreviated by numbers: 1, sh; 2, sw; 3, ah; 4, aw; 5, sh:sw; 6, ah:aw; 7, sh:ah; 8, sw:aw. E S Vv W Eichhornquelle (E) — 6 1,2,6 1-4 Schénbiihel (S) — — 2,3,7 Vienna (V) = — Weidlingbach (W) —— DISCUSSION TAXONOMY The generic allocation of the Austrian species fol- lows Giusti and Pezzoli (1982), who synonymized Bythiospeum Bourguignat, 1882, and Paladilhiopsis Pavlovic, 1913. Bernasconi’s (1990) distinction between these nominal genera is not convincing. Bythiospeum pfeifferi (originally Vitrella pfeifferi Clessin, 1890) is usually cited with the date 1887 (Ehrmann, 1933; Klemm, 1960; Reischiitz, 1981; Bole and Velkovrh, 1986). However, Clessin’s opus appeared in sev- eral parts, the last one (1890) containing the description of this species (Boeters, 1967). In the same year, Pfeiffer (1890) reported observations on living specimens he kept in an aquarium, describing the habitus of the snails. Pfeiffer, who left the first description of shells of this species to Clessin and who knew the page proofs of Clessin’s opus, attributed the taxon to Clessin, but his description is inde- pendent and does not cite Clessin. Also in 1890 Westerlund (1890) published a description of “Paludina (Bythiospeum) 6 5 4 3 2 1 @) distance Fig. 4. UPGMA dendrogram based on Euclidean distances in Table 4 con- sidering all eight shell parameters; the series are numbered as in Table 4. ——1——10 ———2 Fig. 5. Minimum spanning tree based on the same data set as Fig. 4; branch lengths are proportional to distances. pfeifferi Clessin,” but he unambiguously copied Clessin’s description and also referred to the latter. Thus, Westerlund is certainly not the author of B. pfeifferi. Both Clessin’s and Pfeiffer’s contributions appeared early in 1890, although the exact dates could not be determined. Following the principle of the first reviser (Article 24; ICZN, 1985), I give precedence to Vitrella pfeifferi Clessin over V. pfeifferi Pfeiffer, because this taxon has never been used in conjunc- tion with Pfeiffer, and because Pfeiffer himself attributed the authorship to Clessin. MORPHOMETRICS It is interesting to note that the coefficients of varia- tion (s*100/x) of the measurements in most cases exceed 6.5%, whereas they are mostly below 6% for the ratios (Table 1). In species of the hydrobiid genera Belgrandiella Wagner, 1828, and Graziana Radoman, 1975, Haase (1994) found significantly less variation in size. For the pulmonate Arianta arbustorum (Linné, 1758), Baur (pers. comm.) sug- gested that a value less than 5% can be considered typical for a population. Higher variation might indicate that the sample consists of individuals of two or more populations. The results for the series measured in this study allow two possible conclusions. Either (most) populations of Bythiospeum species are more variable in size than those of other gastropods, or (most or some of) the series contain individuals of at least two populations or species, which are very similar in shape. At least for the alluvial samples, the latter explanation seems quite plausible. In fact, Fuchs’ (1929) material of B. noricum (NHMW/K 48 785) is appar- 130 AMER. MALAC. BULL. 11(2) (1995) Table 3. Significantly different shell parameters extracted from Fig. 3; parameters abbreviated by numbers 1 to 8 as in Table 2. bo ci el ge no pf re sp ts B.bormanni(bo) ———===== 2-5,7 2-7 1,3-6 6 2-6 2-7 1,5-8 1,3,6 B.cisterciensorum (Ci) wee 2-8 4-7 2-8 5-7 4-7 2-8 5,6 B.elseri(el) 1-4 2 1-5,7 2-5,7 1-4 1-4,7 B.geyeri(ge) 4 1-4 1,5,7 3,4,7 1,3,4 B.noricum (mo) 1-4 4,5,7 7 1,3,4 B. pfeifferi (pty nn 1-3,5,7. 2-4,7,8 —----- B.reisalpense (tre) 3-5,7,8 1,3,7 B.sp.(St.Martin) (sp) 3,4,7 B.tschapecki (ts) eee ently composed of two species, of which one, represented 11 by two shells, was excluded from the analysis along with one specimen from Reischiitz’s (1983b) material of “B. tschapecki” (here B. sp., the specimen excluded is probably 3 a true B. tschapecki). [The two shells separated from the B. | noricum series and B. sp. from St. Martin probably repre- 7 sent undescribed species. However, I refrain from describ- | ing new species based solely on shell morphology, because | this only increases the number of poorly defined taxa in this 6 5 genus (see also discussion below).] However, the anatomi- | cal investigation of the series attributed to B. geyeri from 12-6 Vienna (s*100/x > 7% for the size parameters) did not indi- | cate the presence of more than one species in the drainage 9 area of well T3. This sample certainly represents a single | 1 : | 10 10 3 4 7 2 6 Fig. 7. Minimum spanning tree based on the same data set as Fig. 4; branch lengths are proportional to distances. 12 9 population, which in turn, supports the first assumption, that populations of Bythiospeum can be quite variable. 8 In his first review of Austrian species of the genus 11 Bythiospeum, Reischitz (1981) ranked all hitherto described taxa as separate species, but suspected that only S B. geyeri and B. pfeifferi were valid. He also speculated that B) the Austrian taxa might be closely related to B. acicula 5 4 3 2 1 O distance Fig. 6. UPGMA dendrogram based on Euclidean distances in Table 4 con- sidering only shape parameters; the series are numbered as in Table 4. (Held, 1837) from the groundwaters of the river Isar near Munich. [Paludina acicula Held, 1837, was not included in this investigation because the syntypes are lost (fide Zilch, 1970) and the name itself is dubious, probably preoccupied by Acmea acicula Hartmann, 1821 (cf. Boeters, 1984b). Hence, it does not make sense to speculate about relation- ships of the Austrian taxa to a German species at the pre- sent state of revision of the German taxa (cf. Boeters, HAASE: AUSTRIAN BYTHIOSPEUM 13] Table 4. Matrix of Euclidean distances calculated from z-transformed means of each variable; above the diagonal: all eight parameters; below: only the ratios (see Table 1); the populations of Bythiospeum geyeri are distinguished by the initials of their localities. (E = Eichhornquelle, S = Schénbiihel, V = Vienna, W = Weidlingbach). 1 2 3 4 5 6 7 8 9 10 11 12 1.B.bormanni —_ ----- 5.391 2.473 2.603 4.221 1.906 1.958 2.323 3.645 2.273 3.389 5.865 2. B. cisterciensorum 3.352. ----- 7.513 5.276 6.603 5.631 6.542 6.653 4.267 3.446 7.421 4.899 3. B. elseri 2.221 5.403 —----- 3.001 4.078 2.183 1.103 2.084 4.590 4.367 2.357 6.705 4. B. geyeriE 1.491 4.475 0.968 — ----- 3.466 1.175 1.994 2.506 1.625 2.869 2.495 3.820 5. B. cf. geyeri S 3.981 5.720 3.501 3.370 ——----- 3.080 3.436 3.586 4.088 4.199 4.097 5.538 6. B. cf. geyeri V 1.478 4.396 1.162 0.571 3.071 — ----- 1.106 1.697 2.571 2.638 2.327 4.743 7. B. cf. geyeri W 1.831 4.866 0.624 0.462 3.225 0.584 — ----- 1.717 3.565 3.468 2.186 5.732 8. B. noricum 2.205 5.277 1.648 1.900 3.474 1.560 1653 ~~ ----- 3.664 3.374 1.700 5.475 9. B. pfeifferi 0.935 3.843 1.566 0.668 3.415 0.669 1.034 1.980 ~— ----- 2.845 3.685 2.342 10. B. reisalpense 1.499 2.309 3.446 2.642 4.087 2.391 2.930 3.006 1.980 ~— ----- 4.513 4.668 11. B. sp.(St. Martin) 3.058 6.373 1.374 2.235 4.030 2.230 1.881 1.488 2.675 4.232 _~—_ ----- 5.241 12. B. tschapecki 1.138 4.147 1.464 0.912 3.258 0.513 0.998 1.370 0.659 2.028 2.311 — ----- Fig. 8. Bythiospeum cf. geyeri from Vienna (SEM). A. Shells, the right one from a subadult specimen. B. Outer lip. C. Apical view. Scale bars = 1 mm (A); 100 pm (B,C). 132 AMER. MALAC. BULL. 11(2) (1995) Fig. 9. Bythiospeum cf. geyeri from Vienna; longitudinal section through the stomach. Arrows indicate dark granules in the connective tissue. (cm, columellar muscle; dg, digestive gland; in, intestine; k, kidney; ss, style sac; st, stomach.) Scale bar = 100 pm. 1984b; Bernasconi, 1990).] In 1983 (Reischtitz, 1983a), B. elseri became a synonym, and B. noricum a subspecies, of B. geyeri without any comment. In the same paper Reischiitz stated that stunted forms of B. cisterciensorum and B. reisalpense approach B. geyeri. Some months later, Reischiitz (1983b) assumed that the Austrian taxa might in fact represent several races of a single species. Yet in the very same year, Reischtitz (1983c) classified B. geyeri, B. noricum, B. reisalpense, and B. cisterciensorum explicitly as subspecies of B. acicula, and B. bormanni became [as originally described (Stojaspal, 1978)] a subspecies of B. pa Fig. 10. Bythiospeum cf. geyeri from Vienna; SEM micrograph of a male, with mantle cavity removed. Arrow indicates dorsomedian row of cilia on the tentacle. (s, snout; p, penis; t, tentacle.) Scale bar = 100 mm. tschapecki. Besides more or less explicit biogeographic considerations, Reischiitz argued primarily based on simi- larities, but he did not define these similarities, i.e. he did not provide measurements of the shells. Thus, the princi- ples behind these classifications and their rapid changes remain unclear. Figs. 3-7 seem to confirm Reischiitz’s conclusions in that Bythiospeum elseri and B. geyeri are very similar. B. noricum and B. sp. from St. Martin also come close to the former taxa. Considering only shape, we also find, in addi- tion to the above-mentioned species, B. pfeifferi and B. pa Fig. 11. Bythiospeum cf. geyeri from Vienna, mantle cavity organs (except gill and osphradium) of a female. A. From dorsal. B. From the left. (aa, anterior albumen gland; ac, anterior capsule gland; be, bursa copulatrix; db, bursal duct; go, genital opening; gp, gonopericardial duct; hg, hypobranchial gland; in, intestine; od, oviduct, pa, posterior albumen gland; pc, posterior capsule gland; rs, receptaculum seminis; vc, ventral channel.) Scale bar = 200 pm. HAASE: AUSTRIAN BYTHIOSPEUM 133 Fig. 12. Bythiospeum cf. geyeri from Vienna, radula (SEM). A. seven transverse rows; B. Rachidian teeth. Scale bars = 10 pm (A); 5 pm (B). tschapecki in a cluster of slender species (Fig. 6). B. cister- ciensorum and B. reislapense, on the other hand, can hardly be regarded as similar to B. geyeri (contra Reischiitz, 1983a, c, 1988b). They are the relatively widest (sh:sw) species (Fig. 3). B. bormanni comes closest to them (Figs. 3-7). A relationship to B. tschapecki based on conchologi- cal similarity can hardly be inferred (contra Stojaspal, 1978; Reischiitz, 1983c). B. sp. and B. tschapecki are obvi- ously two distinct, sympatric species. Because Reischitz (1983b) did not realize that his series from St. Martin con- tained two species (see above), it is unclear whose anatomy he described. The four series of or attributed to Bythiospeum gey- eri are quite distinct in size, but relatively similar in shape (Figs. 2, 4-7). Only the alluvial sample from Schénbihel, which is almost equally distant to all other series (Table 4), does not cluster with its (presumably) conspecific popula- tions (Figs. 4 and 6), but it has its nearest neighbor on the minimum spanning trees of Figs. 5 and 7 in B. cf. geyeri from Vienna. Whether these four series really belong to one species can hardly be determined from these morphometric comparisons. The same holds for whether Bythiospeum elseri should or should not be considered synonymous with B. geyeri. In general, shell morphology is hardly appropri- ate for the analysis of relationships among hydrobiids. Because the shells are, as in the present case, mostly very simple, it is often impossible to assess whether similarities are due to common descent or to convergence. Therefore, systematic statements must be based primarily on anatomi- cal investigations. There are also cases of sibling species, i.e. Species with identical shell shape but distinct anatomy [e.g. Hemistomia caledonica Crosse, 1872, and H. fluminis Cockerell, 1930 (pers. obs.), or two Austrian species of Belgrandiella Wagner, 1928 (Haase, in prep.)], as well as cases where conspecific populations reveal significant dif- ferences [e.g. Graziana pupula (Westerlund, 1886) (Haase, 1994)]. For the present contribution it should be considered again that especially the alluvial samples might contain shells of more than one population or species, which would have distorted the analysis severely. Until more anatomical data are available for Austrian Bythiospeum, it is suggested to classify each sample as separate species. Only then is it meaningful to discuss relationships and biogeographic implications. ECOLOGY OF BYTHIOSPEUM CF. GEYERI FROM VIENNA Danielopol et al. (1992) measured parameters such as temperature, dissolved oxygen, dissolved organic car- bon, and bacterial activity, in several monitoring wells around the Eberschiittwasser. Most noteworthy is the low concentration of oxygen. During more than six of nine months of continuous observation, the conditions were hypoxic with oxygen concentrations below | mg/I. Only in winter did the situation seem to improve. Despite this defi- ciency of oxygen, Pospisil (1989) found quite a rich stygo- fauna with a number of amphipod, cyclopoid, isopod, ostra- cod, and harpacticoid crustaceans, and a second hydrobiid gastropod, Lobaunia danubialis Haase, 1993 (Haase, 1993a). How these animals are adapted to these hypoxic con- ditions is hardly understood. Danielopol et al. (1992) observed behavioral differences between the isopods Proasellus slavus (Remy, 1948) and Asellus aquaticus (Linné, 1758), the former a stygobiont and the latter from epigean waters, when exposed to different oxygen concen- trations. P. slavus kept the frequency of pleopod beats con- stant in contrast to A. aquaticus. Although the high toler- ance of oxygen deficiency of P. slavus can certainly not be explained by behavior alone, the underlying physiological 134 AMER. MALAC. BULL. 11(2) (1995) Fig. 13. Bythiospeum cf. geyeri from Vienna, genital and digestive systems (without salivary glands) of a female. A. Aspect from the right; different stip- plings serve only for clarity and do not imply structures or colors. B. Section of the digestive system, slightly turned. (aa, anterior albumen gland; an, anus; bc, bursa copulatrix; dg, digestive gland; e, esophagus; hg, hypobranchial gland; in, intestine; od, oviduct; ov, ovary; pa, posterior albumen gland; pc, poste- rior capsule gland; ph, pharynx; ra, radula sheath; rs, receptaculum seminis; ss, style sac; st, stomach; vc, ventral channel.) Scale bar = 200 pm. adaptations remain unclear. The adaptations of Bythiospeum geyeri must be primarily physiological, because the gill is definitely reduced in size and in number of filaments. ANATOMY OF BYTHIOSPEUM CF. GEYERI FROM VIENNA The lack of pigmentation and eyes is a consequence of the subterranean habitat of Bythiospeum geyeri. Seibold (1904) claimed to have found rudiments of the eye in B. quenstedti (Wiedersheim, 1873), but his figure is not con- vincing. He probably mistook a large cell for the anlage of an eye. The black granules in the connective tissue around the stomach and the distal lobes of the digestive gland were also found in Bythiospeum quenstedti (fide Seibold, 1904). Their function is not known. Bythiospeum geyeri has a number of plesiomorphic features, that is, features that are believed to be primitive within the family Hydrobiidae: the large hypobranchial gland, the rachidian tooth with only a single pair of basal cusps, the short but distinct cerebropleural connectives, the sac-shaped female and male gonads, the simple penis lack- ing any muscular or glandular appendages, and the duct connecting the prostate and mantle cavity. The hypobranchial gland is often reduced in hydro- biids. In Belgrandiella and Graziana, its size can be used to diagnose species (Haase, 1994). In many hydrobiid species cerebral and pleural gan- glia are fused. The retention of a distinct connective is cer- tainly primitive. The forms of the rachidian tooth, the penis, and the HAASE: AUSTRIAN BYTHIOSPEUM es) Fig. 14. Bythiospeum cf. geyeri from Vienna, genital system of a male. (dv, distal vas deferens; mp, duct connecting mantle cavity and prostate; pr, prostate; Pv, posterior vas deferens; te, testis; vs, vesicula seminalis.) Scale bar = 200 pm. gonads are believed to be primarily simple and not reduced (as to the gonads contra Davis, 1980). The prostatic duct opening into the mantle cavity has not been reported for any other species of the genus Bythiospeum. In dissections it is hardly detectable, but Seibold (1904) and Krull (1935), who also made histologi- cal serial sections of B. quenstedti, did not mention it either. But it is present in Hydrobia Hartmann, 1821 (Johansson, 1948; Haase, 1993b) and Heleobia dobrogica (Negrea and Grossu, 1989) (pers. obs.). From an evolutionary point of view this duct represents the incomplete closure of the open, pallial, prostatic groove found in basal caenogas- tropods (Fretter and Graham, 1962; Ponder, 1985, 1988). According to Fretter and Graham (1962) this duct might have a compensatory function for the rapid retraction of the male in case of disturbance during copulation. Whether the peculiar formation of the connective tis- sue surrounding the stomach and digestive gland, the posi- tion of stomach and digestive gland, and their connection through a duct, represent ancestral states or autapomorphies cannot be judged at present. There is anatomical information on a number of species of the genus Bythiospeum (Seibold, 1904; Krull, 1935; Bole, 1970; Boeters, 1971, 1984a, b; Giusti and Pezzoli, 1980; Radoman, 1983; Reischtitz, 1983b, 1988a; Bernasconi, 1990 ). Unfortunately, many of these descrip- tions are incomplete or do not depict the organs in their natural positions, which is probably due to the small num- ber of specimens available in most cases, and because most of these authors did not have the opportunity to prepare his- tological serial sections. Thus, comprehensive comparisons are hardly possible at present. It can only be concluded that practically all organs are potentially important for differen- tial diagnoses. In addition, this is only a very small fraction of the nominal species described in this genus so that we are far from tracing the evolution of and establishing rela- tionships among Bythiospeum species. ACKNOWLEDGMENTS The financial support of the Kulturamt der Stadt Wien is gratefully acknowledged. The living animals were found by Drs. D. L. Danielopol (Mondsee) and P. Pospisil 136 AMER. MALAC. BULL. 11(2) (1995) (Vienna) in the course of a groundwater project financed by the Austrian Science Foundation (P7881 BIO). I thank Dr. E. Wawra (Vienna), Mr. H.-J. Niederhdfer (Stuttgart), and Mag. P. L. Reischiitz (Horn) for providing material, and Dr. H. L. Nemeschkal and two anonymous reviewers for help- ful comments on the manuscript. LITERATURE CITED Bernasconi, R. 1990. Revision of the genus Bythiospeum (Mollusca Prosobranchia Hydrobiidae) of France, Switzerland and Germany. Privately published. Miinchenbuchsee. 44 pp., 19 figs. Boeters, H. D. 1967. Die Publikationsdaten der Clessin’schen Molluskenfaunen. Mitteilungen der deutschen malakozoologis- chen Gesellschaft 10: 210 - 212. Boeters, H. D. 1971. Iglica pezzolii n. sp. und ein neues Merkmal zur Unterscheidung zwischen Bythiospeum und Paladilhia (Prosobranchia, Hydrobiidae). Archiv fiir Molluskenkunde 101:169-173. Boeters, H. D. 1984a. Zur Ideatitat des Bythiospeum-Typus (Prosobranchia: Hydrobiidae). Heldia 1:6-8, pl. 1a. Boeters, H. D. 1984b. Gedanken zu einer Revision der Gattung Bythiospeum in Deutschland. Mitteilungen der deutschen malako- zoologischen Gesellschaft 37:142-171. Boettger, O. 1905. Die Konchylien aus den Anspiilungen des Sarus- Flusses bei Adana in Cilicien. Nachrichtenblatt der deutschen Malakozoologischen Gesellschaft 37:97-123, | pl. Bole, J. 1970. Prispevek k poznavanju anatomije in taksonomije podzemeljskih hidrobiid (Gastropoda, Prosobranchia). Razprave, Slovenska Akademija Znanosti in Umetnosti, Razred za Prirodoslovene in Medicinske Vede, Ljubljana \3:85-111. Bole, J. and F. Velkovrh. 1986. Mollusca from continental subterranean aquatic habitats. In: Stygofauna mundi, L. Botosaneanu, ed. pp. 177-208. E. J. Brill/ Dr. W. Backhuys, Leiden. Clessin, S. 1887-1890. Die Mollusken Oesterreich-Ungarns und der Schweiz. Bauer and Raspe, Niimberg. 858 pp. Danielopol, D. L., J. Dreher, A. Gunatilaka, M. Kaiser, R. Niederreiter, P. Pospisil, M. Creuze des Chatelliers, and A. Richter. 1992. Ecology of organisms living in hypoxic groundwater environment at Vienna (Austria); methodological questions and preliminary results. In: Proceedings of the First International Conference on Ground Water Ecology, J. A. Stanford and J. J. Simons, eds. pp. 79-90. American Water Resources Association, Bethesda, Maryland. Danielopol, D. L. and R. Niederreiter. 1987. A sampling device for groundwater organisms and oxygen measurements in multi-level monitoring wells. Stygologia 3: 252-263. Davis, G. M. 1980. Snail hosts of Asian Schistosoma infecting man: evo- lution and coevolution. Malacological Review, Supplement 2:195- 238. Ehrmann, P. 1933. Mollusken (Weichtiere). In: Die Tierwelt Mitteleuropas II (1), P. Brohmer, P. Ehrmann, and G. Ulmer, eds. 264 pp. + 13 pls. Quelle and Meyer, Leipzig. Felsenstein, J. 1993. PHYLIP (Phylogeny Inference Package) version 3.5c. Department of Genetics, University of Washington, Seattle. Fretter, V. and A. Graham. 1962. British Prosobranch Molluscs: their functional anatomy and ecology. Ray Society, London. xvi + 755 pp. Fuchs, A. 1925. Lartetia geyeri nov. spec. Archiv fiir Molluskenkunde 57: 282-284. Fuchs, A. 1929. Beitrag zur Molluskenfauna Oberosterreichs. Archiv fiir Molluskenkunde 61:139-142. Giusti, F. and E. Pezzoli. 1980. Guide per il riconoscimento delle species animali delle acque interne Italiane, 8: Gasteropodi, 2. (Gastropoda: Prosobranchia: Hydrobioidea, Pyrguloidea). AQ/1/47, Consiglio Nazionale delle Richerche, Italy. [Collana del progetto finalizzato “Promozione della qualita dell’ambiente.”]. 67 pp. Giusti, F. and E. Pezzoli. 1982. Notes on the small Hydrobioidea in Italian subterranean waters: catalogue, biogeography and some systemat- ic problems. Malacologia 22: 463-468. Haase, M. 1992. A new, stygobiont, valvatiform, hydrobiid gastropod from Austria (Caenogastropoda: Hydrobiidae). Journal of Molluscan Studies 58:207-214. Haase, M. 1993a. Hauffenia kerschneri (Zimmermann 1930): zwei Arten zweier Gattungen (Caenogastropoda: Hydrobiidae). Archiv fiir Molluskenkunde 121:91-109. Haase, M. 1993b. The genetic differentiation in three species of the genus Hydrobia and systematic implications (Caenogastropoda, Hydrobiidae). Malacologia 35:389-398. Haase, M. 1994. Differentiation of selected species of Belgrandiella and the redefined genus Graziana (Gastropoda: Hydrobiidae). Zoological Journal of the Linnean Society 111:219-246. International Commission of Zoological Nomenclature. 1985. International Code of Zoological Nomenclature [ICZN], 3rd ed. International Trust for Zoological Nomenclature, London, xx + 338 pp. Johansson, J. 1948. Uber die Geschlechtsorgane der Hydrobiiden und Rissoiden und den urspriinglichen Hermaphroditismus der Prosobranchier. Arkiv for Zoologi 40A: 1-13. Klemm, W. 1960. Mollusca. In: Catalogus Faunae Austriae, Teil VIla. Osterreichische Akademie der Wissenschaften, Osterreichische Staatsdruckerei, 59 pp. ed. Wien. Krull, H. 1935. Anatomische Untersuchungen an einheimischen Prosobranchiern und Beitrage zur Phylogenie der Gastropoden. Zoologisches Jahrbuch, Abteilung fiir Anatomie und Ontogenese 60:399-464. Kuijper, W. J. and E. Gittenberger. 1993. De grondwaterslak Bythiospeum husmanni (Boettger, 1963) in Nederland (Gastropoda: Prosobranchia: Hydrobiidae). Basteria 57:89-94. Pfeiffer, A. 1890. Ein Beitrag zur oberésterreichischen Gastropoden- Fauna. In: Neunzehnter Jahresbericht des Vereins fiir Naturkunde in Osterreich ob der Enns zu Linz. pp. 1-22. Der Verin fiir Naturkunde zu Linz, Linz. Ponder, W. F. 1985. A review of the genera of the Rissoidae (Mollusca: Mesogastropoda: Rissoacea). Records of the Australian Museum, Supplement 4:1-221. Ponder, W. F. 1988. The truncatelloidean (= rissoacean) radiation - a pre- liminary phylogeny. Malacological Review, Supplement 4:129- 164. Pospisil, P. 1989. Austrocyclops gmeineri n. sp. (Crustacea, Copepoda) aus dem Grundwasser von Wien (Osterreich): Bemerkungen zur Zoogeographie und zur Sauerstoffsituation des Grundwassers am Fundort. Zoologischer Anzeiger 223:220-230. Radoman, P. 1983. Hydrobioidea a superfamily of Prosobranchia (Gastropoda), I: Sistematics. Serbian Academy of Sciences and Arts, Monograph 547, 57: ti + 256 pp., 12 pls. Reischiitz, P. L. 1981. Die rezenten Wasserschneckenarten Osterreichs (Moll., Gastropoda). Mitteilungen der Abteilung Zoologie des Landesmuseums Joanneum 10:127-133. Reischiitz, P. L. 1983a. Beitrége zur Molluskenfauna Niederdésterreichs, 4. Neue Taxa niederésterreichischer Hydrobioidea (Gastropoda). Malakologische Abhandlungen Staatliches Museum fir Tierkunde HAASE: AUSTRIAN BYTHIOSEPUM 137 in Dresden 8:149-153. Reischiitz, P. L. 1983b. Ein Beitrag zur Anatomie von Bythiospeum tschapecki (Clessin, 1878) (Moll., Gastropoda, Prosobranchia). Mitteilungen der Abteilung Zoologie des Landesmuseums Joanneum 30:79-82. Reischiitz, P. L. 1983c. Die Verbreitung der Gattung Bythiospeum in Osterreich. In: Abstracts of the Eighth International Malacological Congress, L. Pinter, ed. p. 118. Hungarian Natural History Museum, Budapest. Reischiitz, P. L. 1988a. Drei bemerkenswerte Vertreter der Hydrobioidea aus Nordgriechenland (Gastropoda, Prosobranchia). Malakologische Abhandlungen Staatliches Museum fiir Tierkunde Dresden 13:105-107. Reischiitz, P. L. 1988b: Contributions to the mollusc fauna of Lower Austria, VII. The distribution of the Hydrobioidea of Lower Austria, Vienna and Burgenland. De Kreukel, Jubileumnummer: 67-87. Seibold, W. 1904. Anatomie von Vitrella Quenstedtii (Wiedersheim) Clessin. Jahreshefte des Vereins fiir vaterlandische Naturkunde Wiirttemberg 60:198-226. Sneath, P. H. and R. R. Sokal. 1973. Numerical Taxonomy. W. H. Freeman and Company, San Francisco. xv + 573 pp. Stojaspal, F. 1978. Zwei neue Grundwasserschnecken aus dem Barenloch bei Mixnitz (Steirmark). Die Hoehle 29:87-90. Westerlund, C. A. 1890. Fauna der in der paldarktischen Region (Europa, Kaukasien, Sibirien, Turan, Persien, Kurdistan, Armenien, Mesopotamien, Kleinasien, Syrien, Arabien, Egypten, Tripolis, Tunesien, Algerien und Marocco) lebenden Binnenconchylien. Zweiter Theil. VII. Malacozoa Acephala: Zusdtze, pp. 1-8. R. Friedlander and Sohn, Berlin. Zilch, A. 1970. Die Typen und Typoide des Natur-Museums Senckenberg, 45: Mollusca, Hydrobiidae (1): Bythiospeum Bourguignat. Archiv fiir Molluskenkunde 100:319-346. Date of manuscript acceptance: 19 September 1994 { ; F 1 HPLC analysis of chloroplast pigments from the marine ascoglossan Tridachia crispata (Morch, 1863) (Mollusca: Opisthobranchia) Richard A. Roller!* and Thomas S. Bianchi2 1Center for Coastal Marine Studies and Department of Biology, Lamar University, Beaumont, Texas 77710 U. S. A. 2 Department of Ecology, Evolution, and Organismal Biology, Tulane University, New Orleans, Louisiana 70118-5698 U.S. A. Abstract. Pigment composition of ingested kleptoplastids (= symbiotic chloroplasts”; Waugh and Clark, 1986) in the marine ascoglossan Tridachia crispata (Mérch, 1863) were analyzed for the first time using reverse-phase high performance liquid chromatography (HPLC). The pigment signatures from the slug closely reflected that of the host plant Caulerpa sertularioides (Gmelin). One noted exception was the detected presence of fucoxanthin in the tis- sues of T. crispata which is attributed to the ingestion of epibenthic microflora (diatoms) from the surface of the host plant. This demonstrates for the first time that plant pigments in the tissues of T. crispata are not exclusively from the host plant. Additionally, the detection of phaeophorbide-a and phaeophytin- a (chlorophyll-a decay products) in 7. crispata suggested that there was some minor breakdown (via digestion) of pigment during the transfer. Tridachia crispata (MOrch, 1863) is a marine ascoglossan (= sacoglossan) which feeds on several species of Siphonales algae including Caulerpa sertularioides (Gmelin). The slug ingests intact chloroplasts from the alga and retains them in its digestive diverticula. The chloro- plasts apparently retain the ability to undergo photosynthe- sis and provide nutrients to T. crispata during periods of food scarcity (Hinde and Smith, 1974; Waugh and Clark, 1986). Several opisthobranch species are capable of surviv- ing 14-70 days without external nutrients (Hinde and Smith, 1974; Taylor, 1970a, b; Keefer, pers. comm.). The degradation products of chloropigments, com- monly referred to as phaeopigments, have been used to infer the availability of different source materials to con- sumers. For example, the three dominant tetrapyrrole derivatives of chloropigments (phaeophorbide, phaeo- phytin, and chlorophyllide) are typically formed during bacterial, autolytic cell lysis, and metazoan grazing activi- ties (Sanger and Gorham, 1970; Shuman and Lorenzen, 1975; Hawkins et al., 1986; Bianchi et al., 1988, 1991). Differences in the production of phaeophorbide during grazing activity has been shown to be reflective of the qual- ity of food resources (Shuman and Lorenzen, 1975; Bianchi et al., 1988, 1991). Thus, higher quality food resources will result in the production of greater amounts of phaeophorbide. Previous studies on symbionic chloroplasts (i.e. *Present Address: Department of Biology, Valdosta State University, Valdosta, Georgia 31698 U.S. A. “kleptoplastids,’ Waugh and Clark, 1986) associated with several species of Ascoglossa have dealt primarily with descriptions of these phenomena, as well as carbon translo- cation, physiological measurements of oxygen production, and photosynthetic rates (Hinde and Smith, 1974; Taylor, 1967, 1968, 1970a, b; Trench, 1969, 1980; Trench et al., 1969). To date, no studies have been published on the quan- titative measurement of chlorophylls and carotenoids of intact symbionic chloroplasts in opisthobranchs using “state-of-the-art” HPLC. The objectives of the present study were to: (1) quantitatively measure, using reverse-phase high perfor- mance liquid chromatography (HPLC), pigment concentra- tions in chloroplasts from intact specimens of Caulerpa sertularioides and from chloroplasts ingested by adult Tridachia crispata; and (2) determine if any observable dif- ferences exist between the pigment content of intact algae and that of chloroplasts from the digestive diverticula of the slug. MATERIALS AND METHODS COLLECTION AND MAINTENANCE Adult Tridachia crispata (> 25 mm length) and samples of Caulerpa sertularioides were collected at Key Largo, Florida, U. S. A., during March 1994. The speci- mens were maintained on a 12:12 light/dark photoperiod in glass containers (1000 ml) containing aerated Instant Ocean® artificial seawater (32 ppt salinity and 18° + American Malacological Bulletin, Vol. 11(2) (1995):139-143 139 140 AMER. MALAC. BULL. 11(2) (1995) 0.5°C). The water in each container was replaced daily with fresh seawater. The slugs were fed C. sertularioides for 2 wk prior to HPLC analysis. After 2 wk, individual T. crispata and C. sertularioides fronds were removed from the apparatus, rinsed with filtered (0.22 pm) seawater, blot- ted dry, and immediately frozen (-80°C) for HPLC analysis. PLANT PIGMENT EXTRACTION AND ANALYSIS Frozen plant and animal tissues were lyophilized, weighed, and extracted for pigments using 100% acetone according to Bianchi et al. (1993). Tissue samples were frozen immediately (-80°C) to prevent the possibility of post-mortem phaeopigment production in the guts of the animals (Millie et al., 1993). Lyophilization of plant and animal tissues for pigment extractions has been shown to be the preferred method for quantification (Millie et al., 1993). To preclude any artifacts from light effects, all extractions were performed under low light conditions, however, meth- ods using methanol tend to be more problematic (Bianchi et al., 1995). The formation of phaeopigments in acetone extracts using the method described in Bianchi et al. (1993) has been shown to be minimal. Canthaxanthin was added to each of the acetone extracts as an internal standard to confirm that there was no loss of pigments during extrac- tion. Reverse-phase HPLC analysis was conducted using Pigment Concentration (mg/g tissue) Fig. 1. Dominant pigment concentrations (mg/g tissue; mean + S.E.) found in tissues of Tridachia crispata and Caulerpa sertularioides. the method of Wright et al. (1991); the gradient used in this method has recently been described as the selected method for pigment analyses (Millie et al., 1993). A Waters® Model 660 solvent-delivery system was coupled with dual- channel detection using a Waters® Model 990 photodiode array detector set at 438 nm for absorbance, and a Milton- Roy® fluorescence detector with excitation set at 440 nm and emission at > 600 nm. The injector was connected via a guard column to a reverse-phase Cig Alltech® Adsorbosphere column (5 pm particle size; 250 mm x 4.6 mm i.d.). After injection (100 pl), a gradient program (1 ml/min) began isocratically at time zero with mobile phase A (80:20 methanol: 0.5 M ammonium acetate, aq.; pH 7.2 v/v) which then ramped to 100% mobile phase B in 4 min (90:10 acetonitrile : HPLC grade water v/v) and then changed to 20% B and 80% mobile phase C (100% ethyl acetate) in 35 min. This was followed by a return to 100% B in 3 min with final ramping to 100% A in 3 min. This method provided sufficient resolution of all pigments of interest in Tridachia crispata and Caulerpa sertularioides (Figs. 1, 2). Although fluorescent chromatograms are not presented here, a fluorescence detector was used as a sim- ple means of identifying chloropigments versus carotenoids. High purity HPLC standards for chlorophylls a and b as well as a and B carotene were obtained from Sigma Chemical Co. (St. Louis, Missouri). Standards of the fol- lowing carotenoids were kindly provided by Hoffman LaRoche Co. (Basil, Switzerland): fucoxanthin, diadinox- anthin, lutein. All standards were spectrophotometrically verified for purity. Peak identifications were based on retention times of standards and confirmed with a diode- array detector. All peaks were quantified with calculated response factors and the Water Millenium software package (Waters Corporation, Inc.; Bianchi et al. 1995). Pigment concentrations of Tridachia crispata and Caulerpa sertularioides tissue were tested for comparison using Pearson’s Product-Moment Correlation Coefficient (Sokal and Rohlf, 1981). A probability level of < 0.001 was considered significant for statistical analysis. RESULTS AND DISCUSSION There was a strong correlation (Pearson’s Product- Moment Correlation Coefficient; P < 0.001) in the occur- rence of dominant plant pigments (carotenoids and chloro- phylls) in Tridachia crispata and Caulerpa sertularioides (Fig. 1). The only pigments found exclusively in T. crispa- ta were phaeophorbide-a and phaeophytin-a (labelled as peaks 11 and 12 in the T. crispata chromatogram; Fig. 2A). All pigment concentrations/g tissue were significantly high- er (P < 0.001) in C. sertularioides than in T. crispata. The ROLLER AND BIANCHI: CHLOROPLAST PIGMENTS FROM TRIDACHIA 141 A Tridachia crispata B Caulerpa sertularioides RELATIVE ABSORBANCE 0 16.00 30.00 46.00 TIME (min.) Fig. 2. Absorbance chromatograms showing plant pigments extracted from the tissues of: A. Tridachia crispata, and B. Caulerpa sertularioides. Identified pigments are: 0 = chlorophyll-c; 1 = siphonaxanthin; 2 = fucoxanthin; 3 = 9-cis-neoxanthin; 4 = violaxanthin; 5 = diadinoxanthin; 6 = lutein; 7 = chlorophyll-b; 8 = chlorophyll-a; 9 = o-carotene; 10 = B- carotene; 11 = phaeophorbide-a; 12 = phaeophytin-a. removal and translocation of chloroplasts, from C. sertular- ioides to T. crispata tissue, occurred with minimal pigment loss. The presence of phaeophorbide-a and phaeophytin-a in Tridachia crispata was due to moderate breakdown (via digestion) during the transfer of pigments. Phaeophorbide is a chlorophyll decay product typically produced by het- erotrophic grazing processes (Shuman and Lorenzen, 1975; Bianchi et al., 1988). Moreover, the presence of fucoxan- thin and chlorophyll-c (pigment markers for diatoms) in Caulerpa sertularioides extracts (Fig. 2B) provides evi- dence that epibenthic microflora were present, and primari- ly composed of diatoms. Fucoxanthin, not normally found in Chlorophyta, such as C. sertularioides, was also found in T. crispata (Fig. 2A). Although we did not use epiphyte- free algae (Fig. 3), we are confident that the fucoxanthin found in the C. sertularioides was from an external source; this is based on the extensive literature of pigments in algae (Rowan, 1989) as well as microscopical evidence of diatoms in the culture water of C. sertularioides (Fig. 3). This suggested that chloroplasts from epibenthic microflora on the surfaces of algae consumed by T: crispata were also transported into their tissues. It was previously thought that only robust chloroplasts of the Siphonales algae contributed to the pigment signature of T: crispata tissue (Giles and Sarafis, 1972), and the majority of the pigment signatures support this belief. However, the presence of fucoxanthin suggested that intact chloroplasts from other sources (i.e. epibenthic diatoms) are likely to have been ingested while the slug was feeding on C. sertularioides. Selective feeding on diatoms has been previously documented in Elysia eveli- nae Marcus, 1957 (Jensen, 1981); however, in the present study we did not directly observe preferential feeding on epibenthic diatoms by T: crispata. It is improbable that the fucoxanthin may have arisen from other sources though, because this pigment decays rapidly once outside of the thylakoid membrane (Bianchi et al., 1991). The convention- al spectrophotometric and fluorometric methods previously used for examining plant pigments in T. crispata would not detect these subtle differences in pigment signatures. While chloroplasts have been shown to be capable of providing resources to Tridachia crispata via photosyn- thate production, they could also prove to be important in shielding these animals from the deleterious effects of UV irradiation. Carotenoids, such as zeaxanthin and B- carotene, function as photoprotectants for living plants, by quenching singlet oxygen which precludes peroxidation reactions (Foote et al., 1970). Thus, the abundance and dis- tribution of photoprotectants (such as pigments) in organ- isms may prove to be very important with the current global trends in stratospheric ozone depletion (via enhanced UV- B). However, before any specific claims can be made con- cerning the photoprotectant nature of these pigments further work is needed to determine the overall abundance and dis- tribution of pigments in the tissues of the organism. CONCLUSIONS In this study, we have examined for the first time (using state-of-the-art HPLC) differences in the concentra- tion and composition of plant pigments in chloroplasts 142 AMER. MALAC. BULL. 11(2) (1995) Fig. 3. A. Phase-contrast micrograph of pennate diatom in culture water of Caulerpa sertularioides. B. Phase-contrast micrograph of pennate diatom adhering (via cellular debris) to frond of C. sertularioides. located in Tridachia crispata and the source plant itself (Caulerpa sertularioides). We have also demonstrated the probable ingestion of epibenthic diatoms by T. crispata, indicated by the presence of fucoxanthin in the slug’s tis- sues. Thus, the pigment composition in the slug is not exclusively derived from the Siphonales algae. This study and others (see review of Millie et al., 1993) illustrate the utility of HPLC analysis in trophic studies of benthic marine invertebrates. ACKNOWLEDGMENTS We thank Corey and Christine Lambert for their assistance in the plant pigment extraction and analysis. This research was funded in part from the National Science Foundation (to TSB) and from the Lamar University Research Enhancement Program (to RAR). LITERATURE CITED Bianchi, T. S., R. Dawson, and P. Sawangwong. 1988. The effects of mac- robenthic deposit-feeding on the degradation of chloropigments in sandy sediments. Journal of Experimental Marine Biology and Ecology 122:243-255. Bianchi, T. S., S. Findlay, and D. Fontvieille. 1991. Experimental degra- dation of plant material in Hudson River sediments. I. Heterotrophic transformations of plant pigments. Biogeochemistry 12:171-187. Bianchi, T. S., S. Findlay, and R. Dawson. 1993. Organic matter sources in the water column and sediments of the Hudson River estuary: the use of plant pigments as tracers. Estuarine, Coastal, and Shelf Sciences 36:359-376. Bianchi, T. S., C. Lambert, P. Santschi, M. Baskaran, and L. Guo. 1995. Plant pigments as biomarkers of high-molecular-weight dissolved organic carbon. Limnology and Oceanography 40(2):422-428. Foote, C. S., Y. C. Chang, and R. W. Denny. 1970. Chemistry of singlet oxygen. X. Carotenoid quenching parallels biological protection. Journal of the American Chemical Society (92):5216-5218. Giles, K. L. and V. Sarafis. 1972. Chloroplast survival and division in vitro. Nature, New Biology 236:56-58. Hawkins, A. J. S., B. L. Bayne, R. F. C. Mantoura, C. A. Llewellyn, and E. Navarro. 1986. Chlorophyll degradation and adsorption throughout the digestive system of the blue mussel Mylilus edulis. Journal of Experimental Marine Biology and Ecology 96:213- 223. Hinde, R. and D. C. Smith. 1974. “Chloroplast symbiosis” and the extent to which it occurs in Sacoglossa (Gastropoda: Mollusca). Biological Journal of the Linnean Society 6:349-356. Jensen, K. 1981. Observations on feeding methods in some Florida ascoglossans. Journal of Molluscan Studies 47:190-199. Millie, D. F., H. W. Paerl, and J. P. Hurley. 1993. Microalgal pigment assessments using high performance liquid chromatography: a synopsis of organismal and ecological applications. Canadian Journal of Fisheries and Aquatic Sciences 50:25 13-2527. Rowan, K. S. 1989. Photosynthetic Pigments of Algae. Cambridge University Press, Cambridge. 334 pp. Sanger, J. E. and E. Gorham. 1970. The diversity of pigments in lake sed- iments and its ecological significance. Limnology and Oceanography 15:59-69. Shuman, F. R. and C. J. Lorenzen. 1975. Quantitative degradation of chlorophyll by a marine herbivore. Limnology and Oceanography 20:580-586. Sokal, R. R. and F. J. Rohlf. 1981. Biometry, 2nd ed. W. H. Freeman and Co., New York. 859 pp. Taylor, D. 1967. The occurrence and significance of endosymbiotic chloroplasts in the digestive gland of herbivorous opisthobranchs. Journal of Phycology 3:234-235. Taylor, D. 1968. Chloroplasts as symbiotic organelles in the digestive gland of Elysia viridis (Gastropoda: Opisthobranchia). Journal of the Marine Biological Association of the United Kingdom 48:1- 15. Taylor, D. 1970a. Chloroplasts as symbiotic organelles. International Review of Cytology 27:29-64. Taylor, D. 1970b. Photosynthesis of symbiotic chloroplasts in Tridachia crispata (Bergh). Comparative Biochemistry and Physiology 38A:233-236. Trench, R. K. 1969. Chloroplasts as functional endosymbionts in the mol- lusc Tridachia crispata (Bergh), (Opisthobranchia, Sacoglossa). Nature 222:1071-1072. Trench, R. K. 1980. Uptake, retention and function of chloroplasts in ani- ROLLER AND BIANCHI: CHLOROPLAST PIGMENTS FROM TRIDACHIA 143 mal cells. Jn: Endocytobiology, Endosymbiosis and Cell Biology, Marine Biology 92:483-487. W. Schwemmler and H. E. A. Swchenk, eds. pp. 703-727. de Wright, S. W., S. W. Jeffrey, R. F. C. Mantoura, C. A. Lllewelyn, D. Gruyter and Co., Berlin. Bjomland, D. Repeta, and N. A. Welschmeyer. 1991. Improved Trench, R. K., R. W. Greene, and B. G. Bystrom. 1969. Chloroplasts as HPLC method for the analysis of chlorophylls and carotenoids functional organelles in animal tissues. Journal of Cell Biology from marine phytoplankton. Marine Ecology - Progress Series 42:404-417. 77:183-196. Waugh, G. R. and K. B. Clark. 1986. Seasonal and geographic variation in chlorophyll level of Elysia tuca (Ascoglossa: Opisthobranchia). Date of manuscript acceptance: 10 February 1995 Population ecology of the endangered Ouachita rock-pocketbook mussel, Arkansia wheeleri (Bivalvia: Unionidae), in the Kiamichi River, Oklahoma Caryn C. Vaughn and Mark Pyron* Oklahoma Biological Survey and Department of Zoology, University of Oklahoma, Norman, Oklahoma 73019, U.S. A. Abstract. The only known remaining viable population of Arkansia wheeleri Ortmann and Walker, 1912, in the world occurs within a 128-km stretch of the Kiamichi River in Pushmataha County, Oklahoma. Within this river, A. wheeleri occurs only within the most species-rich mussel beds. In its optimal habitat, this species is always rare; mean relative abundance varies from 0.2-0.7% and the mean density is 0.27 individuals/m2. The youngest individual A. wheeleri encountered was approximately 12 years of age. Forty-three percent of the historically known subpopulations of A. wheeleri below where inflow from an impounded tributary enters the Kiamichi River have apparently been extirpated, and no new subpopulations have been located. A. wheeleri survives at 75% of the historically known locations above the impounded tributary and five new subpopulations have been located. Arkansia (syn. Arcidens) wheeleri Ortmann and Walker, 1912 (Bivalvia: Unionoidea), the Ouachita rock- pocketbook mussel, is a federal endangered species (Federal Register, 1991). Originally named Arkansia wheeleri by Ortmann and Walker (1912), Clarke (1981, 1985) recognized Arkansia as a subgenus of Arcidens. The species was considered by Clarke to be distinct. However, Turgeon et al. (1988) have continued to use the binomial Arkansia wheeleri. The historical distribution of Arkansia wheeleri was in the Ouachita and Little Rivers in Arkansas and the Kiamichi River in Oklahoma, all south-flowing rivers out of the Ouachita Mountains (Fig. 1). A survey by Clarke (1987) indicated that the species was probably extirpated from the Ouachita River and severely depleted in the Little River in Arkansas. In 1992 and 1993, relict shells of A. wheeleri were found in the Little River in Oklahoma below Pine Creek Reservoir (Vaughn, 1993). Arkansia wheeleri was first reported from the Kiamichi River by Isely (1924) who conducted a survey of the river in 1911. Clarke (1987) and Mehlhop and Miller (1989) both conducted recent status surveys for this species in the Kiamichi River. They found that A. wheeleri was patchily distributed and rare in the Kiamichi River from above Hugo Reservoir to Whitesboro (Fig. 2). Since the construction of a dam and reservoir in the lower reaches of the Kiamichi in the 1970s, some of the backwater areas *Present Address: El Verde Field Station, P.O. Box 1690, Luquillo, PR 00773 Oklahoma , 94° 29' W af Kiamichi River Ouachita River Little River Red River Arkansas 32° 35'N Louisiana Fig. 1. Rivers in which Arkansia wheeleri historically occurred. Dashed box indicates area depicted in Fig. 2. where it was known to occur have been destroyed (Valentine and Stansbery, 1971), and connection with potential habitats on the Red River and other tributaries to it has been blocked. Based on the above information A. wheeleri was listed as endangered in October 1991 (Federal Register, 1991). The objectives of this study were to determine the distribution and abundance of Arkansia wheeleri in the Kiamichi River, characterize the species’ microhabitat, and determine movement, growth, and survivorship of individu- als. We also examined the impact of a reservoir on the pop- ulation. American Malacological Bulletin, Vol. 11(2) (1995):145-151 145 146 AMER. MALAC. BULL. 11(2) (1995) Sardis Reservoir Whitesboroe 1 e Antlers Hugo Reservoir Fig. 2. Population monitoring sites for Arkansia wheeleri on the Kiamichi River. STUDY SITE The Kiamichi River is a major tributary of the Red River. It flows for a total of 180 km through a 4,800 km2 drainage area across the Ridge and Valley Belt of the Ouachita Mountain geologic province and the Dissected Coastal Plain province (Curtis and Ham, 1972). The aver- age gradient of the river is 0.47 m/km. Two reservoirs influence the river: Sardis Reservoir is an impoundment of Jackfork Creek, a tributary of the Kiamichi River. Hugo Reservoir is an impoundment of the lower Kiamichi River. The vegetation cover in the watershed can be described as a patchwork of forest made up of short-leaf and loblolly pine, mesic oak forests, and diverse bottomland habitats in various stages of maturity. Another large component of the watershed coverage is made up of pasture and other agri- cultural lands (Vaughn et al., 1993). The Kiamichi River is located near the western edge of mussel species richness in the United States (Williams et al., 1992; Warren and Burr, 1994). Therefore, because of historical biogeographic factors, one would not expect diversity in the Kiamichi River to be as high as that in rivers in more eastern states. Nevertheless, the Kiamichi River has high mussel diversity for its size and geographic location. Fifty-five species of mussels are known from Oklahoma (Williams et al., 1992), and 29 of these currently occur in the Kiamichi River (Vaughn et al., 1993). Only two species that were known historically from the Kiamichi River (Isley, 1924) no longer occur there. Several species of mussels from the Kiamichi River are endemic to rivers in the Ouachita Mountains, including Arkansia_ wheeleri, Ptychobranchus occidentalis (Conrad, 1836) and Villosa arkansasensis (Lea, 1862) (Pyron and Vaughn, 1994). P. occidentalis, the Ouachita kidneyshell, is a candidate for federal listing. METHODS Our quantitative survey efforts were restricted to areas that contained concentrations of mussels and thus could be defined as “beds.” Mussel relative abundance and habitat data for 22 sites in the Kiamichi River were collect- ed during July-August 1990. These sites included 11 areas defined as “pools,” six areas defined as “backwaters,” and five areas defined as “runs” (see Discussion). Mussel sur- veys (timed to standardize sampling effort) (Green et al., 1985; Kovalak et al., 1986) were conducted by hand- searching with the aid of SCUBA in deeper areas and by hand searches in shallow areas in the following manner: (1) a shoal was identified for surveying; (2) the entire area was searched by at least two people for one hr; (3) all mussels encountered were removed to shore; (4) all mussels were immediately identified; (5) mussels were put back into the water as close to where they were removed as possible. At each of the 22 sites, we measured water depth, water temperature, current velocity, conductivity, dissolved oxygen, and pH. Five measures of water depth and current velocity were taken across the mussel bed and averaged. Current velocity was measured 10 cm above the stream bot- tom with a Marsh-McBirney model 201 portable flow meter. Conductivity and dissolved oxygen were measured with Yellow Springs Instruments conductivity and dis- solved oxygen meters, respectively. pH was measured with a Fisher Accumet portable pH meter. At each site, we recorded the proximity of the site to entering tributaries, islands, and macrophyte cover. Three replicate substratum samples were collected at each site. These were brought back to the laboratory and allowed to dry. Samples were dry sieved, weighed, and individual pro- portions of samples were assigned to the appropriate sub- stratum size class (in mm) following Hynes (1970:24). Standard sieving techniques do not segregate particles greater than about 2 mm in diameter (i.e. gravel from peb- ble from cobble). To determine the proportion of fine grav- el, coarse gravel, pebble, and cobble in samples we ran- domly measured the diameter of 100 particles in the sub- sample greater than 2 mm in particle diameter (Dunne and Leopold, 1978). In 1991, we selected ten of the 22 sites as long-term population monitoring sites for Arkansia wheeleri (Fig. 2). The ten sites were chosen to be as evenly distributed as possible along the Kiamichi River above Hugo Reservoir, but still be reasonably accessible, and included sites where A. wheeleri had been located by us in 1990, sites where it VAUGHN AND PYRON: ARKANSIA WHEELERI IN THE KIAMICHI RIVER 147 had been found historically (Mehlhop and Miller,1989; Clarke, 1987), and sites above and below the Sardis Reservoir and Jackfork Creek. Density and relative abun- dance data for mussel species at the ten monitoring sites were collected during July-August 1991. Densities were calculated from quadrat samples and relative abundances were estimated from timed searches, as described above. Quadrat sampling was done with quarter-meter-square PVC pipe quadrats. Fifteen random quadrats were sampled for each site. Quadrats were searched by hand, with the aid of SCUBA in deeper areas, until all mussels had been recov- ered to a depth of 15 cm. Individual mussels were returned to the mussel bed as in timed searches. All A. wheeleri were measured using digital calipers (height, width, and length), and individually marked using numbered, laminat- ed plastic fish tags. All specimens were returned to the same location from which they were captured. To obtain additional information on Arkansia wheeleri size and age distribution, we measured relict shells of A. wheeleri deposited in the Oklahoma Museum of Natural History (OMNH, University of Oklahoma, Norman) or that we found on the Kiamichi River between 1990 and 1992. We counted external annuli on shells we had collected and those in OMNH (Neves and Moyer, 1988; McMahon, 1991). From the above data, we calculat- ed shell length, width, and height vs. number of annuli regression lines. Shell height vs. number of annuli produced the best fit, and the resulting equation was used to predict the number of annuli for live mussels that had been mea- sured in the field. We used Replicated Goodness of Fit tests (Gy) (Sokal and Rohlf, 1981) to compare size distributions of relict shells and live mussels, and to compare predicted age distributions of mussels. We used several statistical techniques to explore rela- tionships between Arkansia wheeleri distribution and abun- dance and measured habitat parameters. For all of these analyses, we used the data collected from the 22 sites in 1990. Associations between A. wheeleri and other species of mussels were calculated using Spearman Rank correla- tions on relative abundance data (Ludwig and Reynolds, 1988). We used discriminant function analysis (Sokal and Rohlf, 1981) to predict the presence or absence of A. wheeleri at a site based on the habitat characteristics of that site. We used an independent-sample t-test (one-tailed) (Sokal and Rohlf, 1981) to compare species richness at sites with and without A. wheeleri. RESULTS Arkansia wheeleri was found in ten of the 22 mussel beds that were sampled. Six of these ten sites were classi- fied as pools and four were classified as backwaters. No Table 1. Results of univariate F-tests of the presence or absence of Arkansia wheeleri at a site using four habitat variables. The multivariate model is significant (F(1,20) = 0.54, P = 0.03). Variable Fo4,17) P Depth 6.87 0.016 Habitat type (pool, backwater, or run) 0.95 0.342 Emergent vegetation (presence/absence) 5.45 0.030 Mussel species richness 10.72 0.004 specimens were found in any of the run areas sampled. A multivariate analysis of variance using all of the habitat variables to predict the presence or absence of A. wheeleri at a site was not significant (F(12,9) = 1.22, P = 0.39). A sig- nificant discriminant model was produced using four habi- tat variables: depth, habitat type (pool, backwater, or run), presence or absence of emergent vegetation at the site, and mussel species richness (Table 1). This model successfully predicted the presence or absence of A. wheeleri 17 out of 22 times (G1) = 7.72, P = 0.005). Mussel species richness at a site was the best individual predictor of A. wheeleri occurrence (Table 1). Mussel sites where A. wheeleri occurred were more species-rich than other mussel sites that we sampled in the Kiamichi River (t(15) = 3.18, P = 0.006). Spearman rank correlations of relative abundance data revealed three significant associations between Arkansia wheeleri and other mussel species. A. wheeleri was positively correlated with the relative abundance of Quadrula cf. apiculata (Say, 1829) (1, = 0.437, P < 0.05) and Megalonaias nervosa (Rafinesque, 1820) (r, = 0.423, P < 0.05), and negatively correlated with Lampsilis teres (Rafinesque, 1820) (r, = -0.368, P < 0.05). In most cases Arkansia wheeleri was located only through timed searches and did not occur in quadrat sam- ples. Mean relative abundance of A. wheeleri at monitoring sites in 1990-1992 is shown in Fig. 3 and varied from 0.2- 0.7%. In 1991, A. wheeleri was found in quadrat samples at sites 6 and 7. This allowed us to estimate the density of A. wheeleri at 0.27 individuals m2, at each of these sites. In 1990 and 1991, we marked and released nine Arkansia wheeleri at the point of capture. In 1991, we recaptured only two marked individuals, although we found nine live individuals. Both recaptured specimens were found at site 3 (Fig. 2). Both of these individuals were found within one meter of where they were released in 1990. No other live A. wheeleri were found at site 3. In 1992, we recaptured the same two A. wheeleri at site 3 that we had recaptured in 1991. The individuals were within a few meters of where they had been released in 1991. The recaptured individuals had not grown discernably (i.e. not more than 1 mm, within the margin of error of our calipers). No other marked specimens were recaptured in 1992. The — RELATIVE ABUNDANCE (%) 9 10 SITE Fig. 3. Mean relative abundance of Arkansia wheeleri at the 10 monitoring sites in 1990-1992. size distribution (means for 1990-1992) for A. wheeleri in the Kiamichi River is shown in Fig. 4. The overall size distribution of relict shells in OMNH (N = 50) was significantly different than the size distribution of live Arkansia wheeleri in the Kiamichi River (N = 43) (Fig. 4, Guys) = 23.1, P < 0.001). The resulting regression equation for number of annuli on shell height was Y = (-0.483)X + 49.62 (N = 24, R2 = 0.467, P < 0.05). We used the above equation to predict the age of A. wheeleri specimens from shell height. Predicted age distri- butions of spent shells vs. live A. wheeleri were also signifi- cantly different (Gyi12) = 57.43, P < 0.001). Using this technique, the youngest predicted age for a live specimen was 12 years. If this estimate is accurate, none of the A. wheeleri we encountered on the Kiamichi River during our study were produced after Sardis Reservoir was filled in 1983. Arkansia wheeleri occurs both above and below the inflow to the Kiamichi River from Sardis Reservoir via Jackfork Creek. Of our ten monitoring sites, three were located above Sardis Reservoir and seven below (Fig. 2). All of these sites historically harbored A. wheeleri. A. wheeleri was found during this study at all three sites (100%) above Sardis Reservoir. A. wheeleri was found at three of seven sites (43%) below the reservoir inflow. The relative abundance of A. wheeleri at sites above Sardis Reservoir was generally greater than the relative abundance of A. wheeleri at sites below the reservoir (Fig. 3), although these differences were not statistically significant. DISCUSSION Prior to this study, the habitat of Arkansia wheeleri was reported to be backwater reaches of rivers where cur- rent is slow and where there are relatively non-shifting deposits of silt/mud and sand (Wheeler, 1918; Isely, 1924; 48 AMER. MALAC. BULL. 11(2) (1995) Clarke, 1987). We found that this species occured in both pools and backwaters in the Kiamichi River, not just in backwaters as was previously believed. The distribution of A. wheeleri may have been underestimated in past surveys because backwaters are relatively easy to survey, whereas pools often require SCUBA. Although pools and backwaters were considered dif- ferent habitat types in this study, in reality they are tightly interconnected and share many characteristics in common. Backwater areas tend to be shallower and have finer sub- strata. As backwaters merge into the main river channel they turn into deeper pools with coarser substrata and slightly higher current velocities. In the Kiamichi River, Arkansia wheeleri occurs in both of these habitats. In addi- tion, we believe A. wheeleri moves back and forth between these habitats either voluntarily or through physical dis- placement of shifting sediments. Locomotory tendencies differ among different mussel species. For example, indi- viduals of Anodonta grandis Say, 1829, migrate up and down with changes in water level (White, 1979) and in this way avoid stranding at low water. Other species such as Uniomerus tetralasmus (Say, 1831) and the introduced Corbicula fluminea (Miller, 1774) remain in position and suffer prolonged exposure to air (McMahon, 1991). Marked individual A. wheeleri in a backwater area (site 3) did not move significantly from July 1990 to July 1992. However at another site (site 5), unmarked individuals moved from a backwater area into the adjacent pool area. This movement was probably the result of physical dis- placement of these individuals through sediment scour and redeposition. Unlike previous surveys (Wheeler, 1918; Isely, 1924; Clarke, 1987), we did not find Arkansia wheeleri to be restricted to areas where the substratum was predominantly mud or fine sand. In the Kiamichi River, A. wheeleri is just as prevalent in gravel/cobble/coarse sand substrata (which predominates in pools) as in finer substrata. Recent studies 40 Live mussels 30 | COShells FREQUENCY (%) nm ra) 71-80 LENGTH (mm) 91-100 >100 Fig. 4. Total lengths of live Arkansia wheeleri (N = 43) compared to relict shells (N = 50) from the Kiamichi River. VAUGHN AND PYRON: ARKANSIA WHEELERI IN THE KIAMICHI RIVER 149 addressing the substratum preferences of unionids have reached different conclusions, and substratum preferences among unionids remain poorly understood. However, mus- sels are generally believed to be most successful in stable, sand/gravel mixtures and are generally absent from substra- ta with heavy silt loads (Salmon and Green, 1983; Stern, 1983; Cooper, 1984; Way et al., 1990). Most unionid species can be found on a number of different substrata, but growth rates of individuals in each microhabitat can be quite different (Kat, 1982; Hinch et al., 1989). Furthermore, many mussel species can occupy a wide range of habitats as a result of extensive larval dispersal over a heterogenous stream environment (Strayer, 1981), but growth and repro- duction can be optimized only under the habitat conditions described above. Arkansia wheeleri only occurred at the most species- rich sites in the Kiamichi River. These shoals represent optimal habitat for most mussel species, as evidenced by the large number of species and their high abundance. These shoals usually contain both pool and backwater areas, have significant gravel-bar development with accom- panying vegetation [dominated by Justicia americana (Linné) Vahl], and are close to a tributary (usually within 0.4 km). Shoals are usually adjacent to a major riffle area, although they can be either up- or downstream of the riffle. Other studies have shown that these mainstream river shoals in shallower areas with slow, steady current and veg- etation and coarse substrate are optimal habitat for lotic unionids because of minimal turbulence, low silt and steady food supply (Salmon and Green, 1983). In the majority of mussel species, the greatest amount of growth occurs in the first few years of life. Shell growth rate then declines exponentially with age, although the rate of tissue biomass accumulation usually remains constant (McMahon, 1991). Our examination of live Arkansia wheeleri in the Kiamichi River and of relict A. wheeleri shells in museum collections indicates that this growth pattern is also followed by A. wheeleri. Early annuli (those near the umbo) are much wider than later annuli near the edge of the shell. Therefore, our finding of no discernable growth in the two marked individuals that we recaptured is not surprising because they were older, adult specimens. Recruitment, growth and survival of mussels can be assessed by monitoring changes in density and size demog- raphy of natural populations (Payne and Miller, 1989). We have no quantitative historical data on densities of Arkansia wheeleri in the Kiamichi River or anywhere else. Past size distribution, however, can be assessed by examining the size distribution of relict shells. The size distribution of live A. wheeleri in the Kiamichi River is skewed to the left (Sokal and Rohlf, 1981) (Fig. 4) with more large individu- als and fewer small individuals than one would expect from a statistically normal distribution. The size distribution of relict shells (Fig. 4) follows a more normal distribution, with a greater proportion of smaller individuals than in the live population. Looking at these shell length data alone, one would conclude that the size distribution of A. wheeleri in the Kiamichi River has changed over time and recruit- ment has decreased. There may be problems with comparing the size dis- tributions of relict shells with live shells. Relict shells from museum collections may have been collected in a size- biased manner because the rate at which dead shells are moved by currents, disintegrate through time, and are col- lected, could all be functions of shell size. Nevertheless, we would predict that these forces would make it more likely for collectors to find large relict shells rather than small relict shells. Because we found that our “population” of relict shells was generally smaller than our population of live mussels, such a size bias would actually make our demographic estimates conservative. External annular rings have long been used to deter- mine mussel age and growth rates. Recently this technique has been criticized as being replete with problems (Neves and Moyer, 1988; Downing ef al., 1992). Natural erosion and corrosion of shells makes it difficult to distinguish true from false annuli. For example, false annuli can be formed by the incorporation of small substrate particles into mussel shells. It is difficult to count closely deposited growth lines near the margins of old shells. This produces an underesti- mate of shell age that becomes more erroneous with shell age. Downing et al. (1992) studied populations of Lampsilis radiata (Gmelin, 1791) and Anondonta grandis in an oligotrophic lake. In these populations, many mussels showed no new external annuli at all, even several years after individual animals had been marked. They concluded that estimates of growth based on shell annuli consistently overestimated real shell growth. In addition, shell size and growth rates are linked to environmental conditions. For example, some species form narrower shells in coarser sub- strates (Hinch et al., 1989) or grow faster in sand than in mud (Hinch et al., 1986). However, taking into account the large margin of error in using this method, most Arkansia wheeleri encountered in the Kiamichi River were old. Using this method, the youngest live A. wheeleri specimen we encountered was approximately 12 years of age. No juveniles were encountered. Both types of data, shell-size distributions and ages predicted from external annuli, demonstrated that most A. wheeleri encountered in the Kiamichi River were old. Because of its rarity, the reproductive biology of Arkansia wheeleri remains unknown. Like other anodon- tines, it is probably bradytictic. The closest relative of A. wheeleri, Arcidens confragosus (Say, 1829), becomes gravid in the fall and releases glochidia in the spring 150 AMER. MALAC. BULL. 11(2) (1995) (Clarke, 1981). We were unable to obtain any gravid A. wheeleri and thus obtained no glochidia. A. wheeleri glochidia are probably similar to other alasmidontine glochidia. Alasmidontine glochidia are asymmetrical and have a stylet covered with microstylets which facilitate attachment to the fish host. Glochidial releases are proba- bly tied to natural water temperature changes in the spring and fall (Jirka and Neves, 1992). It appears that historically, Arkansia wheeleri sur- vived equally well above and below the impounded tribu- tary to the Kiamichi River (Clarke, 1987). Historically, A. wheeleri occurred in at least seven sites below the tributary. However, in five years of combined sampling effort by Mehlhop and Miller (1989), 1988-1989, and ourselves, 1990-1992, only three subpopulations of A. wheeleri have been found below Jackfork Creek. Therefore, only three out of seven (43%) of the known subpopulations of A. wheeleri survive below Jackfork Creek. In contrast, three out of four (75%) of the historical locations of A. wheeleri above Jackfork Creek have been confirmed and five new locations have been discovered. No new locations have been discov- ered below Jackfork Creek despite intensive survey efforts. In addition, the relative abundance of A. wheeleri is slightly higher above Jackfork Creek than below, although these differences are not statistically significant. Unfortunately, we have no historical abundance data for A. wheeleri in the Kiamichi River. The greatest threats to the continued existence of Arkansia wheeleri in the Kiamichi River are land use changes, including further impoundment of the river, water transfers, timber harvesting, and pollution from agricultural and industrial development (Neves, 1993; Mehlhop and Vaughn, 1994). This species is also threatened by the inva- sion of exotic bivalve species, particularly the zebra mussel, Dreissena polymorpha (Pallas, 1771). D. polymorpha _ is now found in the Arkansas River system in Oklahoma. The high dispersal capabilities of this species make it highly probable that it will invade the Red River system, including the Kiamichi River, in the near future (French, 1990). ACKNOWLEDGMENTS We thank John Alderman, Tambra Browning, David Certain, Julian Hilliard, David Martinez, Estelle Miller, David Partridge, and Matthew Winston for help with field work at various times. Matthew Craig performed the annuli counts on Arkansia wheeleri and Christopher Taylor commented on the manuscript. This study was funded by the U.S. Fish and Wildlife Service and the Oklahoma Department of Wildlife Conservation through Endangered Species Act funding (project E-12). LITERATURE CITED Clarke, A. H. 1981. The tribe Alasmidontini (Unionidae: Anodontinae), Part I: Pegias, Alasmidonta and Arcidens. Smithsonian Contributions to Zoology 326:85-89. Clarke, A. H. 1985. The tribe Alasmidontini (Unionidae: Anodontinae), Part II: Lasmigona and Simpsonaias. Smithsonian Contributions to Zoology 399: 75 pp. Clarke, A. H. 1987. Status Survey of Lampsilis streckeri Frierson (1927) and Arcidens wheeleri (Ortmann & Walker, 1912). Unpubl. Report no. 14-16-0004-86-057 to U.S. Fish and Wildlife Service, Jackson, Mississippi. 67 pp. Cooper, C. M. 1984. The freshwater bivalves of Lake Chicot, an oxbow of the Mississippi in Arkansas. The Nautilus 98: 142-145. Curtis, N. M. and W. E. Ham. 1972. Geomorphic provinces in Oklahoma. In: Geology and Earth Resources of Oklahoma, K. S. Johnson, ed. p. 3. Oklahoma Geological Survey, Educational Publication. Downing, W. L., J. Shostell, and J. A. Downing. 1992. Non-annual exter- nal annuli in the freshwater mussels Anodonta grandis grandis and Lampsilis radiata siliquoidea. Freshwater Biology 28:309- 317. Dunne, T. and L. B. Leonard. 1978. Water in Environmental Planning. W. H. Freeman and Co., New York. 818 pp. Federal Register. 1991 (23 October). Endangered and threatened wildlife and plants: final rule to list the Ouachita rock-pocketbook (mus- sel) as an endangered species. Federal Register 56(205):54950- 54957. French, J. R. P. 1990. The exotic zebra mussel - a new threat to endan- gered freshwater mussels. Endangered Species Technical Bulletin 15:3-4. Green, R. H., S. M. Singh, and R. C. Bailey. 1985. Bivalve molluscs as response systems for modelling spatial and temporal environmen- tal patterns. The Science of the Total Environment 46: 147-169. Hinch, S. G., R. C. Bailey, and R. H. Green. 1986. Growth of Lampsilis radiata (Bivalvia: Unionidae) in sand and mud: a reciprocal trans- plant experiment. Canadian Journal Fisheries and Aquatic Sciences 43:548-552. Hinch, S. G., L. J. Kelly, and R. H. Green. 1989. Morphological variation of Elliptio complanata (Bivalvia: Unionidae) in differing sedi- ments of soft-water lakes exposed to acidic deposition. Canadian Journal of Zoology 67:1895-1899. Hynes, H. B. N. 1970. The Ecology of Running Water. University of Toronto Press, Toronto. 555 pp. Isely, F. B. 1924. The freshwater mussel fauna of eastern Oklahoma. Proceedings of the Oklahoma Academy of Science 4:43-118. Jirka, K. J. and R. J. Neves. 1992. Reproductive biology of four species of freshwater mussels (Mollusca: Unionidae) in the New River, Virginia and West Virginia. Journal of Freshwater Ecology 7:35- 44. Kat, P. W. 1982. Effects of population density and substratum type on growth and migration of Elliptio complanata (Bivalvia: Unionidae). Malacological Review 15:119-127. Kovalak, W. P., S. D. Dennis, and J. M. Bates. 1986. Sampling effort required to find rare species of freshwater mussels. In: Rationale for Sampling and Interpretation of Ecological Data in the Assessment of Freshwater Ecosystems, B. G. Isom, ed. pp. 46-59. American Society for Testing and Material, Special Technical Publication No. 894. Ludwig, J. A. and J. F. Reynolds. 1988. Statistical Ecology. John Wiley & Sons, New York. 337 pp. McMahon, R. F. 1991. Mollusca: Bivalvia. In: Ecology and Classification of North American Freshwater Invertebrates, J. H. Thorp and A. P. Covich, eds. pp. 315-390. Academic Press, New York. Mehlhop, P. and E. K. Miller. 1989. Status and Distribution of Arkansia wheeleri Ortmann & Walker, 1912 (syn. Arcidens wheeleri) in the Kiamichi River, Oklahoma. Unpublished report no. 21440-88- VAUGHN AND PYRON: ARKANSIA WHEELERI IN THE KIAMICHI RIVER 151 00142 to U.S. Fish and Wildlife Service, Tulsa, Oklahoma. 101 Ppp. Mehlop, P. and C. C. Vaughn. 1994. Threats to and sustainability of ecosystems for freshwater mollusks. /n: Sustainable Ecological Systems: Implementing an Ecological Approach to Land Man- agement, W. Covington and L. F. Dehand, eds. pp. 68-77. U. S. Forest Service, U. S. Department of Agriculture, Fort Collins, Colorado. General Technical Report Rm-247 for Rocky Mountain Range and Forest Experimental Station. 363 pp. Neves, R. J. 1993. A state-of-the-unionids address. Jn: Conservation and Management of Freshwater Mussels, K. S. Cummings, A. C. Buchanan, and L. M. Koch, eds. Proceedings of a UMRCC Symposium, 12-14 October 1992, St. Louis, Missouri. pp. 1-10. Upper Mississippi River Conservation Committee, Rock Island, Illinois. 189 pp. Neves, R. J. and S. N. Moyer. 1988. Evaluation of techniques for age determination of freshwater mussels (Unionidae). American Malacological Bulletin 6:179-188. Ortmann, A. E. and B. Walker. 1912. A new North American naiad. The Nautilus 25:97-100. Payne, B. S. and A. C. Miller. 1989. Growth and survival of recent recruits to a population of Fusconaia ebena (Bivalvia: Mollusca) in the lower Ohio River. American Midland Naturalist 121:99-104. Pyron, M. and C. C. Vaughn. 1994. Ecological Characteristics of the Kiamichi River, Oklahoma. Unpublished report submitted to the U.S. Fish and Wildlife Service, Tulsa, Oklahoma. 68 pp. Salmon, A. and R. H. Green. 1983. Environmental determinants of union- id clam distribution in the Middle Thames River, Ontario. Canadian Journal of Zoology 61:832-838. Sokal, R. R. and F. J. Rohlf. 1981. Biometry, 2nd ed. W. H. Freeman and Co., San Francisco. 859 pp. Stern, E. M. 1983. Depth distribution and density of freshwater mussels (Unionidae) collected with scuba from the lower Wisconsin and St. Croix Rivers. The Nautilus 97:36-42. Strayer, D. L. 1981. Notes on the microhabitats of unionid mussels in some Michigan streams. American Midland Naturalist 106:411- 415. 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, and J. D. Williams. 1988. Common and Scientific Names of Aquatic Invertebrates from the United States and Canada: Mollusks. American Fisheries Society, Bethesda, Maryland, Special Publication 16. 277 pp. Valentine, B. D. and D. H. Stansbery. 1971. An introduction to the naiads of the Lake Texoma region, Oklahoma, with notes on the Red River fauna (Mollusca: Unionidae). Sterkiana 42: 1-40. Vaughn, C. C. 1993. Survey for Arkansia wheeleri in the Little River, Oklahoma. Unpublished report submitted to the U.S. Fish and Wildlife Service, Tulsa, Oklahoma. 24 pp. Vaughn, C. C., M. Pyron, and D. Certain. 1993. Habitat Use and Repro- ductive Biology of Arkansia wheeleri in the Kiamichi River, Oklahoma - Final Report. Unpublished report to the Oklahoma Department of Wildlife Conservation, Oklahoma City, Oklahoma. 104 pp. Warren, M. L. and B. M. Burr. 1994. Status of the freshwater fishes of the United states: overview of an imperiled fauna. Fisheries 19:6-18. Way, C. M., A. C. Miller, and B. S. Payne. 1990. The influence of physi- cal factors on the distribution and abundance of freshwater mus- sels (Bivalvia: Unionacea) in the lower Tennessee River. The Nautilus 103:96-98. White, D. S. 1979. The effect of lake-level fluctuations on Corbicula and other pelecypods in Lake Texoma, Texas and Oklahoma. Proceedings of the First International Corbicula Symposium, J. C. Britton, ed. pp. 82-88. Texas Christian University Research Foundation, Ft. Worth, Texas. Wheeler, H. E. 1918. The Mollusca of Clark County, Arkansas. The Nautilus 31:109-125. Williams, J. D., M. L. Warren, K. S. Cummings, J. L. Harris, and R. J. Neves. 1992. Conservation status of freshwater mussels of the United States and Canada. Fisheries 18:6-22. Date of manuscript acceptance: 3 November 1994 Infestations of glochidia on fishes in the Barren River, Kentucky Jeffrey L. Weiss* and James B. Layzer National Biological Service, Tennessee Cooperative Fishery Research Unit, Tennessee Technological University, Cookeville, Tennessee 38505, U.S. A. Abstract. We collected fish monthly from the Barren River, Kentucky, to assess glochidial infestations. Glochidia were encysted on 4.1% of the 2,510 fish of 43 species examined. Twenty-five fish species in 11 families were infested; 14 of these species are not known to be hosts of any of the 27 mussel species (Unionidae) occurring in the Barren River. Amblemine glochidia occurred on 19 species of fish. Eight species of fish were infested with anodontine glochidia, while lampsiline glochidia occurred on only five species. Differences in the degree of host specificity were striking among the Ambleminae. Glochidia of Amblema plicata (Say, 1817) occurred on 12 species of fish, whereas those of Quadrula pustulosa (1. Lea, 1831) were found only on channel catfish [/ctalurus punctatus (Rafinesque, 1818)]. Overlap in host fishes occurred between the Ambleminae and the other subfamilies, but not between the Anodontinae and Lampsilinae. Potential new hosts are identified for Lasmigona complanata (Barnes, 1823), Lasmigona costata (Rafinesque, 1820), Megalonaias nervosa (Rafinesque, 1820), A. plicata, and Pleurobema spp. The life cycle of most freshwater mussel species (Unionidae) includes a parasitic larval stage (glochidium). Glochidia develop in the gills of female mussels and are discharged into the water. Once free, they must attach to a suitable host or die. Glochidia of most mussel species par- asitize fish. Some species are host specific, using a single host; others parasitize as many as 25 fish species (Gordon and Layzer, 1989). Glochidia that contact a suitable fish host usually attach to gills or fins and become encysted in fish tissue. Although host fish play a vital role in the life cycle of freshwater mussels, hosts of most mussel species are unknown (Fuller, 1974; Hoggarth, 1992). Some mussel species can maximize the likelihood of glochidial attachment by attracting fish. For instance, the mantle of Lampsilis cardium closely resembles a small fish (Kraemer, 1970). Presumably, this “lure” attracts their host fishes which are piscivorous. Glochidia of other mussel species are released in packets (conglutinates) which can resemble aquatic insects often preyed upon by fish (Stein, 1971; Gordon and Layzer, 1989). Conglutinates can result in higher intensities of glochidial infestations (Neves and Widlak, 1988). Released anodontine glochidia are sus- pended in the water column by mucus strands and left to drift into contact with a host (Stein, 1971). This means of dispersal is augmented by the tremendous number of glochidia (up to 3.5 million) which can be released by a *Present address: Minnesota Department of Natural Resources, Area Fisheries Office, Route 2, Box 85, Lanesboro, Minnesota 55959, U.S. A. single female (Surber, 1912). Despite these adaptations, apparently few glochidia successfully attach to a host. Neves and Widlak (1988) found only 14% of 4,800 fish examined had encysted glochidia. Holland-Bartels and Kammer (1989) reported glochidial infestations on just 4% of 2,000 fish examined from June through August. Although not all species examined may have been hosts, these low rates of infestation suggest that host abundance can play an important role in determining reproductive suc- Gess; This study examines glochidial infestations by a diverse assemblage of freshwater mussels in the presence of a diverse fish fauna. Differences in host specificity within and among mussel subfamilies are examined. Potential new hosts are identified but will require confirmation by induced infestation experiments. MATERIALS AND METHODS The study was conducted on a 5-km reach of the Barren River immediately downstream from Lock and Dam 1, Warren County, Kentucky. Five 150-m long sites were established in riffle, run, and pool habitats. Mussel species composition was evaluated at each site in 1991 by timed diving. Two divers, either snorkeling or using SCUBA equipment, collected mussels at each site for 30 min. Mussels were identified to species and returned to the site. Except for Pyganodon grandis, nomenclature follows Turgeon et al. (1988). American Malacological Bulletin, Vol. 11(2) (1995):153-159 153 154 AMER. MALAC. BULL. 11(2) (1995) Fish were collected at each site from October 1990 to September 1991 in all months except November 1990. A variety of gear was used depending upon water levels. Experimental gill nets were 45.7 m long with six equally sized panels of mesh from 1.9-6.4 cm bar measure. Two sizes of hoop nets were used: large nets had 92-cm diame- ter hoops, a front mesh of 5.1 cm, and back mesh of 3.8 cm; small nets had 51-cm diameter hoops and 3.8 cm mesh throughout. Large nets were used in areas of sufficient depth to completely submerge the hoops. In shallower sites, the small nets were used. Both baited and unbaited sets were made and the time of each recorded. A 30.5-m long seine with 6.4 mm mesh was used on the four shallowest sites when water levels allowed. The number of seining locations established within a station was determined by the habitat present. A 6.1-m long seine with 6.4 mm mesh was used on several dates when water levels prevented use of the large seine. Electrofishing was conducted with a boat- mounted direct-current unit. Each site was sampled by making three passes upstream. One pass was made along each bank and one in the center of each site. All fish col- lected were retained. Small individuals were fixed in 10% buffered formalin in the field and transferred to 70% ethanol in the laboratory. Large individuals were placed on ice and then frozen within 48 hr. Common and scientific names of fishes follow Robins et al. (1991). The maximum number of fish examined for glochidial infestations was restricted to 20 individuals per species for each gear type used at a station on each sam- pling date. Fish were examined under a dissecting micro- scope at 10-70X magnification. All gill arches and fins were removed from fish greater than approximately 100 mm in length and examined individually. Smaller fish were examined whole. Glochidia were removed with a probe and preserved in 70% ethanol. The locations of glochidia on each infested fish were recorded. Length, height, and hinge length of each glochidium were measured to the nearest 5 pm with an ocular micrometer in a compound microscope (100 X). Length was the maximum distance between the anterior and posterior shell margins. Height was measured from the hinge to the ventral margin. Hinge length was defined as the distance between the points where the hinge intersected with the anterior and posterior shell margins. Shape was used to identify glochidium to subfamily; misidentification of subfamily was unlikely because differences in shape were usually distinct. Dimensions of glochidium were compared to those of glochidia obtained from gravid mussels and to measure- ments reported in the literature for identification of species. This method did not account for potential differences in glochidia sizes among populations; however, when species dimensions were not different, identification was restricted to subfamily. RESULTS Twenty-seven species of mussels including the fed- erally endangered Pleurobema plenum were collected from the study area (Table 1). Thirteen species of Ambleminae were collected; Amblema plicata comprised 40% of the total number of mussels collected and Megalonaias nervosa comprised 11%. Actinonaias ligamentina (8%) and Ptychobranchus fasciolaris (7%) were the most common of the 10 species of Lampsilinae collected. Of the four species of Anodontinae present, Lasmigona costata and L. complanata were most abundant; together they comprised 5% of the mussels collected. We collected and examined only 281 fish from December through May because the high discharge from the Barren River Lake, located 40 km upstream, affected our sampling efficiency. During the summer and fall, we examined 164-748 fish per month. Overall, 6,736 fish of 43 species were collected and 2,510 of these were exam- ined for glochidial infestations. Twenty-five species of fish in 11 families were infested (Table 2). Fourteen of these species have not been identified as hosts for any mussel species occurring in the Barren River. Glochidia were not found on seven fish species that are known hosts for one or more mussel species present; however, six of these fish species were rarely collected. Although hooks on glochidial valves are characteris- tic of Anodontinae, they were not easily observable on closed glochidia. Nonetheless, anodontine glochidia were distinctive because of their triangular shape. Moreover, anodontine glochidia were nearly always larger in all dimensions than amblemine and lampsiline glochidia (Table 3). Glochidia of Lampsilinae and Ambleminae overlapped considerably in size of each dimension mea- sured; however, individual species in these subfamilies did not overlap in all three dimensions. For instance, valve lengths and heights of Megalonaias nervosa and Ellipsaria lineolata were similar, but hinge lengths differed consider- ably. In general, glochidia of Ambleminae tended to be rounder in shape than glochidia of Lampsilinae. Amblemine glochidia occurred on 19 species of fish in 10 families (Table 2). Anodontine glochidia occurred on eight species of fish with those of the Lampsilinae infesting just five species. Four fish species, longnose gar (Lepiso- steus osseus), gizzard shad (Dorosoma cepedianum), gold- en redhorse (Moxostoma erythrurum), and flathead catfish (Pylodictis olivaris) were used by anodontine and amblem- ine species. Infestations of lampsiline and amblemine species occurred on channel catfish (Ictalurus punctatus) and spotted bass (Micropterus punctatus), however, infesta- tions by anodontine and lampsiline species did not occur on the same fish species. In three instances, anodontine glochidia occurred along with glochidia of Megalonaias WEISS AND LAYZER: GLOCHIDIA ON FISHES IN KENTUCKY Table 1. Mussel species (Unionidae) collected from the Barren River during 1991. 155 Relative Subfamily Species N Abundance (%) Anodontinae Arcidens confragosus (Say, 1829) 3 0.6 Pyganodon grandis (Say, 1829) 1 0.2 Lasmigona complanata (Bares, 1823) 11 2.2 L. costata (Rafinesque, 1820) 13 2.6 Subfamily Total 28 5.5 Ambleminae Amblema plicata (Say, 1817) 201 39.6 Cyclonaias tuberculata (Rafinesque, 1820) 1 0.2 Elliptio crassidens (Lamarck, 1819) 12 2.4 Fusconaia flava (Rafinesque, 1820) 2 0.4 F. subrotunda (1. Lea, 1831) 4 0.8 Megalonaias nervosa (Rafinesque, 1820) 56 11.0 Pleurobema cordatum (Rafinesque, 1820) 17 3.4 P. plenum (1. Lea, 1840) 8 1.6 P. pyramidatum (1. Lea, 1840) 2 0.4 P. coccineum (Conrad, 1834) 2. 0.4 Quadrula pustulosa (1. Lea, 1831) 13 2.6 Q. quadrula (Rafinesque, 1820) 14 2.8 Tritogonia verrucosa (Rafinesque, 1820) 12 2.4 Subfamily Total 344 67.7 Lampsilinae Actinonaias ligamentina (Lamarck, 1819) 38 IS Ellipsaria lineolata (Rafinesque, 1820) 18 3.5 Lampsilis cardium (Rafinesque, 1820) 4 0.8 L. ovata (Say, 1817) 5 1.0 Leptodea fragilis (Rafinesque, 1820) 3 0.6 Ligumia recta (Lamarck, 1819) l 0:2 Obliquaria reflexa (Rafinesque, 1820) 2 0.4 Potamilus alatus (Say, 1819) 24 4.7 Ptychobranchus fasciolaris (Rafinesque, 1820) 37 13 Truncilla truncata (Rafinesque, 1820) 2 0.4 Subfamily Total 136 26.8 Overall Total 506 nervosa on the same individual (longnose gar, gizzard shad, and flathead catfish). An individual golden redhorse and a sauger (Stizostedion canadense) were infested with glochidia of Anodontinae and those of Amblema plicata. Glochidia of A. plicata and another amblemine species occurred on one mooneye (Hiodon tergisus). Ambleminae displayed varying levels of host speci- ficity (Table 4). Glochidia of Amblema plicata were found on 12 fish species and Megalonaias nervosa on eight fish species. In contrast, glochidia of Quadrula pustulosa occurred only on channel catfish and glochidia of Pleurobema spp. only on mooneye and steelcolor shiner (Cyprinella whipplei). Seven known amblemine hosts were infested while 12 fish species previously not identified as hosts of the Ambleminae occurring in the Barren River, also were infested (Table 4). Glochidia of both Lasmigona complanata and L. costata occurred on gizzard shad and river redhorse (Moxostoma carinatum) while glochidia of either Pyganodon grandis or Arcidens confragosus occurred on four fish species. None of the fishes infested with anodon- tine glochidia were known hosts of an anodontine species occurring in the Barren River. Lampsiline glochidia were rarely found on the fish examined. Glochidia of Potamilus alatus infested freshwa- ter drum (Aplodinotus grunniens), a known host. Glochidia of other Lampsilinae occurred on four fish species, none of which are known hosts of any lampsiline species collected. Most infested fish carried few encysted glochidia (generally 1-5); however, one freshwater drum, captured in April, was infested with 232 Potamilus alatus glochidia. This was the only instance of such a high intensity of infes- tation. Glochidia of Megalonaias nervosa occurred as fre- quently on fins as on gills of fish, particularly longnose gar and gizzard shad. Anodontine glochidia occurred more fre- quently on gills (92%) than on fins while lampsiline glochidia occurred only on gills. 156 AMER. MALAC. BULL. 11(2) (1995) Table 2. Prevalence of glochidial infestations on fishes collected from the Barren River from October 1990 to September 1991. Asterisks indicate fish species which are known hosts of one or more mussel species in the Barren River. Family Species Ambleminae Anodontinae Lampsilinae Lepisosteidae *Lepisosteus osseus (Linné, 1758) L. oculatus (Winchell, 1864) Hiodontidae Hiodon tergisus (Lesueur, 1818) Clupeidae *Dorosoma cepedianum (Lesueur, 1818) Cyprinidae *Pimephales notatus (Rafinesque, 1820) *Cyprinus carpio Linné, 1758 Notropis atherinoides Rafinesque, 1818 N. rubellus (Agassiz, 1850) Cyprinella spilopterus (Cope, 1868) C. whipplei Girard, 1856 Erimystax dissimilis (Kirtland, 1840) Catostomidae Moxostoma duquesnei (Lesueur, 1817) M. erythrurum (Rafinesque, 1818) M. carinatum (Cope, 1870) M. macrolepidotum (Lesueur, 1817) M. anisurum (Rafinesque, 1820) Hypentelium nigricans (Lesueur, 1817) Minytrema melanops (Rafinesque, 1820) Ictiobus bubalus (Rafinesque, 1818) Carpiodes carpio (Rafinesque, 1820) C. cyprinus (Lesueur, 1817) *C. velifer (Rafinesque, 1820) Esocidae Esox masquinongy Mitchell, 1824 Ictaluridae *Ictalurus punctatus (Rafinesque, 1818) *Pylodictis olivaris (Rafinesque, 1818) Cyprinodontidae Fundulus notatus (Rafinesque, 1820) Poeciliidae Gambusia affinis (Baird and Girard, 1853) Atherinidae *Labidesthes sicculus (Cope, 1865) Cottidae *Cottus carolinae (Gill, 1861) Percichthyidae *Morone chrysops (Rafinesque, 1820) Centrarchidae *Lepomis gulosus (Cuvier, 1829) *L. macrochirus Rafinesque, 1819 L. megalotis (Rafinesque, 1820) *Ambloplites rupestris (Rafinesque, 1817) *Poxomis annularis Rafinesque, 1818 *P. nigromaculatus (Lesueur, 1829) Micropterus punctatus (Rafinesque, 1819) *M. salmoides (Lacepede, 1802) Percidae Etheostoma blennioides Rafinesque, 1819 Percina caprodes (Rafinesque, 1818) P. phoxocephala (Nelson, 1876) * Stizostedion canadense (Smith, 1834) Sciaenidae *Aplodinotus grunniens Rafinesque, 1819 N 29 Percent Infested (by Subfamily) 24 3 - 45 - - 9 2 - <1 = a <1 = - 1 = ae 2 ee a 5 = = 5 2 - = 24 Bee = 2 = 6 ae = = 25 = 12 - 16 67 33 - - - <1 100 - - 67 - - <1 - - = = 2: 75 - - 2 - 5 4 = = se 20 x 12 - - WEISS AND LAYZER: GLOCHIDIA ON FISHES IN KENTUCKY 157 Table 3. Dimensions of glochidia measured in this and other studies as cited (ranges are mean + 1 SD). Dimensions (pm) Family Subfamily Length Height Anodontinae Pyganodon grandis 350-362 352-358 366-380 371-396 Arcidens confragosus 354-364 353-355 Lasmigona complanata 290-296 295-309 L. costata 341-347 363-375 353-377 380-414 Ambleminae Amblema plicata 200 200 185-200 195-215 Cyclonaias tuberculata 348-362 288-300 Elliptio crassidens 130 150 150 160 Fusconaia flava 150 150 F. subrotunda 130 150 Megalonaias nervosa 254-268 341-351 Pleurobema cordatum 140 150 160 175 P. coccineum 160 160 Quadrula pustulosa 233-245 292-303 230 300 Q. quadrula 85 90 Tritogonia verrucosa 87-93 98-102 119-125 106-112 Lampsilinae Actinonaias ligamentina 245-263 221-235 220 243 Ellipsaria lineolata 248-258 316-354 232-242 316-326 Lampsilis cardium 244-254 277-289 L. ovata 232 270-278 303-315 257-271 Leptodae fragilis 71-73 80-82 Ligumia recta 205-217 257-263 Obliquaria reflexa 214-220 213-223 Potamilus alatus 213-225 370-386 Ptychobranchus fasciolaris 170-176 180-194 Truncilla truncata 60 70 Hinge Length Citation 250-258 Hoggarth (1988) 272-281 Current Study 239-253 Hoggarth (1988) 195-208 Hoggarth (1988) 239-243 Hoggarth (1988) 249-271 Current Study - Surber (1912) - Howard (1914) 127-139 Jirka and Neves (1992) - Ortmann (1912) - Surber (1915) - Ortmann (1912) = Ortmann (1912) 147-153 Hoggarth (1988) - Yokley (1972) - Surber (1915) = Surber (1915) 98-106 Current Study = Lefevre and Curtis (1912) = Surber (1915) 43-45 Hoggarth (1988) 46-52 Jirka and Neves (1992) 123-135 Jirka and Neves (1992) 125 Hoggarth (1988) 87-109 Curent Study 87-94 Hoggarth (1988) 107-115 Hoggarth (1988) 113-119 Hoggarth (1988) 117-125 Jirka and Neves (1992) 30-36 Hoggarth (1988) 106-112 Hoggarth (1988) 118-126 Hoggarth (1988) 96-108 Hoggarth (1988) 78-88 Hoggarth (1988) Surber (1912) DISCUSSION Most glochidia encysted on fish were those of the most abundant mussel species in the Barren River. Glochidia of Amblema plicata, the most frequently collect- ed mussel, occurred on 12 species of fish in 7 families. Only two of these species have been reported as hosts (see Gordon and Layzer, 1989; Watters, 1994). Although labo- ratory trials of induced infestations are needed to confirm host suitability, the diversity of known hosts for A. plicata suggests that many additional fish species could be suitable hosts. In contrast, we did not find glochidia of Potamilus alatus encysted on any species other than freshwater drum, its only known host (Howard, 1913). The small number of fish collected in the winter and spring most likely resulted in the low prevalence of lampsi- line infestations on fish we examined. The relatively high frequency of fish infested with glochidia of Megalonaias nervosa during the winter probably resulted from the long encystment period (2-6 months) typical of this species (Howard, 1914). The low prevalence of infestations on fins by anodontine glochidia was unexpected. The hooks of anodontine glochidia are thought to be an adaptation for 158 AMER. MALAC. BULL. 11(2) (1995) Table 4. Fish species examined and percent infested by mussel species in the Barren River from October 1990 to September 1991. Asterisks indi- cate known hosts. % Mussel species Fish Species N Infested Anodontinae Lasmigona complanata Lepisosteus osseus 29 3 Dorosoma cepedianum 234 <1 Moxostoma carinatum 17 18 Stizostedion canadense 5 20 L. costata Dorosoma cepedianum 234 <1 Moxostoma carinatum 17 6 Anodontinae spp. M. erythrurum 55 2 M. macrolepidotum 46 2 Carpiodes carpio 4 25 Pylodictis olivaris 3 33 Ambleminae Megalonaias nervosa Lepisosteus osseus 29 24 *Dorosoma cepedianum 234 6 *Pylodictis olivaris 3 67 *Morone chrysops 1 100 Lepomis gulosus 3 67 *Pomoxis annularis 4 75 Micropterus punctulatus 66 2 *Aplodinotus grunniens 73 1 Amblema plicata Hiodon tergisus 11 36 Notropis atherinoides 307 <1 Cyprinella splioptera 357 attachment to fins (Lefevre and Curtis, 1912). However, the higher prevalence of infestations on gills observed in this study and by Wiles (1975) suggests that the hooks are simply an adaptation for attachment, regardless of the loca- tion on the fish. Catostomids have been infrequently identified as glochidial hosts; however, we found glochidia of Amblema plicata, Lasmigona complanata, L. costata, and Anodontinae spp. encysted on several species of Catostomidae. Because steelcolor shiners and mooneyes were infested with glochidia of Pleurobema spp., they should be tested as potential hosts for the federally endan- gered Pleurobema plenum. ACKNOWLEDGMENTS The authors wish to thank Thomas G. Cochran for his invaluable assistance with fish and mussel sampling. We appreciate the clerical assistance of Mrs. Betty Harris. We are grateful to two anonymous reviewers who provided valuable comments. Funding for this study was provided by the National Biological Service, Upper Mississippi Science Center. LITERATURE CITED Fuller, S. L. H. 1974. Clams and mussels (Mollusca: Bivalvia). Jn: Pollution Ecology of Freshwater Invertebrates, C. W. Hart and S. L. H. Fuller, eds. pp. 215-273. Academic Press, New York. Gordon, M. E. and J. B. Layzer. 1989. Mussels (Bivalvia: Unionoidea) of the Cumberland River: review of life histories and ecological rela- tionships. United States Fish and Wildlife Service, Biological Report 89(15). 99 pp. Hoggarth, M. A. 1988. The use of glochidia in the systematics of the Unionidae (Mollusca: Bivalvia). Doctoral dissertation, Ohio State University, Columbus, Ohio. 340 pp. Hoggarth, M. A. 1992. An examination of the glochidia-host relation- ships reported in the literature for North American species of Unionacea (Mollusca: Bivalvia). Malacology Data Net 3(1-4):1- 30. Holland-Bartels, L. E. and T. W. Kammer. 1989. Seasonal reproductive development in Lampsilis ventricosa, Amblema plicata, and Potamilus alatus (Pelecypoda: Unionidae). Journal of Freshwater Ecology 5:87-92. Howard, A. D. 1913. The catfish as a host for fresh-water mussels. Transactions of the American Fisheries Society 42:65-70. Howard, A. D. 1914. Experiments in propagation of fresh-water mussels of the Quadrula group. Bulletin of the United States Bureau of Fisheries, Document 801. 58 pp. Jirka, K. J. and R. J. Neves. 1992. Reproductive biology of four species of freshwater mussels (Mollusca: Unionidae) in the New River, Virginia and West Virginia. Journal of Freshwater Ecology 7:35- 43. Kraemer, L. R. 1970. The mantle flap in three species of Lampsilis (Pelecypoda: Unionidae). Malacologia 10:225-282. Lefevre, G. and W. C. Curtis. 1912. Studies on the reproduction and propagation of fresh-water mussels. Bulletin of the United States Bureau of Fisheries 30:195-201. Neves, R. J. and J. C. Widlak. 1988. Occurrence of glochidia in stream drift and on fishes of the upper North Fork Holston River, Virginia. American Midland Naturalist 119:111-120. Ortmann, A. E. 1912. Notes upon the families and genera of the najades. Annals of the Carnegie Museum 8:222-365. Robins, C. R., R. M. Bailey, C. E. Bond, J. R. Brooker, E. A. Lachner, R. N. Lea, and W. B. Scott. 1991. Common and Scientific Names of Fishes from the United States and Canada. American Fisheries Society Special Publication 20. 183 pp. Stein, C. B. 1971. Naiad life cycles: their significance in the conserva- WEISS AND LAYZER: GLOCHIDIA ON FISHES IN KENTUCKY 159 tion of the fauna. In: Proceedings of a Symposium on Rare and Endangered Molluscs (Naiades) of the United States, S. E. Jorgensen and R. W. Sharp, eds. pp. 19-25. United States Department of the Interior, Fort Snelling, Twin Cities, Minnesota. Surber, T. 1912. Identification of the glochidia of freshwater mussels. Bulletin of the United States Bureau of Fisheries, Document 771. 13 pp. Surber, T. 1915. Identification of the glochidia of freshwater mussels. Bulletin of the United States Bureau of Fisheries, Document 813. 10 pp. 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, and J. D. Williams. 1988. Common and Scientific Names of Aquatic Invertebrates from the United States and Canada: Mollusks. American Fisheries Society Special Publication 16. 277 pp. Watters, G. T. 1994. An Annotated Bibliography of the Reproduction and Propogation of the Unionoidea (Primarily of North America). Ohio Biological Survey Miscellaneous Contribution 1. 158 pp. Wiles, M. 1975. The glochidia of certain Unionidae (Mollusca) in Nova Scotia and their fish hosts. Canadian Journal of Zoology 53:33- 41. Yokley, P., Jr. 1972. Life history of Pleurobema cordatum (Rafinesque 1820) (Bivalvia: Unionacea). Malacologia 11:351-364. Date of manuscript acceptance: 5 July 1994 SYMPOSIUM: GULF OF MEXICO MOLLUSCA Organized by JOSEPH C. BRITTON TEXAS CHRISTIAN UNIVERSITY AMERICAN MALACOLOGICAL UNION HOUSTON, TEXAS 10 - 12 JULY 1994 161 Ecology of infaunal Mollusca in south Texas estuaries Paul A. Montagna and Richard D. Kalke University of Texas at Austin, Marine Science Institute, PO. Box 1267, Port Aransas, Texas 78373 U.S. A. Abstract. The ecology of Texas estuaries is strongly influenced by latitudinal ecotones that exist along the northwestern Gulf of Mexico coastline. Long-term studies were conducted in four of the seven major estuarine ecosystems in Texas. The objective was to determine the role of climatic variability and concordant differences in freshwater inflow among the ecosystems in structuring benthic infaunal communities and maintaining secondary production. Mollusks are prominent members of the infauna in all benthic habitats of Texas estuaries. The abundance, biomass, and community structure of mollusks was measured along salinity gradients within the four south Texas estuaries. Infaunal samples were collected by divers using small (6.7 cm diameter) cores (so larger epibenthic mollusks were not collected). Overall, these Texas estuaries had a mean of 14 species of infaunal mollusks, with mean abundance of 7,500 individuals/m2, and mean biomass of 2.4 g/m2. Freshwater inflow is the dominant factor regulating variability of molluscan communities. Salinity is a surrogate for inflow, therefore, there are zoogeographic patterns within and among estuaries related to salinity patterns. There are seasonal, interannual, and latitudinal patterns of inflow, and these patterns are apparently regulating community structure, population dynamics, and secondary production in Texas estuaries. Recent water projects to enhance the amount of freshwater flowing into estuaries appear to have had an effect and have increased the number of mollusks in those areas. However the projects occurred during a naturally wet period, so it is difficult to differentiate natural versus anthropogenic changes. The response of mollusks to natural gradients and man-induced changes of freshwater inflow demonstrate the importance of this factor in regulating benthic communities. A major component of benthic ecosystems in Texas 1982). These distinct patterns are very important, because estuaries, as is true elsewhere, is the Mollusca. Molluscan growth, reproduction, and migration of many species are biomass dominates the macroinfauna in Lavaca, San keyed to seasonal events. The timing and magnitude of Antonio, Corpus Christi, and Nueces Bays (Kalke and inundation is believed to regulate finfish and shellfish pro- Montagna, 1991; Montagna and Kalke, 1992). During duction (Texas Department of Water Resources, 1982). peak recruitment events, Mollusca can also dominate popu- We have been conducting long-term studies in four lation abundance. However, differences in population size of the seven estuaries to determine the role of freshwater and community structure exist within and among Texas inflow in maintaining benthic productivity. The primary Bays. purpose of the current study was to determine the degree of There are seven major estuarine systems along 600 influence of freshwater inflow in regulating zoogeographic linear kilometers of coastline. The estuaries of Texas are differences of molluscan population size and community remarkably diverse in spite of similar physiography (Fig. structure within and among Texas estuaries. The secondary 1). This is due to a climatic gradient, which influences purpose of this study was to assess the effects of two major freshwater inflow. The gradient of decreasing rainfall, with water projects designed to increase freshwater inflows to concomitant freshwater inflow from north to south, is the estuaries to maintain or enhance productivity. One project most distinctive feature of the coastline (Table 1). Along was a mandated freshwater release schedule from a dam this gradient, rainfall decreases by a factor of two, but and the other was a diversion of river water to an estuary. inflow balance decreases by almost two orders of magni- The focus in this manuscript is on the infaunal mollusks. tude. The inflow patterns appear to group into four distinct types of estuaries that vary by about an order of magnitude each (Table 1). Each estuary-type also has distinctly differ- METHODS ent timing of peak inflow events. The northern estuaries receive peak inflow during the spring, the central estuaries All seven Texas estuaries have similarities in their are bimodal receiving peak inflows during the spring and structure and physiography (Fig. 1). Barrier islands are fall, and the southernmost estuaries receive peak inflows parallel to the mainland along most of the coast. Between during the fall (Texas Department of Water Resources, the islands and the mainland there are lagoons. The American Malacological Bulletin, Vol. 11(2) (1995):163-175 163 164 AMER. MALAC. BULL. 11(2) (1995) Colorado - River “7 . Tres Palacios River .\-.* Lavaca River Matagorda Bay Lavaca Bay Pass Cavallo Guadalupe : -. sit, River <4: San Antonio ~.- River -' , - one San Antonio Bay Gulf of Mexico Aransas River :-* dj) °° erin as Aransas Pass =f Nueces Bay |. Corpus Christi Bay Laguna Madre Baffin Bay Fig. 1. Location of south Texas estuaries and sampling stations. Table 1. Gradients in Texas estuaries. Listed from north to south: area at mean low tide (Diener, 1975), mean annual rainfall (1951-1980; Larkin and Bomar, 1983), mean annual freshwater inflow balance (1941-1976; Texas Department of Water Resources, 1982), and mean annual commer- cial harvest (1962-1987; Texas Parks and Wildlife Department, 1988). Commercial Harvest Estuary Area Rainfall Inflow Finfish Shellfish (km2) (cm/yr) (106 m3/yr) (103 kg/yr) (103 kg/yr) Sabine-Neches 183 142 16,107 5 332 Trinity-San Jacinto 1,416 112 12,284 190 4,060 Lavaca-Colorado 1,158 102 3,242 100 2,076 Guadalupe 551 91 2,545 80 1,545 Mission-Aransas 453 81 190 207 1,453 Nueces 433 76 509 151 544 Laguna Madre 1,139 69 -947 834 147 lagoons are interrupted by drowned river valleys that form the bay and estuarine systems. There are Gulf inlets through the barrier islands, which connect the sea with the lagoon behind the island. The lagoon opens to a large pri- mary bay. There is a constriction between the primary bay and the smaller secondary bay. The river flows into the secondary bay. Primary bays have greater marine influence and secondary bays have greater freshwater influence. So, as well as a latitudinal climatic gradient, there is a longitu- dinal salinity gradient within each estuary. The similarity of the Texas estuaries allowed us to design a sampling program where we could use statistical control on confounding factors, e.g. Gulf exchange, circula- tion patterns, and alterations by humans. Four to six sta- tions were chosen in each estuary (Table 2, Fig. 1) employ- ing the same spatial sampling design that has been employed in previous studies of Texas estuaries (Montagna and Kalke, 1992). Two replicate stations (A and B) were in the secondary bay where freshwater influences are greatest. Two other replicate stations (C and D) were in the primary bay where marine influences are greatest. By using two stations in the freshwater-influenced zone and two stations in the marine-influenced zone, we were replicating effects at the treatment level and avoiding pseudoreplication (Hurlbert, 1984). There has been a diversion of the Colorado River into the east arm of Matagorda Bay, so we located two additional stations (E and F) there. The sta- tions in Laguna Madre are located using a similar strategy. Two stations were located in Baffin Bay (6 and 24), and two stations were located in Laguna Madre in a seagrass bed (189G) and an unvegetated sand patch (189S) (Fig. 1). Two major water projects were initiated during the course of this study. The purpose of both projects was to increase freshwater inflows to bays to enhance secondary productivity. In 1990, the Texas Water Commission MONTAGNA AND KALKE: TEXAS INFAUNAL MOLLUSCA 165 Table 2. Location of sampling stations and sampling periods. Estuary Bay Name Stations Period Lavaca-Colorado Secondary Lavaca Bay A 1984-1994 Secondary Lavaca Bay B 1988-1994 Primary Matagorda Bay C,D 1988-1994 Diversion East Matagorda Bay E,F 1993-1994 Guadalupe Secondary Upper San Antonio Bay A,B 1987-1994 Primary Lower San Antonio Bay C,D 1987-1994 Nueces Secondary Nueces Bay A,B 1988-1994 Primary Corpus Christi Bay C,D 1988-1994 Primary Corpus Christi Bay E 1990-1994 Laguna Madre _— Secondary Baffin Bay 6,24 1989-1993 Primary Laguna Madre 189G, 1989-1993 189S ordered the City of Corpus Christi to release 151,000 ac- ft/yr (1.86 x 108 m3/yr) to the Nueces Estuary from the Choke Canyon/Lake Corpus Christi reservoir system. The releases were mandated, because the City had not been releasing water. Stations A and B in Nueces Bay were used to assess the effects of this project (Fig. 1). The Colorado River was diverted into the eastern arm of Matagorda Bay by the creation of a flood diversion channel in 1991 and a dam in the river channel below the point of diversion in 1992. This project has diverted Colorado River water from the Gulf of Mexico into the eastern arm of Matagorda Bay. Stations E and F were sampled to assess the effect of this diversion into Matagorda Bay (Fig. 1). The current study is not a complete assessment of the efficacy of these two pro- jects. Replicate sediment samples (3) were taken within a 2 m radius at each of the stations in each estuary four times per year. Abundance and community structure were mea- sured using the standard techniques that we (Montagna and Kalke, 1992) have been using since 1984. This includes sectioning 6.7 cm diameter cores (at 0-3 cm, and 3-10 cm) to examine the vertical distribution of infauna. Animals were then extracted using a 0.5 mm sieve, enumerated, and identified. Taxonomic authorities used were Abbott (1974), Turgeon et al. (1988), and Andrews (1992). Principal com- ponents analysis (PCA) was performed on all data sets to determine the relationship among stations in terms of species composition. Species composition was then pooled by bay within each estuary, and PCA was performed on these data to determine relationships among bays. Hydro- graphic data were recorded at each station using a Hydrolab Surveyor II (Hydrolab, Inc.). These measurements includ- ed: salinity, conductivity, temperature, dissolved oxygen, oxidation-reduction potential, pH, and depth. Salinity is reported as practical salinity units (psu). RESULTS SALINITY REGIMES There were large differences in salinity from year to year in all estuaries (Fig. 2). The years 1985-1986 and 1992-1993 were wet periods with concordant low salinities. These wet periods occur during periods when an El Nino event is occurring in the western Pacific Ocean. The inter- vening time between El Nifio events is dry. Texas suffers through a series of flood and drought periods regulated by global climatic events. There are generally lower salinities in the spring and higher salinities in the summer, because of seasonal inflow and evaporation patterns. Salinity in the Lavaca-Colorado Estuary ranged from 0-36 psu (Fig. 2). The lowest salinities always occurred in the secondary bay at stations A and B. After the diversion of the Colorado River, Stations E and F exhibited low salin- ities that are more typical of the secondary bay. Salinity in the Guadalupe Estuary ranged from 0-32 psu (Fig. 2). During flood periods, this estuary is uniformly low (0-10 psu) in salinity. This is unusual compared to other Texas estuaries. It is caused by the high rate of inflow into a relatively small estuary (Table 1). The high turnover rate and low rate of exchange of marine water with the Gulf of Mexico exacerbate this trend. During extreme flooding the entire estuary can be at or near 0 psu. During drought periods, there can be a gradient of salinity. Salinity in the Nueces Estuary ranged from 2-45 psu 60 50 Salinity 0 | eee ee 1984 1986 1988 1990 1992 1994 1996 Fig. 2. Mean bottom salinity (psu) at all stations during each sampling period. (GE = Guadalupe Estuary; LC = Lavaca-Colorado Estuary; LM = Laguna Madre-Baffin Estuary; NC = Nueces Estuary). 166 AMER. MALAC. BULL. 11(2) (1995) (Fig. 2). Prior to 1991, salinities in the estuary were uni- form and high. In 1991, after a series of mandated fresh- water releases, salinities declined in the secondary bay. Salinities in the secondary bay were much lower than in the primary bay, where they had been similar in 1987-1988. Heavy rain in 1992-1993 reduced salinity further. Salinity in the Laguna-Baffin Estuary ranged from 10-60 psu (Fig. 2). Seasonal fluctuations are less evident in this system. Changes occur system-wide when there are large climatic events, e.g. the 1992-1993 El] Nino event. There is little salinity gradient in this ecosystem, because freshwater inflow and exchange with the Gulf of Mexico is restricted. COMMUNITY STRUCTURE In the Lavaca-Colorado Estuary, stations A and B were almost identical (Fig. 3A). Station F, at the mouth of the river diversion was also similar to A and B. Stations C and E were similar, and both these stations are nearly equi- distant from freshwater input and Gulf exchange. Station D, near the pass, was the most different station of all. The pattern elucidated in the PCA was driven by the greater number of species that were found in station D, near the Gulf pass (Table 3). Also, species dominance patterns were different. The dominant bivalves were from the genus Periploma at stations C and D, whereas Mulinia lateralis A. Lavaca-Colorado Estuary B. Guadalupe Estuary 1.0 1.0 Boo ae 4a F | s 054 0.5 4 z= | o 4 6 00-4 D v: a | | fo} s) 0.5 4 0.5 1.0 bee te. aq 1.0 0.5 0.0 0.5 1.0 -1.0 0.5 0.0 0.5 1.0 C. Nueces Estuary D. Laguna Madre-Baffin Bay Estuary 1.0 .——__ 5 ——_—F7] 1.0 — | e DE 19865 05 4 c 0.5 | g 0.05 0.0 E q fo} rs) os 4 -0.5 4 Be4 re eee -1.0 -0.5 0.0 0.5 1.0 -1.0 -0.5 0.0 0.5 1.0 Component 2 Component 2 Fig 3. Principal components analysis of molluscan communities at each station over the entire study period within estuaries. (Say, 1822) was dominant at stations A, B, E, and F where there is freshwater influence. Gastropod species were more uniformly distributed throughout the estuary. The domi- nant gastropods were Nassarius acutus (Say, 1822) and Acteocina canaliculata (Say, 1826). Bivalves were always dominant over gastropods. Gastropods were most common in Lavaca Bay where they constituted 32% of the popula- tion at station A and 49% at B. In contrast, gastropods rep- resented only 24% in C, 14% in D, 16% in E, and 20% in F. In the Guadalupe Estuary, all of the stations were somewhat alike in terms of community structure (Fig. 3B). There was more of a gradient from stations A to B to C to D in terms of abundance of individual species (Table 3). This was true for the dominant species, e.g. Texadina sphinctostoma (Abbott and Ladd, 1953), Acteocina canalic- ulata, and Mulinia lateralis. The brackish water species, Rangia cuneata (Sowerby, 1831) only occurred in stations A and B. In general, there were more species in the marine end of the estuary where stations C and D are located. However, there were much higher abundances of species in the freshwater end of the estuary where stations A and B are located (Table 3). In spite of these trends, the PCA indicated that there may be more affinity between stations A and C, and stations B and D may be more alike (Fig. 3B). This trend may be explained by the unusual circulation pat- tern in San Antonio Bay. Freshwater enters the estuary near station A, and travels southwest along the shoreline toward station C. Marine water enters the bay near station D and travels north toward station B. The species commu- nity pattern in the PCA is driven by the number of gas- tropods versus the number of bivalves. Gastropods were most common at station A (80% of the population) and sta- tion C (58%). In contrast, gastropods represented only 41% in B and 30% in D. In the Nueces Estuary, stations A and B in Nueces Bay were almost identical (Fig. 3C). Stations D and E were very similar, and station C was somewhat different from all other stations. Stations D and E are nearest the Aransas Pass in Corpus Christi Bay. Station C is in the upper part of Corpus Christi Bay. The pattern elucidated in the PCA was driven by the greater number of species that were found in stations D and E, near the Gulf pass (Table 3), and some species unique to station C. In stations A and B, the dominant species were the bivalves, Mulinia lateralis and Macoma mitchelli Dall, 1895. In contrast, at stations D and E, gastropods were always dominant, where they con- stituted 46% of the population at station D and 33% at E. Gastropods represented only 7% in A and B, and 16% in C. Station C was different from the rest in that the dominant bivalves were from the family Nuculanidae (Table 3). The Laguna Madre-Baffin Bay system exhibited the most varied molluscan communities within an ecosystem (Fig. 3D). This trend was due to the difference caused by 167 : TEXAS INFAUNAL MOLLUSCA MONTAGNA AND KALKE (panunuod) (978] ‘AeS) DIDJNIIJDUDI DUID0aIIY 66 6 0 0 0 €ll 0 9€% 6 9IZ 681 681 $6 p8St 165 681 S6 799 I?E aepupueydeos IIg 6z7¢ 0 0 ely S€Il 0 ) 0 0 0 0 0 0 0 0 0 0 0 (OP8I ‘SwWepYy “g °D) SmvIUsoloUNd sixDIo1Yy depluos}oV vapidseyeydag 0 O 8gIl 681 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 (S881 ‘leqd) vuvassos vjJakvg 0 0 Str Ore BLE 8% 681 0 0 Sol 681 0 0 0 0 66 68th 0 ‘ds snosidakg 0 0 66681 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 (6S8] ‘SPWIOH) MID[nNUas D]JapnuDvAK 0 0 0 0 0 0 0 0 0 0 $6 0 0 0 0 0 0 681 82 ‘ds piuojsopO 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 SSI (6061 ‘YsNg) vsogqis DiUOIsopE 0 0 0 0 0 9% O 0 66 971 0 $6 = L9S 0 681 0 0 0 0 ‘ds pwojspuijng 0 0 0 0 0 0 0 66 0 trl «6S6——(iS—t«CSG $6 +8 O 681 LOE ZHI (S8gI ‘Ysng) sasaj Duojsounjngy 0 0 66 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 (ZZ8I ‘ABS) vssasdiu Dauoog aepry[oprueskg eploljaprueskd 0 0 0 66 0 0 0 0 0 0 0 0 0 0 0 681 0 0 0 (ZZBI ‘ABS) xaqiA SniavssoN 0 0 0 66 66 6S6—Ss«SBT 0 0 0 66 0 0 66 6681) «= 681.~—Sss ETS SSTSC8T (ZZBI ‘AeS) Sninov sniavssoN SePILesseN, 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 66 0 0 (9781 ‘Avg) vivun) DIA 0 0 0 8SI 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ELSI ‘SuUIRAIS DIDIIVdNUAS SIysDUY 0 0 0 6 0 0 6 0 0 0 0 0 0 0 0 0 0 0 0 ZBI ‘AQUamog DjoI01a.4S0 SIyDUY aepl[[aquinjod epodonsesoan 0 66 trl s9z 0 $6 = $6 0 0 66 «S6 0 0 0 0 8% O 0 0 (SSLI ‘guUIT) BIDIIUAOf DjNpIda4D aeproenddeg 0 0 0 0 0 6 0 0 0 0 0 0 0 0 0 0 0 0 0 ‘ds wnruojidy sepriuojdg 0 0 1% €0€ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 BZBI ‘SYUD] wnsomny wniys4aD 0 0 6 = 6SE 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 (OP8I ‘JasJlaJq) wnludA DMOIsDIG aeplyIa 0 0 S6EI 8sIsz 0 0 0 0 0 66 0 66 0 66 0 0 0 0 0 I¢gl ‘uosduing wnjjayojnd unsavg 0 0 0 0 $6 Str 9€% O 0 66 S6 0 0 0 $6 S6 S9I 0 0 8061 ‘AaTxUIM Muosuyol unsav) deploae_) 0 0 0 0 0 799 O 0 0 aa 0 0 0 0 0 0 0 0 0 (paynuapiun) sepl{[auLyt A 0 0 0 0 66 0 0 0 0 0 0 0 0 0 0 0 0 0 0 = 9P6I ‘AIUIQNSW 2 Arqs|ig vunpisopf DjjaulatiA dept [fou A 0 0 66 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 (16LI ‘UlfaWI) 202012 DUL4ONNIT de PIULIO}IT 0 0 0 0 0 0 0 0 0 v€6 I8Lp B8EIS E89rl 0 0 0 0 681 681 (ES6I ‘PPT NOggY) Dwossojouids DUIpDxa I aepliqoipAH epodonsesosayy 66 0 66 ~=:« 1% 66 681 0 0 0 y8t =O 0 8LPrzI 0 0 0 6 S6 6 (paynuaprun) epodonsey epodonsey vz 9 S68I D681 a d ) d Vv d fe) d Vv d q d fe) d Vv saloadg¢/A|IWey /JapiO/SSPD Aeg uljjeg-oipey] eun3ey] Arenysq sooony Arenisq odnjepeny Arenysy Opelojod-eoear’y] “uoneys yore Ie (| a]qRJ, 298) pouad Apis astqua Joy W/laquinu uRa “SaLIeNysa sexe], UI UONINGINSIP uLosN]Jow euNvsU] *¢ a[quL, (penunuod) 0 0 56 $6 0 ) 0 0 0 0 0 0 0 ) 0 $6 60 0 ) 0061 ‘Ted vupxai ula 0 0 56 $6 0 ) 0 0 0 0 0 0 0 ) 0 0 ) 0 0 9981 ‘pelUOD sisuapdwmoy vuiya[ 0 0 0 681 gle 681 Str 0 0 56 60 $6 6 OO 0 0 56 0 8% B8LE ‘ds puna 0 ) ) 0 ) ) 0 0 0 0 0 0 0 0 0 0 0 0 ZBI ‘ZNJIVY VDIS149 DLOPYIAL 0 0 ) 0 681 681 681 0 0 $6 60 0 0 0 0 $6 OO 0 0 (pEgl ‘AeS) Diua; DUODW 0 ) ) 0 0 ) $6 LL6 SOEl 06h rt Ihh I€h 6501 681 9€% $6 82% HOF S681 “eC 1/4491 Duos 0 ) 0 0 0 56 0 0 0 0 0 0 0 0 0 0 0 0 0 (pest ‘Aeg) suodfiaasg Duosv py oeplury [aL ) 0 0 ) 0 681 O 0 0 0 0 0 0 0 0 0 0 ) 0 IZ8I ‘ABg siplaia uajog ) 0 0 0 0 0 0 56 60 clp $6 S6 O 0 0 0 0 0 ose O06T ‘Ile 4ounw sisuq depius[os 0 0 0 0 ) 0 0 0 0 0 0 tI gZs 0 0 0 0 0 ) (TE8I ‘Aqiamog) pivauns viduvy OSII ~@SE SIE 6 S19 IZLT L69 7O6Z LLEI HOI O%6Z 7089 SPI9 £66 Z9IL O9% 6%rhl vrs PIL (7781 ‘AeS) sypD4ajD] DIuINW 0 0 56 6SE 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I6LI ‘UNeWD siis04 DaoDy deploy a ) ) $6 971 ) 0 0 0 0 0 ) 0 0 0 0 0 0 0 0 (OE8I ‘PRIUOD) Mosou wnIpsD21AavT S aepripxeg = ) 0 0 ) 0 0 68! O 0) 0 0 0 0 0 0 0 ) 0 0 (pEgT ‘pelUOD) vIDjnUN] D]JauIssD4D ay aepl{[ayesselc = 0 0 0 0 0 Oo tf S6 0 er9 0 0 0 0 0 $6 S6 Trl 8LE (1ggI ‘uosduing) pinjnunjd njjaskp = sepnnoejuopy = 0 0 0 ) gle soo 0 S66 =O 9% S6 0 0 0 0 56 $6 0 ) 696] ‘AueH Dunispxal Duasyy aepmley os Bplolgua A, Cj 0 0) 0 ) 0 0 0 0 0 0 0 0 0 0 0 0 0 56 0 (T6LI ‘UPWID) DIIWId.IA DausOssD4D 0.05, t-test, N = 10 snails) in pore diameter between the two regions (Fig. 27). The secretory product released from a single pore varied in appearance. Some pores con- tained numerous small, globular “spheres” (Fig. 23), whereas others released long “toothpaste-like” exudates (Figs. 25-26). The latter secretions, when examined under higher magnification, appeared to be composed of smaller spheres approximately | pm in diameter (Fig. 26). These spheres correspond in size to the large granules observed with the light microscope (Figs. 13-20). There were no smaller granular-like secretions directly observed with the SEM. There was also no apparent correlation between the type of exudate released and the region of the gland in which it was found; therefore, it is possible that the appar- ent differences may be due to specimen preparation. However, regardless of the location, all of the secretory products were observed to accumulate in the depression of region II prior to release. In other words, the snail evidently secretes the hypobranchial product directly into the mantle cavity. There was no significant difference (P > 0.05, t-test, N = 10 snails) in the density of pores between regions I and III (Fig. 28). It was impossible to estimate the pore density opening into region II because this area was covered with mucus secretions that could not be completely removed. Some pores and cilia were observed at the margins of region II; however, it could not be accurately discerned whether or not these were part of region II or the extreme margins of regions I and III. After release into the mantle cavity, the secretory product appears to be swept out of the cavity under the anterior margin of the mantle by ciliary action. In several of the dissected specimens, some of the hypobranchial secre- tion was observed being released from under the anterior margin of the mantle (Fig. 4). Additionally, several snails were observed releasing a copious amount of secretion into the seawater while feeding on oysters (Fig. 6). Transmission electron microscopy revealed hypo- branchial cells that exhibited ultrastructural detail charac- teristic of active secretory cells (Fig. 29a). Secretory activi- . BULL. 11(2) (1995) Figs. 8-9. 8. Light micrograph of a cross-sectional view through the man- tle, buccal mass, and foot of Stramonita haemastoma canaliculata. (B, buccal mass; Ct, ctenidia; F, foot; H, hypobranchial gland (region II); Os, osphradium; SG, salivary gland.) 9. Higher magnification of cross-section illustrating regions I, II, and III of the hypobranchial gland of S. haemas- toma canaliculata. (B, buccal mass; M, mantle cavity; R, rectum; SE, dor- sal shell epithelium; SG, salivary gland; V, ventral mantle cavity epitheli- um.) ty was indicated by the presence of numerous Golgi, rough endoplasmic reticulum, mitochondria, and distinct electron- dense granules in the cytoplasm. The nuclei of these cells possessed a prominent nucleolus and significant amounts of euchromatin, indicating active transcription of RNA. The electron-dense vesicles of the cytoplasm exhibited a crys- talline or granulated composition (Fig. 29b) similar to that seen in vesicles of other organisms (Kilejian, 1974). Numerous secretory vesicles were observed separating from the trans-face or fusing with the cis-face of the Golgi ROLLER ET AL.: HYPOBRANCHIAL GLAND OF STRAMONITA HAEMASTOMA 183 Figs. 10-12. 10(a-b). SEM of two cross-sections of hypobranchial gland through regions I (a), II (a), and III (b). (R, rectum; Va, vas deferens.) 11. SEM of ventral surface of the hypobranchial gland of Stramonita haemastoma canaliculata illustrating the three distinct regions (I, I], and II). 12. SEM of ventral surface of hypobranchial gland tilted to emphasize the depression (arrows) located in region II. (M, mucus.) (Fig. 29c) (Stevens and Lowe, 1992; Cormack, 1993; Fawcett, 1994). The fate of these vesicles is unknown; how- ever, it is probable that they may ultimately be released from the cell. TEM also confirmed that the cilia covering the ventral surface of the hypobranchial gland possess the typical 9 x 2 + 2 arrangement of microtubules (Fig. 29d). DISCUSSION The hypobranchial gland of Stramonita haemas- toma canaliculata is composed of eight distinct types of cells, some of which evidently release secretions through pores directly onto the gland’s ciliated ventral surface and thus into the mantle cavity. The cilia create currents respon- sible for drawing fresh seawater in through the siphon, over the osphradium and ctenidia, across the hypobranchial gland, and out through the anterior margin of the mantle (Fig. 2b). Thick ciliation of osphradial and ctenidial lamel- lae in this species (Garton et al., 1984) could serve to gen- erate water currents, and thus ventilate the mantle cavity. In addition to assisting olfaction and respiration, this water circulation would also function in the elimination of the hypobranchial secretion and fecal material from the rectum. Bolognani-Fantin and Ottaviani (1981) used light microscopy and histochemistry to develop a clear concept of the histological organization of the hypobranchial gland in Murex brandaris Linné, 1758. They described seven dis- tinct cell types: (1) ciliated cells, (2) empty cells, (3) cells with secretion in big “granula” or irregularly oval-shaped masses, (4) cells with secretion in fine “granula,” (5) cells without secretion granula (homogeneous cytoplasm), (6) mucocytes, and (7) ’purple-producing” cells. The results of 184 AMER. MALAC. BULL. 11(2) (1995) the present investigation reveal similar findings for the hypobranchial gland of Stramonita haemastoma canalicu- lata. The ciliated cells, empty cells, cells containing large granules, cells containing small granules, homogeneous cells, and mucocytes identified in S. haemastoma canalicu- lata are similar to those identified by Bolognani-Fantin and Ottaviani (1981) in M. brandaris. It is possible that cells they labeled as “purple-producing” are the same or similar to cells that we have identified as dark-staining basophilic cells; however, they showed no micrographs of these pur- ple-producing cells, and therefore comparison is impossible without further information. We do agree with their sugges- = y ~y = ~ 4 tion that the empty cells represent cells from which the secretory product had been washed out during histological preparation. Srilakshmi (1991) reported neurosensory cells in the hypobranchial gland of Morula granulata (Duclos, 1832). However, his one micrograph illustrated only “gland cells, clear cells, eosinophilic cells, and mucous cells” and no identification of these purported neurosensory cells. Additionally, the micrograph from that study identified a structure that was labeled as the “hypobranchial nerve” located dorsal to the epithelium. The structure in question does not resemble a nerve but rather more closely resem- bles the dorsal blood vessel containing hemocytes identi- Figs. 13-16. 13. Light micrograph of a cross-section of region I of the hypobranchial gland illustrating the dorsal (D) and ventral (V) mantle cavity epitheli- um, as well as secretory material (arrow) released into mantle cavity (Mc). (E, empty cell; H, hemocytes in blood vessel.) 14. Light micrograph of region I epithelium illustrating the release of secretory material through pores (P) by large-granulated (G) eosinophilic cells. (Mc, mantle cavity.) 15. Light micro- graph of region III epithelium illustrating the exocytosis of mucus (M) and the release of other secretory material. (G, large-granules released by eosinophilic cells; Mc, mantle cavity.) 16. Light micrograph of cells with homogeneous cytoplasm (H) from region I epithelium interspersed between large-granulated eosinophilic cells (G). (E, empty cell; Mc, mantle cavity.) ROLLER ET AL.: HYPOBRANCHIAL GLAND OF STRAMONITA HAEMASTOMA 185 Figs. 17-20. 17. Light micrograph of small basophilic cells from region I epithelium, containing homogeneous material (arrows). (M, mucocyte; Mc, mantle cavity.) 18. Light micrograph of ventral mantle cavity epithelium illustrating the release of secretory product into the mantle cavity (Mc) by small-granulat- ed (G), eosinophilic cells. (C, ciliated cells; M, mucocytes.) 19. Light micrograph of small, non-granulated eosinophilic cells (E) of ventral mantle cavity epithelium. (M, mucocyte; Mc, mantle cavity.) 20. Light micrograph of sectioned dorsal shell epithelium (SE) and dark staining cells (#*) overlying region II. These pigmented cells presumably contribute to the dark coloration overlying region II which can be observed from the dorsal aspect (Fig. 3). fied in the present study (Fig. 13). Muricid gastropods are known from antiquity for their production of a brilliant purple pigment, “Tyrian Purple” (6,6’-dibromoindigotin) (Sadler, 1956; Verhecken, 1989, 1990; Hoffmann, 1990; Michel et al., 1992), which was used in the dye process for the robes of Roman royalty and high-ranking officials. The secretions that give rise to this pigment are released by the hypobranchial gland (pre- sent investigation) and undergo an enzyme-catalyzed, photo-oxidative reaction (Sadler, 1956; Hoffmann, 1990; Michel et al., 1992) involving the following color change: clear — yellow — green — blue — purple (Verhecken, 1989, 1990). In the present investigation the pigment color change had progressed to a white-yellow phase by the time the snails were dissected, which required only 30-60 sec after the shells were cracked. The remaining color change to a brilliant purple required an additional 30-90 min depending on the light intensity (i.e. a fiber-optic light source stimulated fastest color change) (Figs. 7a-d). St. Amant (1938) identified the hypobranchial gland as the “rectal gland” but stated that the structure was not glandular in appearance. However, he offered no histologi- cal evidence of this in his thesis. Later studies (Ottaviani, 1978; Bolognani-Fantin and Ottaviani, 1981; Srilakshmi, 1991; present investigation) have shown that this structure is glandular in nature. Fretter and Graham (1962:23) report- 186 AMER. MALAC. BULL. 11(2) (1995) ont SESW SS Figs. 21-26. 21. SEM of ventral surface of region III of the hypobranchial gland of Stramonita haemastoma canaliculata illustrating presence of globular secretory material (arrows) on surface. 22. SEM illustrating the demarcation between the ventral surface of region I of the hypobranchial gland and the pos- terior mantle surface. (Arrow, secretion being released from pores.) 23. SEM of the ventral surface of region I of the hypobranchial gland illustrating the dense cilia, and secretion of material (S) through the pores. 24. SEM of ventral surface of region III of the hypobranchial gland illustrating a prominent pore (P) and dense cilia (C). 25. SEM illustrating secretion (S) being released from the pores on the ventral surface of the hypobranchial gland (region I). (C, cilia.) 26. SEM illustrating globular composition (arrow) of secretory material being released from a pore in region III of the hypobranchial gland. (C, cilia.) ROLLER ET AL.: HYPOBRANCHIAL GLAND OF STRAMONITA HAEMASTOMA 187 T=0.339 P=0.739 (18 df.) PORE DIAMETER (yum) | Hl HYPOBRANCHIAL GLAND REGION Fig. 27. Pore diameter (ym) (mean + S.E.) of regions I and III of the hypobranchial gland of Stramonita haemastoma canaliculata. There was no significant difference in diameters of pores between the two regions (P > 0.05). (d.f., degrees of freedom; T, t value; P, probability value.) ed that gastropods and the protobranchiate bivalves are the only two groups of mollusks in which a well-developed hypobranchial gland occurs, and that it functions in the pro- duction of a mucus secretion for “trapping and cementing particulate matter sucked into the mantle cavity in the respi- ratory water current, prior to its expulsion on the right.” Secretion of mucoid material by mantle cells is common in many mollusks (Fretter and Graham, 1962; Prezant, 1979), and mucus-producing cells have been identified in the hypobranchial glands of other prosobranchs (Ottaviani, 1978; Bolognani-Fantin and Ottaviani, 1981; Srilakshmi, 1991). It is apparent that this mucus production must, in part, also be an important function of the hypobranchial gland in Stramonita haemastoma canaliculata because mucocytes were observed in this study (Figs. 15-19), and copious amounts of clear mucus were observed when the mantle cavity was exposed (Figs. 5, 12). Additionally, the yellow secretion (which gives rise to the Tyrian Purple dye) observed in region II is viscous and presumably composed in part of mucus secretions. The dark-staining basophilic cells located dorsal to region II (Fig. 20) possibly comprise the rectal gland (Kool, pers. comm.), as they appear glandular in nature; however, we observed no direct connection between these cells and the rectum (Figs. 3b, 8-9). It is possible that the connection was missed in histological sectioning and that the structure is indeed the rectal gland described previously (St. Amant, 1938; Fretter and Graham, 1962). Previous histochemical and pharmacological inves- tigations attempted to reveal the structure, function, and secretions of the hypobranchial gland in gastropods. Some of these studies reported the release of paralytic substances including physiologically-active choline esters by the hypo- branchial glands and salivary glands in several species of muricids (Keyl et al., 1957; Whittaker, 1960; Roseghini et al., 1970; Ottaviani, 1978; Bolognani-Fantin and Ottaviani, 1981; Srilakshmi, 1991) including Thais (= Stramonita) haemastoma (Linné, 1767) (Huang and Mir, 1971, 1972; Roseghini, 1971). Presumably, these putative toxins would be used in the feeding process after a snail has drilled a borehole into the shell of its prey (an oyster for example) (Carriker, 1981; Roller et al., 1984). The choline esters would then be released into the mantle cavity of the prey through the borehole, thus paralyzing the adductor muscles, and allowing the snail to wedge the valves of the oyster open, exposing the soft flesh inside. Roseghini (1971) reported the occurrence of large amounts of urocanyl- choline (i.e. murexine) and imidazolepropionylcholine (dihydromurexine) in the hypobranchial gland of T. 1400 T=1.090 P=0.290 (18 df.) m —_ NO oO oO = je) oO (o) 800 600 NUMBER OF PORES/ sq. | Hl HYPOBRANCHIAL GLAND REGION Fig. 28. Pore density (number/mm2) (mean + S.E.) of regions I and III of the hypobranchial gland of Stramonita haemastoma canaliculata. There was no significant difference in pore density between the two regions (P > 0.05). (d.f., degrees of freedom; T, t value; P, probability value.) 188 AMER. MALAC. BULL. 11(2) (1995) Fig. 29. a. TEM of active secretory cell of region I of the hypobranchial gland of Stramonita haemastoma canaliculata. (A, artifact (lead precipitate); E, euchromatin; G, Golgi; H, heterochromatin; M, mitochondria; N, nucleolus; V, electron-dense vesicle; arrows, rough endoplasmic reticulum.) b. TEM of secretory vesicles illustrating granular composition (arrow). c. TEM illustrating secretory vesicles (arrows) associated with Golgi (G) in the hypobranchial cell of S. haemastoma canaliculata. (M, mitochondria.) d. TEM of cross-section of cilia on the ventral surface of the hypobranchial gland, illustrating typi- cal 9 x 2 + 2 arrangement of microtubules. (Purpura) haemastoma. Huang and Mir (1971) examined the pharmacological properties of the hypobranchial gland of T: haemastoma and the effect of the secretions on verte- brate tissues. They reported marked increase in blood pres- sure accompanied by tachycardia in anesthetized cats and an LDso of 215 mg/kg in mice. They also reported increased contractions of various muscle tissues from other mammals. Most predator toxins are specialized for a partic- ular prey species, or group of species. While these previous pharmacological studies provide valuable information, they did not involve the normal prey species of the oyster drill. In a separate series of experiments we exposed indi- vidual bivalves to the combined hypobranchial secretions from four snails with no significant effect relative to untreated controls (P > 0.05; N = 10). In all instances the bivalves (Rangia cuneata and Crassostrea virginica) failed to gape appreciably and survived the treatment with no apparent harm. Injection of the hypobranchial extract directly into the mantle cavity of these bivalves (with syringe/needle) likewise had no effect. Urocanic acid, a histidine derivative, is a precursor to the presumed toxin urocanylcholine (Roseghini et al., ROLLER ET AL.: HYPOBRANCHIAL GLAND OF STRAMONITA HAEMASTOMA 189 1970). Numerous histidine-rich granules have been previ- ously identified in the malarial parasite Plasmodium lophu- rae Coggeshall, 1938 (Kilejian, 1974), and these granules bear a striking structural resemblance, when examined by TEM, to the larger hypobranchial cell granules identified in the present study. But the apparent lack of toxicity of the hypobranchial secretions from Stramonita haemastoma canaliculata on bivalves (present investigation) seems to negate a possible toxin-producing role. Whether or not the hypobranchial gland functions in toxin production is still unresolved. We suggest that the hypobranchial gland is probably not a “poison gland,” based on the following histological evidence: (1) We were unable to find any duct connecting the gland to the buccal mass, salivary glands, or rectum (Figs. 8-9). Such a duct would be conspicuous, having to pass through the mantle cavity to reach either the buccal mass (i.e. proboscis) or salivary glands. Therefore, the snail apparently has no way to “inject” the toxin through the borehole and into the man- tle cavity of the prey. (2) SEM analysis revealed the pres- ence of pores and dense cilia on the ventral surface of the gland, as well as the actual secretion of material through these pores into the mantle cavity (Figs. 21-26). Therefore, the drill evidently secretes the hypobranchial secretion directly onto “itself” and then flushes it out of the anterior Opening into the mantle cavity (again, not an optimal strate- gy for a poison-producing gland). (3) Snails of this species have been observed to cannibalize each other in our respec- tive laboratories, and one instance of “auto-drilling” has previously been reported (Prezant, 1983). The question of how this toxin affects conspecifics without adversely harm- ing the snail releasing the secretion is still unknown; this fact also makes the hypobranchial gland an unlikely candi- date as the producer of the paralytic secretion. Huang and Mir (1972) also questioned the lack of a hypobranchial duct, but offered no histological evidence for any analogous structure. They did report a more pronounced physiological effect of salivary gland extracts from Thais haemastoma on cats and mice (LDso = 43 mg/kg) than what they had previ- ously observed (Huang and Mir, 1971) with the hypo- branchial extracts. They therefore proposed the salivary gland as a more likely toxin-producing gland. This hypoth- esis agrees with the present investigation and with reported toxin-producing glands in other species of venomous snails such as Conus sp. (Kohn, 1959, Kohn et al., 1960; Songdahl, 1973). The colored secretion(s) of the hypobranchial gland of Stramonita haemastoma canaliculata apparently con- tains the same or very similar compounds to the Tyrian Purple of antiquity. Modern oyster drills occasionally release this dye when feeding (Fig. 6); however, this has not always been observed during the feeding process and in many instances no trace of the dye was observed during feeding or on the empty bivalve shell. It is doubtful that the dye functions in a similar “visual” manner as the “ink” of cephalopods, because it is rarely released in the copious amounts shown in Fig. 6. It is possible that the function(s) of the hypobranchial secretion can involve an olfactory response during the feeding process. Because the secretions possess such a strong odor after the photo-oxidative reac- tions are completed, perhaps the secretion masks the “odor” of the prey once it has been opened. Or it could actively repel or deter other opportunistic marine organisms (e.g. blue crabs, fish) from dislodging the snail from the oyster and usurping its meal. This would be a highly adaptive strategy for the snail, because it has expended considerable energy opening the prey. Therefore, the function(s) of this dye, and other hypobranchial secretions (i.e. choline esters) in predation, and the significance of the dramatic color change are still presently unknown and require further research. In conclusion, this investigation has revealed the general structure of the hypobranchial gland of Stramonita haemastoma canaliculata and has identified the pathway for the release of the secretory product into the mantle cavi- ty of the snail. Based on previous studies and the present investigation, it seems likely that the combined secretion of the hypobranchial gland is composed of several substances each with a potentially distinct function. Further histochem- ical, biochemical, toxological, and behavioral studies are required to accurately determine the functions of the gland in this species. ACKNOWLEDGMENTS We thank Silvard Kool, John Sullivan, Tom Bianchi, Earl Weidner, and Richard Harrel for their comments and suggestions during this study. T. F. Roller provided invaluable assistance with the illustra- tions. Appreciation is extended to the Louisiana Universities Marine Consortium (LUMCON) for the use of its facilities. The following stu- dents assisted during portions of the study: Robert Zimmerman, James Chamberlain, and Lisa Wnuk. We also thank M. Socolofsky for allowing us the use of LSU’s Life Sciences Microscopy Facility. This research was funded in part through grants from Lamar University’s Research Enhancement Program to RAR. LITERATURE CITED Abbott, R. T. 1974. American Seashells, 2nd ed. Van Nostrand Reinhold Company, New York. 663 pp. Bolognani-Fantin, A. M. and E. Ottaviani. 1981. The hypobranchial gland of some Prosobranchia (Mollusca: Gastropoda) living in differen- habitats: a comparative histochemical study. Monitore Zoologico Italiano 15:63-76. Brown, K. M. and T. D. Richardson. 1987. The effect of predator size and 190 AMER. MALAC. BULL. 11(2) (1995) density on the foraging ecology of the southern oyster drill Thais haemastoma. Journal of Experimental Marine Biology and Ecology 114:123-141. Burkenroad, M. D. 1931. Notes on the Louisiana conch, Thais haemas- toma Linn., in its relation to the oyster Ostrea virginica. Ecology 12:656-664. Carriker, M. R. 1981. Shell penetration and feeding by naticacean and muricacean predatory gastropods: a synthesis. Malacologia 20(2):403-422. Cormack, D. H. 1993. Essential Histology. J. B. Lippincott Company, Philadelphia. 430 pp. Fawcett, D. W. 1994. Bloom and Fawcett, A Textbook of Histology, 12th ed. Chapman and Hall, New York. 964 pp. Fretter, V. and A. Graham. 1962. British Prosobranch Molluscs. Ray Society, London. 755 pp. Flegler, S. L., J. W. Heckman, and K. L. Klomparens. 1993. Scanning and Transmission Electron Microscopy - An Introduction. W. H. Freeman and Company, New York. 225 pp. Garton, D. W. and W. B. Stickle. 1980. Effects of salinity and temperature on the predation rate of Thais haemastoma on Crassostrea vir- ginica spat. Biological Bulletin 158:49-57. Garton, D. W., R. A. Roller, and J. Caprio. 1984. Fine structure and vital staining of the osphradium of the southern oyster drill, Thais haemastoma canaliculata (Gray) (Prosobranchia: Muricidae). Biological Bulletin 167:310-321. Hoffmann, R. 1990. Blue as the sea. American Scientist 78:308-309. Huang, C. L. and G. N. Mir. 1971. Pharmacological properties of hypo- branchial gland of Thais haemastoma (Clench). Journal of Pharmaceutical Science 60(12):1842-1846. Huang, C. L. and G. N. Mir. 1972. Pharmacological investigation of sali- vary gland of Thais haemastoma (Clench). Toxicon 10:111-117. Humason, G. L. 1972. Animal Tissue Techniques. W. H. Freeman and Company, San Francisco. 641 pp. Hyman, L. H. 1967. The Invertebrates: Vol. 6, Mollusca I. McGraw-Hill, New York. 792 pp. Keyl, M. J., I. A. Michaelson, and V. P. Whittaker. 1957. Physiologically active choline esters in certain marine gastropods and other inver- tebrates. Journal of Physiology 139:434-454. Kilejian, A. 1974. A unique histidine-rich polypeptide from the malarial parasite, Plasmodium lophurae. The Journal of Biological Chemistry 249(14):4650-4655. Kohn, A. J. 1959. The ecology of Conus in Hawaii. Ecological Monographs 29:47-90. Kohn, A. J., P. R. Saunders, and S. Weiner. 1960. Preliminary studies on the venom of the marine snail Conus. Annals of the New York Academy of Sciences 90:706-725. Kool, S. P. 1986. Systematic revision of thaidid genera based on anatomy. American Malacological Bulletin 4(2):233. Kool, S. P. 1987. Significance of radular characters in reconstruction of thaidid phylogeny (Neogastropoda: Muricacea). The Nautilus 101(3):117-132. Michel, R. H., J. Lazar, and P. E. McGovern. 1992. The chemical compo- sition of the indigoid dyes derived from the hypobranchial glan- dular secretions of Murex molluscs. Journal of the Society of Dyers and Colourists 108:145-150. Ottaviani, E. 1978. A gland of the hypobranchial type in the mantle of Paludina vivipara Draparnaud (Gastropoda: Prosobranchia): a morphological and histochemical study. Monitore Zoologico Italiano 12:77-86. Pollard, J. F. 1973. Experiments to re-establish historical oyster seed grounds and to control the southern oyster drill. Louisiana Wildlife and Fisheries Commision Technical Bulletin 6, 82 pp. Postek, M. T., K. S. Howard, A. H. Johnson, and K. L. McMichael. 1980. Scanning Electron Microscopy, A Student’s Handbook. Ladd Research Industries, Burlington, Vermont. 305 pp. Prezant, R. S. 1979. The structure and function of the radial mantle glands of Lyonsia hyalina (Bivalvia: Anomalodesmata). Journal of the Zoological Society of London 187:505-516. Prezant, R. S. 1983. Auto-drilling in the oyster drill Thais haemastoma (Muricidae). The Nautilus 97(2):81-84. Roller, R. A., D. W. Garton, and W. B. Stickle. 1984. Regeneration of the proboscis, radula and odontophoral cartilage of the southern oys- ter drill Thais haemastoma canaliculata (Gray) (Prosobranchia: Muricidae) after amputation. American Malacological Bulletin 2:63-73. Roller, R. A. and W. B. Stickle. 1988. Intracapsular development of Thais haemastoma canaliculata (Gray) (Prosobranchia: Muricidae) under laboratory conditions. American Malacological Bulletin 6(2): 189-197. Roller, R. A. and W. B. Stickle. 1989. Temperature and salinity effects on the intracapsular development, metabolic rates, and survival to hatching of Thais haemastoma canaliculata (Gray) (Prosobranchia: Muricidae) under laboratory conditions. Journal of Experimental Marine Biology and Ecology 125:235-251. Roseghini, M. 1971. Occurrence of dihydromurexine (imidazolepropi- onylcholine) in the hypobranchial gland of Thais (Purpura) haemastoma. Experientia 27(9): 1008-1009. Roseghini, M., V. Erspamer, L. Ramorino, and J. E. Gutierrez. 1970. Choline esters, their precursors and metabolites in the hypo- branchial gland of prosobranchiate molluscs, Concholepas conc- holepas and Thais chocolata. European Journal of Biochemistry 12:468-473. Sadler, P. W. 1956. Absorption spectra of indigoid dyes. Contributions from the Courtauld Institute of Biochemistry 21:316-318. Songdahl, J. H. 1973. The venom and venom apparatus of the Atlantic cone, Conus spurius atlanticus (Clench). Bulletin of Marine Science 23(3):600-612. Srilakshmi, G. 1991. The hypobranchial gland in Morula granulata (Gastropoda: Prosobranchia). Journal of the Marine Biological Association of the United Kingdom 71:623-634. St. Amant, L. P. 1938. Studies on the biology of the Louisiana oyster drill Thais floridana haysae Clench. Master’s Thesis, Louisiana State University and Agricultural and Mechanical College, Baton Rouge. 116 pp. St. Amant, L. S. 1957. The southem oyster drill. Louisiana Wildlife and Fisheries Commission, Seventh Biannual Report, 1956-57:81-85. Stevens, A. and J. Lowe. 1992. Histology. Glower Medical Publishing, New York. 378 pp. Verhecken, A. 1989. The indole pigments of Mollusca. Annals of the Royal Zoological Society of Belgium 119:181-197. Verhecken, A. 1990. Experiences with mollusc purple. La Conchiglia 22(1):32-46. Webb, R. S. and A. S. M. Saleuddin. 1977. Role of enzymes in the mecha- nism of shell penetration by the muricid gastropod Thais lapillus (L.). Canadian Journal of Zoology 55(11):1846-1857. Whittaker, V. P. 1960. Pharmacologically active choline esters in marine gastropods. Annals of the New York Academy of Science 90:695- 705. Wischnitzer, S. 1981. Introduction to Electron Microscopy, 3rd ed. Pergamon Press, New York. 405 pp. Date of manuscript acceptance: 31 March 1995 The estuarine clam Rangia cuneata (Gray) as a biomonitor of heavy metals under laboratory and field conditions Marc A. McConnell* and Richard C. Harrel** Department of Biology, P.O. Box 100037, Lamar University, Beaumont, Texas 77710 U.S. A. Abstract. Accumulation of copper, chromium, cadmium, and lead by Rangia cuneata (Gray, 1831) was investigated under laboratory and field condi- tions. Bioaccumulation rates and concentrations of metals in tissues and the exposure water were determined in the laboratory for calculation of bioconcen- tration factors (BCF). The BCF ratios ranged between 0 and 422 with significant correlations between exposure time and tissue metal concentration for each metal. Field exposures were conducted below two industrial outfalls that contained some of the metals. High concentrations of the metals were found in the sediments, low concentrations in the water, and intermediate concentrations in the tissues of R. cuneata. No significant relationships were found between exposure time and tissue concentrations during the field exposures. However, in most cases a significant increase in tissue burden of the metals occurred after the 40 d exposures, indicating that R. cuneata accumulated the most bioavailable forms of the metals under natural conditions. In addition, laboratory- retained clams demonstrated that gametogenesis could be inhibited, thus avoiding seasonal variations in body weight and percent gonadal biomass that affect survival and metal uptake. Concentrations of toxins in the tissues of aquatic organisms, particularly heavy metals and refractory organic compounds, are widely used to indicate contamination of aquatic systems. This method is especially useful when substances are present intermittently or in very low concen- trations. The analysis of water samples will determine the concentration of an analyte only at the time of sampling, whereas bioaccumulators reflect concentrations over a longer time period (Mason, 1991). Bioaccumulators can also bring levels within detection limits of most instruments and operators (National Research Council, 1980). The use of bivalves as biomonitors of contaminants in aquatic systems is well documented, e.g. Ellis, 1937; Bedford et al., 1968; Mathis and Cummings, 1973; Lord et al., 1975; Phillips, 1976, 1977; Manly and George, 1977; Curry, 1978; Foster and Bates, 1978; Wright, 1978; Heit et al., 1980; Leard et al., 1980; Imlay, 1982; Farrington et al., 1987; Wade et al., 1988a; Lakshmanan and Nabisan, 1989; Naimo et al., 1992). Since the organization of the International Mussel Watch program in 1978, species of the genus Mytilus have been used as standard biomonitors in temperate waters of the Pacific and Atlantic coasts (Stephenson et al., 1980; NOAA, 1989). Several species of *Present Address: Department of Preventive Medicine and Community Health, Division of Environmental Toxicology, University of Texas Medical Branch at Galveston, 301 University Boulevard, Galveston, Texas 77555-1110 U.S. A. ** Author to whom reprint requests should be sent. oysters have been used in warm waters of the southern Atlantic and the Gulf of Mexico (Wade et al., 1988b; NOAA, 1989). However, no suitable species has been iden- tified as a biomonitor of lower rivers and low salinity estu- aries, which are often heavily industrialized and populated (NOAA, 1989). The estuarine clam Rangia cuneata (Gray, 1831) fulfills many of the criteria considered necessary for bio- monitor species as proposed by Neff et al. (1976), the National Research Council (1980), Mason (1991), Havlik and Marking (1987), and Farrington et al. (1987). R. cuneata is a true estuarine species and occurs naturally where salinity varies between 0-18 ppt (LaSalle and de la Cruz, 1985). It occurs in riverine, bay, and marsh lake estu- aries from Campeche, Mexico, to Florida along the Gulf of Mexico and from Florida to Maryland along the Atlantic coast (Gallagher and Wells, 1969; Hopkins and Andrews, 1970; Hopkins et al., 1973). Salinity is limiting only dur- ing the planktonic larval stage and adult R. cuneata can be transplanted into freshwater or higher salinity water (Hopkins et al., 1973; Cain, 1975). R. cuneata has a long life span and shell length can be related to age (Fairbanks, 1963; Wolfe and Petteway, 1968; Gooch, 1971; Hopkins et al., 1973), allowing a reasonable estimate of exposure time. It is a non-selective filter feeder and ingests large quantities of detritus and plankton and is eaten by many consumers that are eaten by humans (Gooch, 1971; Hopkins ef ail., 1973; LaSalle and de la Cruz, 1985). Another desirable characteristic of R. cuneata as a biomonitor is a wide toler- American Malacological Bulletin, Vol. 11(2) (1995):191-201 19] 192 AMER. MALAC. BULL. 11(2) (1995) ance for many environmental factors, including temperature (Gallagher and Wells, 1969), substratum type (Tenore et al., 1968; Tarver, 1972; Swingle and Bland, 1974), oxygen concentration (Chen and Awapara, 1969; Harrel et al., 1976; Harrel and Hall, 1991), certain heavy metals (Olsen and Harrel, 1973; USEPA, 1984a, b; Eisler, 1988), petrole- um-derived aromatic hydrocarbons (Neff et al., 1976), cer- tain insecticides and herbicides (Chaiyarach et al., 1975; Lunsford and Blem, 1982), and polychlorinated dioxins and furans (Harrel and McConnell, 1995). The purpose of this research was to investigate the use of Rangia cuneata as a biomonitor of the heavy metals copper, chromium, cadmium, and lead by determining accumulation rates and concentrations under laboratory and field conditions. Evidence from the literature indicates that the uptake and loss of pollutants, including metals, by bivalves is influenced by the reproductive phase of the test organism (Phillips, 1976; National Research Council, 1980; Simpson, 1979; Van Hattum et al., 1991). Therefore, the reproductive state and percent gonadal biomass were deter- mined throughout this study in order to relate survival of test animals and differences in accumulation of metals to reproductive phase. A detailed account of this study is the subject of another paper, but some of the data will be used to explain the results of this study. Neff et al. (1976), Lunsford and Blem (1982), and Harrel and McConnell (1995) have shown that R. cuneata is an effective bioaccu- mulator of certain organic compounds. This study will fur- nish information needed to determine if R. cuneata is an effective accumulator of metals, and could serve as a stan- dard biomonitor species for inland coastal waters and estu- aries along the southern Atlantic and the Gulf of Mexico. MATERIALS AND METHODS All clams used in this study were collected from Fence Lake, Jefferson County, Texas. This coastal marsh lake is located between the Intracoastal Waterway and the Gulf of Mexico, and has no known history of environmen- tal contamination. Only clams between 40-60 mm (4-6 yrs old) were utilized to minimize size or age differences. The clams were transported in Fence Lake water to the laborato- ry where the water temperature was allowed to adjust to 16°C to inhibit gametogenesis (Hopkins et al., 1973). They were then transferred to aquaria containing aerated, recon- stituted (NaHCO3, CaSO4*H20, MgSOu, KCl) reagent grade II water (Peltier and Weber, 1985) for 48 hrs of accli- mation and depuration before being used in laboratory or field exposure experiments. Laboratory exposures were conducted between April 1991 and August 1992. Between 75-100 clams were placed in glass aquaria in 40 | of reconstituted reagent grade II water and one of the metal salts. Concentrations for chromium and copper were based on an approximate 5% solution of the 96-hr LCso determined by Olsen and Harrel (1973). The Texas Natural Resource Conservation Commission considers this to be a safe concentration of a persistent toxin in receiving waters. Cadmium and lead concentrations were based on the USEPA’s (1984a, b) crite- ria for safe concentration of specific toxic materials for freshwater exposure and calculated as follows: Lead = (1.273[In hardness]-1.460); cadmium = (1.128[In hard- ness]-1.674). The metals were delivered in the following forms and target concentrations; copper (26 pg/l) as copper sulfate, chromium (19 pg/l) as potassium dichromate, cad- mium (23 pg/l) as cadmium chloride, and lead (32 g/I) as lead acetate. Aquaria containing the same number of clams in reconstituted reagent grade II water without any added metals served as controls. Clams and water were collected after 0, 5, 10, 20, and 40 d exposures for analysis of metals. Depuration rates of lead were determined using 60 clams exposed for 40 d to 35 pg/l lead. These clams were placed in 40 | of lead-free water and after 10, 20, and 40 d the clams were collected for lead analysis. Aeration and circu- lation for all aquaria used in the laboratory study were pro- vided by high volume aerator pumps and diffusion stones. A lab retention study was performed to test the pos- sibility of sustaining Rangia cuneata in captivity over long periods in order to avoid complications associated with reproduction. The clams were collected in January during the spent reproductive phase. From January to September 1992, 40 clams ranging in size from 40-60 mm were main- tained in 45 1 of reconstituted reagent grade II water (salini- ty < 0.5 ppt) at 16°C. The clams were fed finely ground Tetra® fish food weekly. In January, March, June, and September, 8-10 clams were excised from the shell, the whole wet tissue was weighed, and the gonadal tissue was dissected from somatic tissue. The tissue was desiccated at 95°C for 24 hrs and then the total tissue biomass, percent gonadal biomass, and percent water were determined. Accumulation of metals under field conditions was investigated during June-July and October-November 1991 by transplanting laboratory-acclimated clams below two industrial effluent outfalls that contained some of the met- als of interest. Both outfalls were located in the Neches River estuary about 5 km upriver from Sabine Lake, a shal- low inlet off the Gulf of Mexico. Field exposure Site 1 was located along the margin of the Neches River approximate- ly 100 m below the confluence with Star Lake Canal which receives treated petrochemical effluent. At this site 200 clams were placed directly on the bottom in two 1 m2 quadrats at a depth of 1 m. Exposure Site 2 was located in an oil refinery effluent canal approximately 200 m below the outfall and 600 m above the confluence with the Neches River. At Site 2, 200 clams were placed in nylon mesh McCONNELL AND HARREL: METAL UPTAKE IN RANGIA 193 bags and suspended | m below the surface (2 m total depth). At both sites, 16-20 clams, water, and sediments were collected after 0, 5, 10, 20, and 40 d exposures for analysis of metals. Both sites were subjected to tidal influ- ences from Sabine Lake and water temperature, salinity, conductivity, pH, and alkalinity concentrations were mea- sured at the end of each exposure period. Water and sediment samples were collected, pre- served, and digested following USEPA Methods 3020 and 3050 (USEPA, 1986; APHA-AWWA-WPCF, 1989). Clam tissue was removed from the shell and frozen until diges- tion. Digestion was conducted according to USEPA Method 3050 and a modification of Boyer (1984). Metal analyses were performed by atomic absorption spectropho- tometry on either a Perkin-Elmer Model 2100 or a Perkin- Elmer Model 5100 spectrophotometer, both with graphite furnace capability, using USEPA Methods 7000 (USEPA, 1986). A quality assurance project plan was implemented prior to beginning sample analysis according to Verner (1990). The data quality indicators precision and bias were estimated with the use of sample blanks (reagent and proce- dural), as well as spiked and duplicate samples. Data points from the measurements were plotted on quality control charts designed to satisfy a 95% confidence level with an interval of + 3 standard deviation (Verner, 1990). Metal concentrations were expressed as pg/I for aqueous samples and g/kg (wet weight) for tissue and sediments. Bioconcentration factors (BCF), defined as the ratio of the metal concentration in the tissue to the metal concentration in the water, were calculated for all laborato- ry experiments. The BCF ratio is regarded as a valid indi- cator of the capacity of a material to accumulate in animal tissue (Davis and Dobbs, 1984). The BCF was calculated using the following equation: BCF = Can - Cao/Cyw, where Can is the concentration of the metal in the tissue at day n, C,o 1s the concentration in the tissue at day 0, and Cy, is the concentration of the metal in the water. Relationships between exposure time and metal con- centration were determined with the Pearson product- moment correlation coefficient test (Glantz, 1992). Significance of accumulation trends were tested with one- way ANOVA, provided that Bartlett’s tests of homogeneity of variance established the data as being parametric (Keystat, 1985). If the data were non-parametric the Kruskall-Wallis test was used. A one-way ANOVA and summary Statistics were used to determine if significant weight loss occurred in laboratory-retained clams. RESULTS LABORATORY EXPOSURES During all laboratory exposures the physicochemi- cal conditions were relatively uniform; salinity < 0.5 ppt, water temperature 17° + 1°C, pH 7.1-7.3, CaCO3 hardness 41-45 mg/l, and alkalinity 30-34 mg/l. Series 1 copper exposure indicated no accumulation in the tissue at S d and marked accumulation between d 5 and d 40 (Table 1). Exposure medium copper concentra- tions declined approximately 50% by d 5, then remained relatively stable. Bioconcentration factors (BCF) increased from 0 at d 5 to 73 at d 40 with a correlation coefficient (r) between metal concentration and exposure time of 0.82. Background concentrations of copper in the control water ranged from 15-19 pg/I and resulted in accumulation of copper in the control animals, but at a decreased rate and lower final concentration (control mean 1980 pg/kg; expo- sure mean 3434 pg/kg). Copper water lines were responsi- ble for the excessive copper in the control water. Series 2 copper exposure resulted in a steady increase in tissue copper concentration over the 40 d period, but at a decreased rate and lower final concentration than in Series 1 (Table 1). A one-way ANOVA of 40 d tissue cop- per concentration values between Series 1 and 2 indicated no significant difference (p > 0.05). BCF values ranged from 9 at d 5 to 26 at d 40 and the correlation between con- centration and exposure time was r = 0.60. The control group water for Series 2 was also found to contain copper due to exposure to copper water lines (18-23 pg/l) and resulted in an increase in tissue copper concentrations in control animals. However, increases in tissues of the expo- Table 1. Laboratory exposures of Rangia cuneata to copper including exposure water and tissue copper concentrations with corresponding bio- concentration factors (BCF). Tissue means in pg/kg + standard error of the mean. EXPOSURE N WATER TISSUE MEAN BCF (days) (pg/1) (pg/kg) SERIES | (9 April to 19 May 1991) 0 1 42.2 1735.5 - 5 2: 20.6 1700.2 + 223 0 10 2: 30.1 2754.4+54 33.9 20 2. 21.7 2764.1 + 187 47.4 40 1 23.3 3434.3 72.9 Series 2 (10 May to 19 June 1991) 0 1 40.3 1893.2 - 5 2 50.1 2256.1 + 447 9.0 10 3 40.5 2518.3 +90 15.4 20 2 30.9 2541.4+ 10 21.0 40 3 30.5 2684.5 + 44 25.9 Series 3 (16 December 1992 to 5 January 1993) 0 6 23.2 1745.8 + 163 - 5 8 23.0 2030.8 + 60 12.4 10 8 20.9 1995.7 + 133 10.9 20 8 19.4 2165.0 + 136 21.6 194 AMER. MALAC. BULL. 11(2) (1995) sure group were again more rapid than in the control group. No mortalities were observed in either series. Series 3 copper exposure, conducted during December 1992-January 1993, was terminated after 20 d due to mortalities in both the control and exposure groups. The mortalities were believed to be related to a high gonadal biomass burden. Gonadal biomass was 39% of total biomass and could have resulted in the inability to osmoregulate at a salinity < | ppt (unpublished data). Series 3 copper exposure resulted in moderate accumulation from an initial mean concentration of 1746 to 2031 pg/kg after 20 d (Table 1). Day 20 mean tissue copper concentration was significantly higher than the initial concentration (p < 0.05), but the r value of 0.36 indicated a weak relationship between exposure time and concentration. The BCF values ranged from 12-22 (Table 1). The Series 3 control group water was below detection level (2.0 pg/l), while control tissue means ranged from 1746 pg/kg (d 0) to 1760 pg/kg (d 20). Water used in this series was obtained from a dif- ferent source to remedy the copper contamination problem. Series 4 chromium exposure tissue concentration means ranged from 559 pg/kg (d 0) to 2670 pg/kg (d 40) and the correlation between exposure time and concentra- tions was r = 0.92 (Table 2). BCF values ranged from 33 at d 5 to 422 at 40 d. The control group water concentration ranged from 3-4 pg/l, but control group tissue concentra- tions ranged from 559 pg/kg (d 0) to 299 pg/kg (d 40) indi- cating previous exposure to chromium and depuration. Series 5 chromium exposure resulted in a slow ini- tial, but increasing accumulation trend (Table 2). BCF val- ues ranged from 0.1 (d 5) to 14 (d 40). The marked differ- ence between the tissue concentrations and the BCF values Table 2. Laboratory exposures of Rangia cuneata to chromium including exposure water and tissue chromium concentrations with corresponding bioconcentration factors (BCF). Tissue means in pg/kg + standard error of the mean. EXPOSURE N WATER TISSUE MEAN BCF (days) (pg/l) (ng/kg) SERIES 4 (9 April to 19 May 1991) 0 1 11.0 559.1 - 5 2 7.0 787.8 + 63 32.7 10 2, 8.0 1508.7 + 365 118.7 20 2 9.0 1581.8+171 113.6 40 3 5.0 2670.4 + 248 422.3 SERIES 5 (10 May to 19 June 1991) 0 1 11.0 128.6 - 5 3 11.0 129.7+6 0.1 10 3 10.0 143.245 1.5 20 3 7.0 154.8+8 3.7 40 2 7.0 226.9 + 36 14.0 for Series 4 and 5 may be due to initial concentration dis- parities (Series 4, 559 pg/kg; Series 5, 129 pg/kg). Series 6 lead exposure exhibited the most rapid accumulation of any of the metals tested in the laboratory. Initial tissue concentrations were below detection level (100 pg/kg) and increased by d 40 to 3170 pg/kg (Table 3). BCF values ranged from 25 at d 5 to 93 at d 40. The corre- lation between exposure time and tissue lead concentration was r = 0.84. The control group lead concentrations were below the detection level in both the water (1.0 pg/l) and Rangia cuneata tissues during the entire exposure. Depuration of lead from Series 6 test animals was conducted by placing clams exposed to lead for 40 d into lead-free water and analyzing the tissues after d 10, 20, and 40. Tissue lead concentration steadily decreased from 3170 pg/kg at d 0 to 984 pg/kg at d 40 (Table 3). Series 7 exposure to cadmium resulted in a continu- ous uptake of the metal throughout the 40 d test (Table 4). Analysis of Rangia cuneata tissues at initial exposure revealed a concentration of 103 pg/kg with accumulation of cadmium after 40 d exposure to 1151 pg/kg. Cadmium exposure water remained relatively constant with individual measurements ranging from 22-24 pg/l. BCF ratios varied from 8 after 5 d to 44 after 40 d. The correlation coefficient (r) between exposure time and cadmium concentration was 0.75. Control group exposure medium concentration remained below the limit of detection (1.0 pg/l) during the entire 40 d exposure, while tissue levels ranged from 102 pg/kg at initial exposure and declined to 88 pg/kg after 40 d. LABORATORY RETENTION Laboratory retention of Rangia cuneata from January-September 1992 in 16°C water resulted in inhibi- tion of gonadal tissue hypertrophy, as indicated by relative- Table 3. Lead laboratory exposure and depuration study of Rangia cunea- ta including exposure water and tissue lead concentrations with corre- sponding bioconcentration factors (BCF) for lead exposure group. Tissue means in pg/kg + standard error of the mean. EXPOSURE N WATER TISSUE MEAN BCF (days) (pg/l) (pg/kg) SERIES 6 (15 April to 25 May 1992) _ 0 6 38.5 < 100.0 = 5 4 36.1 900.0 + 162 25.0 10 9 37.7 1863.8 + 87 49.3 20 7 33.1 2218.3 + 188 67.0 40 7 33.8 3170.0 + 331 93.8 DEPURATION STUDY (25 May to 3 July 1992) 10 7 < 1.0 2243.2 + 113 - 20 8 < 1.0 1988.4 +79 - 40 7 < 1.0 983.9 +54 - McCONNELL AND HARREL: METAL UPTAKE IN RANGIA 195 Table 4. Laboratory exposure of Rangia cuneata to cadmium including exposure water and tissue cadmium concentrations with corresponding bioconcentration factors (BCF). Tissue means in pg/kg + standard error of the mean. EXPOSURE N WATER TISSUE MEAN — BCF (days) (ug/l) (ng/kg) 0 5 22.3 102.9 + 15 2 5 6 25.4 299.5 + 29 7.8 10 8 21.8 798.9 +94 31.9 20 9 23.1 988.9 + 118 38.4 40 11 23.6 1150.7 + 88 44.4 ly stable percent gonadal biomass for the duration of the nine month laboratory retention. Percent gonadal biomass in January was 12%, and declined to 11% in March, when Fence Lake clams had a gonadal biomass of about 18%. Laboratory-retained clams percent gonadal biomass increased slightly by June to 14%. In September when Fence Lake R. cuneata were ripe and contained over 40% gonadal biomass, the laboratory-retained clams contained only 16% gonadal biomass (Table 5). A significant differ- ence occurred between percent gonadal biomass of labora- tory-retained clams and Fence Lake clams (p < 0.05). In addition, no significant change was observed in somatic tis- sue weight between the laboratory and Fence Lake clams (p < 0.05), indicating that nutritional deficiency was probably not a factor. However, an essential nutrient required for gonadal development may not be present in the supplement and could have contributed to these findings. Nonetheless, these data suggest that R. cuneata can be successfully maintained in the laboratory under controlled conditions over extensive periods of time with no apparent adverse effects. In addition, gametogenesis can be successfully inhibited, thus avoiding complications associated with changing reproductive status. FIELD EXPOSURES Physicochemical conditions varied considerably between the two field exposure sites (Table 6). Most of these differences were due to freshwater conditions in the river during the exposure at Site 1 and the occurrence of salt water intrusion during the exposure at Site 2. At Site 1 mean copper concentrations in Rangia cuneata tissues increased from 2240 pg/kg (d 0) to 2734 pg/kg (d 40) (Fig. 1). The initial and the 40 d tissue con- centrations were significantly different (p < 0.05), but there was a weak relationship between exposure time and tissue copper concentration (r = 0.46). Water samples collected at this site had copper levels below the level of detection (0.01 pg/l). Copper concentrations in the sediments ranged from 3791-5647 g/kg, indicating prior contamination. Copper accumulation by R. cuneata at Site 2 followed a similar trend compared to Site 1, but with slightly less uptake (Fig. 2). Tissue copper concentration means for the initial and 40 d periods were significantly different (p < 0.05). Sediment concentrations were highly variable (1641-4045 Table 5. Seasonal changes in biomass (dry weight) as related to temperature and salinity for Fence Lake and laboratory retained Rangia cuneata. TIME N MEAN GONADAL BIOMASS (%) Sept. 1991 3 49 Oct. 5 24 Nov. 7 18 Dec. 10 15 Jan. 1992 8 12 Feb. 10 18 Mar. 10 18 June 9 23 July 10 38 Sept. 12 40 Oct. 10 27 Dec. 20 39 Laboratory Retention Study Jan. 1992 8 12 Mar. 8 1] June 9 14 Sept. 10 16 MEAN TEMPERATURE SALINITY WT. (°C) (ppt) (g) 1.33 30 21 0.70 20 3.0 0.61 16 5.0 0.43 16 3.8 0.60 16 1.5 0.66 16 1.0 0.68 19 1.5 0.76 23 5.5 1.15 28 9.5 121 30 10.5 0.85 23 12.0 0.95 15 7.0 0.60 16 1:5 0.57 17 < 1.0 0.63 16 < 1.0 0.66 16 < 1.0 196 AMER. MALAC. BULL. 11(2) (1995) Table 6. Physicochemical conditions for the field exposure Sites | (June- July) and 2 (September-November). TEMPERATURE pH SALINITY ALKALINITY (°C) (ppt) (mg/1) Site 1 28-34 7.1-7.6 <0.5 29-36 Site 2 20-25 8.4.8.9 5-9 50-73 pg/kg), probably due to heterogeneous sediment composi- tion. The pattern of chromium uptake by Rangia cuneata was similar at Sites 1 and 2 (Figs. 3 and 4). The 40 d tissue concentration at Site 1 was 99 pg/kg, and at Site 2, 242 pg/kg. Significant differences were demonstrated between the initial and 40 d concentrations at both sites (p < 0.001), however no correlation existed between exposure times and concentrations (Site 1, r = 0.18; Site 2, r = 0.21). Chromium concentrations in the water ranged from 0.07- 0.52 g/l at Site 1 and 2-3 jg/I at Site 2. At both sites the concentration of chromium in the sediment was high and variable, again probably due to differences in sediment type. Cadmium tissue concentrations were variable at both sites (Figs. 5 and 6). Comparison of initial and 40 d tissue concentrations indicated significant difference (p < 0.05) at Site 1, but not at Site 2. The correlation coeffi- cients between exposure time and _ tissue cadmium concen- GM Tissue (ug/kg) [__] Sediment (yg/kg) 6000 6000 4000 Concentration 3000 2000 1000 Exposure Time (Days) "ig. 1. Uptake of copper by Rangia cuneata at field exposure Site | locat- ed below Star Lake Canal in the Neches River. 30 6000 MMM Tissue (ug/kg) [__] Sediment (ug/kg) 26 v Media=Y2 (ug/1) 4000 16 3000 Concentration 2000 o L¥? 1000 0 6 10 20 40 Exposure Time (Days) Fig. 2. Uptake of copper by Rangia cuneata at field exposure Site 2 locat- ed in an oil refinery effluent canal approximately 200 m below the outfall and 600 m above the confluence with the Neches River. Day 20 sediment sample could not be obtained. trations were 0.36 (Site 1) and 0.13 (Site 2). At Site 1, cadmium concentrations in the water ranged from 0.06- 0.22 g/l and varied from 120-560 pg/kg in the sediments. At Site 2, cadmium water concentrations ranged from 0.17- 0.49 g/l and sediment concentrations ranged from 31-138 pg/kg. Lead concentrations in Rangia cuneata tissues var- ied at both sites (Figs. 7 and 8). An analysis of tissue con- centrations at Site | indicated no significant change in lead body burden during the exposure (p > 0.05). At Site 2 a significant change in tissue lead concentration occurred between the initial and 40 d periods (p < 0.001), but the correlation coefficient between exposure time and lead con- centrations indicated no relationship (r = 0.18). Water lead concentrations at Site 1 ranged from < 0.01-1 g/l and sed- iment concentrations were 9890-15429 pg/kg. DISCUSSION AND CONCLUSION By performing laboratory exposures to metals many environmental variables were eliminated and metal specia- tion was predictable (Pankow, 1991). Field exposures test- ed the organism’s performance as a biomonitor in complex and highly labile environments. Even though the test ani- mals were collected from an isolated lake with no known history of contamination, analysis of tissue metal concen- trations at exposure d O showed that a significant body bur- den of all of the metals, except lead, occurred. McCONNELL AND HARREL: METAL UPTAKE IN RANGIA 197. 0.8 600 HM Tissue (ug/kg) Cc] Sediment (yg/kg) v Media=Y2 (yug/1) 0.6 600 0.4 400 0.3 300 Concentration 0.2 200 0.1 100 ool %2 Oo Exposure Time (Days) Fig. 3. Uptake of chromium by Rangia cuneata at field exposure Site 1 located below Star Lake Canal in the Neches River. 6, 1000 HMB Tissue (g/kg) (_] Sediment X 0.1 (yug/kg) 5 v Media = Y2 (yg/1) Concentration o ~ te) 6 10 20 40 Exposure Time (Days) Fig. 4. Uptake of chromium by Rangia cuneata at field exposure Site 2 located in an oil refinery effluent canal approximately 200 m below the outfall and 600 m above the confluence with the Neches River. Day 20 sediment sample could not be obtained. Laboratory and field exposures to copper demon- strated uniform metal uptake (Table 1, Figs. 1 and 2). Little difference was found in the uptake rate of copper from all exposures even though different seasonal periods were rep- resented. However, the laboratory Series 3 copper expo- sure, conducted during December and January had to be terminated due to mortalities in both the exposure group and the control group. These mortalities occurred when the clams had a gonadal biomass equivalent to 39% of their total body biomass, possibly interfering with osmoregulato- ry processes. This hypothesis is consistent with later obser- vations which evaluated seasonal trends in Rangia cuneata osmoregulation capacity in salinities ranging from < 0.50- 20 ppt (unpublished data). Conditions in the laboratory favored the ionic Cut+2 form which is usually the most bioavailable and toxic form of copper (Pankow, 1991; Giesy et al., 1983). The species of copper R. cuneata was exposed to in the field is unknown. However, the cupric ion is highly reactive and forms complexes and precipitates with many inorganic and organic constituents of natural waters (USEPA, 1980). Thus, the proportion of ionic cop- per in the field was probably low. This suggests that sever- al mechanisms of copper uptake occurred, because both laboratory and field exposures contained similar results. Bray et al. (1983) reported an inducible metal binding pro- tein similar to metallothioneins in R. cuneata, while Roesijadi (1980) found increased tolerance to copper asso- ciated with synthesis of metallothioneins. Thus, long term exposure to copper in Fence Lake could have induced the synthesis of metallothionein-like proteins resulting in increased copper tolerance and high copper concentrations in Fence Lake R. cuneata. However, even with an already high body burden of copper, R. cuneata consistently accu- mulated copper regardless of exposure time and season. Accumulation of chromium during the 40 d labora- tory exposures was 4.7X and 1.8X the initial tissue concen- trations for Series 4 and 5, respectively (Table 2). There was a Significant difference in the chromium uptake rates and the final concentrations between the two series (p < 0.05) under comparable physicochemical conditions and reproductive phases of the test animals. The Series 4 clams had 4.3X higher initial chromium concentration than Series 5 clams due to prior exposure, which may account for the disparity in uptake results. In the field exposures chromi- um was not consistently taken up at either site however, Site 2 chromium was significantly higher than at Site 1 (p< 0.05) after 40 d of exposure (Figs. 3 and 4). The cause of this inconsistency is not known, but may be related to body weight variations associated with seasonal reproductive state as reported by Phillips (1976). The difference may also be related to the higher concentration of chromium in the sediments and water found at Site 2 and/or physico- chemical conditions which favored the species of chromi- um with greater potential for uptake. Chromium in natural waters can exist in a number of oxidation states with differ- ent activities ranging from -2 to +6, but is most often found in the trivalent (+3) and hexavalent (+6) forms (USEPA, 1985). Rapid evolution of chromium in the sediment can occur. Precipitated Cr+3 hydroxides in the sediment may solubilize and remain as Cr+3 or can be oxidized to Cr+® by 198 AMER. MALAC. BULL. 11(2) (1995) HM Tissue (ug/xg) [_] Sediment (ug/kg) v Media = Y2 (ywg/1) Concentration Exposure Time (Days) Fig. 5. Uptake of cadmium by Rangia cuneata at field exposure Site | located below the Star Lake Canal in the Neches River. aeration. This dynamic transformation was probably a con- tinual process at Sites 1 and 2 where tidal influence, cur- rents, high rainfall and terrestrial runoff, and boat traffic constantly altered sediment redox. In laboratory chromium exposures, the +6 oxidation state was probably favored due to oxygenated conditions and low organic matter, although the trivalent form was likely present as well. Laboratory exposure to cadmium resulted in rapid accumulation and consistent uptake, with a 11X increase in tissue concentration from d 0 to d 40 (Table 3). Field expo- sure results did not resemble the laboratory exposure results. At field exposure Site 1, Rangia cuneata accumu- lated cadmium, but uptake was variable during the interme- diate exposure period (Fig. 5). At Site 2, no significant uptake of cadmium occurred even though cadmium concen- tration in the water was usually higher (Fig. 6). These vari- ations were probably due to the disparate physicochemical conditions that existed between the laboratory exposure and the two field exposures. In the laboratory the predominate- ly dissolved species of cadmium was probably free cadmi- um (Cd+2) which is usually the most toxic and bioavailable form (Winner, 1984; Stackhouse and Benson, 1989). Conditions in the laboratory and at field exposure Site 1 may have favored a larger fraction of cadmium in the Cd+2 form as salinity, pH, alkalinity, and water hardness were comparable. Another factor is the role of dissolved organic matter (DOM), which is known to readily alter cadmium toxicity and bioaccumulation properties by changing the concentration of Cd+2 in solution through adsorption and desorption processes (Winner, 1984; Eisler, 1985; Stackhouse and Benson, 1989). The DOM concentration at Site 2 may have been higher due to its location (adjacent to marshland), thus rendering less cadmium in the bioavail- able form. Also, pH, salinity, alkalinity, and temperature were vastly different between the field exposures and are all known to influence the biological uptake of cadmium (Phillips, 1976; Hung, 1982; Eisler, 1985; Van Hattum er al., 1988). As with copper, cadmium uptake by Rangia cunea- ta may have been influenced by metal binding proteins. Bray et al. (1983) suggested that a structural difference existed between metallothionein species for cadmium and copper, with the possible implication of metal specific induction mechanisms. Induction of a metallothionein-like protein in the benthic oligochaete Limnodrilus hoffmeisteri (Laparéde, 1862) was reported to have contributed to cad- mium resistance at a metal-polluted site by Klerks and Bartholomew (1991). The resistance was not achieved by a reduction in cadmium but by elevated protein levels and metal sequestering in sulphur rich granules. In the laboratory, accumulation of lead by Rangia cuneata was rapid and highly consistent (Table 4). The BCF values increased from 25 after 5 d to 94 after 40 d. However, in the field exposures lead uptake was slight and erratic despite lead contamination of the sediments (Figs. 7 and 8). The major proportion of lead in the laboratory exposure was probably in the ionic Pb+2 form, as revealed from physicochemical data and pH predominance dia- grams from Pankow (1991). Wong et al. (1978) reported 0.75 HB Tissue (g/kg) [__] Sediment (ug/kg) vy Media = Y2 (ug/1) 0.60 a & 0.45 » G be » S o a £ 0.30 oO 0.15 0.00 Exposure Time (Days) Fig. 6. Uptake of cadmium by Rangia cuneata at field exposure Site 2 located in an oil refinery effluent canal approximately 200 m below the outfall and 600 m above the confluence with the Neches River. Day 20 sediment sample could not be obtained. McCONNELL AND HARREL: METAL UPTAKE IN RANGIA 199 1.50 1800 MMMM Tissue (ug/kg) [__] Sediment X 0.1 (ng/kg) v Media = Y2 (ug/l) 1.25 1600 1.00} 1200 A ° -_ » 3 sey 2 0.76 900 oO 1?) A °o o 0.50 600 0.25 300 o.oo LY? 9 Exposure Time (Days) Fig. 7. Uptake of lead by Rangia cuneata at field exposure Site | located below the Star Lake Canal in the Neches River. Days 0-10 media values were below the limit of detection (0.01 pg/I). HMB Tissue (yg/kg) [J Sediment X 0.1 (ug/kg) 4 v Media = Y2 (pg/1) v. _ Concentration te) 5 10 20 40 Exposure Time (Days) Fig. 8. Uptake of lead by Rangia cuneata at field exposure Site 2 located in an oil refinery effluent canal approximately 200 m below the outfall and 600 m above the confluence with the Neches River. Days 5-20 sediment values could not be obtained. that only soluble waterborne lead is toxic to aquatic biota, and for most cases the ionic forms are more toxic and bioavailable than complexed forms. Lead usually exists in surface waters in three forms: (1) dissolved labile - Pb+2, PbOH?+, PbCO3, (2) dissolved bound - colloids or complex- es, and (3) particulate (Benes et al., 1985). The predomi- nate species of lead occurs in the dissolved bound form or particulate form when surface waters contain substantial concentrations of DOM (Eisler, 1988). Precipitated lead in the sediment is often tightly bound and characterized by slow release time (Prause et al., 1985). This may explain the low water and high sediment lead concentrations during the field exposures. A lead depuration study was conducted following Series 6 laboratory exposure. After 10 d, a 29% reduction in tissue lead level was observed. This decrease may have been due to non-bound lead removal from the gills. Production of mucus, primarily in the gills, is a common toxicological response in bivalves and could have influ- enced removal of trapped lead once transferred to clean water. After 20 d of depuration 63% of the lead remained, indicating a more tightly bound proportion of lead. The half-life of lead in Rangia cuneata tissues occurred between the 20 and 40 d depuration periods. Thus, R. cuneata is capable of reflecting changing concentrations of lead in the water over time, indicating that metabolic processes do not interfere in the evaluation of lead bioavail- ability. Recommendations have been offered to minimize variations in bioaccumulation associated with the reproduc- tive cycle in bivalves (Phillips, 1976; National Research Council, 1980). Phillips (1976) suggested collecting speci- mens in the same season so as to avoid seasonal changes in body weight. This, however is impractical and places tem- poral limitations on monitoring strategies. By collecting Rangia cuneata during the spent phase and retaining them in a thermally controlled (15-17°C) environment, gameto- genesis was inhibited and the gonadal biomass maintained at < 20% of total biomass. Laboratory retained clams were fed regularly and showed no significant decreases in weight. While the possibility remains that an essential nutrient required for gametogenesis was not present in the food supplement, at present, the most rational explanation is that of controlled temperature. It has long been known that temperature is the primary factor regulating gametoge- nesis in R. cuneata (fide Hopkins et al., 1973). Thus, by understanding the mechanisms which regulate reproduc- tion, strategies can be implemented to circumvent seasonal complications associated with the use of R. cuneata as a biomonitor. Laboratory and field exposures seemed to indicate that Rangia cuneata is a suitable biomonitor of the most biologically active forms of copper, chromium, cadmium and lead. In addition, its extreme tolerance to varying envi- ronmental conditions, wide geographical distribution, occurrence in high densities, and knowledge of factors that regulate its life cycle all support the use of R. cuneata as a biomonitor of hazardous substances. R. cuneata can be conveniently transplanted into inland waters or estuaries, thus extending the Mussel Watch effort to sensitive areas along our coastlines which previously have been neglected 200 AMER. MALAC. (NOAA, 1989). However, more work is required to evalu- ate the many factors that control metal uptake during bioac- cumulation trials. ACKNOWLEDGMENTS This research was supported by grants from the Gulf Coast Hazardous Substance Research Center which is supported under coopera- tive agreement with the U.S. Environmental Protection Agency and the State of Texas as part of the Texas Hazardous Waste Research Center, both at Lamar University-Beaumont. We wish to thank William Robbins, Brad Raider, Eric Postula, and Dave White of Earth Analytical Science, Inc., for their technical support of this project. LITERATURE CITED APHA-AWWA-WPCE (American Public Health Association-American Water Works Association-Water Pollution Control Federation). 1989. Standard Methods For Examination of Water and Wastewater, 17th ed. American Public Health Association, New York. 1097 pp. Bedford, J. W., E. W. Roelofs, and M. J. Zabik. 1968. The freshwater mussel as a biological monitor of pesticide concentrations in a lotic environment. Limnology and Oceanography 13:118-126. Benes, P., M. Cejchanova, and B. Havlik. 1985. Migration and speciation of lead in a river system heavily polluted from a smelter. Water Resources 19:1-6. Boyer, K. W. 1984. Metals and other elements at trace levels in foods. In: Official Methods of Analysis of the Association of Official Analytical Chemists, 17th ed., S. Williams, ed. pp. 444-476. Association of Official Analytical Chemists, Richmond, Virginia. Bray, J. T., L. A. Webb, and F. J. Reilly, Jr. 1983. Multielement analysis of metal-binding proteins in the cytosol fractions. The Science of the Total Environment 28:367-374. Cain, T. D. 1975. Reproduction and recruitment of the brackish water clam Rangia cuneata in the James River, Virginia. Fishery Bulletin 73:412-430. Chaiyarach, S., V. Ratananun, and R. C. Harrel. 1975. Acute toxicity of the pesticides Toxaphene and Carbaryl and the herbicides Propanil and Molinate to four species of aquatic organisms. Bulletin of Environmental Contamination and Toxicology 14:281-284. Chen, C. and J. Awapara. 1969. Effects of oxygen on the end-products of glycolysis in Rangia cuneata. Comparative Biochemistry and Physiology 31:39-401. Curry, C. A. 1978. The freshwater clam (Elliptio complanata), a practical tool for monitoring water quality. Water Pollution Research 13:45- 52. Davis, R. P. and A. J. Dobbs. 1984. The prediction of bioconcentration in fish. Water Resources 18:1253-1262. Eisler, R. 1985. Cadmium hazards to fish, wildlife, and invertebrates: a synoptic view. Contaminant Hazard Reviews Report No. 2. United States Fish and Wildlife Service, Biological Report 85(1.2): 60 pp. Eisler, R. 1988. Lead hazards to fish, wildlife, and invertebrates: a synop- tic view. Contaminant Hazard Reviews Report No. 14. United States Fish and Wildlife Service, Biological Report 85(1.14): 134 pp. Ellis, M. M. 1937. Detection and measurement of stream pollution. BULL. 11(2) (1995) Bulletin of the United States Bureau of Fisheries 22:365-437. Fairbanks, L. D. 1963. Biodemographic studies of the clam Rangia cuneata Gray. Tulane Studies in Zoology 10:3-47. Farrington, J. W., A. C. Davis, B. W. Tripp, D. K. Phelps, and W. B. Galloway. 1987. “Mussel Watch” - measurements of chemical pollutants in bivalves as one indicator of coastal environmental quality. New Approaches to Monitoring Aquatic Ecosystems ASTM STP 940:125-139. Foster, R. B. and J. M. Bates. 1978. Use of freshwater mussels to monitor point source industrial discharges. Environmental Science and Technology 12:958-962. Gallagher, J. L. and H. W. Wells. 1969. Northern range extension and winter mortality of Rangia cuneata. The Nautilus 83:22-25. Giesy, J. P., A. Newell, and G. J. Leversee. 1983. Copper speciation in soft, acid, humic waters: effects on copper bioaccumulation by and toxicity to Simocephalus surrulatus (Daphnidae). The Science of the Total Environment 28:23-36. Glantz, E. N. 1992. Primer for Bio-Statistics. Rockwell, Inc., New York. 178 pp. Gooch, D. M. 1971. A Study of Rangia cuneata Gray, in Vermillion Bay, Louisiana. Graduate Thesis, University of Southwestern Louisiana, Lafayette, Louisiana. 48 pp. Hartel, R. C., J. Ashcraft, R. Howard, and L. Patterson. 1976. Stress and community structure of macrobenthos in a Gulf Coast riverine estu- ary. Contributions in Marine Science 20:69-81. Harrel, R. C. and M. A. Hall. 1991. Macrobenthic community before and after pollution abatement in the Neches River estuary (Texas). Hydrobiologia 211:241-245. Hartel, R. C. and M. A. McConnell. 1995. The estuarine clam Rangia cuneata as a biomonitor of dioxins and furans in the Neches River, Taylor Bayou, and Fence Lake, Texas. Estuaries 18:265-270. Havlich, S. H. and L. L. Marking. 1987. Effects of contaminants on naiad mollusks (Unionidae): a review. United States Fish and Wildlife Service, Resource Publication 164: 20 pp. Heit, M., C. S. Klusek, and K. M. Miller. 1980. Trace element, radionu- clide, and polynuclear aromatic hydrocarbon concentrations in Uniodidae mussels from northern Lake George. Environmental Science and Technology 14:465-468. Hopkins, S. H. and J. D. Andrews. 1970. Rangia cuneata on the east coast: thousand mile range extension, or resurgence? Science 167: 868. Hopkins, S. H., J. W. Anderson, and K. Horvath. 1973. The Brackish Water Clam Rangia cuneata as an Indicator of Ecological Effects of Salinity Changes in Coastal Waters. United States Army Engineer Waterways Experimental Station, Vicksburg, Mississipi, Contract Representative DACW39-71-C-0007. 250 pp. Hung, Y. 1982. Effects of temperature and chelating agents on cadmium uptake in the American Oyster. Bulletin of Environmental Contamination and Toxicology 28:546-551. Imlay, M. J. 1982. Use of shells of freshwater mussels in monitoring heavy metals and environmental stress. Malacological Review 15:1-14. Keystat. 1985. Keystat Statistical Software Version 1.3. Oakleaf Systems, Iowa. Klerks, P. L. and P. R. Bartholomew. 1991. Cadmium accumulation and detoxification in a Cd-resistant population of the oligochaete Limnodrilus hoffmeisteri. Aquatic Toxicology 19:97-112. Lakshmanan, P. T. and P. N. K. Nabisan. 1989. Bioaccumulation and depuration of some trace metals in the mussel, Perna viridis (Linnaeus). Bulletin of Environmental Contamination and Toxicology 43:131-138. LaSalle, M. W.-and A. A. de la Cruz. 1985. Species profiles: life histories and environmental requirements of coastal fishes and invertebrates McCONNELL AND HARREL: (Gulf of Mexico): common Rangia. United States Fish and Wildlife Service, Biological Report 82(1 1.32): 16 pp. Leard, R. L., B. J. Grantham, and G. F. Pessony. 1980. Use of selected freshwater bivalves for monitoring organochlorine pesticide residues in major Mississippi stream systems, 1972-73. Pesticides Monitoring Journal 14:47-52. Lord, D. A., W. G. Breck, and R. C. Wheeler. 1975. Trace elements in molluscs in Kingston Basin. Water Quality Parameters, American Society for Testing Materials, Special Technical Publication 573:95-111. Lunsford, C. A. and C. R. Blem. 1982. Annual cycle of kepone residue and lipid content of the estuarine clam Rangia cuneata. Estuaries 5:121-130. Manly, R. and W. O. George. 1977. The occurrence of some heavy met- als in populations of the freshwater mussel Anodonta anatina (L.) from the river Thames. Environmental Pollution 14:139-154. Mason, C. E. 1991. Biology of Freshwater Pollution. Longman Inc., New York. 351 pp. Mathis, B. J. and T. J. Cummings. 1973. Selected metals in sediments, water, and biota in the Illinois River. Journal of Water Pollution and Contamination Federation 45: 1573-1583. Naimo, T. J., D. L. Waller, and L. E. Holland-Bartels. 1992. Heavy met- als in the threeridge mussel Plicata plicata (Say, 1817) in the upper Mississipi River. Journal of Freshwater Ecology 7:209-217. National Oceanic and Atmospheric Administration (NOAA). 1989. A Summary of Data on Tissue Contamination from the First Three Years (1986-1988) of the Mussel Watch Project. NOAA, Rockville, Maryland. 22 pp. National Research Council. 1980. The International Mussel Watch. National Academy of Science, Washington, D.C. 248 pp. Neff, L. M., B. A. Cox, D. Dixit, and J. W. Anderson. 1976. Accumulation and release of petroleum-derived aromatic hydrocar- bons by four species of marine animals. Marine Biology 38:279- 289. Olsen, K. R. and R. C. Hartel. 1973. Effects of salinity on acute toxicity of mercury, copper, and chromium for Rangia cuneata (Pelycypoda, Mactridae). Contributions in Marine Science 17:9-13. Pankow, J. F. 1991. Aquatic Chemistry Concepts. Lewis Publishers, Michigan. 683 pp. Peltier, W. H. and C. I. Weber. 1985. Methods for Measuring the Acute Toxicity of Effluents to Freshwater and Marine Organisms. U. S. Environmental Protection Agency, Cincinnati, Ohio. 215 pp. Phillips, D. J. H. 1976. The common mussel Mytilus edulis as an indica- tor of pollution by zinc, cadmium, lead, and copper. I. Effects of environmental variables on uptake of metals. Marine Biology 38:59-69. Phillips, D. J. H. 1977. The use of biological indicator organisms to mon- itor trace metal pollution in marine and estuarine environments: areview. Environmental Pollution 13:281-317. Prause, B., E. Rehm, and M. Schulz-Bades. 1985. The remobilization of lead and cadmium from contaminated dredge spoil after dumping in the marine environment. Environmental Technology 6:261-266. Roesijadi, G. 1980. Influence of copper on the clam Protothaca sta- minea: effects on gills and occurrence of copper-binding proteins. Biological Bulletin 158:233-247. Simpson, R. D. 1979. Uptake and loss of zinc and lead by mussels (Mytilus edulis) and relationships with body weight and reproduc- tive cycle. Marine Pollution Bulletin 10:74-78. Stackhouse, R. A. and W. H. Benson. 1989. Interaction of humic acid with selected trace metals: influence on bioaccumulation in daphnids. Environmental Toxicology and Chemistry 8:639-644. Stephenson, M. D., M. Martin, and J. H. Martin. 1980. California state Mussel Watch: trace metal concentrations in selected shellfish from METAL UPTAKE IN RANGIA 201 California bays and estuaries. California State Water Resources Control Board, Water Quality Report 79-22: 56 pp. Swingle H. A. and D. G. Bland. 1974. Distribution of the estuarine clam Rangia cuneata Gray in coastal waters of Alabama. Alabama Marine Resource Bulletin 10:6-16. Tarver, J. W. 1972. Occurrence, distribution and density of Rangia cuneata, in Lake Pontchartrain and Lake Maurepas, Louisiana. Louisiana Wildlife and Fisheries Commission Technical Bulletin 1: 8 pp. Tenore, K. E., D. B. Horton, and T. W. Duke. 1968. Effects of bottom substrate on the brackish water bivalve Rangia cuneata. Chesapeake Science 9:99-102. United States Environmental Protection Agency (USEPA). 1980. Ambient Water Quality Criteria for Copper. EPA 440/5-80-036. U. S. Environmental Protection Agency, Washington, D.C. 162 pp. United States Environmental Protection Agency (USEPA). 1984a. Ambient Water Quality Criteria for Copper. EPA 440/5-84-031. U. S. Environmental Protection Agency, Washington, D.C. 142 pp. United States Environmental Protection Agency (USEPA). 1984b. Ambient Water Quality Criteria for Chromium. EPA 440/S-84-033. U. S. Environmental Protection Agency, Washington, D.C. 99 pp. United States Environmental Protection Agency (USEPA). 1985. Ambient Water Quality Criteria for Chromium. National Technical Information Service PB85-227478. U.S. Environmental Protection Agency, Washington, D.C. 138 pp. United States Environmental Protection Agency (USEPA). 1986. Test Methods for Evaluating Solid Waste, Physical/Chemical Methods. SW 846. 3rd ed. U.S. Environmental Protection Agency, Cincinnati, Ohio. Van Hattum, B., K. R. Timmermans, and H. A. Govers. 1988. Bioavailability of cadmium, lead, zinc, and copper in sediments and water to the freshwater isopod Asellus aquaticus (L.). In: Proceedings of the International Conference on Environmental Contamination, Venice, Italy. pp. 439-441. CEP-Consultants, Edinburgh, U.K. Van Hattum, B., K. R. Timmermans, and H. A. Govers. 1991. Abiotic and biotic factors influencing in situ trace metal levels in macroin- vertebrates in freshwater ecosystems. Environmental Toxicology and Chemistry 10:275-292. Verner, S._ 1990. Handbook for Preparing Quality Assurance Project Plans for Environmental Measurements. Technical Resources, Inc., Rockville, Maryland. 123 pp. Wade, T. L., B. Garcia-Romera, and J. M. Brooks. 1988a. Tributylin con- tamination in bivalves from United States coastal estuaries. Environmental Science and Technology 22:1488-1493. Wade, T. L., E. L. Atlas, J. M. Brookes, M. C. Kennicutt II, R. G. Fox, J. Sericano, B. Garcia-Romero, and D. DeFrettas. 1988b. NOAA Gulf of Mexico Status and Trends Program: trace organic contaminant distribution in sediments and _ oysters. Estuaries 11:171-179. Winner, R. W. 1984. The toxicity and bioaccumulation of cadmium and copper as affected by humic acid. Aquatic Toxicology 5:267-274. Wolfe, D. A. and E. N. Petteway. 1968. Growth of Rangia cuneata Gray. Chesapeake Science 9:99-102. Wong, P. T. S., B. A. Silverburg, Y. K. Chau, and P. V. Hodson. 1978. Lead and the aquatic biota. /n: The Biogeochemistry of Lead in the Environment. Part B. Biological Effects. J. O. Nriagu, ed. pp. 279- 342. Elsevier/North Holland Biomedical Press, Amsterdam. Wright, D. A. 1978. Heavy metal accumulation by aquatic invertebrates. Applied Biology 3:33-394. Date of manuscript acceptance: 21 February 1995 Ecological notes and patterns of dispersal in the recently introduced mussel, Perna perna (Linné, 1758), in the Gulf of Mexico David W. Hicks and John W. Tunnell, Jr. Center for Coastal Studies, Texas A & M University - Corpus Christi, 6300 Ocean Drive, Corpus Christi, Texas 78412, U.S. A. Abstract. Invasive mussels, Perna perna (Linné, 1758), were first detected in south Texas on the jetties at Port Aransas (27° 50’ N) in February 1990 (Hicks and Tunnell, 1993). Within four years the species has colonized jetties, navigation buoys, petroleum platforms, wrecks, and other artificial hard sub- strata as well as natural rocky shores between Matagorda Peninsula (28° 35’ N), Texas, and Playa Escondida (18° 35’ N), southern Veracruz, Mexico, a dis- tance of over 1,300 km. Densities of 27,200/m2 small individuals (mean = 16 mm + 0.3 SE) have been recorded from south Texas jetties. Spat settlement densities of up to 300/25 cm2 (120,000/m2) have been recorded from algal substrata in southern Veracruz, Mexico. On oil production platforms, 6-27 km offshore from Port Aransas, the species occurs from the intertidal zone down to depths of 9 m. Dispersal patterns interpreted from discovery data in the Gulf of Mexico indicate a primarily southward expansion from the initial recording. The occurrence of P. perna in euryhaline environments, such as river mouths and bay systems, demonstrates the ecological adaptability of this species. Perna perna (Linné, 1758) (Mytilidae) was first detected in south Texas on the jetties at Port Aransas (27° 50’ N) in February 1990 (Hicks and Tunnell, 1993). Within four years jetties, navigation buoys, petroleum plat- forms, wrecks, and other artificial hard substrata as well as natural rocky shores between Matagorda Peninsula (28° 35’ N), Texas, and Playa Escondida (18° 35’ N), southern Vera- cruz, Mexico, a distance of over 1,300 km, had been colo- nized. Although the species was first detected at Port Aransas, the establishment of high-density beds there and to the north has been relatively slow compared to recently colonized areas to the south. Worldwide records of geo- graphic distribution for P. perna include: Aden, the Red Sea, Madagascar, the east coast of Africa from central Mozambique to False Bay in the Cape, the African west coast from Luderiz Bay northwards, the Mediterranean from Gibraltar to the Gulf of Tunis, Brazil, Uruguay, Venezuela, the West Indies, and the Straits of Magellan (Berry, 1978). This paper documents the current status of this recently introduced species and demonstrates its dis- persal through an examination of the chronology of discov- ery and settlement patterns from locations along the west- ern Gulf of Mexico coastline. Perna perna is opportunistically colonizing hard substrata in the southwestern Gulf of Mexico, settling on jetties and debris (natural and artificial) in the surf zone as well as on offshore oil production platforms and navigation buoys. Locally (Port Aransas to Brazos Santiago Passes), individuals dominate the lower mid-intertidal zone of gran- ite jetties, forming a distinct “mussel belt” (Fig. 1). Densities of up to 27,200/m2 small individuals ( mean = 16 mm + 0.3 SE) have been recorded from jetty rock. Jetties commonly support densities of 10,000-12,000/m2. The species has also been found colonizing the skeletal struc- tures of petroleum platforms and navigation buoys along the south Texas coast as far as 27 km offshore from Port Aransas, Texas. On petroleum platforms, P. perna colo- nizes structures from the intertidal zone down to depths of 9m. This aggressive new member of the fouling communi- ty in the Gulf of Mexico has the potential to dramatically increase the maintenance and or replacement interval of offshore navigation aids. The concern is that heavy infesta- tions could partially sink navigation buoys and affect ship- ping safety. In Brazil, a native locality, the navy is respon- sible for maintaining navigation buoys, and has to remove fouling layers due to P. perna infestations every six months (E. C. Rios, pers. comm., 1994). The U.S. Coast Guard’s “Aids to Navigation Team” is currently responsible for maintaining approximately 150 navigation buoys in the nearshore waters of the Gulf of Mexico annually. Mytilids are known to be responsible for the forma- tion of complex biological substrata (Paine, 1976). Mussels attach to the substratum and to one another by a dense network of byssal threads, forming a trap for particu- late debris. The mussel matrix and trapped material encrust the underlying substratum, preventing the attachment of other sessile organisms, including their own recruits. On artificial substrata in south Texas, beds can be composed of American Malacological Bulletin, Vol. 11(2) (1995):203-206 203 204 AMER. MALAC several layers and attain thicknesses in excess of 20 cm. Mussels taken from the interior of clumps are often deformed and twisted as reported by Harger (1972) for two species of Mytilus in California. This modification of the primary substratum results in both the displacement and facilitation of other sedentary species. Sessile organisms dominating this zone prior to the Perna introduction include cirripedes, Balanus spp. and Chthamalus fragilis Darwin, 1854; red algae, Hypnea musciformis (Wulfen) Lamouroux, 1813, and Centroceros clavulatum (C. Agardh) Montagne, 1846; anemones, Bunodosoma cavernata (Bosc, 1802); and other mussels, Brachidontes exustus (Linné, 1758). Lower mid-intertidal hard substrata available for Perna perna colonization in south Texas are primarily groins and jetties, and have existed for only a century (Britton and Morton, 1989). Therefore, the invasion by P. perna in south Texas is not disrupting a co-evolved, natural assemblage. However, to the south, the species is rapidly invading the unique volcanic outcroppings of central Mexico that occur from approximately 24° N latitude to the Tuxtlas Mountain Province southeast of the city of Veracruz. The presence of mussel beds will inevitably bring about physical modifications to the natural substra- tum and therefore impact these natural communities. Dominant sessile molluscan species in the lower intertidal zone of this natural rocky shore area, prior to the introduc- tion, were Petaloconchus varians (Orbigny, 1841), Brachidontes exustus (Linné, 1758), and Isognomon bicolor (C. B. Adams, 1845) (Wiley et al., 1982). rth Port Mansfield jetty in August 1994. . BULL. 11(2) (1995) METHODS Discovery data for Perna perna in the Gulf of Mexico were organized from various sources including lit- erature, personal communications, University collections, and field trips. Data were then used to interpret dispersal patterns since the initial discovery. Density and length-fre- quency data, when available, were determined from collect- ed mussel clumps where clump size and individual lengths were measured to the nearest centimeter and millimeter, respectively. RESULTS The dates of first discovery of Perna perna at loca- tions in the Gulf of Mexico are given in Fig. 2. The locali- ties depicted represent only those sites for which a history of collection data were available and do not represent the only sites from which the species is known. Thorough col- lecting efforts on the south jetty at Port Aransas (27° 50’ N) in February 1990 yielded only two specimens, each 20 mm in length. Growth data for P. perna from Durban, South Africa (29° 52’ S) (Berry, 1978), as well as from the Gulf of Mexico (unpublished) indicate that these first specimens had settled approximately two months earlier. First discov- ery of individuals on the north jetty at Port Mansfield (26° 33’ N), approximately 230 km south of Port Aransas, occurred in September 1991. At that time, individuals were widely spaced (< 200 individuals/m2), did not form exten- sive beds, and appeared to be of the same size class. Three HICKS AND TUNNELL: PERNA IN THE GULF OF MEXICO 205 specimens representing the range in length were collected (50, 51, and 55 mm). In July 1992, the species was first discovered at the Brazos Santiago Pass jetties (26° 04’ N) (M. K. Wicksten, pers. comm., 1992). At that time, bed formation was observed on both sides of the north jetty. Three representative specimens were collected (12, 32, and 32 mm; M. K. Wicksten, pers. comm. 1992). No mussels were found at the Brazos Santiago Pass jetties during a pre- vious visit in March 1990. From the jetties at La Pesca, Mexico (23° 47’ N), in March 1990, and the natural rocky shores at Boca Andrea (19° 44’ N) and Playa Escondida (18° 35’ N), Mexico, in October 1990, no mussels were collected. However, the species was found at these same localities during visits exactly three years later. Extensive mussel bed formation was observed on the jetties at La Pesca, Mexico, in March 1993. Individuals at La Pesca ranged in size from 2-71 mm containing two distinct size classes (means = 9.8 mm + 0.33 SE and 57.3 + 0.67 SE). In October 1993, the species was first discovered on the natural rocky shores at Punta Boca Andrea. At that time, bed formation was observed and den- sities as high as 5,216 individuals/m2 (mean = 26 mm + >» @ November 1993 @® February 1990 @) September 1991 @ July 1992 Fig. 2. Locations and dates of first discovery of brown mussels, Perna perna, in the Gulf of Mexico: (1) Colorado River; (2) Port Aransas Pass; (3) Port Mansfield Pass; (4) Brazos Santiago Pass; (5) La Pesca; (6) Boca Andrea; (7) Playa Escondida. 0.58 SE) were recorded. Sizes of mussels at Boca Andrea ranged from 5-71 mm, representing at least two size class- es. The first discovery of Perna perna at Playa Escondida also occurred in October 1993; here they were widely spaced and did not form beds. Twenty-five specimens were collected ranging from 18-46 mm (mean = 37 mm + 1.5 SE). In March 1994, densities of small individuals (2-10 mm) as high as 300/25 cm2 (120,000/m2) were observed on algal substratum [Laurencia papillosa (C. Agardh) Greville] at the Playa Escondida site. Similar recruitment events were observed in Texas prior to the formation of well-established beds. The individuals collected at Playa Escondida represent the southern end of the species’ known distribution in the Gulf of Mexico; however, localities fur- ther to the south and east have not been visited. To the north, Perna perna was collected from the jet- ties adjacent to the Colorado River on Matagorda Peninsula (28° 35’ N), Texas (Davenport, 1994), and from the jetties adjacent to the Gulf Intracoastal Waterway at Port O’Conner (28° 26’ N) (Davenport, 1995). A single speci- men, 52 mm in length, was collected in November 1993 from the east side of the easternmost jetty on Matagorda Peninsula. During a study of marine macroalgae conducted at the same jetty in 1993, Boyd and Wardle (1994) reported a consistent gulfside salinity of 25 ppt for three months sampled (January-March). Two specimens, each 20 mm in length, were collected in February 1995 from the northern- most jetty at Port O’Conner. The discovery of P. perna at Port O’Conner was the first record of the species inhabiting any Texas bay system. DISCUSSION The chronology of collection events depict a south- ward expansion of Perna perna from the initial point of dis- covery, Port Aransas, Texas. Colonization success in areas south of the initial point of discovery is likely the result of hydrographic factors, such as longshore and nearshore-sur- face currents. Accordingly, the slow establishment of mus- sel populations to the north is likely due to a lack of incom- ing recruits, rather than environmental constraints posed by decreasing temperature and salinity gradients from south to north that exist along the Texas coast. Longshore currents flow southeasterly along the upper Texas coast and north- easterly along the southern Texas and northern Mexico coasts, converging at approximately 27° N latitude (Britton and Morton, 1989). The persistent southeasterly direction of longshore and nearshore-surface currents to the north of 27° N is probably responsible for the slow establishment of mussel beds here. Cold fronts, while having only moderate effects on longshore systems to the north have dramatic effects on longshore and nearshore current systems 206 AMER. MALAC. BULL. 11(2) (1995) southward along the coastal bend (Watson and Behrens, 1970; Britton and Morton, 1989). The rapid expansion of P. perna to the south is likely due to the coincidence of mass spawning events with seasonal changes in current direction. Temperature is one of the more important factors controlling gametogenesis in temperate and subtropical bivalve mollusks (Giese, 1959). Peak spawning periods of P. perna have been found to coincide with the season of lowest temperature in South Africa, Angola, Venezuela, and Brazil (Berry, 1978). Spawning peaks of P. perna pop- ulations in Venezuela and Brazil occur when temperature declines to 22° C after a maximum of 28° C (Lunetta, 1969; Velez and Epifanio, 1981). Conversely, elevated temperature is routinely used in conditioning bivalves to spawn out of season (Loosanoff and Davis, 1963; Velez and Epifanio, 1981). The largest recruitment events in the Gulf of Mexico have been observed during fall and early spring. Major recruitment events were observed at La Pesca in March 1993, Port Mansfield in December 1993 and 1994, and at Boca Andrea and Playa Escondida in March 1994. However, no major recruitment was observed at the Port Aransas or Fish Pass jetties (29 km south of Port Aransas) during these same periods. Only minor recruitments (< 64 individuals/m2) were observed at these localities during December 1993 and 1994, and May 1994. It is believed that recruitment in areas to the north of approximately 27° N is largely dependent upon deviations from seasonal pat- terns in current direction and spawning cycles. However, the placement and hard-substratum-type habitat, provided by oil production platforms, could facilitate spread to the north. The discovery of specimens at the mouth of the Colorado River (157 km north of Port Aransas) and the jetty at Port O’Conner demonstrates the ecological adapt- ability of Perna perna. The length of the individual collect- ed from the Colorado River (52 mm) indicates that it had resided at that location for 7-12 mo (Berry, 1978, and unpublished growth data from the Gulf of Mexico) and is thus tolerant of those conditions (salinity and temperature) characteristic of the northern Texas coast. According to lit- erature (Vakily, 1989), P. perna tolerates fairly large fluctu- ations in salinity, adapting well to ranges of 19-44 ppt. The type locality of the species, the Straits of Magellan (Siddall, 1980), demonstrates its ability to tolerate lower tempera- tures. We expect the dispersal of Perna perna to the north of Port Aransas to continue, driven by deviations from sea- sonal patterns in current direction and spawning cycles, and increasing sources of recruitment, as offshore oil produc- tion platforms become colonized, but at a much slower rate than has been observed to the south. ACKNOWLEDGMENTS We thank B. Hardegree for assistance in the field; N. Sohn, R. Smith, N. Barrera, and S. Alvarado for assistance in the laboratory; and M. K. Wicksten and N. Clements for relating findings of mussels. The manuscript benefited from review by two anonymous reviewers. Current and continuing research is being funded in part by a grant from the Texas A & M Sea Grant College Program. LITERATURE CITED Berry, P. F. 1978. Reproduction, growth and production in the mussel, Perna perna (Linnaeus), on the east coast of South Africa. Investigational Report, Oceanographic Research Institute, South African Association for Marine Biological Research 48:1-28. Britton, J. C. and B. Morton. 1989. Shore Ecology of the Gulf of Mexico. University of Texas Press, Austin, Texas. 387 pp. Boyd, T. M. & W. J. Wardle. 1994. Occurrence of the red alga Lomentaria baileyana and associated rocky intertidal macroalgae from the mouth of the Colorado River, Texas. Texas Journal of Science 46(2): 190-193. Davenport, R. 1994. Additional records of Perna perna (Linne, 1758) on the Texas coast. Texas Conchologist 30(1):3-4. Davenport, R. 1995. Perna perna enters the bays. Texas Conchologist 31(3):92. Giese, A. C. 1959. Comparative physiology: annual reproductive cycles of marine invertebrates. Annual Review of Physiology 21:547- 576. Harger, R. J. 1972. Co-existence: maintenance of interacting associations of the sea mussels Mytilus edulis and Mytilus califorianus. The Veliger 14(4):387-410. Hicks, D. W. and J. W. Tunnell, Jr. 1993. Invasion of the South Texas coast by the edible brown mussel Perna perna (Linnaeus, 1758). The Veliger 36(1):92-94. Loosanoff, V. L. and H. Davis. 1963. Rearing of bivalve molluscs. Jn: Advances in Marine Biology. F. S. Russel, ed. pp. 1-136. Academic Press, New York. Lunetta, J. E. 1969. Reproductive physiology of the mussel Mytilus perna. Biologia Faculdade de Filosofia, Ciécias Universidade de So Paulo 26:33-111. Paine, R. T. 1976. Size-limited predation: an observational and experi- mental approach with Mytrilus-Pisaster interaction. Ecology 57:858-873. Siddall, S. E. 1980. A clarification of the genus Perna (Mytilidae). Bulletin of Marine Science 30(4): 858-870. Vakily, J. M. 1989. The Biology and Culture of Mussels of the Genus Perna. ICLARM Studies and Reviews 17. International Center for Living Aquatic Resources Management, Manila, Philippines, and Deutsche Gesellschaft fiir Technische Zusammenarbeit (GTZ) GmbH, Eschborn, Federal Republic of Germany. 63 pp. Velez, A. and C. E. Epifanio. 1981. Effects of temperature and ration on gametogenesis and growth in the tropical mussel Perna perna (L.). Aquaculture 22:21-26. Watson, R. L. and E. W. Behrens. 1970. Nearshore surface currents, southeastern Texas Gulf coast. Contributions in Marine Science 15:133-143. Wiley, G. N., R. C. Circé, and J. W. Tunnell. 1982. Mollusca of the rocky shores of east central Veracruz State, Mexico. The Nautilus 96(2):55-61. Date of manuscript acceptance: 19 May 1995 AMERICAN MALACOLOGICAL UNION FINANCIAL REPORT General Accounts 1994 Income and Expenses BEGINNING BATE AING (127 3 193 csicsicsrra valscyensavanineeess sys viaas sted eaaetenn seasass sani vqetsanad wideavean avenestanveeniveseuntyeeds $96,570.12 TING OME roccrecoectece toe cucen cans et eaaaisncatewsbveiiensd vdicadedste yak saienes aeesedesscaduai beau diviancsogs Gganed eibcanaasaie seautesustusevenssuaies $27,852.16 IMAG tie SLi a DOS acess seca ath xcs cde av pn adc naese ews seth steecevtece cancsteveenuece 15,658.99 DB Ce) 0 oh ee a aE PEPPY rn fe eM na eT re eR Ey Serf ere erC 2,682.28 General Bank ACCOUMNL..................cccccccssscccssssscssssscsssstsrcceccescesessescescssscees 1,238.11 Symposium Endowment Fund Accountts...........cesessceseeceseeseeeeeeeseeeeeeees 1,444.17 Publications: INCOME... ..cccsseeccssesssedosscucessessaucdesssucaccessliocesvsncescdesessalacbsebacessavesess 7,110.17 BAITS S WOSCLIP UL ONS edie 05 sus -assscasdodsevess cvedu se tosvevstipeactave save tcevessaassusintanenees 2,853.00 AME Back ASSUCS cticescsss ssveveasesdacseiristsatitsstaesicvassenevaasliniir eh, eta iaenoiniuast: 674.00 AVES P ABC CHALGES seuseishrsssoensunsussasiessasssdesiussosasisstesueanssacdasaesilieluatsassnadbass 1,898.00 AMBRE print CMAP ES fo xccyicdsdet veces ccasacescenescacaniessdensics tas usshcbaceccnecvetenuseaes 1,656.12 Salessor Other Publicanioms civav.citirvesviesscwenisinioaceen inane 29.05 1903: Animal Mice tlie sees ccvsten sec cost shaven: cbeaiccwncerpasadthasccsiavetest cat scaseeniaeeawessecsasess 2,653.85 ING GOCE a2 20 cna acca oe cscs fs Foca Gu bamae ena deanenaen eee cleccnaitecdapeovetetenin 58.62 Auction (to Symposium Endowment Fund)........... ces seseeseeseeeseeeneeeeees 1,808.10 Donation, Student Award ..........ccccccccccccccccesscsceesssesssssscsscsesccccossecceessseseeeens 250.00 Donation, Symposium Endowment Fund............scesssscssssessesecssoeenceseeeseseesaeeaees 272.00 Miscellaneous: TiC One cco cacexceocxcucs cas Seuace ses tocevaseucdvadinestevsscccnsasdzacavessseviseiosesrocenss 12.00 EXPENSES oso cs cei esse cece ps tas v asco os cca cea ecte oaat en teae cei tah cate eh deexchxoasavidad aceues ups tease ssaboaencan} eapiaedtasesseusniencinessiyens $22,596.34 OTEICESEX PCH SES ceed sees cree isons 288 teas ccpivn test yt sc scccsnsen athss cauueiossp coy eesssecaasancrsglens 569.76 Bvt I SC OSES eipiccecenecs see cecetcc gee speaace st cepsastacs tenses sets maesnwaeagoevaraluasacue sous aupeummcavaves 3,147.20 | okiesy{oucay. Viv domes eavveremererieeayecereeprr ets wrrrrer tere pereetre ere Toeeeret eererrr rer ery reee 1,729.42 POStA RC = RS mr Unt as cuieesyssesoes cases iaesa cto savasesTeeeansepecadacatauaieertseea satin oeaneaas 273.34 Postage, = AIvi) ING wW SIOte re s.r Saxe encase ws ees ads aatektocisaes 675.24 Postage = Miscellame CUS 22 .occsesccvccsteassenabeieexoecastesacet sSerexeteneaseseaceesaetivanes 283.05 MSTIVC LOGS cc ssai esses ectders crear rccet Vana oarkiovey vite teitn ys 1 aacaaaae sven ae ney udaaavaNaTe 186.15 PTH Fate IWECTIDETSINDS es eorscseausscastatceustaasieceasulviuiellacecevac setae es sccthennaismeepn ees 230.00 Poa COS soca secagaseeceee cece cd rece cc core econ eee ss gles sob osawaecudaedanesiiuoasleusedsoeoversctestecs! 384.25 BI COLPOL AU OM Oeics sanaseasast a + ’ ~ 1. °o oe ¥ 7 ” ' 7 4 ~ . 4 er “d A t i n . Pca oa - an a ‘s £44 | i A . s + o 4 = a 4 ° c = < 4 \WQ =< GE B48 S < . 5 <(e& 2) 4 S oc S WY = & S| ac Ss a \ 1] 5 = SY 8 wy 5 zs 5 eS zZ =) ee a) z 2 Zz - Zz > SJIMYVUGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVYSIT LIBRARIES = - = - z - z Lic 4 ° = 2 aes) ° = ° = 2 5 2 > 2 3 a a 2 5 t = b - F = ; = b z o See z i; 2 B Z 1 INSTITUTION NOILALILSNI NVINOSHLINS S3IYVYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI z a 77) za no i ar 2) =a ee n =z Zz &. =z < = a Cor =< &. = S = SS = ay = Zz AS Ss Zz SX = z O § © MR rz LY Ce) <= ODO Wd) &®& L O ; Y Re ae [e) 2NXK 3 GFF 2 g ZR 8 ZN 8 2 = 2 “UY = 2 2 » 2 E 2 E 7 > ‘ p> — 7 *. > : a = 5 s 2 = a. = ? = > S3IMVYGIT. LIBRARIES SMITHSONIAN _INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVYGIT LIBRARIES | ” 2 a 2 fd 2 ” 2 nt oc a o . 5 a = = = z < ay md > = _P = > 2 = 2 : ad a Za = 2 mi on a wo ma n° : =n wn fal 7) = wo ag = 7 = 7 = oO $5 S3IYVUGIT_LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31y¥VvygIt 1) ae 2 Zz “ z no ; Zz | = . < = < = < = , > = > 2 ae D> > = FE = 5 = 5 = i: = a Z o z , o “Zz o =z o z \ INSTITUTION NOILALILSNI NVINOSHLINS S3IYVYGIT LIBRARIES SMITHSONIAN t = Ne: no z n z sys 77) z (77) za < = < Ss r = \S = < = = z lie yee, Wade a 4 tf fy =F WS 4 z= 4 z = 2 UW : 2%y = MG S > : = _ > = > = (72) : rd (77) =a (72) - Pr (Tp) z ” S pS atYVUGIT_ LIBRARIES SMITHSONIAN INSTITUTION NOILMLILSNI_NVINOSHIWS Sal¥VYGIT_LIBRARIES = 3 S 6 i a # & a oc = a. an oc a ww S = o < 4 Pass 4 > = 2 Ae =) Wz is > = > 2 = z E : © HL? e 5 (7p) Ww —— n* rj (7) o z on™ -z a z A z He S sazluvugi7_LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS Sa1uvUugIT_LIBRARIES § 7) ; a a7) = 3 wn = n : Zee no = gs 5 Zz a 2 INIAN INSTITUTION NOILONLILSNI NVINOSHLINS S3IYVYUGIT LIBRARIES SMe INSTITUTION NOILNLILSI - = - z = = 5 = ° = Ae Oo a ° ~o ° by “3 = 2 WS E 2 = LG My “2 E OC: OO gi: x E pa, = aime - UG, pes = m 2 m fe m 9 «= E = 7) . Z wo = z ILIWS S3 1uvyad ryt BRARI ES SMITHSONIAN PCOS NT NINDS ES luvda ee 1B RAR Ihe = ff = Qs = = ‘= = ae ae = ANS 2 Gifp iN. Z 2 il yy 5 Bip BSS 8YG4S RW 3 2 y SU ge 2 ORR 2 Wy = WS 2 2 Uy = 2 “gy = SS - 7 . 2 2 ee = 2 : SJIYVYSIT LIBRARIES SMITHSONIAN INSTITUTION NOILALILS INIAN INSTITUTION NOILNLILSNI NVINOSHLIWS LIBRARIES SMITHSONIAN — w —_ — w > 2) .o us & b wo AG = iy o = a ma a «. t% ff a a pi < 3 z Sig z : ya be 5 a «OF 4 = rs on ro) rs) be UP re a = o Pra pe Fat as Zz ae IWS SS3tavyuygiT LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNI_NVINOSHLINS S3IYVYUdIT LIBRARIE 6 = ae ae S ao z c = ow oat “M ow pene ow > — ow > Be) as Ze 2 os B) \, es = VE : 2 LC : an. Ee S Se: e EAP? f = \S"'5 = | a (7) = 77) = n ‘ = w INIAN INSTITUTION NOILALILSNI NVINOSHLINS LIBRARIES SMITHSONIAN INSTITUTION NOILALILS:! "e 2) = wn = i. ” Pa ie wn x =] Fa =] z = z \, qj 4p, XX 2 8 z eS 2 1s NQd WF a ‘3 sie! oar oO Ba re) rT WONG 6 hed Mee, WO" Z S Z FE 2 E Nw 227 7 Yr CS = >" = = Ss NY 5 =z ” a ” Fa a e S S3tYVYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILOLILSNI_ NVINOSHLINS S3IYVYGIT LIBRARIE Zz < = re) 77) x e = 7) ” = = = 7) = & s é é : j co = w p a a 5 oa = < 4 < ia < pea << 4 oc = ac ¢ oc = a c ry =! oo. = ry = ra) = — ro) oa re) — ° aS — re) - Fa 3 2 a 2 a = INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVYUGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILS e z - aa z a = is z ‘oO = = NS Oo wo = ‘ pees S) 2D ‘ SO ox = 41 Oe NSO HOR 7 = r = = Gap = = m = m » G m dy “fl m .D: ” wo = 7) . = wo = w z= IWS SAlEWYATI_LIBRARIES SMITHSONIAN gy NOILALILSNI NVINOSHLIWS SaluWuall_LIBRAR lf ey Bh 56 WR = = 4 S = BD hy S WO 0D VF DMD W™~ no a Lv ys a WH Oo Mi <= \.. oO oO ; = (e) Witty y 3& \ \. 2% E » 2 2 Uy * 2 “yp = Ss Beene _ 3 Sen ee 2 a INIAN INSTITUTION NOILALILSNI NVINOSHLINS S3IYVYUGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLIL = < a ro) D bs 8g = = ” — ” can ” — = wo a8 us 2 ui 5 uw AG us hyp > a = = = a a e = = c = c <> c Pa an 0 a rs = - = © o = ro) Gan ro) ca Yi rs) ee 2 -! Fa 4 z = z - HUINS SSIYVYGIT LIBRARIES NOILNLILSNI NVINOSHLINS S3IYVYEIT LIBRARII rm re = is = "+ a a = ow = o — ies] > = ive) S 2 = _ wv 5 2 Ww 5 = ANGE = = .> = SF \AQe = a \ = 2 — 7 — a WS Ps) SE = a = ms = WY G = ie 2 Z 2 Z aw 2 NIAN INSTITUTION NOILALILSNI_ NVINOSHLINS S31uvugIT LIBRARIES SMITHSONIAN INSTITUTION NOILONLILS a ae = = < = a yom se & 2 gon ha Sl adied) 2H. Oe epinow rt fp sida Po Auer tthe Pee Mlaean ti wedtial i us hell ee ibd fh ca fetiy b i tui fe ake iN OF (ttt aa vote kee Ey het H it Map nae i Por Urea ne sans state uk eco chon dahuh st hil i. a4 ot be a seneuked hint rate su fi aie iae Ning aticeapatert iit pevek Pearse Pee a eh Bon ee Bod Me Bo Hi nflak ot 1! 8 jaf verge Pal nbai bel "4 bdo BEY By nt eae) Lae rapes nats trated ine ; Fpreas ibe 3 i LPitaated i Rod chapad ain ars cia bespioua ben 3 i ititkeumaces Site Rea vetialey yd we ved, otteh: ¥ BDihemets Flbsanigbub siete td ihesaaeagtebl han i My PP ab di alk fied Wubi ferries aioe iis Sinbau i ES vipisn pea tpg adi ete inte Wat Do 4 Mheaeabe th iaieaeseth Breer voit We ee Matatioterpiats: wh ye (ne ae ‘ Poy se a ay Miuiettut oat ayn ne en. pf one we 1 f en A edd 3) gel oA F Mie paid oe He § f ay i bathers ay Oe a mar dee wider Regen ede getten foe a aC AVRO Re at Sent ee ite ee vow vw hbet on CRS PME Won ate tyne tie eae Hh Bisbeat) te Oi rere ee ee Or AT OM ae Tomeqewcin tp ove 4 epee ncn b er er ae Cae dbl os RON Sat re ee ST Ese Falie a TaF Des et Sees OSI aan Drag tate eines Poe hee ESD iiuesati et bs Ets > Ree Syste wae WP 7 Re * : te Lysine nts in ies ED: eee orion mary curie BD RRMA, LT MTU SI TIE tors Peta tanths Seance prem te chapee ty pape etn Sere mys Fore eet ates eect as Sn Pre or u ly an eget nes twem tym NS et eal hatatibsh site aeRO ENNIS * Matinee = Ane Hy ore Puss Salita? ig = rane yihs a met ey tarp shel Canty f PEN . Ry Pyeh et vyinn a) Benuinal Db wane Bento! , rea ae Ea rath Saini ty ties Try titty Mt sep ee tii BO ge aes yee aot Tete sest key SO CEnn ATS mcr Men Praia vhs ote oF i nV Peyleyact Ge Witte wl 3 9088 Tt l Sieapen tr : DVrytasn ge ae Orne akc vs i yee shaty + a typhtus eer ecu Nyrka ere Ey eurwunntea es pier ww ogee de rane sgene satu Sh 50 eaten are eae soa . lates meena uneeyes pon Menten pasate ete y PORE NICS : si dyrergre sere ns wreaTuna _ Sine casts Su ie BMH aracnnewrdenenree nniTatt 5 on nieratas eo) v \ wie : i Lesnar Mies ttf We ate : Suncare a ysanaurieynaneencinsMryen ses M uate weuens’ Dhar acuatetdane wet aioascrasiie WAAL AV eME RS HERE \ , Soaneueserssmenryny nese A Ahan a Wei Satetatarame eon ! math uray Vatmaterke at Paes Way moving us 288 5 tra fen Le re S CaN tase oni ie Vanya yew X ay ; tiruayresiys d aise! MP te a camacrean atts Sart wk HE * Sy Momebe nea, tu wivew Megadeth) Pune IT “ ; : linn mH MEN TOU MD Tawerpens wereld) t Tur sat \ uit i ; ¥ % Penn Ree RE Hopes were rery SCeeckeye ke yliyr vets ern iy weaned ena uraricwiueysess sw b csuare sarah “ ct Shit . iene es a hy Fe Ye Pay aa UHL WIPE ant ¥ i PODER Via Grew ote EOS tue Pa Aa foyg Meet ururNay arUnAthe teen stay satpunentacu cn mney eumeeae tke enti Rageuetetneathe lV Alay hd Ta foe Rt Ue A Oe De a eS AE at Kher Us entrain a Munn pepe ater titers ee CEUEU ea neha g Manda ESL TRAYS TEAL LEW IEE RE Ea atc aA res : ra 4) a fh ni : tp yenee eon agers Ait tele ty a f a i nee eh ee Cee LUE R ae ; Mita tates fehed i a nibs i) es yeettiturem surneative gy oeatutatucareur cn ta ie Fane i ba neat tiennie eee urna usta LAE GEES DC St ; : Pe pean Peo ify : Rona nicer eiueyeny ie ; Sie! Wivatayunteat sue Piatatita kot : Ort Wren 4, VED ae BY Real) uiitusitar papoua HiTtUsstarurag ca" SUS HE teesb mat Ce eihubyrnre am vena EE ett i 2 i whe t : Woiryr arte ty oi ae i Hi ! h AHA wae HS TR Ae a , Ut « H Poavar ten sists te CSU nev aUsoe cc Peau kr eae 0" Asana Or aahder aa Parse ie saci rie Vesoer bid try i » Ngee ty RAHN ERE inti atk! » eon t) he Chi A H ate iy Perigen BM rie hn Ata (ea 3 ROR CHE : 3 i AE Ee VANE, p) hres sh ine Tek. He | i i maa i zy Ue i x i i i f fa ait ty il get hen PDA Rh Hn a sy fe \ Baa ed » (est A Rata va pti if 4, i orlizih ei uerst that ; Yh Ta Webi hae \ fy