BP PS Myer 4 Tene F » pall INN CE SAP Ma Peve! Yee Pine paneres Tocca westiny Saothe Med & : * 3 tH iavetied } p Ad ay ween ssi eepacirni : : : : pa a ea Ne Keke 8 7 NOE 1h 4 Tit " cre Re ath Stat 0S AD gh ay Ne) ey Be hetd r Se : . 5 : 5 1B .att GNM Pav OTN S ay ChE WAVER We oun b esti ni 4 x ‘ FEN thes regs bug 2 Nata HAiniccneese eee ; f : : penne belctey bash Dalit Rares ‘ : e Ceeanae ee Dea Hews) Raped re . Mee SAS ey EN STS RENTS ON Nm Vow Ile act GF - eT Cg. ot ; A ". Se a Seri X 3 2 : i iW aay pirate sTHBhS TY Sv Fi * y : : tite a fe ORM a sis MENT FET ee ae 3 heties ai neers ; : F th Oe F “ ORS A 5 anc 4 ae pes eh pias ‘ : Taotdnayee (uN git HUE EAI Dan Metin deg bre PIAA My ON Fy my, : 2 me ¥ eveniman Cones aa ose Paar aaa 5 wipainareey eee Me tree tm e - F 2 : ees Cran a eS Seg ‘ F : Sty ts? iN: rents 5 heels A Oe i a - Fe Pe ee An Gy WAR eh Ure IU Sidhe My, TSN GSES SSR em ent ret t Wty : : Hi yee v8 PS Se Mv vat SWS AT rc Sais TS - - ae iN Pyne ‘ eet es “ y 4 x Rivtee"es: eA Mate GE Nahe : : a : Milt BASS 2a HM a OE 2 ¢ i : 2 ; my Me a : i : : i : Spe niastetny enters Mal q c fe ete, iNilietia yt Bsiby 4 ( Br dete 6 By Ry PY SES ES atats Ga ; ; oes shoul ators : shawn - ig aid tory Sey (irvine BNET AYDE ae We ester Weeden ts Z AD ect tabi fie eal Se eae eel etal hy Be dot ps ae eis su eal Ut ks een es 3 Dip ae : LONG : AR tears net y ees fe Pasta os wa PN ges mane 5 soe = aeS Moho re wandnstads Ween re & Bey Peeteetit oe Tee TSE tee Sree te. eA ; ONIN of ped an gee Netessan 5 Prior Te Aa re ot : d - Jae : Palast et 2 4 2 ol 5, fo ANS Yel eyo Btn oe oun eb. srnigin Wanye J ae Nese na Oty gly aad wt SN Riy erecta eat ne Mt ae Med ce APS a Fen? s Sete ete POP seth tee gre bs, Foe Ef OD aD en nary PROCEEDINGS of the Biological Society of Washington VOLUME 108 1995 Vol. 108(1) published 24 March 1995 Vol. 108(3) published 19 September 1995 Vol. 108(2) published 22 June 1995 Vol. 108(4) published 28 December 1995 WASHINGTON PRINTED FOR THE SOCIETY EDITOR C. BRIAN ROBBINS ASSOCIATE EDITORS Classical Languages Invertebrates GEORGE C. STEYSKAL JON L. NORENBURG FRANK D. FERRARI RAFAEL LEMAITRE Plants Vertebrates DAVID B. LELLINGER THOMAS A. MUNROE Insects WAYNE N. MATHIS All correspondence should be addressed to the Biological Society of Washington, Smithsonian Institution Washington, D.C. 20560 ALLEN PREss INC. LAWRENCE, KANSAS 66044 OFFICERS AND COUNCIL of the BIOLOGICAL SOCIETY OF WASHINGTON FOR 1994-1995 OFFICERS President JANET W. REID President-Elect STEPHEN D. CAIRNS Secretary CAROLE C. BALDWIN Treasurer T. CHAD WALTER COUNCIL Elected Members ROBERT J. EMRY SUSAN L. JEWETT RICHARD C. FROESCHNER LYNNE R. PARENTI ALFRED L. GARDNER F. CHRISTIAN THOMPSON TABLE OF CONTENTS Volume 108 Albert, James Sperling and Robert Rush Miller. Gymnotus maculosus, a new species of electric fish (Chordata: Teleostei: Gymnotoidei) from Middle America, with a key to species: of Gymnotus 22 ee Ee a ee ee Se ee eee Bayer, Frederick M. A new species of the gorgonacean genus Narella (Anthozoa: Octocorallia) from: Haw aiwan< waters ss ed ee eee ent Brown, Rafe M., John W. Ferner, and Luis A. Ruedas. A new species of lygosomine lizard (Reptilia: Lacertilia: Scincidae: Sphenomorphus) from Mt. Isarog, Luzon Island, Philippines ..22 bo a ee Brown, Rafe M., John W. Ferner, and Rogelio V. Sison. Rediscovery and redescription of Sphenomorphus beyeri Taylor (Reptilia: Lacertilia: Scincidae) from the Zambales Mountains of [uzon; Philippines!2 = 2s ee ee Brown, Walter C. A new lizard of the genus Sphenomorphus (Reptilia: Scincidae) from Mt. Kitanglad, Mindanao, Philippine Islands ________--_-___--_-___---- Brown, Walter C. and Ely L. Alcala. A new species of Brachymeles (Reptilia: Scincidae) from’ Catanduanesylsl anes) P baal p pony cs eee ce eae ee ae eee ee Bruce, Niel L. and Jgrgen Olesen. Natatolana nukumbutho, a new species (Crustacea: Isopoda: Cirolanidae) from deep water off Suva, Fiji — Cairns, Stephen D. New records of azooxanthellate stony corals (Cnidaria: Scleractinia and Stylasteridae) from the Neogene of Panama and Costa Rica Campos, Martha R. A new species of freshwater crab of the genus Strengeriana from Colombia (Crustacea: Decapoda: Pseudothelphusidae) -__-__------_---_----_--_----_------- Campos, Martha R. and Gilberto Rodriguez. Two new species of freshwater crabs of the genus Hypolobocera from Colombia (Crustacea: Decapoda: Pseudothelphusidae) Campos-da-Paz, Ricardo. Revision of the South American freshwater fish genus Sternarchorhamphus Eigenmann, 1905 (Ostariophysi: Gymnotiformes: Apteronotidae), with: notes on its relationships 222° 6) Sees eee a ie ee eee Carleton, Michael D. and Guy G. Musser. Systematic studies of oryzomyine rodents (Muridae: Sigmodontinae): definition and distribution of Oligoryzomys vegetus RB earns 1D) an IS ST RIT AN le Couri, Marcia Souto. Contribution to the knowledge of Reynoldsia Malloch (Diptera: VAS eae) 2 a 2 SoH Nd Tcl eh So a a each A eM ee LL Cuatrecasas, Jose. A new genus of the Compositae: Paramiflos (Espeletiinae) from (Gro) (05101 5) b: Pane Ne ON en ee Oy Ae On Bey ure aA VE is ee ee he Cumberlidge, Neil. Remarks on the taxonomy of Sudanonautes chavanesii (A. Milne- Edwards, 1886) (Brachyura: Potamoidea: Potamonautidae) from Central Africa __. Cumberlidge, Neil. Redescription of Sudanonautes faradjensis (Rathbun, 1921), a fresh- water crab from Central Africa (Brachyura: Potamoidea: Potamonautidae) Dean, Harlan K. A new species of Raricirrus (Polychaeta: Ctenodrilidae) from wood collected in the Tongue of the Ocean, Virgin Islands Erséus, Christer and Olav Giere. Otavius nicolae, a new gutless marine tubificid species (Oligochaeta) ‘frome li Ze i ICN Felder, Darryl L. and Raymond B. Manning. Neocallichirus cacahuate, a new species of ghost shrimp from the Atlantic coast of Florida, with reexamination of N. grandimana and N. lemaitrei (Crustacea: Decapoda: Callianassidae) —__---.--- Ferrari, Frank D. Six copepodid stages of Ridgewayia klausruetzleri, a new species of copepod crustacean (Ridgewayiidae: Calanoida) from the barrier reef in Belize, with COMMENTS onFappEendagerde velop mve mt ee ee tae ee Goldman, David A. A juvenile of the scaled squid, Pholidoteuthis adami Voss, 1956 (Cephalopoda: Oegopsida), from the Florida Keys Harold, Antony S. and Richard Winterbottom. Gobiodon acicularis, a new species of gobioid fish (Teleostei: Gobiidae) from Belau, Micronesia iV 662-678 147-152 18—28 6-17 388-391 392-394 212-219 533-550 98-101 649-655 29-44 338-369 281-291 748-750 238-246 629-636 169-179 491-495 477—490 180—200 136-146 687-694 Heyer, W. Ronald. South American rocky habitat Leptodactylus (Amphibia: Anura: Leptodactylidae) with description of two new species —____----________ Hobbs, Horton H., Jr. and H. H. Hobbs III. Procambarus (Ortmannicus) nueces (Decapoda: Cambaridae), a new crayfish from the Nueces River Basin, Texas ____ Hobbs III, H. H. and Horton H. Hobbs, Jr. Macrobrachium catonium, a new troglobitic shrimp from the Cayo District of Belize (Crustacea: Decapoda: Palaemonidae) ___. Hotchkiss, Frederick H. C. Lovén’s law and adult ray homologies in echinoids, ophiuroids, edrioasteroids, and an ophiocistioid (Echinodermata: Eleutherozoa) ____. Hsieh, Hwey-Lian. Laonome albicingillum, a new fan worm species (Polychaeta: Sabpellidae-sSaveliinae) trond dhabw an ore eatemntwa ee hs Slee Ee a Fe Jiménez, Maria, G. San Martin, and E. Lopez. Pionosyllis maxima Monro, 1930, P. anops Hartman, 1953, and P. epipharynx Hartman, 1953, redescribed as Eusyllis maxima (Monro, 1930), a new combination (Polychaeta: Syllidae: Eusyllinae) ___ Komai, Tomoyuki. A new species of the genus Discorsopagurus (Crustacea: Decapoda: Paguridae) from Japan, previously known as D. schmitti (Stevens) —------------_-- Kornicker, Louis S. and Dale R. Calder. Hydroids colonizing the carapaces of the ostracode Philomedes brenda from the Beaufort Sea, Arctic Ocean _-_-_------ Kritsky, Delane C. and Walter A. Boeger. Neotropical Monogenoidea. 26. Annulotrematoides amazonicus, a new genus and species (Dactylogyridae: Ancyrocephalinae), from the gills of Psectrogaster rutiloides (Kner) (Teleostei: Characiformes: Curimatidae) from the Brazilian Amazon —_- Kropp, Roy K. Lithoscaptus pardalotus, a new species of coral-dwelling gall crab (@iustacea> Brachyura; Cryptochiridac) from’ Belaw = Lemaitre, Rafael. Charybdis hellerii (Milne Edwards, 1867), a nonindigenous portunid crab (Crustacea: Decapoda: Brachyura) discovered in the Indian River lagoon system ORE LECTERNS ee i A ee RE TN. a Lépez-Gonzalez, Pablo J. and Mercedes Conradi. Heteranthessius hoi, a new species (Copepoda: Pseudanthessiidae) from a sea-anemone in the Straits of Gibralter with Be AAR SR OME CHS EC MU Ses waiter cae AL cat au el Lc ure ee Baa A a Bs Ve RE Lotufo, Guilherme R. and John W. Fleeger. Description of Amphiascoides atopus, a new species (Crustacea: Copepoda: Harpacticoida) from a mass culture system ___ Malabarba, Luiz R. and Andreas Kinder. A new species of the genus Bryconamericus Eigenmann, 1907 from southern Brazil (Ostariophysi: Characidae) —__--__-_-_-____- Martin, Joel W. and Jennifer C. Christiansen. A new species of the shrimp genus Chorocaris Martin & Hessler, 1990 (Crustacea: Decapoda: Bresiliidae) from hydrothermal vent fields along the Mid-Atlantic Ridge —-_____ McClure, Matthew R. Alpheus angulatus, a new species of snapping shrimp from the Gulf of Mexico and northwestern Atlantic, with a redescription of A. heterochaelis Savemlolondoccapoda:, Candea: Alpheidac)) sa se McLaughlin, Patsy A. and Janet Haig. A new species of Goreopagurus McLaughlin (Decapoda: Anomura: Paguridae) from the Pacific, and a comparison with its Atlantic SOTTO as cP A MR A SR. RE a Re” Messing, Charles G. Alloeocomatella, a new genus of reef-dwelling feather star from the tropical Indo-West Pacific (Echinodermata: Crinoidea: Comasteridae) = Messing, Charles G. Redescription of a unique feather star (Echinodermata: Crinoidea: Comatulida: Comasteridae) with the diagnosis of a new genus __... Miura, Tomoyuki and Daniel Desbruyéres. Two new species of Opisthotrochopodus (Polychaeta: Polynoidae: Branchinotogluminae) from the Lau and the North Fiji Bac- Alcea aSinS WSOUthWeESterlybaCiiC, OCEAN yw ee ea Ng, Peter K. L. and Takeharu Kosuge. On a new Somanniathelphusa Bott, 1968, from Vietnam (Crustacea: Decapoda: Brachyura: Parathelphusidae _____»_»_____ Norris, James N. and David L. Ballantine. Two new species of the red alga Chrysymenia J. Agardh (Rhodymeniales: Rhodymeniaceae) from the tropical western Atlantic __- Olson, Storrs L. and Helen E James. Nomenclature of the Hawaiian Akialoas and INU PIS CAVeSss Dre pani Gini) == eset 2 cine eine ee erode ee ge a es ee Ott, Jorg A., Monika Bauer-Nebelsick, and Veronica Novotny. The genus Laxus Cobb, 1894 (Stilbonematinae: Nematoda): Description of two new species with cctosyimbiotic: Chemoautotrophic bacteria 228 eave Ty See 1) ee 695-716 54-60 50-53 401-435 130-135 496-501 617-628 125-129 528-532 637-642 643-648 107-116 117-124 679-686 220-227 84-97 68-75 436—450 656-661 583-595 61-67 153-165 373-387 508-527 Vi Palma, R. Eduardo. The karyotypes of two South American mouse opossums of the genus Thylamys (Marsupialia: Didelphidae), from the Andes, and eastern Paraguay Pamplona, Denise and Marcia Souto Couri. Morellia dendropanacis, a new species, and other species with spotted wings: characterization and comparison (Diptera: Miuscidae: Mii Sciniae)) settee 2 ios et cas Rd oe Patton, James L. and Maria Nazareth F da Silva. A review of the spiny mouse genus Scolomys (Rodentia: Muridae: Sigmodontinae) with the description of a new species from. the ‘western Amazon: of: Brazil. 2.22: ee ee Pettibone, Marian H. New genera for two polychaetes of Lepidonotinae _________ Pohle, Gerhard and Fernando Marques. First zoa of Dissodactylus glasselli Rioja and new range and host records for species of Dissodactylus (Brachyura: Pinnotheridae), with a discussion of host-symbiont biogeography -_-----------------------_______________------ Rambla, Juan Pablo Blanco, Ildefonso Lifiero Arana, and Luis Cal Beltran Lares M. A new calianassid (Decapoda: Thalassinidea) from the southern Caribbean Sea ______. Rausch, R. L. and V. R. Rausch. The taxonomic status of the shrew of St. Lawrence Island Bering, Seax(ViammaliaS Sorncidac) Roccatagliata, Daniel and Richard W. Heard. Two species of Oxyurostylis (Crustacea: Cumacea: Diastylidae), O. smithi Calman, 1912 and O. lecroyae, a new species from the Gulf of Wile xtc ee a ae ea ee tone ee Rodriguez, Gilberto and Austin B. Williams. Epilobocera wetherbeei, a new species of freshwater crab (Decapoda: Brachyura: Pseudothelphusidae) from Hispaniola _______ Rédriguez-Almaraz, Gabino A. and Thomas A. Bowman. Sphaerolana karenae, a new species of hypogean isopod crustacean from Nuevo Leon, Mexico —_.________. Ross, Charles A., Gregory C. Mayer, and Roger Bour. Designation of a lectotype for Crocodilus siamensis Schneider, 1801 (Reptilia: Crocodylia) —----------------_------------ Ruedas, Luis A. Description of a new large-bodied species of Apomys Mearns, 1905 (Mammalia: Rodentia: Muridae) from Mindoro Island, Philippines ___-_-_-----_--_--____- Russell, David E. Description of a new viviparous species of Dentatisyllis (Polychaeta: Syllidae) from Belize with an assessment of growth and variation, and emendation of CEN fA UNS ae We es AN re el 8 BN La co ao Schultz, George A. Sinoniscus cavernicolus, a new genus and species of terrestrial isopod crustacean from a cave in China (Styloniscidae: Oniscidea) _____-____________ Simone, Luiz Ricardo L. Rissoella ornata, a new species of Rissoellidae (Mollusca: Gastropoda: Rissoelloidea) from the southeastern coast of Brazil —----______ Stark, Bill P. A new species of Neoperla (Insecta: Plecoptera: Perlidae) from Mississippi Tan, Cheryl G. S. Dittosa, a new genus of leucosiid (Crustacea: Decapoda: Brachyura) from southern Australia and New Zealand Thomas, James Darwin and Manuel Ortiz. Leucothoe laurensi, a new species of leucothoid amphipod from Cuban waters (Crustacea: Amphipoda: Leucothoidae) __ Thompson, Fred G. A new freshwater snail from the Coosa River, Alabama (Gastropoda: Prosobranchia: Hydrobiidae) Todaro, M. Antonio. Paraturbanella solitaria, a new psammic species (Gastrotricha: Macrodasyida: Turbanellidae), from the coast of Californian Turner, Richard L. and Robyn M. Heyman. Rediagnosis of the brittlestar genus Ophiosyzygus and notes on its type species O. disacanthus (Echinodermata: Ophiuroidea: Ophiomyxidae) based on the type specimens from Japanese waters and new, materialstromithesGullfso fMex 1c @ yee ee en Vazquez-Bader, Ana Rosa and Adolfo Gracia. A new crab species of the genus Pseudorhombila H. Milne-Edwards, 1837 (Crustacea: Decapoda: Goneplacidae) __ Wicksten, Mary K. Wlthin-species variation in Periclimenes yucatanicus (Ives), with taxonomic remarks on P. pedersoni Chace (Crustacea: Decapoda: Caridea: Palaemiomidae) 25st ode oe le Re A ON Pe a i eS Williams, Austin B. and Fred C. Dobbs. A new genus and species of caridean shrimp (Crustacea: Decapoda: Bresiliidae) from hydrothermal vents on Loihi Seamount, Hawait __.22008 tts Secale she Er Rea Sd 8 Rr ee pe A, A PTE Ot Zug, George R. and Ivan Ineich. A new skink (Emoia: Lacertilia: Reptilia) from the forést of Fajitas. 3 ust, dante ie aor SUN ie lito Sean aE ee ads eo Zunio, Gabriel E., Olga B. Vaccaro, Marcelo Canevari, and Alfred L. Gardner. Taxonomy of the genus Lycalopex (Carnivora: Canidae) in Argentina —-_- 451-457 319-337 577-582 247-253 102-106 717-729 596-612 76-83 207-211 298-301 302-318 568-576 201-206 560-567 45—49 465-476 613-616 502-507 553-559 292-297 254-265 458-464 228-237 395—400 729-747 INDEX TO NEW TAXA VOLUME 108 (New taxa are indicated in italics; new combinations designated n.c.) CNIDARIA Anthozoa PMA INGN SA CITY S102 7; C2 CL LES pees eee as es ea Seth Sag ee a J a 536 ETO SSTIMUT, WAR GIA CT AIS a ge ee re a 541 INE GN OIA LATD as ASE SV SIO ee 148 ESE AG AGN UES NCECL CLO Spot eine Pater ee ee IN oe OE la ee) 538 Sg STIMt ABH OC C111 CCl mca aaa Da a Ti sa a a aS eg ee Se 539 SEOUL, LILES TGS es a ee a a 534 PLATYHELMINTHES Trematoda PANPIRTEL OD LTCTIULLOUCLC Settee ree rasa Te ee a a ee 529 CLO LLY DI OVID. Sa BEI IIIA, APD LED ELLEN DILL ALL DE LOO ELE AD ALTE AEA, EN 529 GASTROTRICHA Reate ct GUNTUL ATV NAS LEE CA gd Co ete ies ea 554 NEMATODA LEASES CDM LY DOT TIA so 1 a I ef A ee SIL7/ OLCUS TLE S puerta re snk RARE Ea Si rel ronan rem se ada min ohare mc Svs a micas, PS PRESEN SL NY, Bae 509 MOLLUSCA Gastropoda O pisthomochOpoGuUsise ONL GCI oe ss ae ee eae a ae 584 AY YIU IS asa SN ts I ENE SES aera SAPS RR OS Ae SU A a i Taya rege Re 588 | PRETO] SINS BT LIAS I GI Ee RA Re el ec 502 BRASS Oe lam (ERAS SOe IA) 1511 CE Ca seers Is ee se ee 561 ANNELIDA Polychaeta PETE TEL CPLA OTL OLUES pete ea aE Nad a esd ee Si, (INSTEON DNA, Sj ha gen ST A a 578 Sra ALIS NNT S Wrst OCLLES pre meatier eee I ee ees ee Ee 569 | EUAN TORRES SUD T AY (1 Chg LN a eo OTe 496 EAC R LDL CLI OTL irr pmtietaa toes aeniemn rsa ele en eee eat Ee 131 ONDA DIL D YONI: a ET eB nae eB Ee 580 | UWP VET, TO Oy eS I I eS a a ve 580 j PEW EF CETURTAINSS LUA ZY UTI tas IE I os I Te aS a 169 Oligochaeta CON ER IES 1K CUL AAG goa 8a TEIN SIE RS IT EP ee vga den 491 Vill ARTHROPODA Crustacea PN Tp SATU QTL ATUL S cp a te 6 ee AmmphiaSCO1deSs’ QLOp uss 222 ie ST i a a ee ee (COLO CATES Tf OMT GL Ge T E Discorsopaguius maclaughlinae —— ee aa ee ee DUI OS Goo oN a OE Rive ptt BEA EME NIE STE AE eae ek FAR Sy EAE AEE ENE OR OE OO RE Ra Sia nS ae Epilobocerawetherbeeiy Soret es er a GOLEOPASUNUS are I Ea Ie EE see F¥eteramthe sss (Oe 22s aa ee a eat nee ee ne Eby POlODOCE Ra CHD CIC TaN ts oe ae NE Ie eco EN Ea en ere ee ene RATES GUAT SL a i an at SER CRA RAL Oe Sn ra Ree ae ea a ee IECUCOENOC QUT CHS a eh eee Ne see en eae ee FvehO@SCAptuS: Parada OL1US ee sae ee ae are tle ee TO gee le ee ee IMlaACE@ Ora Lanta CLONE LET ITS mise I ee ce ee ae nan ene Ce oer ee Oxyurosty lis WecrOy ae oe eee ce ee Procambarus (Ortmannicus e7ueces si eee 0 ee eat Eseudornombilayoriet arti ee re oe a ee RIA 8 CWAyias KIQUSTUCTZVC TE 2-2 Sin LL NE NG aS ee ee a eed SEL SION SUOLGUCTE, Hk SO ss ad SE RN ee VILCV UCC AC nN AR Ee ea Dl SETLOMLES CUS Ae Docc 8 Es eh SS TED AE AR Ee se se Soman miatine lps apa xe eI ce Sphaerolanma karen <0 ea eae ge StKEMP ERAN A /OTEM Cia mL A EE Teg eae ae, Morelliadendropanacising. 28 8 06Gs Dies Se eden RU rare 2 ON eae ee ee INe@o@perl aC niet iis ae (Ue Ik 9 eg SS ee SE a erie ee VA LOC OCOniate Ta 2 te x 5 Beet eG eal al A ale il st TO alca A a PS CET UL Trea ot: Ce a I eee Sees Ne NE PCN oc VOWS CLG chp to tae NIE a lc ad oi ae SLO ZAP ILAN OC OTIGS ECT, = he A Sa ea SAC al S18 Tat Ut NO Nae, re RE Sea eo CHORDATA Pisces Bryconamericus lambari Gobiodon acicularis Keptodactylus*lithonaetes: wi sesarince. Sy Nineties eine ee ee eel ee LOLA AY Se Ra eRe ti eet a ee A ene) eee en eee og. fe eae ee AN ee en Bach y ime les” 172trerrii ves mse ene eke ee ee LP CE RR TEE ERE elie OE ee EMOla INOKOSATINIVCIKGU ee ee a eee ee tet ere SMe MOM Op SACL SLA CUS 1S ee we LB Ie PN UA OUST DLL LG ssc 20 Ss Nt ADS Nae RE nT ee eS ANI OWID GY ca atl Bs EOI VE ESS NP eC SO PRIS AT eR ee ESS TRE TOO oS NS TE Pa aT a ed oe a dl A ON STO en Ung PUL | ENTS TNO al SE OR Ll ee RL gd COOLS CU Trea NF epee ee eerste ee ns BE Le a OE A ng a RS Ee © SG Tel IMs © spree oe ae ce a ee (Us BUNA SST Sy a Ie ee a eee Mammalia ENSOLEDSIS IRGC MITOSIS AE at eg he ed ee SCO MOMMY SW LEMLECLCIUS Cima sie wena IRE Lee a ene EB CG intgy SW MTN TNA ELE C11 CTU Cas te eee eal a a ee a TU OO CLUL ONS (a me ne De SIA SAL PS SARI AE Ca ea I a I a de seo a i PECDIROLT TG DS, Na Tel WON OR A ae aL ete EE Ten ce STRING UNOS UI S UMC nypeeereto cers Dae nce PN a ee ee ee 384 385 385 1X ee (oO bu, VOW a * oy , & 0 é & Rae VOLUME 108 NUMBER 1 24 MARCH 199 ISSN 0006-324X THE BIOLOGICAL SOCIETY OF WASHINGTON 1994-1995 Officers President: Janet W. Reid Secretary: Carole C. Baldwin President-elect: Stephen D. Cairns Treasurer; T. Chad Walter Elected Council Robert J. Emry Susan L. Jewett Richard C. Froeschner Lynne R. Parenti Alfred L. Gardner F. Christian Thompson Custodian of Publications: Austin B. Williams PROCEEDINGS Editor: C. Brian Robbins Associate Editors Classical Languages: George C. Steyskal Invertebrates: Jon L. Norenburg Frank D. Ferrari Plants: David B. Lellinger Rafael Lemaitre Insects: Wayne N. Mathis Vertebrates: Thomas A. Munroe Membership in the Society is open to anyone who wishes to join. There are no prerequisites. Annual dues of $25.00 (for USA and non-USA addresses) include subscription to the Proceed- ings of the Biological Society of Washington. Library subscriptions to the Proceedings are: $40.00 for USA and non-USA addresses. Non-USA members or subscribers may pay an addi- tional $25.00 to receive the Proceedings by Air Mail. The Proceedings of the Biological Society of Washington (USPS 404-750) is issued quarterly. Back issues of the Proceedings and the Bulletin of the Biological Society of Washington (issued sporadically) are available. Correspondence dealing with membership and subscriptions should be sent to the Biological Society of Washington, P.O. Box 1897, Lawrence, Kansas 66044, U.S.A. Payment for membership is accepted in US dollars (cash or postal money order), checks on US banks, or MASTERCARD or VISA credit cards. Manuscripts, corrected proofs, editorial questions should be sent to the Editor, Biological Society of Washington, National Museum of Natural History, Smithsonian Institution, Wash- ington, D.C. 20560. Known office of publication: National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. Printed for the Society by Allen Press, Inc., Lawrence, Kansas 66044 Second class postage paid at Washington, D.C., and additional mailing office. POSTMASTER: Send address changes to PROCEEDINGS OF THE BIOLOGICAL SOCI- ETY OF WASHINGTON, P.O. Box 1897, Lawrence, Kansas 66044. This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(1):1-5. 1995. The karyotypes of two South American mouse opossums of the genus Thylamys (Marsupialia: Didelphidae), from the Andes, and eastern Paraguay R. Eduardo Palma Department of Biology and Museum of Southwestern Biology, University of New Mexico, Albuquerque, New Mexico 87131-1091, U.S.A. Abstract. —The karyotypes of Thylamys macrura, from eastern Paraguay and adjacent Brazil, and 7. pallidior, from the Andean Altiplano in Argentina and Bolivia, are reported for the first time. The karyotypes of somatic tissues from female 7. macrura and female T. pallidior consist of 2N = 14. However, the diploid number found in somatic cells in males of 7. pallidior exhibits 2N = 13 because the Y chromosome is missing. The missing Y phenomenon might represent another case of chromosome mosaicism in American marsupials, although this time detected in didelphids. Tate (1933) recognized five species groups in the genus Marmosa Gray, 1821 (sensu lato) based on phenetic characters. Subse- quent morphologic, chromosomic, and se- rologic studies have shown that these as- semblages approximate genera (Reig et al. 1985, 1987; Gardner & Creighton 1989). Among these, Thylamys Gray, 1843, has the southernmost distribution of mouse opossums in South America encompassing elevations as high as 3500 m, as well as lower temperate and subtropical habitats (Cabrera 1958). Thylamys differs from oth- er mMarmosines in having a characteristic tri- colored dorsal pattern, and the capacity to store fat in the tail (Tate 1933, Mann 1978) Recent revisions of the genus recognize five species (Gardner & Creighton 1989, Gard- ner 1993): Thylamys pusillus (Desmarest 1804), Thylamys macrura (Olfers 1818), Thylamys elegans (Waterhouse 1839), Thy- lamys velutinus (Wagner 1842), and Thy- lamys pallidior (Thomas 1902). Thylamys pallidior occurs on the rocky slopes of the Altiplano of Bolivia and Argentina (Tate 1933; fig. 1), whereas, the subtropical form T. macrura 1s found in the humid forests of eastern Paraguay and adjacent Brazil (Gard- ner 1993; fig. 1). This constitutes the sixth published record of this species for Para- guay, since the original description by Olfers (1818) based on Azara’s (1801) ““Micouré a queue longue,” or ““Colilargo” (Azara 1845). Tate (1933) referred to this species as Mar- mosa marmota, and Cabrera (1958) as Mar- mosa grisea. The name macrura (Olfers 1818) is available and 7. macrura is used by Gardner (1993). In this study, the kar- yotypes of 7. macrura and T. pallidior are reported for the first time. This is the first study documenting the absence of one of the sex chromosomes of somatic cells in didelphid marsupials. Methods Chromosomal preparations were ob- tained directly from bone marrow using the standard velban technique described in An- derson et al. (1987). A minimum of 10 metaphase spreads were counted for each specimen. Nomenclature for chromosome morphology and fundamental number (FN) follows Patton (1967). Five specimens of 7. pallidior from two localities in Bolivia were analyzed. The karyotype of 7. macrura was 2 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 1000 km 80 70 60 50 100 90 Fig. 1. Map showing the distribution of T. pallidior along the Andes of Argentina and Bolivia, and 7. ma- crura in eastern Paraguay. The boxes and dot represent the collecting site given in the text. obtained from a specimen from eastern Par- aguay. Voucher specimens, chromosome slides, and cell suspensions are deposited in the Museum of Southwestern Biology (MSB) and the American Museum of Natural His- tory (AMNB). Collection localities are (Fig. 1): Bolivia: Department of Chuquisaca, Ca- margo, 68 km (by road) N of Camargo, 3400 m, 20°09’S, 65°17'W (3 males MSB 57003, AMNH 262406, and AMNH 262407; 1 fe- male AMNH 262405); Department of Tar- ya, Serrania Sama, 3200 m, 21°27’S, 64°52’W (1 female, AMNH 263555). Par- aguay: Department of Concepcion, Escuela Agropecuaria, 7 km (by road) NE from Con- cepcion, 23°21'S, 57°23’W (1 female, MSB NK 27536). Results and Discussion The autosomes of 7. macrura (2N = 14, FN = 20; Fig. 2a) consist of three pairs of large submetacentrics (1-3), one pair of me- dium-sized metacentrics (4), and two pairs of small acrocentrics (5-6). The X chro- mosome is a small acrocentric. The auto- somes of T. pallidior (Fig. 2b) are not dis- tinguishable from those of 7. macrura, al- though the three males of the Andean spe- cies present 2N = 13, FN = 20; the Y chromosome was absent in all counted plates. The female 7. pallidior exhibited the complete set of chromosomes, 2N = 14. A male of 7. macrura was not available for karyotype. The autosomic complement of the species of Thylamys reported here is similar to those documented previously for other species of the genus, such as T. elegans from Chile and Bolivia, which possess the identical three group pattern of autosomes and morphol- ogy (Reig et al. 1972, Palma & Yates 1995). Micoureus cinereus and M. constantiae have similar diploid and fundamental number (2N = 14, FN = 20; Palma & Yates 1995). Marmosa (sensu stricto), Marmosops, and Gracilinanus, have 2N = 14, however the fundamental number in these taxa is FN = 24 (Reig 1968, Palma & Yates 1995). The mouse opossum karyotypes present- ed in this paper reinforce the concept of chromosomic conservatism in marsupials, and support the fact that marsupial species that occur in remarkably different habitats share a common karyotype (Reig et al. 1977, Hayman 1990). The common 2N = 14 is shared by most of the marmosines in the Neotropics. Marmosa canescens (2N = 22), is the only known exception (Engstrom & Gardner 1988). — The conclusive evidence of the absence of the Y chromosome in T. pallidior is dif- ficult to determine under the methodology followed in this study. It is possible that the Y has been translocated to another chro- mosome, or this condition may be another example of chromosome mosaicism, 1.e., a difference in sex-chromosome presence be- tween the germ line and cells of the somatic tissues (Hayman 1990). Similar patterns of absence of the Y chromosome for somatic VOLUME 108, NUMBER 1 b) Fig. 2. a, Standard karyotype of a female Thylamys macrura from Concepcion, Paraguay, 2N = 14, FN = 20. b, Standard karyotype of a male Thylamys pallidior from Camargo, Chuquisaca, Bolivia, 2N = 14, EN 20: cells have been found for Chironectes min- imus (Palma & Yates 1995) and Dromiciops australis (Gallardo & Patterson 1987). Ad- ditional reports of chromosome mosaicism have been also made for Australian mar- supials of the family Peramelidae, where one of the X chromosomes is missing in somatic cells of females, and in Petauroides (Pe- tauridae), where the Y chromosome is miss- ing from the majority of cells obtained from bone marrow (Murray et al. 1979, Hayman 1990). The missing Y chromosome found in Dromiciops australis (Gallardo & Patterson 1987), caused these authors to suggest that the microbiotheriid would be more related to Australasian marsupials than to Ameri- can metatherians, supporting Szalay’s (1982) contention that Dromiciops and Australian marsupials constitute the cohort Australi- delphia. Data from this study and from Pal- ma & Yates (1995), prove that the Y chro- mosome is missing not only in Dromiciops and Australasian marsupials, but in Amer- ican marsupials as well. This scenario fits a typical case of parallelism or represents a plesiomorphic condition in the evolution of metatherian sexual chromosomes of both geographic regions. Hence, this character cannot be used as evidence for inferring phylogeny between Australian and Ameri- can marsupial lineages. Acknowledgments Mike Bogan, Jennifer K. Frey, Jorge Sa- lazar, and Terry L. Yates provided valuable comments on a draft of the manuscript. Special thanks to Lucy Aquino from CITES- Paraguay, and Carl Shuster; to Flaviano Colman and Luis Moran of the National Museum of Natural History of Paraguay, Maribé Robles, and to the National Parks Office of the Ministry of Agriculture and Livestock of Paraguay. Funding support was provided by a NSF Research Improvement Grant PDS-105-774, the Latin American Institute, University of New Mexico, and Sigma Xi. Field work in Bolivia was sup- ported by NSF grants BSR-83-16740 to the American Museum of Natural History (S. Anderson), and BSR-84-08923 to the Uni- versity of New Mexico (T. L. Yates). The map was prepared by Beth Dennis. Literature Cited Anderson, S., T. L. Yates, & J. A. Cook. 1987. Notes of Bolivian mammals 4: the genus Ctenomys (Rodentia, Ctenomyidae) in the eastern low- lands.—American Museum Novitates 2891:1- 20. Azara, F., d’. 1801. Essais sur l’histoire naturelle des quadrupédes de la Province du Paraguay. Trad- uits sur le manuscript inédit de l’auteur, Pra. M. L. E. Moreau-Saint-Meéry. Charles Pougens, Paris, 366 pp. 1845. Viajes por la América del Sur. Bibli- oteca de Impresos Raros Americanos. Univer- sidad de la Republica, Facultad de Humani- dades y Ciencias, Instituto de Investigaciones Historicas, Montevideo 1982, 318 pp. Cabrera, A. 1958. Catalogo de los mamiferos de América del Sur.— Revista del Museo Argenti- no de Ciencias Naturales ‘““Bernardino Riva- davia”, Ciencias Zooldgicas 4:1-307. Desmarest, A. G. 1804. Taubleau méthodique des mammiféres. Pp. 5-38 in Nouveau dictionnaire d@histoire naturelle, appliquée aux arts, princi- palement a l’agriculture, a l’@conomie rurale et domestique: Par une société de naturalistes et d’agriculteurs: Avec des figures tirées des trois regnes de la nature. Vol. 24. Paris, Deterville. Engstrom, M. D., & A. L. Gardner. 1988. Karyotype of Marmosa canescens (Marsupialia: Didelphi- dae): a mouse opossum with 22 chromo- somes.— The Southwestern Naturalist 33:230- 233. Gallardo, M. H., & B. D. Patterson. 1987. An ad- ditional 14-chromosome mosaicism in South American marsupials. Pp. 111-115 in B. D. Pat- terson & R. M. Timm, eds., Studies in Neo- tropical mammalogy: essays in honor of Philip Hershkovitz.—Fieldiana: Zoology, new series 39:1-5S06. Gardner, A. L. 1993. Order Didelphimorphia. Pp. 15-23 in D. E. Wilson & D. M. Reeder, eds., Mammal species of the world: a taxonomic and geographic reference. Second edition. Smith- sonian Institution Press, Washington, 1206 pp. ——., &G.K. Creighton. 1989. A new generic name for Tate’s (1933) microtarsus group of South American mouse opossums (Marsupialia: Di- delphidae).— Proceedings of the Biological So- ciety of Washington 102:3-7. Gray, J. E. 1821. On the natural arrangement of the vertebrose animals.— London Medical Reposi- tory 15: 296-310. . 1843. List of the specimens of mammalia in the collection of the British Museum of London. British Museum (Natural History), 216 pp. Hayman, D.L. 1990. Marsupial cytogenetics. — Aus- tralian Journal of Zoology 37:331-349. Mann, G. 1978. Los pequenos mamiferos de Chile. — Gayana, Zoologia 40:1-342. Murray, J. D., G. M. Mckay, & G. B. Sharman. 1979. Studies on metatherian sex chromosomes. IX. Sex chromosomes of the greater glider (Mar- supialia: Petauridae).—Australian Journal of Biological Sciences 32:375-386. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Olfers, I. 1818. Bemerkungen zu Illiger’s Ueberblick der Sdugthiere, nach ihrer Vertheilung tuber die Welttheile, riicksichtlich der Sudamerikanisch- en Arten (Species). Pp. 129-137 in Journal von Brazilien, oder vermischte Nachrichten aus Brasilien, auf wissenschaftlichen Reisen ges- ammelt, W. L. Eschwege, 129-137 in F. J. Ber- tuch, ed., Neue Bibliotek der wichtigsten Reise- beschreibungen zur Erweiterung der erd- und Volkerkunde; in Verbindung mit einigen an- deren Gelehrten gesammelt und herausgegeben. Weimar: Verlage des Landes-Industrie-Comp- toirs. Vol. 15, 304 pp. Palma, R. E., & T. L. Yates. 1995. The chromosome of Bolivian didelphid marsupials. — Occasional Papers, The Museum, Texas Tech University, Lubbock (in press). Patton, J. L. 1967. Chromosome studies of certain pocket mice, genus Perognathus (Rodentia, Het- eromyidae).— Journal of Mammalogy 48:27-37. Reig, O. A. 1968. The chromosomes of the didelphiid Marmosa robinsoni Bangs.—Experientia 24: 185-186. — , R. Fernandez, & A. Spotorno. 1972. Further occurrence of a karyotype of 2n = 14 chromo- some in two species of Chilean didelphoid mar- supials.— Zeitschrift fir Saugetierkunde 37:37- 42. —, J. A. W. Kirsch, & L. Marshall. 1985. New conclusions on the relationships of the opos- sum-like marsupials with an annotated classi- fication of the Didelphimorphia. —Ameghi- niana 21:335-343. —, , & . 1987. Systematic relation- ships of the living and Neocenozoic American ““opossum-like”” marsupials (suborder Didel- phimorphia), with comments on the classifica- tion of these and of the Cretaceous and Paleo- gene New World and European metatherians. Pp. 1-89 in M. Archer, ed., Possums and opos- sums: studies in evolution. Surrey Beatty and Sons and the Royal Society of New South Wales, Sydney, 1 and 2, 800 pp. — . A. L. Gardner, N. O. Bianchi, & J. L. Patton. 1977. The chromosomes of the Didelphidae (Marsupialia) and their evolutionary signifi- cance.— Biological Journal of the Linnean So- ciety 9:191-216. Rofe, R., & D. Hayman. 1985. G-banding evidence for a conserved complement in the Marsupi- alia. — Cytogenetics and Cell Genetics 39:40-50. Szalay, F. S. 1982. A new appraisal of marsupial phylogeny and classification. Pp. 621-640 in M. Archer, ed., Carnivorous marsupials. The Royal Biological Society of New South Wales, Sydney, Australia, 802 pp. Tate, G. H. H. 1933. Systematic revision of the mar- supial genus Marmosa, with a discussion of the VOLUME 108, NUMBER 1 adaptive radiation of the murine opossums (Marmosa).—Bulletin of the American Muse- um of Natural History 66:1-—250. Thomas, O. 1902. On Marmosa marmota and ele- gans with descriptions of new subspecies of the latter.—Annals and Magazine of Natural His- tory, series 7, 9:408-—410. Wagner, J. A. 1842. Diagnosen neuer Arten brasi- lischer SAugethiere. — Archiv Naturgesch 8:356- 362. Waterhouse, G. R. 1839. Mammalia. Pp. 49-97 in C. Darwin, ed., The zoology of the voyage of the H. M. S. Beagle under the command of Cap- tain Fitzroy, R. N., during the years 1832-1836. Smith, Elder and Company, London, 97 pp. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(1):6-17. 1995. Rediscovery and redescription of Sphenomorphus beyeri Taylor (Reptilia: Lacertilia: Scincidae) from the Zambales Mountains of Luzon, Philippines Rafe M. Brown, John W. Ferner, and Rogelio V. Sison (RMB) Department of Zoology, Miami University, Oxford, Ohio 45056, U.S.A.; (JWF) Department of Biology, Thomas More College, Crestview Hills, Kentucky 41017, U.S.A.; (RMB, JWF) Department of Vertebrate Zoology, Cincinnati Museum of Natural History, 1720 Gilbert Avenue, Cincinnati, Ohio 45202-1401, U.S.A.; (RVS) Zoology Division, National Museum of the Philippines, Executive House, P. Burgos Street, Manila, Philippines Abstract. —Sphenomorphus beyeri Taylor, 1922, long known from a single specimen, is redescribed on the basis of newly acquired material from the herpetologically unsurveyed area of the Zambales Mountains, west central Lu- zon Island, Republic of the Philippines. Phenotypically most similar to its distantly allopatric congener S. diwata, the holotype and our new series are distinguished from other Philippine Sphenomorphus by the combination of their small to moderate size (SVL = 46.6-67.1 mm), fused frontoparietals, 88— 96 paravertebrals, 38-42 scales at midbody, 19-21 subdigital fourth toe la- mellae, and unique coloration. In his monograph on the lizards of the Philippine Islands, E. H. Taylor (1922a) recognized 16 Philippine species in the ge- nus Sphenomorphus (Lacertilia: Scincidae). He later supplemented this work on Phil- ippine lizards with descriptions of two more scincid species in a more general herpeto- logical contribution (Taylor 1922b). One of these was Sphenomorphus beyeri, a skink described on the basis of a single specimen found on Mt. Banahao in the Laguna Prov- ince of southern Luzon. Brown & Alcala (1980) later recognized 22 species of Phil- ippine Sphenomorphus. One of these was S. diwata Brown & Rabor (1967), which the authors hypothesized to be the closest rel- ative of S. beyeri. Brown & Alcala reported S. diwata from the Diwata mountains of northern Mindanao Island, and placed it and S. beyeri into the two-species non-phy- logenetic couplet “Group I Sphenomor- phus’’ (1980:154). As presently understood, the genus Sphenomorphus contains over 120 species world-wide (Myers & Donnelly 1991) with 60 species in the Oriental and Australian zoogeographic regions alone (Brown & Alcala 1956, 1961a; Greer & Par- ker 1974; Greer 1979). Sphenomorphus beyeri has, to date, only been known from the incomplete holotype, yet it invariably has been recognized as a valid species. In the course of a recent in- ventory of the fauna of the Philippines un- dertaken by the Cincinnati Museum of Nat- ural History (CMNH) and the National Mu- seum of the Philippines (PNM), we cap- tured 16 additional specimens of S. beyeri in the mossy cloud forests of the Zambales Mountains of west central Luzon. Collec- tion of this new material provided us with a unique opportunity to analyze intraspe- cific morphological variation of this endem- ic Philippine skink and to reconsider the validity of its specific rank. VOLUME 108, NUMBER 1 Methods We conducted field studies in the Zam- bales Mountain range from 17 February to 18 March 1992. Drift fences and pitfall traps (25 m of 0.65 ml black plastic, stretched to 15 cm above ground, supported with wood- en stakes, with =50 cm deep plastic-lined pits spaced every 5 m for a total of six pit- falls per 25 m of drift fence at each 100 m interval on slope) and time-constrained searches were used to collect specimens. Specimens were photographed, then fixed in 10% buffered formalin; notes on color- ation, behavior, and habitat (including el- evation) were recorded at time of capture. Upon return to the U.S.A. (approximately one month later), specimens were trans- ferred to 70% ethanol. Detailed examination of all material was conducted at the Cincinnati Museum of Natural History and at the National Mu- seum of Natural History. When possible (see character definitions below), we took mea- surements and scale counts following tech- niques detailed in Brown & Alcala (1980); illustrations of head scalation were made (by RMB) with a Wild microscope equipped with a camera lucida attachment. Characters (measured to the nearest 0.1 mm) are defined as: snout-to-vent length (SVL), from tip of animal’s snout to caudal margin of anal scale; tail length (TL), from caudal margin of anal scale to tail’s tip (specimens with regenerated tails are not included in statistical analyses); axilla-groin distance (AGD), from caudalmost point where forelimb meets body to anteriormost point where hind limb meets body; hind leg length (HLL), from point where rear limb meets body to tip of longest (=4th) toe; head length (HL), from tip of snout to caudal edge of tympanum; head breadth (HB), width of © head at its widest point (=ocular region) when viewed from above; snout length (SL), from anterior edge of bony orbit to tip of snout; eye diameter (ED), horizontal di- ameter across bony orbit; and tympanum 7 diameter (ITD), horizontal distance across the tympanic annulus. Lateral head scales (e.g., labial scales) were examined on both sides of the head and numbers for each side are given separately with a dashed line (—) designating left from right respectively. Meristic and mensural data are given as means + standard devi- ations (SD) and range. Statistical analyses were carried out using the Statistical Analysis Software Program (SAS), version 6.03 (SAS Institute Inc. 1988a, 1988b), using UNIVARIATE pro- cedure for standard statistics. Significance of moment statistics (skewness, g,, and kur- tosis, g,) was calculated by hand (Sokal & Rohlf 1981:174—-175). Specimens are deposited in the California Academy of Science (CAS), Cincinnati Mu- seum of Natural History (CMNH), and Na- tional Museum of the Philippines (PNM). Material Examined includes: Holotype (Lu- zon Island, Laguna Province, Mt. Banahao) CAS 61183, immature male, collected on a rock ledge on Mt. Banahao at 1500 m by E. H. Taylor (Taylor 1922b:285). Six fe- males (CMNH 3652, 3653, and 3658; PNM 2307, 2301, and 2304), nine males (CMNH 3655, 3657, and 3659; PNM 2300, 2302, 2303, 2305, 2306; USNM 337768) and one immature juvenile or hatchling (CMNH 3654), all collected by RMB and JWF. All were taken from Luzon Island, Zambales Province, Municipality of Masinloc, Bar- angay of Coto, Zambales Mountain range, Mt. High Peak. Study Sites Until the present study, (PNM/CMNH Philippine Biodiversity Inventory) the Zambales Mountains (Zambales Province, Municipality of Masinloc) were completely unsurveyed herpetologically due to a com- bination of major insurgency in this area and its close proximity to the Subic Bay Naval Base, 70 km south in the town of Olongapo. Following the eruption of Mt. Pinatubo in 1991 and subsequent closing of 16° N— SOUTH CHINA SEA IS? IN = Fig. 1. The Zambales Mountains. Shaded area rep- resents the major geologic formation constituting the Zambales Mountain range (see text). Darkened circle represents Mt. High Peak, and encompasses both lo- calities sampled. the United States military facilities at Olon- gapo, an opportunity was made available to naturalists to survey this largely unknown region. The Zambales Mountains (Fig. 1) are an isolated coastal mountain range encom- passing an area of approximately 6960 km? on west central Luzon; the Zambales range extends from west of the Lingayen gulf to the western border of Manila Bay at an ori- entation of 20° West of due North. To the east, the Zambales Mountains are isolated from the nearest mountain range, the Cor- dillera Central, by the Plains of Tarlac (also known as the Pampanga Plains). To the west and north, these mountains end in the South China Sea; to the south they separate Subic Bay from Manila Bay. The Zambales Mountain range currently is contiguous with mainland Luzon, connected to the latter by the wide Plains of Tarlac, believed to have PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON been submerged below the Luzon sea during the Pliocene and Pleistocene (Rutland 1968; Hashimoto 1981la, 1981b; Auffenberg 1988). Geological evidence suggests that the Zam- bales Mountains were formerly isolated much farther west of their present position and were surrounded by water—not con- nected to mainland Luzon as they are today (Dickerson 1924, Auffenberg 1988). The Zambales currently represent a large “‘in- sular’’ massif, geologically isolated from the three other principal montane regions of Luzon (Bureau of Mines 1963; UNESCO/ ECAFE 1971; Hashimoto 198la, 1981b; Auffenberg 1988). Site 1.—Zambales Mountains, 1100 m, 15°35'N, 120°09’W. Sampled from 17 to 26 February, the site is a tropical moist decid- uous forest type (Whitmore 1984); virgin timber (predominantly Myrtaceae, Laura- ceae, and Tiliaceae) begins at an elevation of ca. 1030 m. Presence of Mt. Pinatubo ash deposits throughout the area combined with noticeable signs of rattan gathering efforts undertaken by local residents precludes des- ignation of the area as “undisturbed.” Site 2.—(Fig. 2) Zambales Mountains, 1500 m., 15°30'N, 120°08’W. Sampled be- tween 11 and 18 March 1992, this area lies within the broad category of tropical moist deciduous forest type but remains, none- theless, more typical of the upper montane (=“‘mossy’’) rain forest (Whitmore 1984) due, presumably, to its somewhat higher el- evation. Virgin timber predominates (Fa- gaceae, Myrtaceae, Magnoliaceae, and Pin- aceae) and very little bamboo or other sec- ondary or disturbance indicator species were encountered. A more extensive description of these study sites and a discussion of their geologic importance is provided by Ruedas et al. (1994). Results Capture data.—A single specimen of S. beyeri collected at Site 1 was a mature male VOLUME 108, NUMBER 1 most specimens of Sphenomorphus beyeri were found. The vegetation and forest floor in this area were coated with approximately 2—4 cm of volcanic ash from the eruption of Mt. Pinatubo. Photo from a color transparency by RMB, courtesy of CMNH. (PNM 2300) collected under leaf litter on the forest floor at 1265 m. When disturbed, this lizard became alert, moved in a rapid serpentine manner and attempted to bur- row under debris and leaf litter. A pair of pitfall trapped specimens (CMNH 3652, a gravid female; PNM 2301, a mature female without eggs) from Site 2 were captured dur- ing the day, at 1510 m, and 1610 m, re- spectively. The rest of the new series was captured by splitting open rotten logs lying horizontally in contact with the forest floor. One specimen (male, PNM 2305) was dam- aged when the machete used to split the log struck it. No specimens were taken at night despite extensive search efforts. Sphenomorphus beyeri Taylor, 1922:283 Fig. 3, 4 Diagnosis.—A small to moderate-sized Sphenomorphus, S. beyeri is readily distin- guished from its congeners by the following combination of characters: (1) prefrontals moderate, usually separate; (2) frontopar- ietals fused except in immature specimens; (3) usually 6-7 labials; (4) four large supra- oculars (5) paravertebrals 88—96; (6) scales around midbody 38-42; (7) fourth toe la- mellae 19-21; (8) body proportion ratios as follows, SL/HL = 0.25-0.37; SL/HB = 0.38- 0.57; HB/HL = 0.60-0.69; HB/SVL = 0.13- 0.17; ED/SL = 0.30-0.43; (9) unique col- oration and color pattern. Description (based on holotype and 16 re- ferred specimens). — Details of the head sca- lation of an adult male (PNM 2302; cap- tured by RMB on 16 March 1992 in mon- tane cloud forest at 1460 m, between 1400 and 1630 hr) are shown in Fig. 3 from dorsal (A), ventral (B) and lateral (C) perspectives. Head scalation.— Rostral 1.2-2.9 (X = 2.2 + 0.4 SD, n = 14) mm wide (holotype 10 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. Dorsal (A), ventral (B), and lateral (C) per- spectives of head scalation of Sphenomorphus beyeri (PNM male 2302; HL = 12.0 mm). Arrow indicates the presence of the azygous interprefrontal scale on snout. = 1.8). Frontonasals 1.7-2.4 (X = 2.2 + 0.3 SD, n= 14) mm wide and 1.1-1.6 (¥ = 1.3 + 0.2 SD; n = 14) mm long (holotype = 1.9 x 1.1). Ten of 17 specimens (including holotype) with separate prefrontals, four specimens with prefrontals in narrow to moderate contact, and three with prefron- tals separated by an azygous interprefrontal (Fig. 3). Interprefrontals wider anteriorly; their caudal tip projecting slightly beyond prefrontal/frontal suture. Frontal triangular, longer than wide, nar- rowed to a point caudally; in contact with two anterior supraoculars; length 2.8-3.7 mm (X = 3.3 + 0.4 SD, n= 14) width 1.7- YS mmm OF = 20) ae 0.4 SD, a= 1) io lotype = 3.1 x 1.9). Frontoparietals fused; D3 .¢\ won OF = D& +2 O.4 SID, m= 1.2) long and 2.5-4.3 mm (X = 3.5 + 0.4 SD, n = 14) wide (holotype = 2.7 x 3.2). Pa- rietals behind interparietal in contact for a distance shorter than the parietal itself. In- terparietals 1.5-2.8 mm (X = 1.9 + 0.5 SD, n = 14) long and 1.2-1.6 mm (X¥ = 1.3 + 0.2 SD, n = 14) wide (holotype = 1.7 x 1.3). Nasal large and single with nostril at cen- ter, bordered posteriorly by two pairs of overlapping (superimposed) loreals; most dorsal larger than ventral. Holotype with only 10 supraciliaries; supraciliaries varying considerably among and between our series of specimens: 1 1 — 12 (two specimens), 12— GD ISO GY), WAS Oy, WAH M3 (@), 13—12 (2), 13—13 (3), 13—14 (1), 14—13 ()yand 14= 141() x66 OnO = 0.9 SD; range = 11-14—11-14; n = 16— 16. All specimens with four large supraocu- lars; anterior most strongly triangular; the second widest. Caudalmost supraocular fol- lowed by three or four curved rows of very small scales, each with two or three scales. Tympanum exposed, not strongly de- pressed or sunken. Holotype with 6—7 up- per labials and 6 —6 lower labials. Our series varies considerably in both upper and lower labial scale counts: Upper labials: 6—5 (one specimen), 6—6 (10), 7—7 (4) or 7—6 (1); X—=(63— 6.6 = 05 11S srances—s6= 7—S-7; n = 16—16. Lower labials: 5—5 (1), 5—6 (1), 6—6 (12), 6—8 (1), 7—7 (1), or 8—7 (1); X = 6.3—6.4 + 1.0—1.5 SD; range = 5-8 —5—-8; n = 16—16. Mental chin scale 2.3-3.4 mm (X = 2.8 + 0.3 SD, n= 14) wide and 1.5-2.4 mm (X = 1.9 + 0.3 SD, n = 14) long (holotype = 2.6 x 1.6); followed by one postmental, the latter in contact with two lower labials. Dorsal scalation.—Nuchal scales undif- ferentiated. Scales (transversely) around midbody = 38 (two specimens), 39 (6), 40 (S, including holotype), 41 (1), and 42 (2); X = 39.9 + 1.17 SD, n= 16. Paravertebrals = 88 (one specimen), 89 (2), 90 (2), 91 (2), 92 (6), 93 (1), 94 (holotype), and 96 (2); X = Wiles) am AS esvOh Tp = 10, Subdigital lamellae and ventral scala- tion. — All specimens with long digits, as re- flected in lamellae scale counts. Both the holotype and our specimens display a_lon- gest to shortest toes rank of 4, 3, 2, 5, 1. Holotype has 19—19 fourth toe lamellae, whereas our series with 19—19 (two spec- imens), 19—20 (1), 20—20 (5), or 21—21 (8); X¥ = 20.3—20.4 + 0.8—0.7 SD; range VOLUME 108, NUMBER 1 11 Table 1.— Measurements (in mm) from all known specimens of Sphenomorphus beyeri (character abbreviations in text). Standard univariate statistics, presented below, are abbreviated as follows: X, mean; SD, standard deviation; W, Shapiro-Wilk test for normality (N = normal); g, = skewness statistic; g, = kurtosis statistic. None of the skewness or kurtosis statistics were found to be significant (P > 0.05). CMNH 3654 and the holotype were excluded from the analysis as they were not sexually mature. Tail length (TL) also was excluded from the analysis as several individuals showed scars indicative of caudal autotomy and regeneration. Character Specimen # Sex SVL TL AGD HLL HL HB SL ED TD CMNH 3652 f 59.4 76.5 31.4 29.5 12.3 7.8 31) 2.3 1.3 CMNH 3653 f 67.1 Udo 35.0 21.9 9.2 12.9 8.9 2.8 tod CMNH 3654 v 358) 44.1 14.8 123 8.1 5.3 Dog 1.8 le CMNH 3655 m 64.9 90.8! 31.3 24.3 14.4 9.8 4.7 3.1 2.0 USNM 337768 m_ 63.4 56.4! 31.8 24.3 1350 8.7 4.0 Srl 15 CMNH 3657 m 66.7 100.0 34.6 25.0 14.1 8.9 4.6 B52 1.7 CMNH 3658 f Sl 33.5! DIS 20.8 Well 7.4 BoP) Dp N72 CMNH 2359 m 62.8 Vol 32.9 24.1 13.4 9.3 4.3 25 1.5 PNM 2300 m 56.6 70.8 DAES 22.9 12.0 7.6 4.6 od 1ES PNM 2301 ff 46.6 66.4 Wie) 16.9 OD 6.9 3.3 1.6 lee PNM 2302 oy. | (a7) 91.6 33.3 D3%5 13.1 9.1 4.6 3.1 1S PNM 2303 m 48.0 49.1! 21.8 20.3 10.8 6.5 Sod) 2 1.1 PNM 2304 fy 55.8 69.8 28.6 20.3 11.8 7.4 4.1 Dee) 1.4 PNM 2305 m 56.5 80.4 29.8 DDS) IDES 8.2 4.5 2.8 1.5 PNM 2306 ol Oa 61.8! 30.3 22.4 W2e7/ 8.2 4.1 2.6 1.3 PNM 2307 fe 50.0 62.4 26.3 18.1 2 13. 3.8 2.4 NZ CAS 61183* m 43.0 _ 20.9 17.1 Wo3) 6.5 3.9 325 2.4 X 58.9 ~ 30.2 22.3 1225 8.2 4.2 Def 1.5 SD 6.1 — 3.6 2.0 1.1 0.9 0.4 0.3 0.2 Range 19.2 — Sot To 3.6 3)3} 1S 0.9 0.9 n 14 — 14 14 14 14 14 14 14 WwW N — N N N N N N N £1 =0.39) — —0.83 —0.60 =0),05 =){05 —0.67 i209 —0.66 Zo —=(0),2)1 — —0.86 —0.02 —0.66 =(0)i331 O07 =I —0.42 1 Tail recently autotomized or showing scars of caudal autotomy and subsequent regeneration. * Holotype. = 19-21—19-21; n = 16—16. Holotype with 5—5 first finger lamellae and ours have 5—5 (five specimens), 6—6 (7) or 7—7 (4); Nee) 2 323) SID. range — 6— 7—5-7; n= 16—16. Preanal scales strongly enlarged. Body size.—Table 1 contains the mor- phological measurements taken from all known specimens of S. beyeri. Body size proportions of holotype and our new series are discussed below. Coloration. —Color plates (Fig. 4) contain lateral views of male PNM 2302 in life. Dorsal surfaces dark umber to lavender brown. A mid-vertebral line of very dark brown to black spots variably obvious; ver- tebral spots (or line) at midbody usually en- compassing one scale only, but occasionally two or three scale rows; spot series (or line) fading caudally on some specimens. Very dark brown to black series of irregular blotches laterally, usually forming a solid band, strongest at the pre-nuchal region, next strongest above hind limbs at pectoral gir- dle; band extending anteriorly through tym- panum to rostral scale and caudally to thick- est portions of tail base (‘4 total tail length). Dorsally, lateral band bordered by a thin golden yellow edge; ventrally, golden yellow edge becoming solid yellow on caudal por- 12 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 4. Sphenomorphus beyeri in life before preservation (PNM male 2302; SVL = 62.7 mm). Note presence of grey volcanic ash on substrate. Photographs by RMB, courtesy of CMNH. VOLUME 108, NUMBER 1 tions of body. Laterally, axilla-groin region bright orange with round yellow markings arranged in spots. Distinctive black tip on tail of all complete specimens. Undersides of arms and legs bright golden yellow. Pos- terior regions of hind limbs with distinct yellow spots surrounded with brown bor- ders fading into a lavender brown back- ground. Ventral side of body from nuchal region to tail golden yellow with pinkish tan, light grey, or very light sky blue flecks on nuchal region, ventral side of head, torso, and tail base. Lavender brown series of spots arranged into irregular, disjunct stripes of four to eight scales in length and one scale in width on ventral side of neck. Chin pink- ish tan or translucent. Labial scales lavender to purplish, each with white spot in center or on ventral border of scale. In alcohol: coloration generally faded throughout; ventral yellow and lateral or- ange fading to dull pinkish tan; ventral nu- chal region devoid of blue tones; purple on labials fading to lavender or brown. Comparisons. — Table 2 compares S. be- yeri with closely related, sympatric, and/or morphologically similar species of Sphen- omorphus. Discrepancies with Taylor’s original de- scription. —Taylor’s (1922b) description of S. beyeri stated that the holotype exhibited separated prefrontals, as do many of ours (10 of 16). However, three of our series have prefrontal scales in narrow to moderate con- tact and three others show the presence of an azygous interprefrontal (Fig. 3). In ad- dition, Taylor (1922b) counted 40 scales at midbody in the holotype but did not record paravertebrals. Brown & Alcala (1980) and Alcala (1986) confirmed his midbody scale counts, and reported also that there were 94 paravertebral scales in the holotype. Our examination of the holotype confirms both of these reports, which fall within the range of variation for the new series (88—96; Table 2). Taylor (1922b) counted 17 lamellae un- der the holotype’s longest (=fourth, rear) toe, but Brown & Alcala (1980) and Alcala 13 (1986) reported that the holotype had 19 fourth toe lamellae. Our examination of the holotype confirms that the specimen has 19—19 subdigital lamellae under the fourth toe. Taylor (1922b) reported a SL/HL ratio for the holotype of 0.34; calculations for our series are remarkably close (¥ = 0.32, range = 0.25-0.37; n = 16). Taylor (1922b) re- ported that the SL/HB ratio was 0.60 for the holotype; our series averaged 0.49 (range = 0.38-0.57; n = 16). Taylor’s specimen had a HB/HL ratio of 0.57, and ours average 0.66 (range = 0.60-0.69; mn = 16). Taylor reported a HB/SVL ratio of 0.15; our spec- imens average 0.14 (range = 0.13-0.17; n = 16). The holotype also had an ED/SL ratio of 0.50, but the eyes on our specimens seemed somewhat smaller on average (X = 0.37; range 0.30-0.43; nm = 16). Disparities in these calculated ratios may reflect the small sample size (n = 1) available to Tay- lor; the distinctive black tip on the tails of all complete specimens undoubtedly would not have been missed by Taylor had he been able to recover the portion of the holotype’s tail which was autotomized and lost in cap- ture. Remarks.—Two of the Zambales speci- mens were gravid females (CMNH 3652 and 3653), each containing two yellowish, thick- ly-shelled, oviductal eggs, the texture of which suggest oviparity. One of our speci- mens appears to be recently hatched (CMNH 3654). It is well within the range of mor- phological variation described here for S. beyeri except that it has unfused frontopar- ietals. As stated above (site descriptions), both areas surveyed in this study contained ash deposits from the June 1991 eruption of Mt. Pinatubo. This rendered the forest floor, vegetation, and much of the remain- ing habitat very dry (Fig. 2). Where water might be expected to collect (i.e., in Pan- danus axils or other depressions in rocks or vegetation), the ash rapidly absorbed water deposited by rain or dew. Taylor (1922b) mentioned that he found the holotype on a rock ledge on Mt. Banahao; none of our PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 14 30e1U09 uoznT ‘S sof SoA UI 10 o}e1Edas O'1S-S°LP OC-LI 6€-VE €8-€L DUDULTJOUY ‘S uOZ -N] WIdYINOS puke UIDISOMpI| sod soA a}eledos Ajjensn ZILL9-19 9 17-61 TH-BE 96-88 ladaq “S 10e1U09 ovuepullN “N SoA ou ut AjoyeJopoul Css Is Vinal Cv-OV €6—-06 DIDMIP 'S’ uozn] ‘Ss sah SoA 10e]U0D peoiq 67S ST Or GE SNIDINIDUL1ABID] “S 39e1U09 Tewles “okoT SoA ou UI 10 poyeiedas 0°06-6 €9 VC—-0T Cb-8E OL-L9 1SOUD]] “S' soutddiryg oy} noysno1y sof ou o1e1edas 8 801-6 0L O€-¥e OV-8t SEAS) uosnl “§ o1opulpy ‘onbnpuLeyy ‘uozny] SoA sof poyeredos Ajjensn 0°06-0'°€9 97-IZ Ov-8E ¢L-+9 gSUassIAIp “2 “S Iewes ‘ayAoT ‘uInsIweD ‘ovuepuly] SoA sof poyeredos Ajjensn 0'S8-0'ES 97-61 SE-ZE ZL-79 1X0I ‘2° oto ‘uoznT SoA ou payeredas Ayyensn C S615 ¢c—-0¢ Ste SLC) esniuojinbv ‘dD ‘s Joyog “‘umnsiueD ‘ovuepuly SoA ou pojeredes Ayyensn O'16-€'°18 S7-IZ Th-9E pL-€9 snjoipqv “D's OIOPUT| ‘yoseulq ‘ueAnqis ‘uoz 1] ‘ovkoT ‘Joyo ‘ovurpulyy SoA ou = poyeredas ATOPIAM TEVI-STI WGaCC pS-8h L8-SL IBUNUND *§ UINSIUIeD ‘jeseUuIq “Joyo, povel aho7T “ory ning ‘ovuepulypy sah ou -edos A[MOLIeU 9°79-6 6 C7-61 Ppr-8Ee 91-99 SNJDBIIUDA *S yeseuIq pur ‘joy -Og Jews ‘a}A97 ‘ovuepulpy SoA SoA 10e]U0D peoIq 0'9L-0'1S OI-8 O£€-97 €9-IS snjnov -S uozn’y] Ajoreq sof yoeR1U0S PLKolq 0'SS—-O'CS 81-91 (G3 89-9 sojidsoona] “S 1081009 uozn’y "N SoA SoA ur AJoyeJopoul 8 SP Slee 87 CO-LS 1U0jMD] 'S' 10e]U09 uoZzn’T] “N ou SoA UI JO 9}¥I1edosS O'8b—-S SE Z1-6 87 11-99 asuaguozn] “§ Aeueg ‘ied -NZY 9p ued ‘olEQsey] ‘SOIZONY SoA ou 10e1U09 perolq €°99-0'SP Z7-B8 I 6€-9E ¢L-+9 SUaAOGID “S sejqey, ‘ovurpuryy ‘uoznT SoA SoA yoe]UOS peolq OSt-I Te 8I-F1 8E-CE OLAS suaidiap “§ 97 Ajorel sourddyryg oy} noysnory L SoA Soh y9e]U09 UT 0'9€-S'9C vI—6 Ce—-OE v9-TS 19499]S ‘S' eee eee ee ee ee ee ee ————EEeEEEEE———E=E= uOTINGLIsIp {posodxo sjejoued s[eUOLsIg AyLinjeul 3e oe [Owe] Apoqptul S[PIQOLIOA sa1s0dg uMOoUyY WeWITxoiddy winuedwh |, -O]UOLJ (Www) TAS 30} punore sajeog -eleg posn{ qunoy eR (S661 Te 19 UMOIG “O86 BIBT -O86I BIFOTV 2 UMOI_ ‘AZZ6I ‘ETZ6I JO[ARL) JUNODOR oINj}eI9}T] JUOUTLIOd WOI pue ‘WINS 38 pourwiexo suowtoads wrod ‘Apnjs sty} Ul poyiodor suowtoods WoL oie Bed "14adaqg “S' UVM omedurAs 10 ‘01 sepIUUIS ATTeorsojoydiou ‘oj pore snydiowouayds surddityd JO satsads s19joeIvYO ONsOUseIP 19Y4}0 PUL S}UNOD g[BdS JO suosLiedwi0og — "7 IIGeL VOLUME 108, NUMBER 1 specimens were collected anywhere near rocks or outcroppings in the Zambales but, instead, from inside moist rotten logs. Only two specimens were caught in our extensive pitfall trap system, suggesting either low lev- els of activity or semi-fossorial existence. However, due to the disturbed nature of the forest on Mt. High Peak, it is impossible to establish the true microhabitat of S. beyeri. The apparent semi-fossorial nature of this species in the Zambales could have been due to animals retreating from the aridity of the immediate area which was blanketed by Mt. Pinatubo’s ash. At present, S. beyeri is known to exhibit an allopatric insular montane distribution. However, little biogeographical informa- tion can be inferred from this fact alone due to lack of reliable survey data for southern Luzon, specifically the areas between the two localities discussed here. While the two mountain ranges included in this species’ locality records are located on separate geo- logic components of Luzon, adequate hab- itat connecting these insular regions prob- ably existed during the last glaciation (Rut- land 1968; Hashimoto 1981la, 1981b; Auf- fenberg 1988). Mountains like Mt. Makiling and Mt. Arayat may support undocumented populations of S. beyeri as suggested by their intermediate placement between the two known localities (Mt. High Peak and Mt. Banahao). Studies of poorly known high el- evation montane environments (e.g., Brown & Alcala 1961b, Custudio 1986, Auffenberg & Auffenberg 1988) are needed in such areas in and between these two disjunct localities to ascertain the true distribution of S. beyeri as well as other relict species. While ele- vational species succession of scincid liz- ards (Custudio 1986) and small mammals (McCoy & Connor 1980, Rapoport 1982, Rickart et al. 1991, Ruedas et al. 1994) has —_— 15 recently been documented in some areas of the Philippines, little is known about the potential effects of elevational gradients on lizard species stratification, diversity, and endemism in the Philippine archipelago. Our unpublished survey data suggest that spe- cies diversity 1s inversely proportional to elevation, but that endemism is positively correlated with elevation at least on the is- lands of Luzon, Panay, Mindanao, and Mindoro. The latter topic is subject of an- other work in progress (Ferner, pers. comm.). Acknowledgments Collecting permits were facilitated by the Protected Areas and Wildlife Bureau of the Philippines Department of the Environ- ment and Natural Resources, especially by A. Alcala, A. Ballesfin, J. Caleda, and C. Catibog-Sinha. J. F. Barcelona, J. R. Dem- boski, R. E. Fernandez, and F. T. Wacdisen assisted in the collection effort and capture of several specimens. We thank R. I. Crom- bie for his help in obtaining the S. beyeri type specimen from the California Acade- my of Science, for facilitating our work at USNM, and for confirming the identity of our S. beyeri specimens along with W. C. Brown. Statistical analyses were carried out by L. A. Ruedas who also provided the map used in the manuscript. This study was funded by a grant from the John T. and Catherine C. MacArthur Foundation (to P. C. Gonzales and R. S. Kennedy) and by the following grants from the Miami University community (Oxford, Ohio) to RMB: The Zoology Student Enrichment Fund Grant, the Roschman Student Enrichment Fund Grant, the Undergraduate Research Fund Grant, and the Hefner Museum of Zoology Independent Studies Grant. JWF acknowl- edges the continued support of Thomas 4 Differs from S. a. abdictus by the presence of four (as opposed to five) supraoculars. > Differs from S. c. coxi by the presence of four (as opposed to five) supraoculars and coloration. © New species (Brown et al. 1995, this issue). 16 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON More College. Equipment was made avail- able to us by H. Eshbaugh and G. Kritsky; outdoor apparel and other gear was provid- ed by Outdoor Adventures in Cincinnati, Ohio. Comments on preliminary drafts of the manuscript were provided by R. I. Crombie, R. F. Inger, S. M. Moody, and L. A. Ruedas. This paper is Contribution No. 3 to the Results of the PNM/CMNH Bio- diversity Inventory of the Philippines. Literature Cited Alcala, A. C. 1986. Guide to Philippine flora and fauna. Vol. X, Amphibians and reptiles. Natural Resource Management Center Ministry of Nat- ural Resources and University of the Philip- pines, xv + 195 pp. Auffenberg, W.G. 1988. Gray’s monitor lizard. Uni- versity of Florida Press, Gainesville, iv + 419 pp. — .,&T. Auffenberg. 1988. Resource partitioning in a community of Philippine skinks (Suaria: Scincidae).— Bulletin of the Florida State Mu- seum, Biological Sciences 32(2):150—219. Brown, R. M., J. W. Ferner, & Luis A. Ruedas. 1995. A new species of lygosomine lizard (Reptilia: Lacertilia: Scincidae; Sphenomorphus) from Mt. Isarog, Luzon Island, Philippines. — Proceedings of the Biological Society of Washington 108:18— Dy. Brown, W.C., & A.C. Alcala. 1956. A review of the Philippine lizards of the genus Lygosoma (Leiol- opisma).—Occasional Papers of the Natural History Museum, Stanford University 3:1-10. —,,& . 1961a. A new sphenomorphid liz- ard from Palawan Island, Philippines. —Occa- sional Papers of the California Academy of Sci- ence 32:1-4. ——.,,& . 1961b. Populations of amphibians and reptiles in submontane and montane forests of Cuernos de Negros, Philippine Islands. — Ecology 42(4):628-636. —,, & 1970. The zoogeography of the herpetofauna of the Philippine islands, a fring- ing archipelago. — Proceedings of the California Academy of Science, fourth series 4(38):105- 130. ——.,& . 1980. Philippine lizards of the fam- ily Scincidae. — Silliman University Natural Sci- ence Monograph Series 2. 264 pp. —,&D.S.Rabor. 1967. A new sphenomorphid lizard (Scincidae) from the Philippine islands. — Proceedings of the Biological Society of Wash- ington 80:69-72. Bureau of Mines, Philippines, in coordination with the Board of Technical Surveys and Maps. 1963. Geological map of the Philippines, edition 1. Single sheet map. Custudio, C. C. 1986. Altitudinal distribution of liz- ards of the Scincidae in Mt. Makiling, Laguna. — Sylvatropical Philippine Forest Research Jour- nal 11(3, 4):181-202. Dickerson, R. E. 1924. Tertiary paleogeography of the Philippines. — Philippine Journal of Science 25(1):10-55. Greer, A. E. 1979. A phylogenetic subdivision of Australian skinks.—Records of the Australian Museum 32:339-371. —, & F. Parker. 1974. The faciatus species group of Sphenomorphus (Lacertilia: Scincidae): notes on eight previously described species and de- scriptions of three new species. — Proceedings of the Papua New Guinea Science Society 25:31- 64. Hashimoto, W. 198la. Geologic development of the Philippines. Pp. 83-170 in T. Kobiyashi, R. To- riyama, & W. Hashimoto, eds., Geology and paleontology of Southeast Asia, CCXVII, Vol. DOP 1981b. Supplementary notes on the geologic development of the Philippines. Pp. 171-190 in T. Kobiyashi, R. Toriyama, & W. Hashimoto, eds., Geology and paleontology of Southeast Asia, CCXVIII, Vol. 22. McCoy, E. D., & E. F. Connor. 1980. Latitudinal gradients in the species diversity of North American mammals.—Evolution 34:193-203. Myers, W. C., & M. A. Donnelly. 1991. The lizard genus Sphenomorphus (Scincidae) in Panama, with a description of a new species. — American Museum Novitates 3027:1-12. Rapoport, E. 1982. Areography: geographical strat- egies of species. Pergamon Press, New York, xvi + 269 pp. Rickart E. A., L. R. Heaney, & R.C. Utzurrum. 1991. Distribution and ecology of small mammals along an elevation transect in Southeast Luzon, Philippines. —Journal of Mammalogy 72:458- 469. Ruedas, L. A., J. R. Demboski, & R. V. Sison. 1994. Morphological and ecological variation in Otop- teropus cartilagonodus Kock, 1969 (Mammalia: Chiroptera: Pteropodidae) from Luzon, Phil- ippines. — Proceedings of the Biological Society of Washington 107:1-16. ; Rutland, R. W. 1968. A tectonic study of part of the Philippine Fault Zone. — Quarterly Journal of the Geological Society of London 123(4):293-325. SAS Institute Inc. 1988a. SAS/STAT user’s guide, release 6.03 edition. SAS Institute Inc., Cary, North Carolina, 1028 pp. VOLUME 108, NUMBER 1 7 1988b. SAS procedures guide, release 6.03 . 1922b. Additions to the herpetological fauna edition. SAS Institute Inc. Cary, North Caroli- of the Philippine Islands, II.— Philippine Jour- na, 441 pp. nal of Science 21(3):253-303. Sokal, R. R., & F. J. Rohlf. 1981. Biometry, second UNESCO/ECAFE. 1971. Geologic map of Southeast edition. W. H. Freeman and Co., New York, Asia. United Nations Publication, No. 69-30632. Xvlil + 859 pp. Whitmore, T. C. 1984. Tropical rain forests of the Taylor, E. H. 1922a. The lizards of the Philippine Far East. Clarendon Press, Oxford, England. xvi islands.— Monograph of the Bureau of Science, + 718 pp. Manila, 17:1-115. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(1):18-28. 1995. A new species of lygosomine lizard (Reptilia: Lacertilia: Scincidae; Sphenomorphus) from Mt. Isarog, Luzon Island, Philippines Rafe M. Brown, John W. Ferner, and Luis A. Ruedas (RMB) Department of Zoology, Miami University, Oxford, Ohio 45056, U.S.A.; (JWF) Department of Biology, Thomas More College, Crestview Hills, Kentucky 41017, U.S.A.; (LAR, RMB, JWF) Department of Vertebrate Zoology, Cincinnati Museum of Natural History, 1720 Gilbert Avenue, Cincinnati, Ohio 45202-1401, U.S.A.; (LAR) Department of Biology, Cayey University College, Cayey, Puerto Rico 00736 Abstract. —Sphenomorphus knollmanae, a new species, is described on the basis of recently collected material from Mt. Isarog, Bicol Peninsula, south- eastern Luzon, Philippines. The small series (7 = 5) differs from its congeners by the combination of its fused frontoparietals, relatively low number of par- avertebrals (73-83) and midbody scales (34-39), the presence of 17—20 sub- digital fourth toe lamellae, distinctive patterns of coloration, and a host of measurements related to its small body size (SVL = 47.5-51.0 mm). To better distinguish between the new species and two closely related congeners, uni- variate and multivariate analyses were performed on a suite of morphological characters. The three species were found to be well differentiated morpholog- ically. Worldwide, the genus Sphenomorphus contains over 120 species and is a “taxo- nomically residual’? plesiomorphic taxon that “remains a convenient repository for . species, pending further phylogenetic analysis’”> (Myers & Donnelly 1991:2). Brown & Alicala (1961b) reported that Ori- ental and Australian zoogeographic regions contain over 60 scincid species in Spheno- morphus. In their key to Philippine Scin- cidae, Brown & Alcala (1980) recognized 22 species of Sphenomorphus, subdividing these into five groups based on external morphology. The Group I species of Phil- ippine Sphenomorphus are S. beyeri (Taylor 1922) and S. diwata (Brown & Rabor 1967, see Brown & Alcala 1980, for review). Until recently, S. beyeri was known only from the holotype, collected by E. H. Taylor on Mt. Banahao, Laguna province, southern Luzon Island (Taylor 1922). During a recent bio- diversity inventory of the Philippines con- ducted by the National Museum of the Phil- ippines (PNM) and the Cincinnati Museum of Natural History (CMNH), we rediscov- ered and redescribed Sphenomorphus beyeri from specimens taken on Mt. High Peak, Zambales Mountains, west central Luzon Island (Brown et al. 1995). Sphenomorphus diwata also is currently known only from a small number of specimens collected in the Diwata Mountains, Surigao del Sur Prov- ince, northern Mindanao Island (Brown & Rabor 1967, Brown & Alcala 1980). While examining material in the United States National Museum of Natural History (USNM), R. I. Crombie brought to our at- tention a small series of Sphenomorphus skinks that appeared very similar to our specimens of S. beyeri from the Zambales. At the time, we were not confident in the assignment of these specimens to our con- cept of S. beyeri (from the type locality or from Mt. High Peak) as several inconsis- VOLUME 108, NUMBER 1 tencies immediately were apparent. Follow- ing detailed examination of these specimens and a host of univariate and multivariate statistical analyses we concluded that dif- ferences between this series and its most closely-related congeners were sufficient to warrant its recognition as a distinct species. Methods Morphological characters and scale counts used here follow definitions and abbrevia- tions in Brown & Alcala (1980) and Brown et al. (1995). Measurements were taken to the nearest 0.1 mm with digital calipers. All measurements are based on specimens pre- served in 70% ethanol. In cases where scales of interest are found on both sides of the head (e.g., labials), scale numbers are given in pairs, separated with a long dash (—), designating left from right respectively. Mensural and meristic character abbrevia- tions (defined in Brown et al. 1995) include: snout-to-vent length (SVL), tail length (TL), axilla-groin distance (AGD), hind leg length (HLL), head length (HL), head breadth (HB), snout length (SL), eye diameter (ED), tym- panum diameter (TD), paravertebrals (PVS), midbody scales (MBS), supralabials (SUL), and infralabials (IFL). Specimens examined are deposited in the California Academy of Science (CAS) the Cincinnati Museum of Natural History (CMNH), the National Museum of the Philippines (PNM), and the United States National Museum of Natural History (USNM). Statistical analyses were performed using the Statistical Analysis System software, version 6.03 (SAS Institute Inc., 1988a, 1988b). Sexually immature specimens (S. beyeri, PNM 2303 and CMNH 3654, CAS 61183; S. diwata, CAS 133514; Spheno- morphus sp., USNM 318343) were exclud- ed from univariate and multivariate anal- yses. A Student-Newman-Keuls multiple range test was performed on both raw and log (base 10) transformed data to determine patterns of significant character variation. Two principal component analyses were 19 performed, both on the correlation matrix of the variables. The first included only raw (untransformed) data; the second was car- ried out on the log (base 10) transformed data, in order to minimize the effects of size differences among the different populations examined herein; in the case of the log (base 10) transformed analysis, the size compo- nent of the variation is restricted to prin- cipal component axis one. In both in- stances, the first and second and the first and third principal component scores were then plotted in order to ascertain morpho- logical differentiation among groups. Results Sphenomorphus beyeri, S. diwata, and S. n. sp., distinctly segregated into discrete groups in the principal component analysis (Fig. 1). In the PC analysis based on raw data (Fig. la, b), principal component one differentiates between S. n. sp. and S. diwata and between the new species and S. beyeri. This component loads heavily on HL, SVL, HB, HLL, AGD, and SL. Principal com- ponent two distinguishes between S. diwata and remaining Group I Sphenomorphus. This component loads heavily on fourth toe lamellae and MBS as well as TD, ED, and PVS. The third principal component differ- entiates between S. beyeri and remaining Group I Sphenomorphus. This component loads primarily on fourth toe lamellae, PVS, and TD. Together, the first three principal components account for 85.7% of the vari- ation (PC I, 55.0%; PC II, 16.7%; PC III, 14%). In the PC analysis based on the log trans- formed data (Fig. lc, d), principal compo- nent one and two discriminate between S. n. sp. and S. diwata, while principal com- ponents two and three discriminate between S. diwata and S. beyeri. Principal compo- nent three also discriminates between S. be- yeri and the new species. The first four prin- cipal components account for 91.6% of the variation (56.4, 16.8, 13.2, and 5.2, respec- tively). Factor loadings along the first prin- 20 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON pe) —s -4 4 2 0 -2 0 PC III PC II Fig. 1. Plots of principal component scores for two species of Group I Sphenomorphus and the new species examined in this study. Component I versus compo- nent II (a); component I versus component III (b); component I versus component III for log (base 10) transformed data (c); component I versus component III for log (base 10) transformed data (d). Symbols are: squares = S. beyeri; triangles = S. diwata; circles = S. knollmanae, new species. cipal component are relatively homoge- neous, indicating that the size variation gen- erally has been isolated to this component, while shape is more important among re- maining principal components. High load- ings 1n principal component two are shown in discrete characters (scale row counts). An additional PC analysis carried out on the correlation matrix of the log (base 10) trans- formed variables but using only measure- ments (not shown), indicates that tympa- num and eye diameters were the heaviest contributors to the variation in PC axes two and three. An important point to make regarding these analyses concerns the orientation of the principal component axes of each pu- tative species group’s dispersion in multi- variate space. Orthogonal orientation of the axes 1n multivariate space has been inter- preted as indicative of differing allometric growth patterns (Voss et al. 1990, Voss & Marcus 1992); a corollary of the foregoing is that different orientations of the principal component axes are thereby good evidence of distinct specific status. In the case of the analyses of the populations of Sphenomor- phus examined herein, it is quite clear that the orientation of the axes of dispersion are quite distinct both in analyses based on raw data as well as log (base 10) transformed data. This particular distinction 1s especially severe between S. beyeri and remaining Sphenomorphus examined and is clearly ob- served in the raw data plots, but more sig- nificantly in the log (base 10) transformed data, which minimizes the contribution of size to principal component axis one. In view of the quantum separation in al- lometric growth patterns, as well as mor- phology between the Mt. Isarog Spheno- morphus, S. beyeri, and S. diwata in both multivariate (PCA) and univariate analyses (ANOVA) of discrete and continuously varying characters, we describe the series from Mt. Isarog as: Sphenomorphus knollmanae, new species Figs. 2, 3 Holotype. —PNM 2311 (formerly USNM 318342), adult male, collected by L. R. Hea- ney on | May 1988, in loose leaf litter along- side a fallen, partially decomposed log on the forest floor in primary mid montane forest at 1125 m on Mt. Isarog (Philippines, S. Luzon, Bicol Peninsula, Camarines Sur Prov.), 4.5 km N, 20.5 km E Naga City, 13°40'N, 123°22’E (map: fig. 1 in Goodman & Gonzales 1990). Paratypes.—(4) USNM 318341 (female) and 318343 Guvenile), same data as above except as follows: USNM 318341, collected by L. R. Heaney on 29 Apr 1988; USNM 318343, collected by S. M. Goodman, 22 Mar 1988. USNM 318344 and CAS 191800 (formerly USNM 318345) collected by S. M. Goodman, 19 Mar 1988, and by R. C. B. Utzurrum on 20 Mar 1988 respectively, at 4 km N, 21 km E Naga City, 13°40’N, 123°22’E, at 1350 m in primary upper mon- tane forest. All specimens collected in loose leaf litter and loose topsoil on forest floor; USNM 318344 and CAS 191800 associated VOLUME 108, NUMBER 1 with fallen, partially decomposed logs on forest floor. Etymology. —Named in honor of the late Margy Knollman, friend and teacher, who guided the senior author through his first scientific experiment at age seven and con- tinued to encourage his herpetological pur- suits until the time of her death in Novem- ber 1989. Diagnosis. —A small to moderate species of Sphenomorphus (SVL, 47.5—51.0 mm) differing from its congeners by a combina- tion of the following characteristics: fron- toparietals fused; prefontals separate, in contact, or with azygous interprefrontal; 73 to 83 paravertebrals; 34 to 39 scales around midbody; 17 to 20 subdigital fourth toe la- mellae; unique coloration (see below). Description of holotype.—(PNM 2311) Total Length, 118.7 mm; SVL, 48.7 mm; nie 0.0mm, 115) mime? SL, 3:9 mm: HB, 7.1 mm; ED, 3.1 mm; lower eyelid scaly with translucent window, oval in shape, ar- ranged horizontally; ear opening and tym- panum exposed, not deeply sunken, verti- cally oval, 1.2 mm in width; ear opening without spines or lobules; limbs pentadac- tyl, well developed; HLL, 18.3 mm; AGD, 22.7 mm; head (viewed from above) ta- pered, snout rounded dorsally and laterally; dorsal, lateral, and ventral scales smooth, unstriated; rostral large, visible from above, broader (1.8 mm) than long (0.7 mm), form- ing a curved suture with frontonasal; latter wider (1.5 mm) than long (1.1 mm); pre- frontals in broad contact; frontoparietals fused (2.8 mm wide, 1.8 mm long) frontal moderate, rhomboidal, pointed caudally, 1.9 mm wide, 3.0 mm long, in contact with two supraoculars; interparietal moderate, point- ed, 1.3 mm wide, 1.9 mm long; parietals in contact behind interparietal; nasals large and single with round nostril at centers, widely separated by frontonasal and nasal bordered caudally by two pairs of overlapping loreals, dorsal pair slightly larger than ventral; 3— 3 large preoculars, most ventral contacts su- ture between third and fourth labial; 4—4 21 oo LK 4 heheh “Sih eG Fig. 2. Sphenomorphus knollmanae holotype (PNM 2311)—Lateral (A), and dorsal (B) head scalation and subdigital lamellae, right hind foot (C). large supraoculars, anteriormost triangular, second widest; last supraocular followed by 3 rows of small scales clustered in postoc- ular region, each row containing 2-3 scales; 3—3 temporals, dorsalmost wraps one- fourth of way around posterior edge of pa- rietal scales; nuchals undifferentiated, ex- cept for most lateral pair which is very slightly enlarged; 2—2 rows of small scales between eye and labials; 12—12 supracili- aries; 15—16 lower ciliaries; 6—6 suprala- bials; 6—6 infralabials; 73 paravertebrals; 34 midbody scales; 20 subdigital fourth toe lamellae; 5 first finger lamellae; toe length (shortest to longest) 4, 3, 2, 5, 1; two strongly enlarged preanal scales apparent; mental chin scale followed by single postmental bordered caudally by two pairs of chin shields; subcaudals only slightly larger than ventrals. The holotype had a live weight of 3.0 g. Body size proportions and coloration discussed below. Coloration. — Field notes recorded by R. C. B. Utzurrum (courtesy L. R. Heaney, Field Museum of Natural History) state that in life CAS 191800 had a “golden venter, dark brown dorsum, mottled on sides.” In alcohol, dorsal surfaces very dark brown with black spots and a darkly pigmented (=2-3 scale rows) black mid-vertebral line. Mid-vertebral line darkest on CAS 191800, 22 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 22 S) sé en NS ez l2 Oe 6! i i é\ ol g| 71 a el il Ol 6 8 i SS) S 7 £ ty Naliidtedlndt hon usdvnbselsusotsnlastunebvittisshsntiscticsteuabsaticatintisalinatscalioutiesdinetinstes sve tiaptavsrlnetiey Bee cemOCe Gc Vcc ccmlcer: Ocgirol ieaeee| 91 i] 7\ gi 2! WW Ol 6 8 Z 9 Ss 7 € mint iyntunluntonhuntantoutwubralidiutonfostysbotusluebuolotonbuntunbaertidstouelvvtuliutte valent aytirdyutaatonlntunlutadaduntystuonaty Fig. 3. Dorsal (top) and ventral (bottom) views of the type series of Sphenomorphus knollmanae placed underneath five specimens of S. beyeri for comparison. VOLUME 108, NUMBER 1 23 Table 1.— Morphological measurements and scale counts taken from all known specimens of Sphenomorphus knollmanae (see text for abbreviations of characters). Standard univariate statistics are presented below as means + one standard deviation (sexually mature specimens only). Character Specimen # Sex SVL TL AGD HLL HL HB SL ED TD PVS’' MBS_ SUL _ IFL 4th Toe MISINIMESISS842 7 -f 49°78 71:5 24-97 18:5. 1010 6.2 3.6 2:5 1.2 76 39 7-7 6-6 19 PNM 23112 ma AS VOW)» 22 MSS lle Wel Be SSE wales 7/8} 34 6-6 6-6 20 USNM 318343 ? 47.5 Sirol 4 Selo) NOLO O.00 395.25) 1E2, 83 39 7-7 6-6 19 WSNM 3183445 of S10 41.4 26.5 17.2 10:0. 6.4 3.5 2.7 1.4 76 37 7-7 6-6 17 CAS 191800 f 50.4 SOD SH Sse eR) 6:4 32 27, ICS 76 35 6-6 7-7 18 Mean 49.1 — DSS 4 Ost OsOm Seige.) eles) OPN 36-5) 36:5) 62317 18-5 SD 1.36) = Did a OMtte ORG, OFS SOSH HOLD 083) PA2he 92845. 0:6 (0:5 1.3 @ Holotype. > Tail autotomized and partially regenerated. but also heavy on PNM 231 and USNM 318341, somewhat lighter (1-2 scale rows) on USNM 318343 and 318344. Mid-ver- tebral line ending abruptly at pectoral girdle where dorsal mottling coalesces into trans- verse bars that fade caudally (this pattern not apparent in USNM 318343). Laterally, with a heavy series of black blotches, form- ing a solid stripe in canthal region, and ex- tending posteriorly from nostril, through eye and typanum, to groin. Lateral black stripe anteriorly bordered ventrally by a distinct white line intersecting the tympanum at one- half its height and extending from caudal edge of eye to region dorsal to forearm. Lat- eral white line on midsection breaking up into series of white spots that continue cau- dally through anterior one-third length of tail. Ventral surfaces in complete specimens pale yellow from chin to tip of tail (USNM 318344 has a regenerated tail that is com- pletely black; USNM 318343 and CAS 191800, with autotomized tails). Throat pale yellow with dark umber flecks (darkest on CAS 191800 and almost invisible on ho- lotype and USNM 318341). In specimens with heavy speckling, pattern wraps around onto lateral portions of neck, extending to approximately one-half the height of the tympanum. Variation. —Our sample includes one male, three females and a sexually undeter- mined juvenile. USNM 31834 and CAS 191800 both were gravid at the time of pres- ervation, each containing two thinly-shelled eggs. USNM 318341 may have been gravid when preserved (remnants of what appear to be eggs remain), but some breakdown of the ovaries has occurred. Table 1 contains morphological measurements of the five specimens of S. knollmanae and characters and measurements differing from holotype description follow below. Variation in head scalation is as follows: FOstraleles= 22) (Xe— 1S = 0.408): 0 — 4) mm wide; frontonasal 1.5-1.8 (1.7 + 0.1; n = 4) mm wide and 1.0-1.1 (1.1 + 0.1; = 4) mm long; frontoparietal 2.5—2.9 (2.7 + 0.2; n = 4) mm wide and 1.7-2.1 (1.9 + 0.1; n = 4) long; frontal 0.9-1.9 (1.5 + 0.4; n = 4) mm wide and 1.5-3.0 (2.4 + 0.67; n= 4) mm long; interparietal 1.1 to 1.4 (1.2 + 0.1; n = 4) mm wide and 1.7 to 1.9 (1.8 + 0.1) mm long. Prefrontals are in broad contact (three specimens) or separated by an azygous interprefrontal (one specimen), the latter somewhat wider anteriorly where contacts frontonasal; increasingly narrow caudally where its most caudal border ex- tends slightly beyond margin delineated by contact between prefrontal and frontal. Su- praciliaries 12-16 (¥ = 13.4 + 1.5 SD; n= 5) on right and 12-15 (2.6 + 1.3; n= 5) on left; lower ciliaries, 5-18 (16.2 + 1.l; n= 24 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 2.—Comparisons of selected measurements and scale counts for Sphenomorphus knollmanae and two closely related congeners. Presented below are standard univariate statistics (means + one standard deviation, sexually mature specimens only) and results of Student-Newman-Keuls multiple range tests. Superscript letter by means indicates group assignment (means with same letter are not statistically different at the P > 0.05 level). See text for definitions of abbreviations used in this table. Character 4th SVL TL AGD HLL HL HB SL ED TD PVS-' MBS_~ SUL _ IFL_ Toe S anova X AGIE = O52 IIBZE NOs bce Sse ZIP 1.3? T/0P B6.5° 65 63 18.5% n=4 SD VAS) We 25 0.1 OG Ws Os OLA Os 47 2:4 » (0:6 055 ay es S. beyeri 2X SBOB = ZO22 WEP MDS Bye ae Bye 5° OBE BOS 6.5 GA 20.3" n= 14 SD S621 7 = 316 2.0 hei O9) OA Osss - O27 1 1.2 O18 Ld S088 S. diwata 26 SAP = 5G DAD Ngj.se OP Ase Bie Die See alse 55.5 5.5 13.5 n= SD 47 — 5.4 8.6 0.4 06 06 06 0.8 1.4 Aes] OL O55) OL7/ 5) on right and 14-17 (16.0 1.27 7 — 5) on left; supralabials 6-7 (6.6 + 0.5; n = 5); infralabials 6-7 (6.2 + 0.4; n = 5) on right and 5-6 (5.8 + 0.4; n = 5) on left. In specimens with regenerated tails (USNM 318343 and CAS 191800), normal subcaudals replaced by a series of narrow scales that cover entire ventral surface of regenerated tail. Besides holotype, only CAS 191800 was weighed before preservation (3.4 g). Ratios of morphological measurements for the series (holotype in parentheses) are as follows: SL/HL, ¥ = 0.34 + 0.03, range = 0.29-0.36 (0.35); SL/HB, ¥ = 0.55 + 0.03, range = 0.50-0.58 (0.55); HB/HL, X¥ = 0.61 + 0.03, range = 0.60-0.64 (0.64); BV OMIL, 26S 0.13 = O01, memes = 0,12 0.15 (0.15); HL/SVL, X¥ = 0.21 + 0.01, range = 0)20=0123(0. 23); BD/SID X0— 077) =: 0.06, range = 0.71-0.84 (0.77); ED/HB, X = 0.42 + 0.01, range = 0.41-0.44 (0.41); AGD/SVL, X = 0.52 + 0.04, range = 0.47- 0.57 (0.47); HLL/SVL, X = 0.36 + 0.02, range = 0.34—-0.38 (0.37). Comparisons. —Table 2 compares S. knollmanae with Group I Sphenomorphus species. Excepting its low number of par- avertebrals, S. knollmanae adheres to the gestalt of Group I Sphenomorphus members (Brown & Alcala 1980); accordingly, the new species appears closely related to S. beyeri and S. diwata. Besides size and characters He I+ listed in Table 2 of this study, S. knoll- manae differs from the former by coloration and disposition of color pattern (Brown et al. 1995). It differs from the latter by char- acters in Table 2 and also in that it invari- ably has only 4 supraoculars (vs. 5-6 in S. diwata) and fused frontoparietals (vs. 2 in S. diwata; Brown & Rabor 1967, Brown & Alcala 1980). As in S. beyeri (Brown et al. 1995) contact, or lack thereof, between the prefrontal scales is not fixed in this species as itis in S. diwata (Brown & Alcala 1980). The azygous interprefrontal scale exhibited by USNM 318344 also is apparent in some specimens of S. beyeri (Brown et al. 1995), but not in any known specimens of S. di- wata (Brown & Rabor 1967, Brown & Al- cala 1980). The low number of paravertebrals in S. knollmanae assigns this species to Group III of Brown & Alcala’s (1980) key; accord- ingly, comparisons with S. /eucospilos and S. laterimaculatus, as well as S. decipiens, are warranted. Differences between S. knollmanae. S. leucospilos, and S. decipiens are as follows: Sphenomorphus knollmanae distinct by its 73-83 paravertebrals (vs. 63- 68 in S. leucospilos and 57-66 in S. deci- piens), 34-39 scales around midbody (vs. 32 in S. leucospilos and 32-38 in S. deci- piens), and 17-20 fourth toe lamellae (vs. 14-18 in S. decipiens). Sphenomorphus de- cipiens also has a smaller overall body size VOLUME 108, NUMBER 1 (SVL = 31-45 mm; Brown & Alcala 1980) than S. knollmanae. Differences in color pattern and body proportions between these three species also are apparent (Brown & Alcala 1980). Body measurements and scale counts of the single known S. /aterimaculatus speci- men are very close to the range of variation of both S. leucospilos and S. knollmanae (see Brown et al. 1995). While the range of paravertebrals and midbody scales in S. knollmanae do not overlap with S. /ateri- maculatus (73-83 vs. 72 and 34-39 vs. 40 respectively), the small number of known S. laterimaculatus specimens (n = 1) pre- cludes classification based solely on these characters. However, others are apparent: Brown & Alcala (1980) describe the fronto- parietal of S. laterimaculatus as “long and pointed, almost as long as frontoparietals and interparietal together” (1980:178), a description which does not accord with the relative size of these scales in S. knollman- ae, especially since the frontal of S. /ateri- maculatus touches three supraoculars, whereas the frontals of S. knol/manae only contact two supraoculars. In addition, the holotype of S. /aterimaculatus has eight in- fralabials and all specimens of S. knoll- manae have six or seven. There are six or seven first finger subdigital lamellae in S. laterimaculatus and five to six in S. knoll- manae. Coloration and body proportion dif- ferences between these species are also ap- parent (see Brown & Alcala 1980). Discussion At the present time, the new species is only known from the type locality on Mt. Isarog (fig. 1 in Goodman & Gonzales 1990). Detailed habitat descriptions (see Brown 1919, and Whitmore 1984 for review of for- est classifications), habitat photographs, and a map of the type locality for S. knollmanae are included in Goodman & Gonzales (1990). Very little is known about the habitat and pe) ecology of S. Anollmanae and its closely- related congeners, S. beyeri, S. diwata, S. laterimaculatus, S. decipiens, and S. leu- cospilos. Excepting S. decipiens and S. di- wata, all are known only from the Luzon faunal region; excepting S. decipiens, all are known only from small series. Taylor (1922) reported that the S. beyeri holotype from Mt. Banahao was collected on a rock ledge at 1500 m, but we collected most of our specimens from the Zambales Mountains by splitting open rotten logs, 1n pitfall traps, or under leaf litter—at high elevations (1265-1610 m; Brown et al. 1995). Brown & Alcala (1980) reported that S. diwata were found under leaf litter between 1600 and 1700 m on Mt. Hilong-hilong, northern Mindanao (see Brown & Rabor 1967). Sphenomorphus decipiens is also semi-fos- sorial at low to medium elevations (100- 1200 m; Brown & Alcala 1980). The known specimens of S. knollmanae were taken from similar semi-fossorial environments on the forest floor on Mt. Isarog. All were captured by L. H. Heaney, A. Alcala, and coworkers (under leaf litter, in loose topsoil, occasion- ally beside rotten logs), while digging for worms to be used as bait for mammal traps. No habitat data are available for S. /ateri- maculatus and S. leucospilos (Brown & A\l- cala 1980). Studies of high elevation scincids and their habitats have been sorely lacking, with the exception of a few instances (Brown & Al- cala 1961la, Custudio 1986). The effects of altitudinal gradients on species richness, abundance, diversity, and distributional patterns have been addressed to a greater extent in birds (Goodman & Gonzales 1990) and mammals (Heaney et al. 1989, Rickart et al. 1991). Efforts to provide a preliminary report of altitudinal effects on scincid lizard distribution in the Philippines currently are under way. While mountain tops have been neglected by many collectors and surveyors in the past, a recent renewal of interest in their unique flora and fauna has produced discoveries (e.g., Gonzales & Kennedy 1990, 26 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Lazell 1992, Ross & Gonzales 1992) and rediscoveries (e.g., Ross & Lazell 1990; Brown et al. 1995; Crombie, pers. comm.) of many taxa endemic to the Philippines. The works of Goodman & Gonzales (1990) and Oliver et al. (1992) have both stressed the importance of continued study of montane regions in order to fuel conser- vation efforts aimed at preserving these fragile centers of endemism and diversity (see Balate et al. 1992, for a bibliography of conservation in the Philippines). We sup- port their invocations to public awareness with respect to this central issue of Philip- pine conservation given that we repeatedly have witnessed and participated in discov- eries of endemic animals new to science which inhabit extremely limited distribu- tions at high elevations in disappearing frag- ments of pristine habitat. The loss of such habitat can and often does have effects det- rimental to populations of amphibians and reptiles restricted to the immediate area (pers. obs.). Finally, while we do not wish to engage in speculation (sensu Lazell 1992) of exactly what species may await biologists in similar environments on Philippine mountains, we do agree with Ross & Gonzales (1992) that the northern Philippines (especially the Lu- zon faunal zone) is zoogeographically com- plex and contains more centers of endem- ism than previously thought. Our recent studies suggest that the higher volcanic peaks of southern Luzon (Mt.’s Bulusan, Mayon, Labo, Banahao, Isarog, Samat, Natib, Cuadrado, Angilo and Maquiling) all war- rant intensive, long-term survey efforts of the kind that have produced (and continue to produce) many new discoveries on their neighbors. Comparative material examined. — Sphenomorphus beyeri holotype, CAS 61183; S. beyeri, PNM 2300-2307, CMNH 3652-3655, 3657-3659, USNM 337768. S. diwata holotype (CAS 2478), S. diwata (CAS 133514 and 133515). Acknowledgments Collecting permits were facilitated by the Protected Areas and Wildlife Bureau of the Philippine Department of the Environment and Natural Resources, especially by A. Al- cala, and C. Catibog-Sinha. The Philippine Bureau of Forestry Development and the Bicol University College of Fisheries, Ta- baco, Albay assisted during the collection of these and other specimens during L. R. Heaney’s group’s field work on Mt. Isarog. R. I. Crombie of the United States National Museum (USNM) facilitated loans and pro- vided many helpful comments and sugges- tions throughout this and related research. W.C. Brown and J. Vindum facilitated loans of specimens in the California Academy of Science (CAS) collections and A. Alcala and L. R. Heaney generously provided access to the specimens they collected on Mt. Isarog with R. de Leon, S. M. Goodman, E. A. Rickart and R. C. B. Utzurrum. RMB thanks the Miami University Zoology Department for its encouragement and JWF acknowl- edges the continued support of Thomas More College. Comments on preliminary drafts of the manuscript were provided by W. C. Brown, R. I. Crombie, R. F. Inger, S. M. Moody, S. Simon, and one anony- mous reviewer. We owe a debt of gratitude to Pedro C. Gonzales (PNM) and Robert S. Kennedy (CMNH) for their continued sup- port of our work with Philippine herpeto- fauna. Literature Cited Alcala, A. C. 1986. Guide to Philippine flora and fauna. Vol. X, Amphibians and reptiles. Natural Resource Management Center Ministry of Nat- ural Resources and University of the Philip- pines, 195 pp. Auffenberg, W. G. 1988. Gray’s Monitor Lizard. University of Florida Press, Gainesville, 419 pp. Balate, D. S., H. C. Miranda, L. R. Heaney, & J. F. Rieger. 1992. Diversity and conservation of Philippine land vertebrates: an annotated bib- liography.— Silliman Journal 36(1):129-149. Brown, R. M., J. W. Ferner, & R. V. Sison. 1995. Rediscovery and redescription of Sphenomor- VOLUME 108, NUMBER 1 phus beyeri Taylor (Reptilia: Lacertilia: Scinci- dae) from the Zambales Mountains of Luzon, Philippines. — Proceedings of the Biological So- ciety of Washington 108:6—17. Brown, W. C., & A. C. Alcala. 1961la. Populations of amphibians and reptiles in submontane and montane forests of Cuernos de Negros, Philip- pine Islands.— Ecology 42(4):628—636. ,& . 1961b. A new sphenomorphid liz- ard from Palawan Island, Philippines. —Occa- sional Papers of the California Academy of Sci- ence 32:1-4. =, 2 1970. The zoogeography of the herpetofauna of the Philippine islands, a fring- ing archipelago. — Proceedings of the California Academy of Science, fourth series 38(6):105-— 130. ——.,& . 1980. Philippine lizards of the fam- ily Scincidae. Silliman University Natural Sci- ence Monograph Series No. 2, 264 pp. —., &D.S. Rabor. 1967. A new sphenomorphid lizard (Scincidae) from the Philippine islands. — Proceedings of the Biological Society of Wash- ington 80:69-72. Brown, W. H. 1919. Vegetation of the Philippine mountains. Bureau of Printing, Manila, 434 pp. Bureau of Mines, Philippines, in coordination with the Board of Technical Surveys and Maps. 1963. Geological map of the Philippines, edition No. 1. Custudio, C. C. 1986. Altitudinal distribution of liz- ards of the Scincidae in Mt. Makiling, Laguna. — Sylvatropical Philippine Forest Research Jour- nal 11(3, 4):181-202. Dickerson, R. E. 1924. Tertiary paleogeography of the Philippines. — Philippine Journal of Science 25(1):10-55. Gonzales, P. C., & R.S. Kennedy. 1990. A new spe- cies of Stachyris babbler (Aves: Timaliidae) from the island of Panay, Philippines.— Wilson Bul- letin 102:367-379. Goodman, S. M., & P. C. Gonzales. 1990. The birds of Mt. Isarog National Park, Southern Luzon, Philippines, with particular reference to altitu- dinal distribution. Fieldiana 60:1-39. Hashimoto, W. 1981la. Geologic development of the Philippines. Pp. 83-170 in T. Kobiyashi, R. To- riyama, & W. Hashimoto, eds., Geology and Paleontology of Southeast Asia, CCXVII, Vol. UD . 1981b. Supplementary notes on the geologic development of the Philippines. Pp. 171-190 in T. Kobiyashi, R. Toriyama, & W. Hashimoto, eds., Geology and paleontology of Southeast Asia, CCXVIII, Vol. 22. Heaney, L. R. 1986. Biogeography of mammals in SE Asia: estimates of rates of colonization, ex- Di, tinction and speciation. — Biological Journal of the Linnean Society 28:127-165. , P. D. Heideman, E. A. Rickart, R. B. Utzur- rum, & I. S. H. Klompen. 1989. Elevational zonation of mammals in the central Philip- pines. — Journal of Tropical Ecology 5:259-280. Lazell, J. 1992. New flying lizards and predictive biogeography of two Asian archipelagos. — Bul- letin of the Museum of Comparative Zoology 152(9):475—505. McCoy, E. D., & E. F. Connor. 1980. Latitudinal gradients in the species diversity of North American mammals.—Evolution 34:193-203. Myers, W. C., & M. A. Donnelly. 1991. The lizard genus Sphenomorphus (Scincidae) in Panama, with a description of a new species. American Museum Novitates 3027:1-12. Oliver, W. L. R., C. R. Cox, P. C. Gonzales, & L. R. Heaney. 1992. Cloud rats in the Philippines— preliminary report on distribution and status. — Oryx 27(1):41-48. Rapoport, E. 1982. Areography: geographical strat- egies of species. Pergamon Press, New York, 269 pp. Rickart, E. A., L.R. Heaney, & R.C. Utzurrum. 1991. Distribution and ecology of small mammals along an elevation transect in Southeast Luzon, Philippines.—Journal of Mammalogy 72:458- 469. Ross, C. A., & P. C. Gonzales. 1992. Amphibians and reptiles of Catanduanes Island, Philip- pines. — National Museum Papers (Manila) 2(2): 50-76. —., & J. D. Lazell, Jr. 1990. Amphibians and reptiles of Dinagat and Siargao Islands, Phil- ippines.—The Philippine Journal of Science 119(3):257-286. Ruedas, L. A., J. R. Demboski, & R. V. Sison. 1994. Morphological and ecological variation in Otop- teropus cartilagonodus Kock, 1969 (Mammalia: Chiroptera: Pteropodidae) from Luzon, Phil- ippines. — Proceedings of the Biological Society of Washington 107:1-6. Rutland, R. W. 1968. A tectonic study of part of the Philippine Fault Zone. — Quarterly Journal of the Geological Society of London 123(4):293-325. Taylor, E.H. 1922. Additions to the herpetological fauna of the Philippine Islands, II.— Philippine Journal of Science 21(3):253-303. SAS Institute Inc. 1988a. SAS/STAT user’s guide, release 6.03 edition. SAS Institute Inc., Cary, North Carolina, 1028 pp. 1988b. SAS procedures guide, release 6.03 edition. SAS Institute Inc., Cary, North Caro- lina, 441 pp. Sokal, R. R., & F. J. Rohlf. 1981. Biometry, second 28 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON edition. W. H. Freeman and Co., New York, ——, , & P. Escalante P. 1990. Morpholog- 859 pp. ical evolution in muroid rodents I, Conservative UNESCO/ECAFE. 1971. Geologic map of Southeast patterns of craniometric covariance and their Asia. United Nations Publication, No. 69-30632. ontogenetic basis in the Neotropical rodent ge- Voss, R. S., & L. F. Marcus. 1992. Morphological nus Zygodontomys. —Evolution 44:1568—1587. evolution in muroid rodents II. Craniometric Whitmore, T. C. 1984. Tropical rain forests of the factor divergence in seven Neotropical genera, Far East. Clarendon Press, Oxford, England, 718 with experimental results from Zygodonto- pp. mys. —Evolution 46:1918—-1934. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(1):29-44. 1995. Revision of the South American freshwater fish genus Sternarchorhamphus Eigenmann, 1905 (Ostariophysi: Gymnotiformes: Apteronotidae), with notes on its relationships Ricardo Campos-da-Paz Secao de Peixes, Museu de Zoologia da Universidade de Sao Paulo, Avenida Nazaré 481, Ipiranga—Sao Paulo—SP, CEP 04263-000, Brasil; Laboratorio de Ictiologia Geral e Aplicada, Departamentos de Zoologia e Biologia Marinha, Instituto de Biologia/CCS, Universidade Federal do Rio de Janeiro, Caixa Postal 68049, Ilha do Fundao— Rio de Janeiro— RJ, CEP 21944-970, Brazil Abstract. —A revision of the apteronotid genus Sternarchorhamphus Eigen- mann is presented. Sternarchorhamphus muelleri (Steindachner) from the Amazonas and Orinoco river systems, is redescribed and a lectotype is des- ignated. Sternarchorhamphus hahni Meinken, a nominal species previously assigned to this genus, is transferred to another gymnotiform genus on the basis of evidence presented herein. As a consequence, Sternarchorhamphus is now considered monotypic. A brief discussion of Mago-Leccia’s apteronotid sub- family Sternarchorhynchinae is provided and its monophyly is tentatively ac- cepted. Relationships of Sternarchorhamphus muelleri within the Sternar- chorhynchinae remain obscure, due to an absence of information on the phyletic history of closely related species, such as Orthosternarchus tamandua (Boulen- ger) and Ubidia magdalenensis Miles. A discussion on the taxonomic status of nominal species previously assigned to Sternarchorhamphus is provided. Resumo. —Uma revisao do género de Gymnotiformes Sternarchorhamphus Eigenmann (familia Apteronotidae) € apresentada. Sternarchorhamphus muel- leri (Steindachner), das bacias dos rios Amazonas e Orinoco, é redescrita, e lectotipo é designado para esta espécie. Uma espécie nominal previamente referida a Sternarchorhamphus, S. hahni Meinken, é removida deste género; como consequéncia, Sternarchorhamphus é agora considerado género mono- tipico. Uma breve discussao sobre a subfamilia de Apteronotidae proposta por Mago-Leccia, Sternarchorhynchinae, é¢ apresentada, e sua monofilia é tentati- vamente aceita. A posicao filogenética de Sternarchorhamphus dentro da sub- familia permanece obscura, 0 que se deve, em parte, a auséncia de informacoes sobre espécies relacionadas aS. muelleri, tais como Orthosternarchus tamandua (Boulenger) e Ubidia magdalenensis Miles. E feita uma discussao sobre o status taxon6mico de espécies nominais previamente incluidas em Sternarchorham- phus. The Neotropical gymnotiform genus Sternarchorhamphus Eigenmann (in Eigen- mann & Ward 1905) was established to in- clude three nominal species with elongate snouts, originally described in Sternarchus Bloch & Schneider (currently a junior syn- onym of Apteronotus Lacépéde; e.g., Fowler 1951), namely: Sternarchus (Rhamphoster- narchus) macrostomus Gunther, 1870 (from Jeberos [=Xeberos], Rio Maranon basin, 30 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Peru), Sternarchus (Rhamphosternarchus) muelleri Steindachner, 1881 (type species of Sternarchorhamphus, from the Rio Ama- zonas, at Para State, Brazil), and Sternar- chus tamandua Boulenger, 1898 (from the Rio Jurua, a tributary of the Rio Amazonas in Brazil). Eigenmann (in Eigenmann & Ward 1905:166), however, noted that this latter species was probably a member of “‘a genus distinct from Sternarchorhamphus as here understood” and, in fact, it was sub- sequently used by Ellis (1913) to establish the monotypic apteronotid genus Ortho- sternarchus. The first appearance of the name Stern- archorhamphus was in a key to genera of Gymnotiformes (““Gymnotidae”’ of earlier authors) presented by Eigenmann & Ward (1905). An unequivocal citation present in that publication (p. 165) makes Eigenmann the sole author of the genus, according to Article 50a of the International Code of Zoological Nomenclature (1985). In that key, the new taxon was first included in the ““Sternarchinae”’ (=Apteronotidae), and distinguished from Sternarchorhynchus Castelnau on the basis of its “snout straight, the gape moderate,” rather than “snout strongly decurved, mouth minute.”’ Addi- tionally, it was stated that Sternarchorham- phus seemed to be “intermediate between Sternarchus and Sternarchorhynchus, hav- ing the long snout of the latter and mouth size approaching the former’ (Eigenmann & Ward 1905:165), a position also held by Ihering (1907). Some years later, Meinken (1937) de- scribed Sternarchorhamphus hahni on the basis of a single specimen from the Rio Pa- rana drainage in Argentina. That was the first record of the genus outside of the Am- azon basin. As discussed below, however, the examination of the holotype of S. hahni showed that it does not belong to the Ap- teronotidae. Mago-Leccia (1976) first noted the pres- ence of Sternarchorhamphus in the Rio Ori- noco system. He also proposed a phyloge- netic scheme for Venezuelan gymnotiform genera where Sternarchorhamphus and Sternarchorhynchus were considered sister- groups within the Apteronotidae (Mago- Leccia 1976, fig. 99). Along with Orthoster- narchus Ellis and Ubidia Miles, those two genera were assigned to the Sternarcho- rhynchinae (not of Hoedeman 1962), a pro- posed apteronotid subfamily whose mem- bers were recognized by their “very elongate snout and reduced mouth gape” (Mago- Leccia 1976:206; see also Mago-Leccia 1978:14 and Mago-Leccia 1994:13). Campos-da-Paz (1992) had recently ex- pressed some doubts about the monophyly of the Sternarchorhamphus plus Stern- archorhynchus clade, but tentatively rec- ognized Mago-Leccia’s subfamily Stern- archorhynchinae. In a recent investigation, Triques (1993) did not find characters supporting a close relationship between Sternarchorhamphus and Sternarchorhynchus. Instead, he pro- posed Sternarchorhamphus as the sister- group to a subunit of the Apteronotidae in- cluding the short-snouted genera Adonto- sternarchus Ellis, plus Sternarchella Eigen- mann and Porotergus Ellis (Triques 1993: 123, fig. 24). Mago-Leecia (1994) recently included two species in Sternarchorhamphus: S. muelleri and S. hahni. He stated in a footnote (p. 36), however, that this latter species could belong to a different gymnotiform genus, but without additional discussion. Sternar- chorhamphus macrostomus was used by him to establish a new monotypic apteronotid genus, Platyurosternarchus Mago-Leccia, whose relationships are currently unknown. The Sternarchorhynchinae appear only as part of a classificatory system proposed by Mago-Leccia (1978) (Mago-Leccia 1994:13). The present study is a revision of Stern- archorhamphus Eigenmann. A detailed morphological description and osteological analysis of the type species, S. muelleri, is presented, along with data on geographic distribution based on recent collections from VOLUME 108, NUMBER 1 the Amazon and Orinoco drainages. Re- examination of the holotype of S. hahni re- sulted in a transfer of this taxon to another gymnotiform genus. Methods. —Most measurements were taken according to Mago-Leccia (1978) and were made point-to-point with dial calipers and ruler on the left side of specimens when- ever possible. Measurements that require further explanation are the following: LEA (Mago-Leccia et al. 1985) is the distance from the tip of snout to posterior end of anal-fin base; interorbital width is the space between the dorsal margins of the eyes; head depth was taken at the occiput; and tail depth was measured at posterior end of the anal fin. ““HL” denotes head length. In the counts of pectoral-fin rays, the an- terior unbranched rays are indicated by low- er-case roman numerals, and branched rays are indicated by arabic numerals. The term “branched rays”’ refers to all rays posterior to the anterior unbranched rays, even if the posterior terminal ray is unbranched at its base or distal segment. All specimens with indications of regeneration and/or injuries at the posterior region of body were exclud- ed from measurements of total length (TL) and tail length (CL), and counts of caudal- fin rays. Morphometric and meristic data for the paralectotype of Sternarchorham- phus muelleri are given between brackets. Osteological illustrations were prepared with the aid of a camera lucida, from a spec- imen cleared and counterstained for bone and cartilage with Alizarin red and Alcian blue respectively, according to the proce- dure of Taylor & Van Dyke (1985). Verte- bral counts were made on this specimen and from radiographs of additional specimens, and follow Mago-Leccia et al. (1985). Bone terminology follows Mago-Leccia et al. (1985), except that the terms mesethmoid (e.g., Fink & Fink 1981) and anguloarticular (see Nelson 1973) were used instead of eth- moid and angular, respectively. In the list of material examined, “‘c&s’’ indicates a cleared and stained specimen. 31 Specimens examined for the present study are deposited in the following institutions: American Museum of Natural History, New York (AMNH), Academy of Natural Sci- ences of Philadelphia, Philadelphia (ANSP), Instituto Nacional de Pesquisas da Ama- zonia, Manaus (INPA); Museu Nacional, Rio de Janeiro (MNRJ), Museu de Zoologia da Universidade de Sao Paulo, Sao Paulo (MZUSP), Naturhistorisches Museum, Wien (NMW), National Museum of Natu- ral History, Smithsonian Institution, Wash- ington D.C. (USNM), and Zoologisches Museum, Berlin (ZMB). In the ‘‘Material Examined”’ section, names of collectors are cited when known, and the term ““EPA”’ re- fers to the ““Expedicao Permanente da Ama- zonia,” a Brazilian field expedition under the direction of P. Vanzolini (MZUSP) fi- nanced by the Fundacao de Amparo a Pes- quisa do Estado de Sao Paulo (FAPESP), Sao Paulo State Government. Sternarchorhamphus Eigenmann, 1905 Sternarchorhamphus Eigenmann in Eigen- mann & Ward 1905:160, 165-166, pl. VIII, fig. 7, partim, type by original des- ignation Sternarchus (Rhamphosternar- chus) muelleri Steindachner, 1881.— Ihering 1907, in key to ““Gymnotidae,”’ key to species of Brazil.—Eigenmann 1910, list of species.—Regan 1911, list- ed.—Ellis 1912, in key to “‘Sternarchin- ae.” —Ellis 1913, in key to “‘Sternarchin- ae,’ described, food items, mutilation and regeneration of caudal peduncle. — Jordan 1920, listed, Eigenmann & Ward consid- ered as authors.—Eigenmann & Allen 1942, in key to “Apteronotinae,”’ briefly described, list of species.— Mules 1945, compared to Ubidia. —Miles 1947, com- pared to Ubidia.—Fowler 1951, Brazil, list of species.—Travassos 1960, synon- ymy, inclusion of S. hahni.—Gery & Vu- Tan-Tué 1964, briefly compared to Gym- norhamphichthys Ellis, present in key to Apteronotidae, upper Amazon.—Britski 32 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. of apteronotid fishes. a. Widespread condition in the Apteronotidae (modified from Meunier & Kirsch- baum, 1984:140, fig. 5a). b. Caudal-fin skeleton of Ster- narchorhamphus muelleri. Diagrammatic view of caudal-fin skeleton 1972, reported from Rio Parana basin. — Mago-Leccia 1976, described from Ven- ezuela, osteology, first record in the Rio Orinoco basin, sister group of Stern- archorhynchus, assigned to the “Stern- archorhynchinae.’’— Mago-Leccia 1978, listed in the “‘Sternarchorhynchinae.” — Bullock et al. 1979, listed. —Kramer 1990, discussion on electric organ discharge patterns.—Campos-da-Paz 1992, com- pared to Sternarchorhynchus. —Triques 1993, relationships, Rio Orinoco. — Mago- Leccia, 1994, diagnosis, brief description, notes on osteology, species. Diagnosis. — According to recent studies on interrelationships of gymnotiform gen- era (Mago-Leccia 1976, Triques 1993; see also ““Note on Relationships,’ below), the following characters are derived for Stern- archorhamphus, among closely related ap- teronotids: caudal-fin rays reduced in num- ber (two to five rays; Fig. 1); presence up to five irregular rows of diminutive dentary teeth (most easily observed on the anterior portion of this bone in adults); presence of a mesocoracoid; and snout slightly turned dorsally. The following combination of characters, either primitive or of uncertain polarity, distinguishes Sternarchorhamphus from all other known apteronotids: snout elongated and laterally compressed; upper jaw pro- duced and pointed, lower jaw somewhat in- cluded; teeth minute, present on both jaws at all ages; mouth small to moderate, 11.2- 16.5% [16.5%] in HL; eyes small, 2.5—4.4% [3.6%] in HL; lateral ethmoids present; scales rare or absent on antero-dorsal, and dorsal regions of body; dorsal fleshy filament orig- inating on anterior third of total length and extending to vertical through posterior end of anal fin. Etymology. —From Sternarchus, an ear- lier nominal genus of the Apteronotidae and currently junior synonym of Apteronotus, and the Greek rhamphos, for beak. Gender masculine. Distribution. —Most examined speci- mens of Sternarchorhamphus were collect- ed at localities near the main channels of the Rio Amazonas and Rio Orinoco, and their large tributaries. Additional speci- mens came from the vicinities of Belém (Para State, Brazil), Amapa (Amapa State, Brazil), and the Rio Ucayali basin (Peru; Fig. 2). Mago-Leccia (1976) was the first to record Sternarchorhamphus from Venezue- la, studying specimens from Rio Portuguesa and Rio Apurito. Recent collections from the Rio Orinoco originated in the deep river channel (examined specimens from the Ori- noco Delta and middle Orinoco, collected by the crew of R/V Eastward, 1978-1979, deposited at AMNH, ANSP, and USNM). Ellis (1913) and Santos et al. (1984) ex- amined specimens from the Rio Tocantins basin. Note on relationships. —Published phy- logenetic hypotheses of the Apteronotidae do not include more than six of the ten cur- rently accepted genera (e.g., Mago-Leccia 1976, Triques 1993) and, therefore, a com- prehensive cladogram for the family is still unavailable (efforts on this subject are in progress elsewhere; J. Albert, and M. Tri- ques, pers. comm.). It is beyond the scope VOLUME 108, NUMBER 1 NN oA x SWE oY 33 : Fig. 2. Geographic distribution of Sternarchorhamphus muelleri. Some symbols represent more than one collection locality and/or lot of specimens. of this study to present a detailed discussion on the intrafamilial relationships of the Ap- teronotidae. Furthermore, specimens of the apteronotids Orthosternarchus tamandua (Boulenger, 1898) and Ubidia magdalenen- sis Miles (1945) are extremely rare in col- lections, making it impossible to examine their internal anatomy and precluding a rig- orous test of Sternarchorhynchinae mono- phyly (O. tamandua MZUSP 2647 [1 ex.] and U. magdalenensis USNM 123795 [1 e€X.; paratype] were examined for external features). The characters used by Mago-Leccia to define the Sternarchorhynchinae (1976), “‘reduced mouth gape’’ and “elongate snout,” and to include Sternarchorhampus in it, are rather vague; their status as sec- ondary homologues (=synapomorphies; see Pinna 1991) depends not much from their overall similarity, but rather on a well-cor- roborated hypothesis of relationships of closely related groups. An attempt to give these characters a more objective approach is presented below. Sternarchorhynchines all have the mouth gape length less than 38% of the snout length. In remaining apteronotids, most gymnoti- forms (except rhamphichthyids), and most closely related non-gymnotiform ostario- physans (Characiformes and Siluriformes; see Fink & Fink 1981), this value is con- sistently greater than 55%. When mouth gape length is compared to postorbital length, the value is always less than 31% in sternarchorhynchines, and greater than 40% in the proposed non-rhamphichthyid out- groups. In most apteronotids, remaining gym- notiforms (but, again, not rhamphi- chthyids), and most closely related non- gymnotiform ostariophysans, snout length represents less than 40% in head length. In Orthosternarchus, Sternarchorhamphus and Sternarchorhynchus this value is always greater than 50%; Ubidia, however, is a 34 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON unique case. Although it also has a con- spicuously elongate snout, the eye in this genus is located well forward in the head (an autapomorphic feature; see Mago-Lec- cia 1994:159, fig. 56B), resulting in the snout length/head length ratio decreasing to around 40%. Comparing snout length to postorbital length results in values of more than 90% for sternarchorhynchines (except Ubidia which, because of eye position, has the snout 70% of postorbital length) and less than 80% in remaining non-rhamphi- chthyid groups cited. Currently available evidence indicates that the Rhamphichthyidae is distantly re- lated to the Sternarchorhynchinae (Mago- Leccia 1976, 1978; Triques 1993), and the “reduced mouth gape” and an “elongate snout” can be hypothesized as being inde- pendently acquired in these taxa. Some authors have called attention (e.g., Schaefer 1987) and discussed (e.g., Pimentel & Riggins 1987) the questionable general utility, in phylogenetic analyses, of propor- tional differences observed between mor- phological characters. It should be noted, however, that the above discussion is based on previous phylogenetic hypotheses con- cerning external (Fink & Fink 1981) and internal (Mago-Leccia 1976, 1978; Triques 1993) gymnotiform relationships. Further- more, the utilization of morphometry in the present case is an attempt to make the terms “‘reduced”’ (the mouth) and “elongate” (the snout) as objective as possible (this kind of strategy is commonly observed in literature; e.g., Weitzman & Fink 1985 [fig. 79, char- acters 18, 32, and 40]). Additional justifi- cation for utilization of morphometry was found in Chappill (1989:231), who stated that “‘[q]uantitative characters should gen- erally only be used [i.e., in phylogenetic analyses] when the choice is between ex- amining them or abandoning the analysis entirely for want of sufficient characters.” In some cases, however, there is evidence that morphometry can explicitly be used for phylogenetic purposes (e.g., Pinna 1989:24 [character 2], Costa 1990 [fig. 32, characters 13, 34, 39, 46, 80 and 89], Schaefer 1991 (fig. 15, character 13], and Vari 1991 [fig. 11, characters 17 and 42]). In conclusion, the monophyly of the Sternarchorhynchinae, including Sternar- chorhamphus, is tentatively accepted. The position of this genus within the subfamily, however, remains uncertain until additional material of related genera is available, which will allow a more detailed analysis of this question. Sternarchorhamphus muelleri (Steindachner, 1881) Figs. 1-5, Table 1 Sternarchus (Rhamphosternarchus) Mulleri Steindachner 1881:99, original descrip- tion, Para State, Brazil.—Steindachner 1882:15, pl. V, fig. 4, described, no exact locality, Rio Amazonas at Para State, Brazil. Sternarchorhynchus mulleri (not S. mulleri Castelnau 1855).—Eigenmann & Eigen- mann 1891:62, listed, Para. —Eigenmann 1894:625, listed, Para. Sternarchorhamphus mulleri. —Eigenmann & Ward 1905:165-—166, pl. VIII, fig. 7, as- signed as type species of Sternarchorham- phus, original designation, Para. — Eigen- mann 1910:449, listed, Para to Peru.— Ellis 1913:142-143, 174, 182-183, figs. 10 and 23-25, described, food habits, re- generation of caudal peduncle, Alcobaca (=‘*‘Alcoboca’’), Rio Tocantins, Para.— Meinken 1937:79, compared to Sternar- chorhamphus hahni.—Fowler 1939:276, Contamana, Peru.—Eigenmann & Allen 1942:320, listed, lower Amazon to the Ucayali.— Fowler 1945:180, fig. 65, Peru (Contamana) and Amazonas.— Miles 1947:184, compared to Ubidia magda- lenensis. Sternarchorhamphus muelleri. —\hering 1907:277, Amazonas, Para.— Mago-Lec- cia 1976:244—249, fig. 78, described from Rio Orinoco, Venezuela. — Ortega & Vari VOLUME 108, NUMBER 1 1986:12, Peru.—Campos-da-Paz 1992: 24, 134, briefly compared to Sternar- chorhynchus. —Mago-Leccia, 1994:35, 36, 71, 107, 156, 203, fig. 52, brief descrip- tion, notes on osteology, listed. Sternarchorhamphus mulleri.—Starks 1913: 23, described, variation of anus position, Para.—Santos et al. 1984:18, 78, figure, in list of species from lower Rio Tocan- tins, common names, Brazil.—Triques 1993:91, listed, discussion of relation- ships to other gymnotiforms. Sternarchus mulleri. —Jordan 1920:512, cited as type species of Sternarchorham- phus. Sternachorhamphus [misspelling] muel- leri. —Magalhaes 1931:178, cited. Sternarchus mulleri.—Travassos 1960:24, cited as type species of Sternarchorham- phus. Sternarchoramphus [misspelling] mul- leri. —Begossi & Braga 1992:107, tb.2, and 11, fig. 5, common name in Rio Tocan- tins, listed as fish avoided as food. Material examined. —Brazil: NMW 65328:1, lectotype (photograph; new des- ignation), 260.5 mm LEA, no exact collec- tion site, Rio Amazonas at Para State, F. Steindachner donation, accession file num- ber “1874.1.299 & 299a”; NMW 65328:2, paralectotype, 249.0 mm LEA, same data of lectotype; INPA 4850, 7 specimens, Ca- maleao, Ilha da Marchantaria, Rio Soli- moes, Amazonas State, 3 Feb 1982, G. M. Soares; INPA 4852, 5 specimens, Cama- leao, Ilha da Marchantaria, Rio Solimoes, Amazonas State, 1 Jun 1981, G. M. Soares; MNRJ 1221, 1 specimen, no collection data; MNRJ 1222, 2 specimens, no collection data; MNRJ 9022, 2 specimens, ‘Ver-o- Peso’ market, Belem, Para State, 10 Feb 1958, L. Travassos & F. Pires leg.; MNRJ 12182, 1 specimen, Amapa, Amapa State, Apr 1981, G. W. Nunan et al.; MZUSP 6983, 7 specimens, Rio Madeira, 25 km below Nova Olinda, Amazonas State, 27 Sep 1967, EPA; MZUSP 9580, 1 specimen, 35 fish market at Manaus, Amazonas State, 17— 19 Sep 1968, EPA leg.; MZUSP 23321, 1 specimen, mouth of Parana do Catito, be- low mouth of Rio Jurua, Rio Solimdées, Amazonas State, 4 Oct 1968, EPA; MZUSP 24675, 1 specimen, Santa Luzia, Rio Purus, Amazonas State, 11 Jan 1975, P. E. Van- zolini; USNM 52547, 2 specimens, Para to Manaus, Rio Amazonas, 1901, J. B. Steere. Peru: ANSP 95834, 2 specimens, Rio Ucayali at Cantamana, Jul-Aug 1937, W. C. Morrow; ANSP 120348, 2 specimens, Pucallpa, on Rio Ucayali, 18-19 Jun 1969, E. J. Huggins. Venezuela: ANSP 149460, 3 specimens, shallow river, N side of river across from Isla Tres Canos, Delta Amacuro, 13 Nov 1979, R/V Eastward (H. Lopez, M. Cor- coran); ANSP 149457, 1 specimen, Rio Ori- noco, below Barrancas, ca. km 140, depth 26 m, Delta Amacuro, 17 Feb 1978, R/V Eastward (Lundberg & Baskin); ANSP 160250, 6 specimens, Rio Guariquito at confluence of Rio Orinoco, Estado Bolivar, 25 Nov 1985, B. Chernoff et al.; ANSP 162297, 2 specimens, Rio Orinoco, near mouth of Rio Caura, Estado Bolivar, 22 Nov 1985, G. J. Lundberg et al.; ANSP 166792, 1 specimen, Caicara, L. Bartolico, Estado Bolivar, Rio Orinoco basin, 20 May 1987, M. Rodriguez & R. Richardson; ANSP 166793, 1 specimen, Caicara, Cas- tillero, Estado Bolivar, Rio Orinoco basin, 19 Apr 1988, M. A. Rodriguez & A. Mar- tinez; USNM 226339, 1 specimen, Rio Ori- noco, deep river channel, Brazo Imataca, 82 n.m. upstream from sea buoy, Territorio Federal Delta Amacuro, 22 Feb 1978, D. Taphorn; USNM 226491, 2 specimens, Rio Orinoco, deep river channel, north side of Isla Tortola, 123 n. m. from sea buoy, Delta Amacuro Fed. Territory, 19 Feb 1978, J. N. Baskin; USNM 226495, 3 specimens, Rio Orinoco, Brazo Imataca, south from Isla Remolino, ca. km 82.9 from sea buoy, Delta Amacuro Fed. Territory, 22 Feb 1978, Baskin/Lundberg; USNM 228805, 6 spec- imens (1 C&S), across from Palua, ca. 182 36 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Morphometric data for the paralectotype (NMW 65328:2) and additional non-type material of Sternarchorhamphus muelleri. TL, LEA, HL, and CL are expressed in mm; measurement | is expressed as % of TL; measurements 2 to 7 are expressed as % of LEA; 8 to 21 as % of HL; 22 as % of CL. SD, standard deviation; and n, number of examined specimens. Paralectotype Range X SD n TL 292.0 261.0-446.0 Ss 15 LEA 249.0 188.0—375.0 - — 17 HL 30.0 20.5-63.3 - — 28 CL 43.0 35.0-77.0 - — tS) 1) Standard length 85.2 81.4-88.2 83.9 + 1.7 15 2) Anal-fin length 90.4 89.1-93.0 Oiie sale 15 3) Body depth 10.1 8.7-12.2 10.2 + 0.8 18 4) Head length 12.1 11.4-14.5 12.4 + 0.9 18 5) Pre-anal distance 9.3 8.0-10.8 Wh se 07 18 6) Pre-pectoral distance Doi 11.3-15.4 WoW ae Ilo 18 7) Tail length 72 16.7—22.7 HOLS =e9 14 8) Snout length 49.5 45.9-52.9 49.3 + 2.0 28 9) Eye diameter 3.6 2.54.4 3.3 + 0.4 28 10) Mouth width 16.5 11.2-16.5 13.3 + 1.4 25 11) Interorbital width 11.9 8.0-13.0 10.6 + 1.4 28 12) Snout to occiput 90.1 83.5-91.7 87.4 + 2.3 28 13) Postorbital distance 53.4 46.9-53.8 50.5 + 1.9 28 14) Pectoral-fin length 45.2 43.5-59.0 SOLO) as S77 28 15) Pre-anal-fin length 715.9 68.8-79.8 VAS) se ZO 26 16) Snout to anus 58.7 48 .5-60.2 54.8 + 3.5 Di) 17) Anterior naris—posterior naris 5.6 3.3-5.6 4.3 + 0.6 26 18) Posterior naris—eye MoT 22.0-30.3 26.8 + 2.1 26 19) Head depth 62.0 52.1-63.9 S75 se Sats} 26 20) Head width 24.1 22.2-29.9 2p eal 28 21) Branchial opening 19.4 15.6-19.4 1 2 IED 28 22) Tail depth 9.7 6.3-10.6 Sie ales 14 n. m. from sea buoy, 9 Nov 1975, H. Lopez & O. Riviero; USNM 228806, 6 specimens, shallow river, downstream from sea buoy 82, near mouth of a small cano, Delta Ama- curo Fed. Territory, 21 Nov 1979, H. Lopez et al.; USNM 228808, 1 specimen, shallow river, north shore, 49 n. m. from sea buoy, Delta Amacuro Fed. Territory, 20 Nov 1979, Lopez et al. Diagnosis. —As for the genus. Description. —Morphometric data for the paralectotype (NMW 65328:2) and addi- tional specimens of S. muelleri are given in Table 1. The lectotype (NMW 65328:1) is shown in Fig. 3. Body strongly compressed laterally, es- pecially posterior to abdominal cavity, deepest in this region; dorsal profile from nearly straight to gently convex. Lateral line complete, extending to caudal peduncle, but not onto caudal fin in some specimens. Head laterally compressed, more mark- edly so anteriorly, deepest at occiput and widest in opercular area; snout pointed and conical, slightly turned dorsally in most specimens; eyes reduced in size, completely covered with skin, located dorso-laterally on head; small (paired) sensory pore pres- ent, dorsal to eye and usually near vertical through anterior border of eyes. Mouth small, sub-terminal (inferior in some specimens), its gape usually not reach- ing vertical through anterior border of an- terior nare; upper jaw projected, lower jaw somewhat included. Single patch of numer- ous diminutive conical teeth present on pre- 31/ VOLUME 108, NUMBER 1 “QUIS UOT}OI{JOO 1OeXd OU ‘a}B1S BIL 1e SPUOZEUY OTY ‘[IZeIg ‘odAy0j99] “WaT WU C097 “TL WU C'PIE ‘T:8Z7€S9 MINN “Hapjjaniu snydupys0yoivUusays ViBCES9 um MW SN EN oe PONG [KGa ‘¢ “31 38 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON maxilla (25-40, ¥ = 34.1, nm = 7 [approx. 35 in paralectotype]). Usually four of five irregular rows of similar teeth on dentary (25-45, X = 36.4, n = 7 [approx. 25 in par- alectotype]). Maxilla and mesopterygoid edentulous. Small rounded papillae cover- ing part of the roof and floor of the oral cavity. Anterior nares tubular; posterior ones rounded, without a tube and remote from eyes. Branchial opening reduced to a short slit immediately anterior to pectoral-fin origin; branchial membranes joined or partially joined to isthmus. Anterior chamber of gas bladder small, posterior elongate. Anus and urogenital papilla adjacent and located ven- tral to opercular region, shifting anteriorly with age. Pectoral fins moderate, elongate and slightly pointed, with 11 + 12-14 rays [ii + 13]. Anal fin with 230-262 rays [231], its origin near vertical through nape; first an- terior anal-fin rays less developed (some un- divided) and smaller than posterior ones. Scales small, cycloid, absent or greatly re- duced in number on anterior region of body and dorsally; lateral line perforated scales usually larger than those immediately dor- sal and ventral to it. Scales above lateral line four to nine. Small scales, sometimes in single series, present on caudal fin in some specimens. Dorsal filament (=“‘dorsal thong”; a mod- ified muscle according to Mago-Leccia 1994) originating on anterior third of total length, its tip reaching vertical through posterior end of anal fin. Tail (region from base of last anal-fin ray to tip of caudal fin) elongate, laterally compressed, ending in reduced caudal fin with two to five rays [four]; con- dition unique among Gymnotiformes. Cau- dal peduncle sometimes with discrete con- striction at caudal-fin base. Osteological features of Sternarchorham- phus muelleri, relevant at the generic level, are as follows: premaxilla broadest anteri- orly, and extending posteriorly to near the maxilla; maxilla elongate and curved pos- teriorly, its anterior portion well-developed (Figs. 4 and 5); dentary large, extending pos- teriorly and covering anterior portion of re- troarticular; Meckel’s cartilage well-devel- oped and elongate, partially associated with anguloarticular and dentary; coronomeck- elian bone reduced in size, compared to ad- jacent bones; retroarticular well-developed, with pointed antero-ventral process; lateral ethmoids present; vomer elongate, its an- terior portion “arrow-shaped” and contact- ing parasphenoid through small cartilagi- nous bridge; posterior end of vomer point- ed; palatine cartilage present, well-devel- oped anteriorly and contacting anterior portion of maxilla; mesopterygoid broad, edentulous, with poorly-developed ascend- ing process; parasphenoid elongate, bifur- cate anteriorly and posteriorly; infraorbital series represented only by canal-bearing portions of bones; mesethmoid elongate, rounded and reduced at its anterior portion; two cranial fontanels present and well-de- veloped (interfrontal larger than interpari- etal); supraoccipital small, with reduced crest; posttemporal fossae absent; opercle ornamented with numerous small trabecu- lae; mesocoracoid present; scapular fora- men absent; coracoid with well-developed postero-ventral process, but not reaching cleithral symphysis; posttemporal fused to supracleithrum; extrascapular present; two postcleithra; four pectoral radials; four branchiostegal rays, first and second almost filamentous, others large and laminar; uro- hyal broad and expanded posteriorly with reduced head, and approaching in size the basihyal and first ceratobranchial; gill rakers not ossified; four infrapharyngobranchials, fourth one cartilaginous; five epibranchials, fifth one cartilaginous; upper pharyngeal tooth plate with seven or eight teeth, con- nected to third epibranchial through a lig- ament; lower pharyngeal tooth plate with 12-13 teeth; Weberian apparatus without claustrum; 16-17 precaudal vertebrae (We- berian complex included); two “‘rib-like bones”’ (modified ribs?) present in posterior wall of abdominal cavity, anterior one en- larged, laterally compressed and turned an- VOLUME 108, NUMBER 1 39 de — Fig. 4. Jaws of S. muelleri, USNM 228805, 188.8 mm LEA. Lateral view of left side. Scale bar = 1 mm. aa, anguloarticular; de, dentary; ma, maxilla; pm, premaxilla; ra, retroarticular. teriorly; proximal pterygiophores of anal fin with expanded, pointed, symmetric projec- tions anteriorly, directed dorsally on the an- terior third of pterygiophore, well-devel- oped posterior to abdominal region; 91—99 vertebrae to base of last anal-fin ray (We- berian complex included); well-developed intermuscular bones, especially conspicu- ous dorsally and ventrally to vertebral col- umn, and posterior to abdominal region; caudal-fin skeleton consolidated into a sin- gle element of reduced size. Color in alcohol.—Body light yellowish brown, covered with diminutive irregular dark spots (chromatophores), most on dor- sal portion of head and snout, and on back. The overall coloration can vary from pallid to melanistic. Pectoral fins hyaline at base, -and usually dark distally (chromatophores on fin membranes); anal fin of some spec- imens with continuous black margin along entire length, with melanophores also pres- ent over fin membranes; specimens from the Rio Orinoco, however, can be nearly completely pallid (J. Lundberg, pers. comm.). Caudal fin hyaline. Food habits. —Stomach-contents of Sternarchorhamphus showed fragments of partially digested insect larvae (tentatively identified as Diptera), along with some un- identified Annelida. These findings agree with those by Ellis (1913:174), who had also ra Fig. 5. de Lower jaw of S. muelleri, USNM 228805, 188.8 mm LEA. Medial view of left side. Scale bar = 1 mm. aa, anguloarticular; cb, coronomeckelian bone; de, dentary; mc, Meckel’s cartilage; ra, retroarticular. 40 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON noted a single unidentified Entomostraca specimen and additional insect larvae, other than Diptera. Remarks. —Santos et al. (1984) listed the names “itui” (a common name for most apteronotid fishes in Brazil) and “‘tuvira”’ (also used for other Gymnotiformes, except the Electrophoridae) for S. muelleri in the lower Rio Tocantins region. The same au- thors also observed that this species is the only apteronotid with some commercial value as a food fish in that area. Begossi & Braga (1992), curiously, listed S. muelleri as a fish avoided as food by local fishermen at Rio Tocantins (in States of Maranhao and Tocantins), and recorded the common name ‘“‘lampréia”’ to this species (that name is also assigned for other non-apteronotid gym- notiform fishes occurring in that area). Langner & Scheich (1978) and Kramer (1990) discussed the electric organ discharg- es (EODs) of some gymnotiforms, including specimens they identified as Sternarchor- hamphus, which showed a fundamental fre- quency of EOD between 1300 and 1700 Hz. A figure presented by Langner & Scheich (1978:236, fig. 1), and later reproduced by Kramer (1990:198, fig. 4.57), however, does not show Sternarchorhamphus, but Stern- archorhynchus. Etymology. —Steindachner proposed the name Mulleri (the original spelling in his 1881 paper) for the new species in honor of Johannes Muller, a famous German ichthy- ologist of the 19th century. Several different spelling forms for this species are currently widespread in literature (see synonymy, above). According to the International Code of Zoological Nomenclature (1985; Article 32d), however, the correct form is muelleri. Taxonomic status of Sternarchus (Rhamphosternarchus) macrostomus Gunther, Sternarchus tamandua Boulenger, and Sternarchorhamphus hahni Meinken Gunther (1870) described Sternarchus (Rhamphosternarchus) macrostomus (“‘ma- crostoma’’ in the original spelling; not Ster- narchus macrostomus Fowler [1943]) on the basis of a single specimen from Jeberos (=Xeberos), Rio Maranon drainage, Peru. This nominal species was subsequently as- signed to Sternarchorhynchus by Eigen- mann & Eigenmann (1891; “Upper Ama- zon’’). Eigenmann (in Eigenmann & Ward 1905) included S. macrostomus in Sternar- chorhamphus, without any detailed expla- nation. Ellis (1913) did not examine spec- imens of S. macrostomus and simply quot- ed Gunther’s original description, following Eigenmann’s placement of the species. This view remained unalterated since then, until the recent work of Mago-Leccia (1994), who used the species to establish the monotypic apteronotid genus Platyurosternarchus. Platyurosternarchus first appeared in a key to apteronotid genera (Mago-Leccia 1994: 26). Later, in the same study, that author stated that the genus was “proposed in order to locate properly the species Sternarchus macrostomus,’”’ that the new genus is “‘clear- ly different from Sternarchorhamphus”’ (p. 37), and provided a photograph of the head of a specimen from the Orinoco basin (p. 160, fig. 57A). A list of distinctive features (including uniquely derived characters, such as the overall morphology of caudal fin) was also given by Mago-Leccia, who definitely demonstrated that P. macrostomus and S. muelleri are different species (but note also characters already pointed out by Gunther 1870). Relationships of Platyurosternar- chus, however, were not discussed in that work, and the closest relatives of that genus remain uncertain. A preliminary view of this problem suggests that Platyurosternarchus and Sternarchorhamphus belong to differ- ent subsets within the family Apteronotidae and are not sister groups. As in S. muelleri, the snout in P. macrostomus is elongate (around 50% in head length), but the latter species does not have the reduced mouth gape length characteristic of sternarcho- rhynchines (approximately 65% in snout length, and 75% in postorbital length vs. less VOLUME 108, NUMBER | than 35% in snout length, and less than 31% in postorbital length, respectively). The pos- sibility of Platyurosternarchus macrosto- mus constitutes the sister-group of the Sternarchorhynchinae cannot be dismissed at this time, but further detailed investiga- tion is needed (see discussion on the Stern- archorhynchinae, above). Sternarchus tamandua was described by Boulenger (1898) on the basis of a single specimen, with an injured tail, from Rio Jurua (Brazil). The species was transferred to Sternarchorhamphus by Eigenmann (in Eigenmann & Ward 1905), who was fol- lowed by Ihering (1907) and Eigenmann (1910). Later, the species was used by Ellis (1913) to establish the monotypic apter- onotid genus Orthosternarchus. The name of this genus, however, appeared first in key to genera to ““Gymnotidae” (=Gymnoti- formes) from British Guiana provided by Ellis (1912; type-species’ name was not mentioned), and that should be considered its original description. Curiously, not a sin- gle specimen of Orthosternarchus was re- corded from Guiana until the present date. Little has been published on this poorly known species since then, mainly because of its scarcity in collections. Detailed infor- mation on its internal morphology and os- teology is unavailable. The overall form of the snout is an autapomorphic feature, clearly suggesting that O. tamandua is not conspecific with S. muelleri or any known apteronotid (see, for example, Ellis 1913: 144, fig. 11, Mago-Leccia 1994:147, fig. 41). Since this genus is currently assigned to the Sternarchorhynchinae, the possibility of Orthosternarchus and Sternarchorhamphus are its sister-groups must be left open until enough material is available to allow this kind of investigation. If a sister-group re- lationship between these two genera can be demonstrated, then an interesting taxonom- ic question will arise, since their included species were once referred to a single genus, Sternarchorhamphus. Current available ev- idence is scarce and cannot definitively show 41 a close relationship between O. tamandua and S. muelleri, so these species are here kept in their own separate genera. A study on the single known specimen of Sternarchorhamphus hahni Meinken (ho- lotype, ZMB 31367), collected near Corri- entes, Rio Parana basin in Argentina, re- vealed that it lacks the diagnostic features of the Apteronotidae (e.g., caudal fin inter- nally supported by a single bony element; a dorsal fleshy filament) and should rather be referred to Rhamphichthys Miller & Tros- chel, 1846 (Rhamphichthyidae). In addi- tion to several features observed only in rhamphichthyids and closely related groups among gymnotiforms (Hypopomidae; see Mago-Leccia 1978, and Triques 1993; e.g., no teeth on both jaws; anterior nares not tubular and located close to upper lip), S. hahni has 323 anal-fin rays (330 recorded by Meiken 1937; vs. a maximum of 290 anal-fin rays in related groups), a condition apparently uniquely derived for Rhamphi- chthys (or a subset of it) among closely re- lated taxa. Mago-Leccia (1976, 1994) pre- viously posed some doubts on the taxonom- ic status of S. hahni but, because of lack of access to type material, did not discuss the question in greater detail (a more detailed approach on this subject is currently being published elsewhere; Campos-da-Paz & Paepke 1994). In conclusion, it seems reasonable not to include the nominal species S. hahni Mei- ken and S. macrostomus (Gunther) in Stern- archorhamphus which, otherwise, would make it non-monophyletic. Sternarchus ta- mandu Boulenger remains in Orthosternar- chus Ellis until a detailed phylogenetic study on the Sternarchorhynchinae reveals its po- sition regarding that genus. The solution presented herein is to consider Sternar- chorhamphus a monotypic subunit of the Apteronotidae, comprising only S. muelleri (Steindachner). As discussed above, how- ever, its phylogenetic relationships to other sternarchorhynchine apteronotid genera re- mains uncertain, depending on more com- 42 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON plete and conclusive investigations than those presented so far. Acknowledgments For conversations and discussions on sys- tematics and taxonomy of gymnotiform fishes, I am grateful to J. Albert, H. A. Brit- ski, W. J. E. M. Costa, C. Cox-Fernandes, J. G. Lundberg, N. A. Menezes, H.-J. Paepke, M. C. C. de Pinna, J. P. Sullivan, S. Toledo and M. L. Triques. I am deeply indebted to the following individuals and institutions for their hospitality, access to collections and/or to specimens under their care, logistic and equipment support, loan of specimens and all kinds of additional as- sistance: N. Feinberg, G. J. Nelson and M. S. T. Piza (AMNH); E. B. Bohlke, P. A. Buckup, W. G. Saul, and S. A. Schaefer (ANSP); M. C. C. de Pinna (Field Museum of Natural History, Chicago); M. Jégu (INPA); C. A. Bizerril, U. Caramaschi, D. F. Moraes, Jr., I. B. Moreira, G. W. Nunan and J. C. de Oliveira (MNRJ); F. A. Bock- mann, H. A. Britski, J. L. Figueiredo, N. A. Menezes, O. T. Oyakawa, L. M. S. Porto, and M. L. Triques (MZUSP); B. Herzig (NMW); J. G. Lundberg (University of Ar- izona, Tucson); E. P. Caramaschi, W. J. E. M. Costa, and R. Sachsse (Universidade Federal do Rio de Janeiro); S. Jewett, R. P. Vari, and S. Weitzman (USNM); and H.-J. Paepke (ZMB). Earlier versions of manu- script benefited from comments and sug- gestions by J. Albert, H. A. Britski, W. J. E. M. Costa, J. G. Lundberg, N. A. Menezes, T. A. Munroe, M. C. C. de Pinna, and R. P. Vari. A photograph of the lectotype of S. muelleri was taken by Mrs. A. Schumacher (NMW), and kindly provided by B. Herzig (NMW). The author was financially supported by the Fundagao de Amparo a Pesquisa do Es- tado de Sao Paulo (FAPESP), Sao Paulo State Government, Brazil. The following in- stitutions also funded visits to study part of their gymnotiform collections: Academy of Natural Sciences of Philadelphia (Philadel- phia), American Museum of Natural His- tory (New York), and National Museum of Natural History, Smithsonian Institution (Washington, D.C.). Literature Cited Begossi, A., & F. M.S. Braga. 1992. Food taboos and folk medicine among fishermen from the To- cantins River (Brazil).—Amazoniana 12:101- 118. Boulenger, G. A. 1898. Onacollection of fishes from the Rio Jurua, Brazil.— Transactions of the Zoo- logical Society of London 14:421-428. Britski, H. A. 1972. Peixes de agua doce do estado de Sado Paulo: Sistematica. Pp. 79-108 in Po- luigao e piscicultura.—Comissao Interestadual da bacia Parana-Paraguai, Sao Paulo. Bullock, T. H., N. Fernandes-Souza, W. Graf, W. Hei- ligenberg, G. Langner, D. L. Meyer, F. Pimentel- Souza, H. Scheich, & A. Viancour. 1979. As- pectos do uso da descarga do orgao elétrico e eletrorrecepcao nos Gymnotoidei e outros peix- es amazonicos.— Acta Amazonica 9:549-572. Campos-da-Paz, R. 1992. Revisaéo taxondmica do género Sternarchorhynchus Castelnau, 1855 (Ostariophysi Gymnotoidei, Apteronotidae). Unpublished MLS. thesis, Universidade de Sao Paulo, 159 pp. , & H.-J. Paepke. 1994. On Sternarchorham- phus hahni, a member of the rhamphichthyid genus Rhamphichthys (Ostariophysi: Gymno- tiformes).—Ichthyological Exploration of Freshwaters 5:155-159. Castelanu, F. 1855. Poissons nouveaux ou rares reé- cuillem pendant l’expedition dans les parties centrales de l’Amérique de Sud, de Rio de Ja- neiro a Lima, et de Lima au Para.—Paris, 1/2: 85-95, pls. 45-47. Chappill, J. A. 1989. Quantitative characters in phy- logenetic analysis. — Cladistics 5:217-—234. Costa, W. J.E.M. 1990. Andalise filogenética da fam- ilia Rivulidae (Cyprinodontiformes Aplochei- loidei)— Revista Brasileira de Biologia 50:65— 82. Eigenmann, C. H. 1894. Notes on some South Amer- ican fishes.— Annals of the New York Academy of Sciences 7:625-637. 1910. Catalogue of the freshwater fishes of Tropical and South Temperate America. Pp. 375-511 in Report of the Princeton University Expedition to Patagonia, 1896-1899, Volume Three. ——, & W. R. Allen. 1942. Fishes of the Western South America, I: the Intercordilleran and Am- VOLUME 108, NUMBER 1 azonian Lowlands of Peru; II: the High Pampas, Bolivia, and Northern Chile; with a revision of the Peruvian Gymnotidae, and of the genus Or- estias. University of Kentucky, Lexington, 494 pp. ,& R.S. Eigenmann. 1891. A catalogue of the fresh-water fishes of South America. — Proceed- ings of the United States National Museum 14: 1-81. —., & D. P. Ward. 1905. The Gymnotidae.— Proceedings of the Washington Academy of Sci- ences 7:157-186. Ellis, M. M. 1912. Order Glanencheli. Family X. Gymnotidae. Pp. 422-442 in C. H. Eigenmann, The freshwater fishes of British Guiana, includ- ing a study of the ecological groupings of species and the relation of the fauna of the Plateau to that of the Lowlands. — Memoirs of the Carnegie Museum 5:1-578. 1913. The gymnotid eels of Tropical Amer- ica.— Memoirs of the Carnegie Museum 6:109- 195. Fink, S. V., & W. L. Fink. 1981. Interrelationships of the ostariophysan fishes (Teleostei).—Zoo- logical Journal of the Linnean Society 72:297- 353. Fowler, H. W. 1939. A collection of fishes obtained by Mr. William C. Morrow in the Ucayali river basin, Peru.— Proceedings of the Academy Nat- ural Sciences of Philadelphia 91:219-289. 1943. A collection of freshwater fishes from Colombia, obtained chiefly by Brother Nicéforo Maria. — Proceedings of the Academy of Natural Sciences of Philadelphia 95:260-264. 1945. Los peces del Peru: Catalogo siste- matico de los peces que habitan en aguas per- uanas.— Boletin del Museo de Historia Natural ‘Javier Prado,” Lima 7:176—185. 1951. Os peixes de agua doce do Brasil (3a Entrega).—Arquivos de Zoologia, Sao Paulo 6:405-628. Géry, J., & Vu-Tan-Tué. 1964. Gymnorhamphi- chthys hypostomus petiti ssp. nov., un curieux poisson gymnotoide arénicole.— Vie et Milieu, Suppl. 17:485-498. Gunther, A. 1870. Catalogue of the fishes in the col- lection of the British Museum. —British Muse- um Ed., 8:1-549, London. Hoedeman, J. J. 1962. Notes on the Ichthyology of Surinam and other Guianas. 11. New Gymno- toid Fishes from Surinam and French Guiana, with additional records and a key to the groups and species from Guiana.— Bulletin of Aquatic Biology 3:97-107. Ihering, R. von. 1907. Os peixes de agua doce do Brasil.— Revista do Museu Paulista 7:258—-336. International Commission on Zoological Nomencla- 43 ture. 1985. International Code of Zoological Nomenclature. Third Edition. International Trust for Zoological Nomenclature & British Museum (Natural History), London. University of California Press, Berkeley, Los Angeles, 338 pp. Jordan, D. S. 1920. The genera of fishes. Part IV. From 1881 to 1920, thirty-nine years, with the accepted type of each. A contribution to the stability of scientific nomenclature.— Leland Stanford Junior University Publications, Uni- versity Series, p. 512. Kramer, B. 1990. Electro-communication in Teleost fishes: behavior and experiments. Springer- Ver- lag, Berlin, 240 pp. Langner, G., & H. Scheich. 1978. Active phase cou- pling in electric fish: behavior control with mi- crosecond precision.—Journal of Comparative Physiology 128:235-—240. Magalhaes, A. C. 1931. Monographia brazileira de peixes fluviaes. Graphicars Ed., Sao Paulo, 260 pp. Mago-Leccia, F. 1976. Los peces Gymnotiformes de Venezuela: un estudio preliminar para la revi- sion del grupo en la America del Sur. Unpub- lished Ph.D. thesis, Universidad Central de Venezuela, Caracas, 376 pp. 1978. Los peces de la familia Sternopygidae de Venezuela.— Acta Cientifica Venezolana 29 (Supl. 1):1-89. . 1994. Electric fishes of the continental waters of America.—Biblioteca de la Academia de Ciencias Fisicas Matematicas y Naturales 29:1- ies ——., J. G. Lundberg, & J. N. Baskin. 1985. Sys- tematics of the South American freshwater ge- nus Adontosternarchus (Gymnotiformes, Apter- onotidae).— Contributions in Science, Natural History Museum, Los Angeles County, 358:1- 19. Meinken, H. 1937. Beitrage zur Fischfauna des mit- tleren Parana.III].—Blatter fiir Aquarien und Terrarienkunde 48:73-80. Miles, C. 1945. Some newly recorded fishes from the Magdalena river system. Gymnotidae.—Cal- dasia 3:461-464. 1947. Los peces del Rio Magdalena (‘‘A field book of Magdalena fishes’’). Ministerio de la Economia Nacional, Seccion de Piscicultura, Pesca y Caza, Bogota, 214 pp. Nelson, G. 1973. Relationships of clupeomorphs, with remarks on the structure of the lower jaw in fishes. — Zoological Journal of the Linnean So- ciety 53 (Suppl. 1):333-349. Ortega, H., & R. P. Vari. 1986. Annotated checklist of the freshwater fishes of Peru. —Smithsonian Contributions to Zoology 437:1—25. 44 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Pimentel, R. A., & R. Riggins. 1987. The nature of cladistic data.— Cladistics 3:201-209. Pinna, M. C. C. 1989. A new sarcoglanidine catfish, phylogeny of its subfamily, and an appraisal of the phylectic status of the Trichomycteridae (Teleostei, Trichomycteridae).— American Mu- seum Novitates 2950:1-39. 1991. Concepts and tests of homology in the cladistic paradigm. — Cladistics 7:367-394. Regan, C. T. 1911. The classification of the Teleos- tean fishes of the order Ostariophysi. I. Cypri- noidea. Division 2. Gymnotiformes.— Annals and Magazine of Natural History, ser. 7, 11:23- 26. Santos, G. M., M. Jégu, & B. Merona. 1984. Catalogo dos peixes comerciais do baixo rio Tocantins. — Projeto Tucurui. Eletronorte/CNPq/INPA. Ma- naus, 83 pp. Schaefer, S. A. 1987. Osteology of Hypostomus ple- costomus (Linnaeus), with a phylogenetic anal- ysis of the loricariid subfamilies (Pisces: Silu- roidei). — Contributions in Science, Natural His- tory Museum, Los Angeles County, 394:1-31. 1991. Phylogenetic analysis of the loricariid subfamily Hypoptopomatinae (Pisces: Siluro- idei: Loricariidae), with comments on generic diagnosis and geographic distribution. —Zoo- logical Journal of the Linnean Society 102:1- 41. Starks, E. C. 1913. The fishes of the Stanford Expe- dition to Brazil.— Leland Stanford Junior Uni- versity Publications, University Series 12:1-77. Steindachner, F. 1881. Beitrage zur Kenntniss der Flussfische Sudamerika’s. IIJ.—Anzeiger der Akademie der Wissenschaften, Wien 18:97—100. 1882. Beitrage zur Kenntniss der Flussfische Sudamerika’s. III.—Denkschriften der Akade- mie der Wissenschaften, Wien 44:1-18. Taylor, R., & C. C. Van Dyke. 1985. Revised pro- cedures for staining and clearing small fishes and other vertebrates for bone and cartilage study. — Cybium 9:107-119. Travassos, H. 1960. Ictiofauna de Pirassununga. IV. Subordem Gymnotoidei Berg, 1940 (Actino- pterygii— Cypriniformes).— Boletim do Museu Nacional, Rio de Janeiro, nova série, Zoologia 217:1-34. Triques, M. L. 1993. Filogenia dos géneros de Gym- notiformes (Actinopterygii, Ostariophysi), com base em caracteres esqueléticos.—Comunica- coes do Museu de Ciéncias da PUCRS, série zoologia, Porto Alegre, 6:85—130. Vari, R. P. 1991. Systematics of the Neotropical char- aciform genus Steindachnerina Fowler (Pisces: Ostariophysi).—Smithsonian Contributions to Zoology 507:1-118. Weitzman, S. H., & S. Fink. 1985. Xenurobryconin phylogeny and putative pheromone pumps in glandulocaudine fishes (Teleostei: Characi- dae).—Smithsonian Contributions to Zoology 421:1-121. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(1):45—-49. 1995. A new species of Neoperla (Insecta: Pleocoptera: Perlidae) from Mississippi Bill P. Stark Department of Biology, Mississippi College, Clinton, Mississippi 39058, U.S.A. Abstract. —Neoperla coxi, a new species, is described from male, female, egg and nymphal stages, and a holotype male is designated. Males differ from all known Nearctic Neoper/a in having most of the aedeagal tube spinulose. The new species is apparently endemic to southwest Mississippi. During a recent survey to determine the status of Alloperla natchez Surdick & Stark, 1980, and Haploperla chukcho (Surdick & Stark 1980), two chloroperlid stoneflies en- demic to southwest Mississippi, specimens of an undescribed Neoperla were collected near the type localities of these species. A few additional specimens were found among unidentified material in my collection and from P. K. Lago of the University of Mis- sissippi, and a larger series was obtained from the Mississippi Entomological Muse- um (MEM), Mississippi State University. The material from the MEM was collected by malaise and blacklight traps at three sites in the Homochitto National Forest with the support of National Science Foundation grant DEB-9200856 awarded to Richard Brown. Recent study of extensive Neoperla col- lections from throughout Mississippi and Alabama (Stark & Harris 1986, Stark & Lentz 1988.) suggests the new species is en- demic to the Homochitto National Forest and surrounding areas of southwest Missis- sippi and perhaps Louisiana. The holotype is placed in the National Museum of Natural History (USNM) on in- definite loan from the Mississippi Ento- mological Museum. Paratypes are in the collections of the author (BPS), the Uni- versity of Mississippi (UM) and the Mis- Sissippi Entomological Museum. Termi- nology follows Stark & Lentz (1988). Neoperla coxi, new species Figs. 1-8 Male. — Forewing length 8-10 mm. Gen- eral color yellow patterned with brown. Wings amber brown with darker veins; cos- ta and subcosta pale. Legs yellow except for diffuse dorsal brown spot near apex of fem- ora and a narrow longitudinal brown band extending from femur for *% of outer tibial margin. Cerci pale. Process of tergum 7 tri- angular, upturned in lateral aspect and armed ventrally with 4-5 prominent sen- silla basiconica. Mesal sclerite of tergum 8 with lateral margins parallel for most of length. Hemiterga slightly tapered apically, finger-like process of hemitergum slender and bent slightly ventrolaterad (Fig. 1). Tube of aedeagus slightly sinuate, 3.1 times as long as bulb; conspicuous spicules cover most of tube (Fig. 2). Sac sparsely armed in basal half by a pair of basolateral patches of slender spines (Figs. 2-3); apical arma- ture of scattered, thickened spines (Fig. 2), mid section unarmed (Figs. 2-3). Female. —Forewing length 11-13 mm. Posterior margin of sternum 8 produced into a small rounded to slightly bilobed tab (Fig. 4). Spermathecal stalk membranous except for small apical section lined with fine brown setae (Fig. 5); stalk grotesquely inflated in some specimens. Egg. —Length 0.31 mm; width 0.19 mm. Sessile collar surrounded by a single irreg- 46 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Figs. 1-5. Neoperla coxi male and female genitalia. 1, Male terminalia; 2, Aedeagal tube, lateral; 3, Aedeagus, sac partially extruded, apical armature not shown; 4, Female sterna 8 and 9; 5, Vagina and spermathecal stalk, dorsal. T = tube; S = sac; B = bulb; H = hemitergum. ular row of reticulation. Striae relatively straight, narrowed at each end and separat- ed by wide sulci. Sulci with 4—5 irregular aeropyle rows. Micropyles in sulci near equator (Figs. 7—8). Mature nymph.—Body length 7—9 mm. General color yellow patterned with brown. Anterior of frons covered with a broad transverse brown pigment band; posterior margin of band sinuate, anterior margin straight. Ocellar area covered with a small diffuse brown spot. Pronotal disk ringed by a sub-marginal pigment band. Each abdom- inal tergum with a narrow transverse pig- ment band (Fig. 6). Etymology. —I am pleased to name this species for my friend and colleague, Prentiss G. Cox, in recognition of his interest and support of research at Mississippi College. Types.— Holotype 6 and 104 6 and 48 2 VOLUME 108, NUMBER 1 paratypes from Middleton Creek, Franklin County, Mississippi, TSN R4E Sec 31S, 1 Jun 1992, T. Schiefer, R. Fontenot (USNM, MEM). Additional paratypes, all from Mis- sissippi: Claiborne Co: Little Sand Creek, Rocky Springs, 14 May 1977, B. Stark, 1 2 reared (BPS). Sand Creek tributary, 0.5 mi W Hunt Road, 18 May 1978, B. Stark, 1 2 (BPS). Same location, 3 Jun 1986, B. Stark, J. Ballard, 1 2 (BPS). Copiah Co: Brushy Creek, Hwy 27, 0.5 mi E Hopewell, 20 May 1978, B. Stark, 1 6, 1 2 reared (BPS). Frank- lin Co: Clear Springs Lake, 4.5 mi SW Meadville, 30 May 1988, P. K. Lago, 1 6 (UM). Middleton Creek, TSN R4E Sec 31S, 9-15 Jun 1992, T. Schiefer, R. Fontenot, 6 6 (MEM). Same data except 29 Jun 1992, 4 6 (MEM). Same data except 30 Jun-—6 Jul 1992, 1 6 (MEM). McGehee Creek tribu- tary, TON R4E Sec 26SW, 1 Jun 1992, T. Schiefer, R. Fontenot, 88 6, 35 2 (MEM). Porter Creek, TSN R4E Sec 8NW, 1 Jun 1992, T. Schieffer, R. Fontenot, 6 6, 3 2 (MEM). Hinds Co: Sand Creek, Hunt Road, 24 Apr 1992, B. Stark, G. Gee, D. Kelly, 2 6 (BPS). Discussion. —This description of N. coxi brings to 14 the number of Nearctic Neo- perla species. Using keys in Stark & Lentz (1988), males of NV. coxi are identified as N. stewarti Stark & Baumann, 1978, or N. os- age Stark & Lentz, 1988, and females are identified as N. stewarti. N. coxi males differ from all known Nearctic species in having most of the aedeagal tube spinulose (Fig. 2) and they also differ from both N. osage and N. stewarti in lacking prominent aedeagal armature in the basal half of the sac (Figs. 2-3). Separation of unassociated females of N. coxi from N. stewarti is more difficult, but the subgenital plate is slightly longer and the spermathecal stalk lining is less devel- oped in N. coxi; females of other Nearctic Neoperla differ from N. coxi in having most of the spermathecal stalk lined with brown setae. Nymphs are virtually identical in col- or pattern to N. robisoni Poulton & Stewart, 1986 (Ernst et al. 1986, Poulton & Stewart 1991), with perhaps a less extensive ocellar 47 Fig. 6. Neoperla coxi nymphal habitus. pigment spot in N. coxi. The egg collar and posterior pole of N. coxi bear less prominent reticulations than do the eggs of N. stewarti. Six Neoperla species are now known from Mississippi but only N. robisoni and N. carl- soni Stark & Baumann, 1978, have been 48 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON a = Ait wie ee “Ai a art m7 ie mere “asi” f 4 wth ertith rend te pansion uv 41,500 14rm WD? Figs. 7-8. Neoperla coxi eggs. 7, Entire egg, lateral; 8, Detail of surface. St = striae, Su = sulci, M = micropyle. collected from the same streams with N. August of 1992, and blacklight collections coxi. At the Middleton Creek site in Frank- were made 1 June, 29 June and 31 August. lin County, personnel of the Mississippi En- Early collections through 8 June contained tomological Museum operated a malaise exclusively N. coxi; N. robisoni males first trap continuously from 2 June through 31 appeared by 9 June and males of the 2 spe- VOLUME 108, NUMBER 1 cies were collected together through 6 July. Thereafter, no males of N. coxi were taken but a single female was present in a mid- August malaise trap. N. robisoni males were collected through July and August and by late August, the first NV. carlsoni specimens were taken. Although elsewhere in Missis- sippi, NV. carlsoni has been collected as early as 22 June, at the Middleton Creek site a temporal emergence sequence which begins with N. coxi and ends with N. carlsoni may occur. Acknowledgments I thank R. L. Brown and T. Schiefer (Mis- sissippi State University) and P. K. Lago (University of Mississippi) for the loans of specimens. K. St. John (University of Mis- sissippi School of Dentistry) helped in the preparation of photomicrographs. Literature Cited Ernst, M. R., B. C. Poulton, & K. W. Stewart. 1986. Neoperla (Plecoptera: Perlidae) of the Ozark and 49 Ouachita Mountain region, and two new species of Neoperla.— Annals of the Entomological So- ciety of America 79:645-661. Poulton, B. C., & K. W. Stewart. 1991. The stoneflies of the Ozark and Ouachita Mountains (Plecop- tera).— Memoirs of the American Entomologi- cal Society 38:1-116. Stark, B. P., & R. W. Baumann. 1978. New species of nearctic Neoperla (Plecoptera: Perlidae), with notes on the genus.—Great Basin Naturalist 38: 97-114. —., & S.C. Harris. 1986. Records of stoneflies (Plecoptera) in Alabama. — Entomological News 97:177-182. —.,& D.L. Lentz. 1988. New species of Nearctic Neoperla (Plecoptera: Perlidae).— Annals of the Entomological Society of America 81:371-376. Surdick, R. F., & B. P. Stark. 1980. Two new species of Chloroperlidae (Plecoptera) from Mississip- pi.— Proceedings of the Entomological Society of Washington 82:69-73. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(1):50—53. 1995. Macrobrachium catonium, a new troglobitic shrimp from the Cayo District of Belize (Crustacea: Decapoda: Palaemonidae) H. H. Hobbs III and Horton H. Hobbs, Jr. (HHH, Jr) Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, U.S.A. (deceased 22 March 1994); (HHH III) Department of Biology, P.O. Box 720, Wittenberg University, Springfield, Ohio 45501, U.S.A. Abstract. —A new troglobitic shrimp, Macrobrachium catonium, is described from the Vaca Plateau, Cayo District of Belize. It is the third albinistic member of the genus known to occur along the Gulf of Mexico-Caribbean versant of Middle America. It may be distinguished from the Oxacan M. villalobosi Hobbs by the eye which is more pigmented and lacks an apical cleft, and by the shorter, often slightly upturned rostrum which does not reach the distal extremity of the antennal scale. It differs from the Tabascan M. acherontium in possessing a more attenuate (less vaulted) rostrum, and from both in that usually there are more than two ventral rostral teeth, and fewer subapical spines on the appendix masculina. The new shrimp described herein is the third albinistic member of the genus Mac- robrachium known to occur in caves on the Gulf of Mexico-Caribbean versant of Mid- dle America (Botosaneanu 1986). Macro- brachium villalobosi Hobbs (1973) has been reported from a single locality, Cueva del Nacimiento del Rio San Antonio, 10 km SSW of Acatan, Oaxaca, Mexico, and M. acherontium Holthuis (1977) from two caves in Tabasco, Mexico. Specimens of the new troglobite have been collected in two caves located approximately 46 km apart on the Vaca Plateau in the Cayo District of Belize close to Guatemala. Macrobrachium catonium, new species Description. —Rostrum (Fig. la, d) mod- erately high, weakly arched, and slightly de- flected or upturned anteriorly, tip not reach- ing distal extremity of antennal scale; dorsal margin with 6 to 9 teeth, as many as 3 bi- spinous (8 in holotype of which 3 bispi- nous), and 1 or 2 epigastric; ventral margin with 2 to 5 teeth. Carapace (Fig. la) with antennal spine arising slightly posterior to ventral part of orbital margin and hepatic spine almost di- rectly ventral to first epigastric tooth. Bran- chiocardiac groove prominent. Abdomen (Fig. 1a) smooth, pleura of fifth abdominal somite with acute posteroven- tral angle, more anterior pleura rounded posteroventrally. Sixth somite 1.5 times as long as fifth, and telson 1.2 times longer than sixth; dorsal surface of telson (Fig. 1), 0) with anterior pair of spines situated at base of posterior third and posterior pair at about base of posterior sixth, slightly pos- terior to midway between anterior pair and median apex of telson; posterior margin of telson sharply contracted, forming broadly acute tip, and bearing 2 pairs of spines ven- tral to margin, more mesial pair decidedly overreaching apex of telson, lateral pair fall- ing short of apex; (holotype with row of 8 plumose setae between mesial pair of spines VOLUME 108, NUMBER 1 and single simple submarginal seta dorsal- ly). Eyes (Fig. la, d) moderately large, round- ed distally and with apical black to purplish pigment spot; cornea over pigmented area lacking facets. Antennule (Fig. la, d) with proximal po- domere of peduncle longer than combined length of distal 2 podomeres, these subequal in length, and distal podomere falling short of base of lateral spine on antennal scale; distolateral spine on basal podomere reach- ing slightly beyond midlength of penulti- mate podomere; holotype with lateral long flagellum about 5 times length of postorbital length of carapace, mesial flagellum about 3 times as long. Antenna (Fig. la, d, 1) with peduncle as illustrated, basal segment with ventrolateral spine, flagellum about 7.3 times as long as postorbital carapace length. Antennal scale 2.5 times as long as broad, with lateral margin almost straight. Gnathal appendages (Fig. 1b, e, h, 1, k, m) as figured. Third maxilliped reaching slightly beyond midlength of antennal scale. First pereiopod (Fig. 1a) overreaching an- tennal scale by length of dactyl, latter sub- equal in length to mesial margin of palm of chela; carpus about twice length of chela and slightly shorter than merus. Second pereio- pod (Fig. la, n) overreaching antennal scale by only slightly more than length of chela; latter with fingers slightly longer than smooth palm; opposable margin of fixed finger with 1 very small corneous tubercle near base, otherwise fingers lacking tubercles, but both fingers with scattered fine setae and sub- apical clusters of curved stiff ones; carpus 1.2 times as long as either propodus or me- rus, and merus 1.4 times longer than ischi- um. Third pereiopod overreaching antennal scale by propodus and 3 length of carpus; latter almost half as long as merus, and me- rus 2.5 times as long as ischium. Fourth pereiopod overreaching antennal scale by dactyl and slightly less than half length of propodus; propodus approximately 2.7 times length of dactyl; carpus about half as 51 long as merus, and latter almost 2.5 times as long as ischium. Fifth pereiopod over- reaching antennal scale by dactyl and % length of propodus; propodus almost 5 times length of dactyl and 1.7 times length of car- pus, latter little less than 74 length of merus; merus 2.6 times length of ischium. First pleopod (Fig. 1c) with exopodite 2.2 times as long as endopodite. Second pleo- pod (Fig. 1f) with exopodite 1.2 times length of endopodite and latter with appendix masculina (Fig. 1f, g) reaching distinctly be- yond its midlength. Lateral ramus of uro- pod (Fig. lo) with straight lateral margin bearing fixed spine and longer, slenderer, movable one at its mesial base. Branchial formula typical of that of many, if not most, other members of Macrobra- chium: 5 pleurobranchs corresponding to pereiopods, 2 arthrobranchs at base of third maxilliped, podobranch on coxa of second maxilliped, and epipodites on second max- illa and first maxilliped. Size. —Carapace length of male holotype 9.9 mm; that of 12 females, none of which Ovigerous, ranging from 8.8 to 14.4 mm. Color. —Lacking pigment except for eye spot which purplish to black. Type locality.—Lake in Actun Chapat (cave), Cayo District, Belize. This cave is located on the northern end of the Vaca Plateau and although it has not been fully explored it is estimated to be about 2 km in length. Shrimp were found in the lake passage, apparently a perched overflow route with a series of small lakes extending more than 150 m. The shrimp were numerous on the silt-covered substrate of the lakes which were as deep as 6 m. The troglobitic crab, Typhlopseudothelphusa acanthochela Hobbs, 1986, occurred in small numbers and the catfish with reduced eyes, Rhamdia laticauda typhla Greenfield, Greenfield, & Woods, were fairly abundant. Bats were roosting over the lakes and probably were insectivorous Natalus sp. and Mormoops sp. and frugivorous Corollia sp. and Glosso- 34 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON if g (i,ksm) Fig. 1. Macrobrachium catonium, new species (all appendages from left side of holotype except c and h from female paratype from type locality): a, Lateral view; b, Third maxilliped; c, f, First pleopods; d, Dorsal view of anterior region of carapace, including cephalic appendages; e, Mandible; g, Appendices masculina and interna; h, First maxilla; i, Second maxilla; j, Dorsal view of caudal end of telson; k, First maxilliped; 1], Ventral view of basal part of antenna; m, Second maxilliped; n, Chela of second pereiopod; 0, Dorsal view of telson and right uropod. (Scales in mm.) VOLUME 108, NUMBER 1 phaga sp. Temperature of the lake water in December 1991 was 25°C. Specimens examined. —In addition to 1 6 and 9 2 from the type locality (Actun Cha- pat, 5 Dec 1992, R. Foster & W. R. Elliott, coll.), specimens were examined from Tunkul Cave, Chiquibal System, Cayo Dis- trict, Belize, 1 2, 24 Mar 1986, D. Coons, coll. and from ‘“‘Chiquibal System,’’ Cayo District, Belize, 2 9, 8 Mar 1986, D. Coons, coll. (This cave is located approximately 46 km south of Actun Chapat on the south end of the Vaca Plateau near Guatemala.) Disposition of types. —The holotype (USNM 260328) and 10 of the paratypic females are deposited in the National Mu- seum of Natural History, Smithsonian In- stitution, Washington, D.C. A paratypic fe- male is in the collection of the Texas Me- morial Museum, Austin, and another in the Rijksmuseum van Natuurlijke Historie, Leiden, Netherlands. Relationships.—Macrobrachium caton- ium 1s Closely related to the two Mexican troglobitic shrimps assigned to the genus, M. villalobosiand M. acherontium. The three are markedly similar suggesting common ancestry. The apically notched eyes and long (reaching to or beyond the distal end of the antennal scale), slender, straight rostrum of M. villalobosi sets it apart from the other two, and the comparatively slender (nearly Straight rather than with a strongly convex ventral margin) rostrum with usually more than two ventral teeth in M. catonium dis- tinguishes it from M. acherontium. In ad- dition, there are fewer subapical setae on the appendix masculina of M. catonium than in the other two species. Holthuis (1977: 191) reported the presence of 5 pleuro- 53 branchs but no other gills, exopods, or epi- pods in M. acherontium whereas M. caton- lum possesses 2 arthrobranchs at the base of the third maxilliped, and epipodites on the second and first maxillipeds. Etymology.—Catonium (L.), the lower world, noting the subterranean habitat of this shrimp. Acknowledgments Thanks are extended to T. Miller for lo- cating and sending us the misplaced shrimp collected from the Chiquibul Cave system in 1986, and to D. Coons who collected three of the specimens. We are also grateful to R. Foster and W. R. Elliott who collected the shrimp from the type locality, and to the latter who provided us with the infor- mation recorded in the description of the type locality. For their critical review of the manuscript appreciation is extended to C. W. Hart, Jr., B. F. Kensley, and A. B. Wil- liams. Literature Cited Botosaneanu, L. 1986. Stygofauna Mundi. A. Fau- nistic, distributional, and ecological synthesis of the world fauna inhabiting subterranean waters (including the marine interstitial). E. J. Brill, Leiden, Netherlands, 740 pp. Hobbs, H. H., Jr. 1973. Two new troglobitic shrimps (Decapoda: Alpheidae and Palaemonidae) from Oaxaca, Mexico. — Association for Mexican Cave Studies Bulletin 5:73-80. Holthuis, L. B. 1977. Cave shrimps (Crustacea De- capoda, Natantia) from Mexico. Part III. Fur- ther results of the Italian zoological missions to Mexico, sponsored by the National Academy of Lincei (1973 and 1975).—Problemi attuali di Scienza e di Cultura, Accademia Nazionale dei Lincei 171(3):173-195. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(1):54-60. 1995. Procambarus (Ortmannicus) nueces (Decapoda: Cambaridae), a new crayfish from the Nueces River Basin, Texas Horton H. Hobbs, Jr., and H. H. Hobbs III (HHH, Jr.) Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, U.S.A. (deceased 22 March 1994); (HHH III) Department of Biology, P.O. Box 720, Wittenberg University, Springfield, Ohio 45501, U.S.A. Abstract. —Procambarus (Ortmannicus) nueces is described from a small, slug- gish tributary to the Nueces River in Atascoso County, Texas. It may be distinguished from its closest relative, P. (O.) zonangulus Hobbs & Hobbs, by the less attenuated distal part of the male first pleopod which lacks a cephal- omesial shoulder and bears a more robust mesial process. Four specimens of this new crayfish, in- cluding a first form male and three juve- niles, were collected for the Smithsonian In- stitution by Bruce B. Collette and Oliver S. Flint from the type locality on 10 September 1960. While its distinct features were rec- ognized several years ago and futile searches made in several collections for additional specimens, we were reluctant to describe this new species until additional material be- came available. Not until July 1993, when the second author, assisted by Howard and Stephen Kronk, collected in the area where the original lot had been taken, were two additional specimens secured, both pro- cured with the aid of a seine in the same segment of stream in which Collette and Flint collected almost 39 years earlier. Al- though other aquatic habitats in the area were sampled in July and a search was made for burrows, no other crayfish were found. We offer the following description of this new species that is based on six specimens (3 adults and 3 juveniles) from the type lo- cality. Procambarus (Ortmannicus) nueces, new species Fig. 1, Table 1 Diagnosis. —Body pigmented, eyes well developed. Rostrum of adults with well de- fined marginal spines but lacking median carina. Carapace with single moderately strong cervical spines. Areola 14.7 to 17.3 times as long as wide and constituting 32.0 to 36.4% of total length of carapace and 44.0 to 47.9% of postorbital length. Suborbital angle very small to virtually obsolete; post- orbital ridge with small, corneous cephalic tubercle; hepatic area weakly tuberculate; branchiostegal spine well developed. An- tennal scale almost twice as long as broad, widest slightly proximal to midlength. Is- chia of third and fourth pereiopods with simple hooks overreaching basioischial ar- ticulation and lacking opposing tubercles on corresponding basis; coxa of fourth pereio- pod with strong bulbiform boss, that of fifth much smaller and flattened. First pleopods of first form male reaching coxa of third pereiopods, symmetrical but not conspic- uously tapering distally; cephalomesial margin lacking even rudiment of hump or shoulder. Terminal elements consisting of: (1) short, tapering mesial process directed caudodistally and inclined laterally, and (2) cephalic process obscuring much of central projection in cephalic aspect, corneous, sub- acute, tapering from broad base, inclined mesially and gently curved caudally, its apex lying slightly mesial to central projection; caudal element consisting of corneous, ta- VOLUME 108, NUMBER 1 pering acute caudal process lying against caudal surface of central projection and in- conspicuous, non-corneous, setiferous cau- dal knob at lateral base of cephalic process; and corneous central projection, largest of terminal elements, inclined mesially and ta- pering to apex which directed caudodistally and slightly laterally. Annulus ventralis about 1.7 times as wide as long, dextral half elevated little more than sinistral; sinus originating on median line, disappearing under dextral wall and emerging on caudal flank of dextrally directed tongue, slightly posterior to midlength of annulus, where crossing median line and, in gentle arc, coursing caudally to margin of annulus. Sternum cephalic to annulus deeply cleft and bearing few prominent tubercles. Una- dorned postannulur sclerite more than half as broad and as long as annulus. First pleo- pod present in female. Holotypic male, form I.—Cephalothorax (Fig. la, 1) subcylindrical in section, only slightly broader than high. Abdomen nar- rower than thorax (20.1 and 24.1 mm). Greatest width of carapace greater than height at caudodorsal margin of cervical groove. Cephalic section of carapace twice as long as areola, length of latter 33.1% of entire length of carapace (44.9% of postor- bital carapace length). Surface of carapace punctate dorsally, granulate to tuberculate laterally. Rostrum not noticeably deflected ventrally, with basally subparallel margins gently converging to level of distal margin of penultimate podomere of antennular pe- duncle where diverging and forming well developed marginal spines marking base of acumen, latter reaching distal extremity of ultimate podomere of antennular peduncle; dorsal surface concave with many small se- tiferous punctations particularly in basal half. Subrostral ridges evident in dorsal as- pect for very short distance at caudal margin of orbit. Postorbital ridges well developed, grooved dorsolaterally and bearing well de- veloped spiniform tubercles at cephalic ex- tremities. Suborbital angle very weak, vir- tually obsolete. Branchiostegal spines com- 55 paratively well developed. Single cervical spine on both sides of carapace rather small but well defined and acute. Abdomen (Fig. 1j) slightly shorter than carapace. Pleura of third through fifth seg- ments subtruncate to rounded with caudov- entral extremities subangular. Cephalic sec- tion of telson (Fig. le) with 2 spines in each caudolateral corner, mesial one on each side movable; caudal margin of caudal section with shallow median excavation at end of dorsomedian groove extending length of section. Cephalic lobe of epistome (Fig. 10) subcylindrical with elevated (ventrally) ir- regular margins and bearing acute antero- median projection; central area subplane except for anteromedian groove; distinct anteromedian fovea and median longitu- dinal groove present on main body. Ventral surface of proximal podomere of antennular peduncle with strong spine slightly proxi- mal to midlength. Antennal peduncle with small spiniform tubercles on basis and is- chium; flagella reaching midlength of telson. Antennal scale (Fig. lh) 2.2 times as long as broad, widest slightly proximal to mid- length; greatest width of lamella more than twice width of thickened lateral part. Third maxilliped extending cephalically to level of ultimate podomere of antennular peduncle; ischium not produced distolater- ally, its ventral surface densely studded with plumose setae. Right chela (Fig. 1m) subovate in cross section, not strongly depressed. Mesial sur- face of palm with row of 6 tubercles sub- tended dorsolaterally by additional row of 6 and few scattered ones and ventrally by row of 4. Entire palm tuberculate but ven- trolateral surface more weakly so. Dorsal and ventral longitudinal ridges weak on both fingers. Opposable surface of fixed finger with dorsal row of 20 tubercles and lower one of 8; dorsal row, with fifth from base largest, extending from base of finger shght- ly beyond level of locking tubercle (more distal ones too small to be included in il- lustration); ventral row much shorter, ex- tending along middle third of finger and 56 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON consisting of locking tubercle, which much larger than others and, not visible in dorsal aspect, 7 proximal and 3 distal to it; band of minute denticles present ventral to dorsal row of tubercles and suddenly broadening at level of locking tubercle, covering entire opposable surface distally to base of cor- neous tip; except for few small tubercles on proximodorsal and proximoventral sur- faces, finger otherwise with setiferous punc- tations. Opposable margin of dactyl with dorsal row of 24 small to minute, rounded tubercles along proximal % of finger and ventral row of 13 along second fourth of finger from base; band of minute denticles situated between tubercles of dorsal row proximally, broadening at level of proximal end of ventral row and covering opposable margin to base of corneous tip of finger; mesial surface with row of 6 tubercles along proximal third flanked by few others prox- imally but, except for few additional tiny tubercles proximally, finger with setiferous punctations. Both fingers with punctations forming rows flanking inconspicuous lon- gitudinal ridges and those with longer setae projecting toward opposable member. Car- pus of chela longer than broad, bearing prominent oblique furrow dorsally; tuber- culate mesially and on mesial half of dorsal surface; ventral surface with usual 2 distal tubercles, otherwise sparsely punctate. Me- rus tuberculate dorsally, distomesially, and ventrally; 2 premarginal tubercles larger than others on dorsodistal surface; ventral sur- face with mesial row of 16 tubercles and less clearly defined lateral one of 10. Ischium with ventromesial row of 5 tubercles. Hooks on ischia of third and fourth per- eiopods (Fig. 1f) simple, both overreaching basioischial articulation and neither op- posed by tubercle on corresponding basis. Coxa of fourth pereiopod with prominent bulbiform boss; that of fifth with much smaller compressed one. Sternum between third, fourth, and fifth pere1opods compar- atively deep with mat of plumose setae ex- tending mesially from ventrolateral mar- gins. First pleopods (Fig. 1b, c, d, g, n) as de- scribed in “Diagnosis.” Uropod (Fig. le) with both lobes of basal podomere bearing small acute spine, both rami with distola- teral spines, and distomedian spine on me- sial ramus situated distinctly proximal to boadly rounded distal margin. Allotypic female. — Differing from holo- type, except in secondary sexual characters, as follows: 2 marginal spines on sinistral side of rostrum; acumen slightly overreach- ing antennular peduncle; cephalic section of telson with 3 spines in each caudolateral corner, more mesial pairs on each side mov- able, mesialmost spines minute; cephalic lobe of epistome more concave; antennal flagellum reaching tergum of sixth abdom- inal segment; 5 tubercles present in row ven- tral to mesial row on palm of chela (Fig. 1p); opposable margin of fixed finger with single dorsal row of 10 small tubercles along basal half, locking tubercle (only member of ventral row present) situated near mid- length of finger and minute denticles on both fingers confined to single row; opposable margin of dactyl with dorsal row of 12 small tubercles extending from base of finger to level of locking tubercle on fixed finger, ven- tral row represented by only 2 tubercles on proximal third of finger; merus with 14 tu- bercles in ventromesial row and 11 in ven- trolateral row (rows not distinct in part, so numbers approximate); ventromesial sur- face of ischium with row of 4 tubercles. Annulus ventralis and adjacent sternal el- ements as described in “‘Diagnosis.” Color notes. —Dorsum of carapace most- ly pale tan but rostrum darker and with red- dish suffusion; rostral and postorbital ridges bluish gray, latter flanked ventrolaterally by reddish splotch; mandibular adductor areas covered by reticulate dark brown pattern; dorsolateral part of branchiostegites with conspicuous black stripe extending from brownish-to-almost-black cervical groove to caudal margin of carapace; remainder of lateral surface of carapace pale gray with darker gray mottlings; cervical spine white and caudal flange of carapace edged in dark VOLUME 108, NUMBER 1 Fig. 1. S// Procambarus (Ortmannicus) nueces, new species (all from holotype except i, p from allotype): a, Lateral view of carapace; b, Mesial view of first pleopod; c, Caudal view of same; d, Cephalic view of same; e, Dorsal view of sixth abdominal segment, telson, and uropods; f, Basal podomeres of third, fourth, and fifth pereiopods; g, Lateral view of first pleopod; h, Antennal scale; i, Annulus ventralis and flanking sternal elements; j, Lateral view of abdomen; k, Caudal view of mandible; 1, Dorsal view of carapace; m, Dorsal view of distal podomeres of cheliped; n, Caudal view of first pleopods; 0, Epistome; p, Dorsal view of distal podomeres of cheliped. gray. Abdomen with broad dorsomedian black stripe extending from carapace through anterior third of sixth abdominal segment, most of which pinkish tan; stripe interrupted by narrow transverse bands of pinkish tan along caudal margins of terga; black stripe flanked laterally by narrower pinkish tan stripe, and it, in turn, separated from mostly reddish pleura by concave splotches of gray-merging-to-black on bases of pleura, dark color continuing along cau- dodorsal margins of pleura; telson and uro- pods, particularly lateral ramus of latter, reddish brown. Antennae and antennules 58 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Measurements (mm) of Procambarus (O.) neuces, new species. Holotype Allotype Paratype Carapace Entire length 48.4 54.0 40.3 Postorbital length 35.6 39.3 40.3 Width 24.1 26.1 20.1 Height 22.0 24.0 19.0 Areola Width 1.0 1.0 1.0 Length 16.0 Wes 14.7 Rostrum Width 19 9.0 7.0 Length 14.5 17.7 11.0 Right chela Length, palm mesial margin 16.1 10.8 12.6 Palm width 13.6 2 10.4 Length, lateral margin 50.9 34.3 41.5* Dactyl length 30.5 hell Dal Abdomen Width 20.1 24.0 Wer Length 47.0 53.0 41.0 * Approximate, tip broken. greenish cream; antennal scales also pale laterally, lamella suffused with pink. Che- liped with distal part of merus and dorsal and distomesial parts of carpus olive tan, remainder of these and more proximal po- domeres pinkish cream; palm of chela pink- ish cream dorsally with olive to dark green- ish brown tubercles, deeper pink ventrally; fingers with greenish suffusion proximo- dorsally, and dactyl also proximomesially, fading quickly distally to very pale cream. Remaining pereiopods with merus and car- pus greenish dorsally, otherwise, cream to pinkish cream. Size. —See Table 1 for measurements of the only three adult individuals available. Type locality. —Sluggish stream tributary of the Nueces River 4.3 miles (6.9 km) south of Jourdanton at junction of State Routes 97 and 173, Atascosa County, Texas. About 5 m at its widest point, the stream was ex- tremely turbid when visited 24 July 1993 and very little flow was noted. The channel was slightly sinuous and tree limbs and oth- er vegetative debris impeded the flow in various sections. The substrate consisted of sand, gravel, and in the pooled areas that were as deep as 1.5 m, the bottom was cov- ered with silt. Sedges and grasses as well as composite plants dominated the riparian vegetation on the upstream (east) side of the bridge along St. Rte. 97 and the water re- ceived direct sunlight for much of the day- light hours. Disposition of types.—The holotypic male, form I, and allotype (USNM 260326 and 260327) are deposited in the National Mu- seum of Natural History, Smithsonian In- stitution, as are the paratypes consisting of 1 6I, 1 juv 6, and 2 juv 2. Relationships. —Procambarus (O.) nueces, new species, belonging to a closely allied group of species often treated as the blan- dingi “group” of the subgenus (see Ortmann 1905:100, Hobbs 1962:286), has its closest affinities with P. (O.) zonangulus Hobbs & Hobbs, 1990. Within this assemblage are the following currently recognized species: P. (O.) blandingii (Harlan 1830), P. (O.) acutus (Girard 1852), P. (O.) acutissimus (Girard 1852), P. (O.) lecontei (Hagen 1870), P. (O.) hayi (Faxon 1884), P. (O.) viaeviridis (Faxon 1914), P. (O.) cuevachicae Hobbs, 1941, P. (O.) caballeroi Villalobos, 1944, P. (O.) verrucosus Hobbs, 1952, P. (O.) lopho- tus Hobbs & Walton, 1960, P. (O.) texanus Hobbs, 1971, P. (O.) geminus Hobbs, 1975, P. (O.) zonangulus, and P. (O.) nechesae Hobbs, 1990. Four of these crayfishes typ- ically have symmetrical first pleopods: P. (O.) viaeviridis, P. (O.) texanus, P. (O.) zon- angulus, and P. (O.) nueces; at least one population of P. (O.) cuevachicae has nearly symmetrical ones. The new species differs from viaeviridis, which is the most marginal member of the group, in many conspicuous respects: among them, P. (O.) nueces has VOLUME 108, NUMBER 1 well developed marginal spines on the ros- trum; the setiferous caudal knob of the first pleopod is laterally rather than cephalically situated, is not conspicuously produced, and the setae partly obscure other terminal el- ements; and the central area of the annulus ventralis is not deeply excavate. Distin- guishing it from P. (O.) cuevachicae are a more strongly tapering, subacute, less com- pressed cephalic process and a less twisted central projection of the first pleopod in first form males, and a straight, not reflexed, hook on the ischium of the fourth pereiopod that is not opposed by a tubercle on the corre- sponding basis. From fexanus it differs in that the cephalic process and central pro- jection are bent caudally at distinctly less than a right angle to the shaft of the ap- pendage, and there is not even a hint of a shoulder on the appendage proximal to the terminal elements; the cephalic lobe of the epistome is subcircular instead of subtrian- gular in outline; in the female there are fewer tubercles adorning the sternum anterior to the annulus, and the postannular sclerite is about *4 as long as broad and 1s as long as the annulus ventralis. The new crayfish dif- fers from P. (O.) zonangulus in that the first pleopod is not distinctly tapering distally and it lacks any trace of a shoulder on the cephalic or cephalomesial surface proximal to the terminal elements; the mesial process is less attenuate and comparatively stockier; in the female there are markedly fewer tu- bercles on the sternum anterior to the an- nulus, and the length of the postannular sclerite is as great as that of the annulus along its median longitudinal axis. Acknowledgments We are grateful to B. B. Collette and O. S. Flint for securing the first specimens of this new crayfish of which we are aware and to H. and S. Kronk for their assistance in obtaining additional specimens. For their critical reading of the manuscript we extend 59 thanks to J. F. Fitzpatrick, Jr., of the Uni- versity of South Alabama, and C. W. Hart, Jr., and R. B. Manning, of the National Mu- seum of Natural History. Literature Cited Faxon, W. 1884. Descriptions of new species of Cam- barus; to which is added a synonymical list of known species of Cambarus and Astacus. — Pro- ceedings of the American Academy of Arts and Sciences 20:107-158. 1914. Notes on the crayfishes in the United States National Museum and the Museum of Comparative Zoology with descriptions of new species and subspecies to which is appended a catalogue of the known species and subspe- cies. — Memoirs of the Museum of Comparative Zoology at Harvard College 40:352—427. Girard, C. 1852. A revision of the North American Astaci with observations on their habits and geographical distribution.— Proceedings of the Academy of Natural Sciences of Philadelphia 6:87-91. Hagen, H. A. 1870. Monograph of the North Amer- ican Astacidae.—Illustrated Catalogue of the Museum of Comparative Zodlogy at Harvard College 3:viiit+ 109. Harlan, R. 1830. Description of a new species of the genus Astacus.— Transactions of the American Philosophical Society 3(15):464—-465. Hobbs, H. H., Jr. 1941. A newcrayfish from San Luis Potosi, Mexico.— Zoologica 26(1):1-4. 1952. A new crayfish from Alabama with notes on Procambarus lecontei (Hagen).—Pro- ceedings of the United States National Museum 102(3297):209-219. . 1962. Notes on the affinities of the members of the Blandingii Section of the crayfish genus Procambarus (Decapoda, Astacidae).— Tulane Studies in Zoology 9(5):273-293. 1971. New crayfishes of the genus Procam- barus from Alabama and Texas (Decapoda: As- tacidae). — Proceedings of the Biological Society of Washington 84(11):8 1-94. 1975. New crayfishes (Decapoda: Cambari- dae) from the southern United States and Mex- ico.—Smithsonian Contributions to Zoology 201:1-34. 1990. On the crayfishes (Decapoda: Cam- baridae) of the Neches River Basin of Eastern Texas with the descriptions of three new spe- cies.— Proceedings of the Biological Society of Washington 103:573-597. —., & H. H. Hobbs III. 1990. A new crayfish 60 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON (Decapoda: Cambaridae) from southeastern persal over the United States.— Proceedings of Texas. — Proceedings of the Biological Society of the American Philosophical Society 44(180):91- Washington 103:608-613. 136. —, & M. Walton. 1960. A new crayfish of the Villalobos, A. 1944. Estudios de los cambarinos mex- genus Procambarus from southern Alabama icanos, III: Una especie nueva de Procambarus, (Decapoda, Astacidae).— Proceedings of the Bi- Procambarus caballeroi n. sp.—Anales del In- ological Society of Washington 73(20):123-129. stituto de Biologia, Universidad Aut6noma de Ortmann, A. E. 1905. The mutual affinities of the México 15(1):175-184. species of the genus Cambarus, and their dis- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(1):61-67. 1995. On a new Somanniathelphusa Bott, 1968, from Vietnam (Crustacea: Decapoda: Brachyura: Parathelphusidae) Peter K. L. Ng and Takeharu Kosuge (PKLN) Department of Zoology, National University of Singapore, Kent Ridge, Singapore 0511, Republic of Singapore; (TK) % Department of Biology, Faculty of Science, University of the Ryukyus, Nishihara-cho, Okinawa 903-01, Japan Abstract. —A new species of parathelphusid freshwater crab, Somanniathel- phusa pax, 1s described from Hanoi, Vietnam. The new species seems to be most closely allied to species from southern China, but differs in the form of the carapace, male abdomen and male gonopods. Specimens of a freshwater crab recently obtained from a market in Hanoi, Vietnam, proved to belong to an undescribed species of Somanniathelphusa Bott, 1968 (family Parathelphusidae Alcock, 1910). The genus Somanniathelphusa s. str. aS now recog- nized (Bott 1970, Ng 1988, Ng & Naiyanetr 1993, Naiyanetr 1994), contains six de- scribed species: S. brevipodum Dai, Song, He, Cao, Xu & Zhong, 1975 [China], S. chongi (Wu, 1935) [China], S. falx Ng & Dudgeon, 1992 [China], S. sinensis (H. Milne Edwards, 1853) (type species) [Chi- na], S. taiwanensis Bott, 1968 [Taiwan], and S. zanklon Ng & Dudgeon, 1992 [Hong Kong]. Naiyanetr (1994) recently reviewed the genus Somanniathelphusa and showed that it should be split into four genera: So- manniathelphusa s. str., Sayamia Naiya- netr, 1994, Esanthelphusa Naiyanetr, 1994, and Chulathelphusa Naiyanetr, 1994. The freshwater crab fauna of Vietnam is poorly known. Many of the Indo-Chinese species described by H. Milne Edwards (1853), A. Milne Edwards (1869), De Man (1904) and Rathbun (1904, 1905), did not have precise collection localities. With re- gards to the parathelphusids belonging or allied to Somanniathelphusa, Potamon (Parathelphusa) prolatus was described from “Mois Chero”? in northern Cochinchina (Rathbun 1902:186), a location which is now supposed to be part of northern Vietnam (Turkay & Naiyanetr 1987:392). Potamon (Parathelphusa) prolatus, incorrectly syn- onymized with Sayamia dugasti (Rathbun 1902) by Bott (1970:112), is a valid species and should be transferred to Esanthelphusa (P. Naiyanetr, pers. comm.). Chulathelphu- sa neisi was described from somewhere in Cochinchina (Rathbun 1902:186), which may be somewhere in or near Vietnam. Balss (1914:408) reported Somanniathelphusa si- nensis from Tonkin in Vietnam, but Ng & Dudgeon (1992:757) noted that this record is doubtful and his specimens must be re- examined. The present paper describes the new spe- cies, Somanniathelphusa pax. The abbre- viations G1 and G2 are for the male first and second pleopods respectively. Mea- surements are given in millimeters, in the sequence carapace width by carapace length. Specimens examined are deposited in the Zoological Reference Collection (ZRC), De- partment of Zoology, National University of Singapore. Systematic Account Family Parathelphusidae Alcock, 1910 Genus Somanniathelphusa Bott, 1968 62 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Somanniathelphusa pax, new species Figs. 1-3 Material examined. —Holotype male (carapace 41.9 by 32.8 mm) (ZRC), Hanoi, Vietnam, from Dong Xuan market, leg. T. Kosuge, 19 Oct 1993. Paratype female (car- apace 34.8 by 26.6 mm) (ZRC), same data as holotype. 2 males (larger, carapace 30.9 by 24.8 mm), 2 females (larger, carapace 31.2 by 25.0 mm) (ZRC), market in Hanoi, Vietnam, leg. A. U. Kara, September 1994. Description of holotype male. —Carapace broader than long, surfaces smooth, trans- versely convex (Figs. 1, 2A). Epigastric cris- tae sharp, distinct, separated by deep lon- gitudinal groove; postorbital cristae sharp, distinct, reaching to beginning of shallow cervical grooves, inner edge of crista reach- ing to slightly below inner edge of epigastric crista; beyond cervical grooves, cristae are rounded, curving gently to meet bases of first epibranchial teeth; median H-shaped depression distinct, relatively deep (Figs. 1, Somanniathelphusa pax, new species. Holotype male, carapace 41.9 by 32.8 mm (ZRC). Dorsal view. 2A, B). Frontal margin gently sinuous, con- fluent with sinuous, smooth supraorbital margin. External orbital angle triangular, inner margin distinctly shorter than outer, outer margin gently convex to straight. An- terolateral margin with 3 strongly devel- oped, sharp epibranchial teeth, first trian- gular, last spiniform, first and second teeth directed forwards, last tooth directed obliquely outwards (Fig. 2A, B). Ischium of third maxilliped rectangular, much longer than wide, with deep subme- dian, longitudinal sulcus; basis separated from ischium by distinct suture; merus squarish, medially depressed; exopod long, reaching to middle of margin of merus, with pronounced subdistal tooth on inner margin and long flagellum (Fig. 2C). Chelipeds strongly asymmetrical, left larger; outer surfaces of both chelae smooth (Figs. 1, 2F). Merus with small but distinct subterminal spine on dorsal margin. Carpus with strong, sharp spine on inner distal mar- gin; inner margin granulated (Fig. 1). Fin- VOLUME 108, NUMBER 1 63 Fig. 2. Somanniathelphusa pax, new species. Holotype male, carapace 41.9 by 32.8 mm (ZRC). A, carapace; B, frontal view; C, left third maxilliped (hairs omitted); D, anterior sternites; E, last left ambulatory leg; F, enlarged (left) chela; G, abdomen. Scales = 5.0 mm. 64 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON gers of both chelae longer than palm, fingers of smaller chela not distinctly gaping when closed, with numerous denticles along cut- ting edges; fingers of larger chela strongly gaping when closed, cutting edges lined with several large and numerous smaller teeth (Riga): Second pair of ambulatory legs longest. Merus with very strong, distinct dorsal sub- terminal spine (Fig. 2E). Male abdominal cavity reaches an imag- inary line joining anterior edges of bases of chelipeds (Fig. 2D). Suture between ster- nites 2 and 3 only present medially, lateral parts obscured; suture between sternites 3 and 4 very shallow, indistinct (Fig. 2D). Male abdomen distinctly T-shaped, telson broad- ly triangular, lateral margins gently con- cave, gradually converging towards round- ed tip; segment 6 subequal in length to tel- son, strongly constricted at subproximal part, with low but distinct submedian tran- verse ridge, length 1.2 times greatest width, distal width 1.3 times proximal width; seg- ments 3-5 progressively more trapezoidal, segments 1 and 2 narrow (Fig. 2G). G1 terminal and subterminal segments not demarcated (Fig. 3A—C); distal part slender, distal half gradually curving later- ally outwards, tip bent laterally; basal part very broad, outer margin distinctly convex when viewed ventrally (Fig. 3D-F). G2 with very short distal segment, 0.07 times length of elongated basal segment (Fig. 3G). Paratype female.—The paratype female is fully mature. One ofits chelae is distinctly larger than the other, but the degree of en- largement of the major chela is not as pro- nounced as in the holotype male. The bran- chial regions of the female are distinctly in- flated and appear swollen, much more than in the holotype male. In some of the large species of Somanniathelphusa (e.g., S. bangkokensis and S. sexpunctata), larger specimens generally appear to be less in- flated than smaller ones. The holotype male of S. pax is distinctly larger than the female (carapace 41.9 by 32.8 mm against 34.8 by 26.6 mm). The paratype female agrees with the holotype male in all other non-sexual characters. The non-type specimens show some variation in the form of the frontal margin, from sinuous to almost straight. Color.—Young males and females are beige to dull brown on all dorsal surfaces, the ventral surfaces being dirty white. Large males vary from brown to purple on the dorsal surfaces. Discussion. —In characters such as cristae of the carapace, shape of the male abdomen and gonopods, S. pax clearly belongs to So- manniathelphusa (sensu Naiyanetr 1994) which occurs in China and Taiwan (Dai et al. 1975, Bott 1970, Ng & Dudgeon 1992). The carapace of S. pax is more rectangular compared to species like S. sinensis and S. zanklon which are more oval. The male ab- dominal segment 6 of S. pax is similar in shape to that of S. sinensis, S. zanklon and S. falx, but its telson is less triangular, with the distal part broader. Compared to S. chongi, the male abdominal segment 6 of S. pax is also proportionately less elongate (length to maximum width ratio 1.2 against 1.4). The deep and distinct longitudinal groove on the pollex of the enlarged male chela is diagnostic for S. pax. The Gls of S. pax however, differs markedly from all known congeners in having the tip bent lat- erally. In congeners, the tip is either straight, slightly folded laterally or hooked down- wards (fide Wu 1935, Dai et al. 1975, Ng & Dudgeon 1992). As for allied genera and species known from Vietnam, S. pax differs from Esan- thelphusa prolatus in that the anterolateral margin of E. prolatus is more convex, with the smaller epibranchial teeth directed for- wards (margin less convex in S. pax, with the epibranchial teeth relatively larger and the last tooth directed obliquely outwards), the outer surface of the pollex of the en- larged male chela is smooth, without a lon- gitudinal groove (longitudinal groove dis- tinct in S. pax) and the male abdominal segment 6 is relatively shorter (length to VOLUME 108, NUMBER 1 65 D if F F: Fig. 3. Somanniathelphusa pax, new species. Holotype male, carapace 41.9 by 32.8 mm (ZRC). A-C, left G1; D-F, distal part of left G1; G, left G2. A, D, ventral views; C, F, dorsal views; B, E, ventro-marginal views. Scales = 1.0 mm. maximum width ratio 0.9 against 1.2) with the distal width much wider than the prox- imal width (distal to proximal width ratio 1.6 in E. prolatus, 1.3 in S. pax) (fide Rath- bun 1905:245, fig. 59, pl. XI fig. 4). The G1 of E. prolatus s. str. is not known. The dorsal surface of the carapace of Chu- lathelphusa neisi is not only distinctly less convex than all other Somanniathelphusa pax, the distal male abdominal segments are not slender and there is no distinct constric- tion on segment 6 (length to maximum width 66 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ratio 1.0; distal to proximal width ratio 1.1) (cf. Rathbun 1905:249, fig. 61, pl. XI fig. 5). The first author has examined the G1 of C. neisi, and it is straighter, with the distal parts not curved or hooked, and the terminal seg- ment distinctly demarcated compared to S. Dax The type specimens of S. pax were ob- tained by the second author from a market in Hanoi which had almost certainly come from rice-fields near the city. The first au- thor subsequently received additional spec- imens (non-types) from Hanoi and the spe- cies 1s apparently frequently sold in mar- kets. The very low prices and abundance of the crabs strongly suggest that they came from the neighborhood of Hanoi. The col- lection of rice-field crabs for food is a com- mon practice in Indo-China. The first au- thor has observed this often in Thailand, and in every instance, the crabs had been collected from slow-flowing streams or rice- fields on the outskirts of the town and sold as a cheap source of protein. In Thailand, crabs are kept dried (the individuals often caked in mud) and can be kept alive for long periods provided they are kept cool (see Ng & Naiyanetr 1993:43). They are eaten in a variety of ways, but are usually fried with spices until they are crispy. The study of the freshwater crab fauna of Vietnam is still very much in the explora- tion and survey stage. The number of taxa known is likely to increase substantially in the coming years as the country becomes more open to the scientific community. Etymology.—The species name is de- rived from the Latin for peace. Used as a noun in apposition. Acknowledgments The first author is grateful to Anna Ursula Kara for passing the Vietnamese crabs to him for study and Phaibul Naiyanetr for information on FE. prolatus. The study has been partially supported by a research grant (RP 900360) to the first author from the National University of Singapore. Literature Cited Alcock, A. 1910. Brachyura I. Fasc. II. The Indian Freshwater Crabs— Potamonidae. Catalogue of the Indian Decapod Crustacea in the collection of the Indian Museum, Calcutta, pp. 1-135, Pls. 1-14. Balss, H. 1914. Potamonidenstudien.—Zoologisch Jahrbucher (Systematics) 37:401-—10, pl. 15. Bott, R. 1968. Parathelphusiden aus Hinterindien (Crustacea, Decapoda, Parathelphusidae). — Senckenbergiana Biologica, Frankfurt 49(5):403— 422. 1970. Die Siisswasserkrabben von Europa, Asien, Australien und ihre Stammesgeschichte. Eine Revision der Potamoidea und Parathel- phusoidea (Crustacea, Decapoda).—Abhand- lungen der Senckenbergischen Naturforschen- den Gesellschaft, Frankfurt 526:1-338, pls. 1- 58. Dai, A.-Y., Y.-Z. Song, L.-Y. He, W.-J. Cao, Z.-B. Xu, & W.-L. Zhong. 1975. Description of several new species of freshwater crabs belonging to the intermediate hosts of lung flukes.—Acta Zoo- logica Sinica, Beijing 21(3):257—264, pls. 1-3. Man, J. G., De. 1904. Décapodes d’eau douce rap- portes par M. A. Pavie, avec 2 pls. Mission Pa- vie, Indo-Chine 1879-1895. Etudes diverses. III.— Recherches sur |’Histoire Naturelle, Paris 1904:316-331, 2 pls. Milne-Edwards, A. 1869. Revision du genre Thelphuse et description de quelques especés nouvelles faisant partie de la collection du Muséum.—Nouvelles Archives du Muséum d’Histoire Naturelle Paris 5:161-191, pls. 8-11. Milne-Edwards, H. 1853. Mémoire sur la famille des Ocypodiens.—Annales du Sciences naturelle, Zoologie (3)20:163-228, pls. 6-11. Naiyanetr, P. 1994. On three new genera of Thai ricefield crabs allied to Somanniathelphusa Bott, 1968 (Crustacea: Decapoda: Brachyura: Para- thelphusidae).— Raffles Bulletin of Zoology, Singapore 42(3):695-700. Ng, P. K. L. 1988. The Freshwater Crabs of Penin- sular Malaysia and Singapore. Department of Zoology, National University of Singapore, Shinglee Press, Singapore, pp. i—viii, 1-156, 4 color pls. , & D. Dudgeon. 1992. The Potamidae and Parathelphusidae (Crustacea: Decapoda: Brachyura) of Hong Kong.—Invertebrate Tax- onomy, Melbourne 6:741-768. . Ng, P. K. L., & P. Naiyanetr. 1993. New and recently described freshwater crabs (Crustacea: Decap- oda: Brachyura: Potamidae, Gecarcinucidae and Parathelphusidae) from Thailand.—Zoolo- gische Verhandelingen, Leiden 284:1-117, figs. 1-68. VOLUME 108, NUMBER 1 Rathbun, M. J. 1902. Description des nouvelles es- péces de Parathelphusa appartenant au Muséum de Paris. — Bulletin du Muséum national d’His- toire naturelle, Paris 1902(3):184—187. 1904. Les crabes d’eau douce.— Nouvelles Archives du Muséum d’Histoire naturelle, Paris (4)6:225-312, pls. 9-18. 1905. Les crabes d’eau douce.— Nouvelles Archives du Muséum d’Histoire naturelle, Paris (4)7:159-323, pls. 13-22. 67 Turkay, M., & P. Naiyanetr. 1987. The identity of Potamon rangoonense Rathbun 1904 and Thel- phusa larnaudii A. Milne-Edwards 1869, with introduction of Neolarnaudia botti n. g. n. sp. (Crustacea: Decapoda: Potamidae).—Sencken- bergiana biologica, Frankfurt 67(4/6):389-396. Wu, H. W. 1935. Enumeration of the river-crabs (Potamonidae) of China with descriptions of three new species. —Sinensia 4(11):338-352. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(1):68-75. 1995. A new species of Goreopagurus McLaughlin (Decapoda: Anomura: Paguridae) from the Pacific, and a comparison with its Atlantic counterpart Patsy A. McLaughlin and Janet Haig (PMcL) Shannon Point Marine Center, Western Washington, 1900 Shannon Point Road, Anacortes, Washington 98221-4042, U.S.A.; (JH) Allan Hancock Foundation, University of Southern California, University Park, Los Angeles, California 90089-0371, U.S.A. Abstract.— Goreopagurus garthi, a new species of hermit crab, is described and illustrated. This Pacific representative of the heretofore monotypic, Atlantic genus Goreopagurus has necessitated minor emendations to the generic diag- nosis. The similarities and differences between the two species are discussed. Goreopagurus was erected by Mc- Laughlin (1988) for Pagurus piercei Wass, 1963, a very distinctive western Atlantic species, known at the time of its original description from a single male specimen. When additional collections were critically examined, it became apparent that Wass’ (1963) assignment of this species to Pagurus was incorrect. A suite of characters includ- ing the presence of a pair of pleopods, mod- ified as gonopods, in females and a very short sexual tube on the right fifth coxa in males, phyllobranchiate gills, and highly distinctive chelipeds set Goreopagurus apart from all other members of the Paguridae. Before publication of McLaughlin’s (1988) paper, one of us (JH) had found several specimens in the collections of the Allan Hancock Foundation that bore a marked resemblance to P. piercei. Originally re- ported as Pagurus sp. (2) (Haig et al. 1970), they were subsequently cataloged into the collection simply as “‘genus?, sp. H.”? We have now had the opportunity to reexamine these specimens and confirm that they be- long to Goreopagurus. McLaughlin’s (1988) diagnosis, based exclusively on G. piercei, was rather restrictive. The discovery of Go- reopagurus garthi, new species, necessitates minor emendations to the generic diagnosis. One male and one female paratype have been deposited in the collections of the Na- tional Museum of Natural History, Smith- sonian Institution (USNM). The holotype and remaining paratypes currently in the Allan Hancock Foundation (AHF), will eventually be housed in the Crustacea col- lection of the Natural History Museum of Los Angeles County (LACM). Comparative material of G. piercei from Oculina samples taken from Jeffs Reef, Florida, have been provided by the Harbor Branch Oceano- graphic Museum (HBOM). The length of the shield (SL), as measured from the tip of the rostrum to the midpoint of the posterior margin of the shield, is an indicator of spec- imen size. The symbols 6, 2, and 2° refer to male, female, and ovigerous female respec- tively. Goreopagurus McLaughlin, 1988 Emended diagnosis. —Eleven pairs of phyllobranch gills. Ocular acicles triangular, with strong submarginal spine; separated basally by basal width or more of 1 acicle. Antennal peduncle with supernumerary segmentation. Maxillule with internal lobe moderately well developed, with 1 stiff bris- tle; external lobe produced, not recurved. VOLUME 108, NUMBER 1 Third maxilliped with well developed crista dentata and prominent accessory tooth; me- rus with or without dorsodistal spine. Ster- nite of third maxillipeds with or without small spine on either side of midline. Right cheliped with chela very elongate, narrow; articulation with carpus generally perpendicular. Carpus strongly produced ventrally, dorsomesial margin weakly to greatly expanded. Left cheliped appreciably shorter than right; chela narrow, weakly tri- angular in cross-section, articulation with carpus perpendicular. Sternite of third per- eopods with anterior lobe semisubovate, subquadrate or subrectangular. Sternite of fifth pereopods with 2 ovate or subcircular lobes. Fourth pereopods with propodal rasp consisting of 1 row of corneous scales; dac- tyl with or without preungual process. Mature males with coxae of fifth pereo- pods generally symmetrical; vas deferens of right produced as short sexual tube, often directed laterally or posteriorly and some- times partially obscured by posterior tuft of setae; coxa of left often with vas deferens slightly produced; with 3 unpaired unira- mous or weakly biramous pleopods. Fe- males with paired gonopores; well devel- oped, paired first pleopods modified as gon- opods; 4 unpaired pleopods, second to fourth with both rami well developed, fifth with endopod rudimentary or absent. Abdomen flexed. Uropods asymmetrical. Telson with transverse suture; posterior lobes symmetrical or somewhat asymmet- rical, separated by small median cleft; ter- minal margins oblique or straight, each armed with few to several small spines; lat- eral margins frequently delimited by narrow chitinous plate. Goreopagurus garthi, new species Figs. 1, 2, 3A, 4A Pagurus, undescribed species (2): Haig et al. 1970:20. Holotype.—1 6 (SL 2.27 mm), 29 mi S. of Punta Abreojos, west coast of Baja Cal- 69 ifornia, Mexico, “Velero IV” sta 1710-49, 95-102 m, 7 Mar 1949, LACM 49-55.9 (AHF 4926). Paratypes.— West coast of Baja Califor- nia, Mexico: 1 6 (SL 1.70 mm), 8 mi W of Isla Cedros, Velero IIT sta 1253-41, 117- 119 m, 26 Feb 1941, LACM 41-33.20 (AHF 4135).—2 92 (SL 2.21, 2.30 mm), 29 mi S. of Punta Abreojos, Velero IV sta 1710-49 95-102 m, 7 Mar 1949, LACM 49-55.8 (AHF 4927.—2 6 (SL 1.31, 1.55 mm), 1 2 (SL 1.25), 1 92 (SL 1.37 mm), 8 mi W of Punta Redonda, Horizon sta A-11 Magbay Expedition, 106-116 m, 29 Jan 1964, LACM 64-237.1 (AHF 6416), USNM 267574.—16(SL 1.76 mm), San Jaime Bank off Cabo San Lucas, Velero III sta 618-37, 137 m, 3 Mar 1937, LACM 37-19.16 (AHF Byala): Description. —Shield (Fig. 1A) broader than long; anterior margins between ros- trum and lateral projections somewhat con- cave; anterolateral margins sloping; poste- rior margin truncate; surface with scattered stiff setae. Rostrum acutely or obtusely tri- angular, terminating sharply or bluntly, and usually with slight, rounded keel. Lateral projections obtusely triangular or broadly rounded, with small submarginal spine or spinule. Ocular peduncles usually as long as shield, stout or slightly constricted medially, and with 2 or 3 tufts of stiff setae; corneae usu- ally slightly dilated. Ocular acicles acutely triangular, with strong submarginal spine; separated basally by at least basal length of 1 acicle. Antennular peduncles overreaching cor- neae by '4 to %4 length of ultimate segment. Ultimate segment with | or 2 long, stiff setae at dorsolateral distal margin and few scat- tered setae. Penultimate segment with few scattered setae. Basal segment with slender spine on lateral margin medially or distally. Antennal peduncles equaling or over- reaching ocular peduncles by nearly half length of ultimate segment. Fifth and fourth segments with few scattered setae. Third 70 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON QO iS ay LTS aS Diy Dy LO A Fig. 1. Goreopagurus garthi, new species. A-D, G, Holotype, AHF 4926: A, shield and cephalic appendages; B, dactyl of second right pereopod (mesial view); C, dactyl of third right pereopod (mesial view); D, anterior lobe of sternite of third pereopods; E, sternite of fifth pereopods of female paratype (SL 2.3 mm), AHF 4927; F, coxae and sternite of fifth pereopod of male paratype, AHF 4135 (SL 1.7 mm); G, telson. Scale equal 1.0 mm (A-C), and 0.5 mm (D-F). segment with strong, acute spine at ven- trodistal margin. Second segment with dor- solateral distal angle produced, terminating in acute spine and usually with strong spine on lateral margin; dorsomesial distal angle with small spine. First segment unarmed. Antennal acicle long, reaching middle of ul- timate peduncular segment, arcuate, ter- minating in small spine; mesial surface with row of tufts of stiff setae. Antennal flagella long; articles each usually with 1 or 2 short ( females, cb 19.3, 18:3) 17.1. 1625127, mm, cl 12.2, 11.3, 10.7, 10.3, 8.1 mm; (ICN- MHN-CR 1351). Diagnosis. —Male first gonopod with short, inwardly turned, bilobed mesial pro- cess and lateral lobe with 2 processes, one semicircular, distal, and the other rudimen- tary, proximal. Description. —Carapace narrow (cb/cl = 1.67). Cervical grooves straight and deep, reaching lateral margins. Anterolateral mar- gin with depression behind orbits, followed by 7 papillae and second shallow depression at level of cervical groove; rest of margins with approximately 11 papillae, regularly spaced. Postfrontal lobes are small, ovally VOLUME 108, NUMBER 1 99 SCPinze4 Fig. 1. Strengeriana florenciae, new species, holotype, first left gonopod: a, total view, cephalic; b, same, mesial view; c, same, caudal view; d, same, detail of apex, cephalic view; e, same, lateral view; f, third maxilliped, left; g, aperture of efferent channel, left. la, lateral lobe; ma, marginal lobe; me, mesial lobe; mp, mesial process; Pi, P2, lateral processes; s,, s,, distal spines of the mesial process; s,, lateral spine of the mesial process. 100 shaped and low. Median groove absent. Surface of carapace between front and post- frontal lobes inclined anteriorly and to- wards mid-line. Upper border of front bi- lobed in dorsal view, with conspicuous tu- bercles. Lower margin strongly sinuous in frontal view, with tubercles. Surface of front between upper and lower borders high. Lower orbital margins each with row of tu- bercles. Surface of carapace smooth, cov- ered by small papillae; the limit between the regions 1s indistinct. Palm of larger chela (left) strongly inflat- ed; fingers not gaping. Walking legs slender, but not unusually elongated, the largest be- ing those of second and third pairs, which are of similar length (total length 1.31 times the breadth of carapace); merus in third pair 3.5 times longer than wide. Dactylus with 5 rows of large spines, diminishing in size proximally; arrangement of spines on dac- tylus of third left pereiopod as follows: an- terolateral and anteroventral rows with 5 spines plus 2 proximal papillae, external row with 5 spines plus 2 papillae and 1 pair of proximal papillae, posteroventral and pos- terolateral rows with 4 spines. Exognath of third maxilliped overreaches lateral margin of ischium of endognath; merus of third maxilliped shows acute angle on distal half of external margin (Fig. 1f). Orifice of ef- ferent branchial channel closed by spine at jugal angle and by extension of lateral lobe of epistome (Fig. 1g). Male first gonopods short, slightly arched in caudo-cephalic plane (Fig. 1b). Apex formed by 3 distinct lobes: mesial (Fig. le, me), marginal (caudal) (Fig. le, ma) and lat- eral (Fig. le, la); mesial and marginal lobes rounded, cup-shaped, with borders strongly demarcated; border of mesial lobe strongly bent at cephalic and caudal ends. Mesial lobe with short, inwardly turned, bilobed mesial process (Fig. 1d, mp), ending in 2 conical spines (Fig. 1d, S,, S,), and another spine located near basis of the lateral surface of process (Fig. lc, S;); mesial lobe forms PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON with lateral lobe a long slit where the genital pore is located (Fig. lc, d). Lateral lobe forming an inflated protuberance covered with spinules (Fig. la, c), and 2 lateral pro- cesses, one semicircular, distal, and the oth- er rudimentary, proximal (Fig. 1d, p,, p>). Internal cavity of mesial lobe densely cov- ered by spines; with strong, dark spines near genital pore, and small spinules over distal border of lateral lobe (Fig. 1d). In addition to strong caudal setae, gonopod bears tiny setae on mesial and lateral sides (Fig. la, b, @)), Etymology. —The name of the species re- fers to the Corregimiento of Florencia, where the type was collected. Remarks. — This species resembles Stren- geriana bolivarensis Rodriguez & Campos, 1989, in the shape of the first male gonopod, but differs in the shape of the mesial lobe. The mesial process is longer and entire in S. bolivarensis, while shorter and bilobed in S. florenciae. Strengeriana bolivarensis has a proximal spine at the lateral lobe, whereas S. florenciae lacks this spine. Acknowledgments I thank J. V. Rueda for collecting the specimens. I am also very grateful to Dr. Rafael Lemaitre and the referees for their constructive comments. The illustrations were prepared by J. C. Pinzon. Literature Cited Campos, M. R., & G. Rodriguez. 1993. Three new species of Strengeriana from Colombia (Crus- tacea, Decapoda, Pseudothelphusidae).—Pro- ceedings of the Biological Society of Washington 106:508-5 13. Pretzmann, G. 1971. Fortschritte in der Klassifizi- erung der Pseudothelphusidae.—Anzeiger der Osterreichischen Akademie der Wissenschaften Mathematische Naturwissenschaftliche Klasse 179(1/4):14—24. Rathbun, M. 1893. Descriptions of new species of American freshwater crabs. — Proceedings of the United States National Museum 16(959):649- 661, pl. 73-77. VOLUME 108, NUMBER 1 Rodriguez, G. 1982. Les Crabes d’eau douce d’Amé- rique. Famille des Pseudothelphusidae. — Faune Tropicale, ORSTOM 22:1-223. ——, & M.R. Campos. 1989. Cladistic Relation- ships of fresh-water crabs of the tribe Strenger- ianini (Decapoda: Pseudothelphusidae) from the 101 northern Andes, with comments on their bio- geography and descriptions of new species. — Journal of Crustacean Biology 9(1):141-156. Smalley, A. 1964. A terminology for the gonopods of the American river crabs.—Systematic Zo- ology 13:28-31. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(1):102-106. 1995. A new callianassid (Decapoda: Thalassinidea) from the southern Caribbean Sea Juan Pablo Blanco Rambla, Ildefonso Linero Arana, and Luis Beltran Lares M. Instituto Oceanografico de Venezuela, Universidad de Oriente, Apdo. Postal 245, Cumana, Estado Sucre, Venezuela Abstract. —A new species of callianassid, Sergio guaiqueri, similar to S. guara (Rodrigues, 1971) is described from the northeastern coast of Venezuela. The new species differs in having larger, more inflated cornea; in the shape of margins of merus and carpus of the major cheliped, and in dentition of the dactyl of the major cheliped. Manning & Felder (1991) placed the ge- nus Neocallichirus Sakai, 1988, in the sub- family Callichirinae, and included six west- ern Atlantic species. Manning (1993) later added two new species to this genus, but noted that it could be divided into two groups of species based on the shape of the telson and uropodal endopod. Subsequent- ly, Manning & Lemaitre (1994) restricted the genus Neocallichirus, and proposed the new genus Sergio for four western Atlantic species forming one of those groups: S. guassutinga (Rodrigues, 1971), S. guara (Rodrigues, 1971), S. mirim (Rodrigues, 1971), and S. trilobatus (Biffar, 1970). With the discovery of S. guaiqueri new species, the genus now contains five species in the western Atlantic. Specimens from Vene- zuela reported by Blanco Rambla & Linero Arana (1994) as Neocallichirus sp. actually represent the new species described herein. Specimens of the new species were col- lected in the northeastern coast of Vene- zuela with a Petersen grab on board R/V Guaiqueri II. Measurements (mm) were made with an ocular micrometer. Carapace length (cl) was measured along the middor- sal line of the carapace from the tip of the rostrum to posterior margin; total length (tl) was measured from tip of rostrum to pos- terior margin of telson. The holotype has been deposited in the National Museum of Natural History, Smithsonian Institution, Washington, D.C. (USNM), and paratypes in the Museum of the Instituto Oceano- grafico de Venezuela, Cumana (IOV). Family Callianassidae Dana, 1852 Subfamily Callichirinae Manning & Felder, 1991 Genus Sergio Manning & Lemaitre, 1994 Sergio guaiqueri, new species Figs. 1-3 Neocallichirus sp. Blanco Rambla & Linero Arana, 1994:20-22, figs. 4—5. Material. — Venezuela: north of Jose, An- zoategui State (10°08’40”N, 64°50'10”W), Petersen grab, 9 May 1991, 24 m, clay-silt bottom: 1 male (cl 4.2 mm, holotype, USNM 265294); 3 males (tl 28.3-49.1 mm, para- types, IOV). Diagnosis.—Carapace with 3 unarmed anterior projections; median extending less than half length of eyestalks. Eyestalk with cornea subterminal, inflated. Propodus of third maxilliped longer than wide, distal margin sinuous. Major cheliped with ven- tral serrations on merus, carpus as long as palm; dactylus with 3—4 small teeth. Telson nearly 2 times wider than long; posterior VOLUME 108, NUMBER 1 103 Fig. 1. Sergio guaiqueri, new species. a-c, f, g, holotype male, USNM 265294, cl 4.2 mm; f, g, paratype male, IOV, cl 2.5 mm. a, Carapace and cephalic appendages (dorsal view); b, Same (lateral view); c, Anterior portion of carapace and eyestalk; d, Major cheliped (inner face); e, Minor cheliped (outer face); f, Abdomen; g, Telson and left uropod. margin divided by a median cleft. Uropodal endopods longer than wide. Description. — Front with 3 anterior pro- jections, median subtriangular, flattened, extending less than half length of eyestalks, with scattered setae on dorsal surface and few setae on ventral surface near the apex. Carapace rounded dorsolaterally; linea thal- assinica distinct, reaching posterior margin of carapace; cervical groove distinct, delim- iting posterior margin of dorsal oval (Fig. la, b). Abdominal somites smooth; somite 1 smallest, with 2 short setae at each side on posterior half; somite 2 largest, with small setae posteriorly; somites 3—5 wider than long, with 1 tuft of dense setae posterola- terally on each side; somite 6 bilobed, an- terior lobe larger than posterior, with long setae on posterior margin (Fig. If). Eyestalk dorsally flattened, not extending to end of first segment of antennular pe- duncle; cornea subterminal, hemispherical, pigmented; anterior margin of eyestalk with angled projection (Fig. 1c). Antennular pe- duncle shorter than antennal peduncle; seg- ment 3 of antennular peduncle longer than segments | and 2. Segment 4 of antennal peduncle as long as segment 5, segments 1|- 3 short. Mandible with | sharp tooth on molar process; incisor process with 10 sharp teeth, median longest; mandibular _ palp 3-segmented, third article longest, bearing 104 f a,c,d,e b f Imm | Imm | Imm PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Sergio guaiqueri, new species, paratype male, IOV, cl 2.5 mm. a, Mandible; b, Maxillule; c, Maxilla; d, First maxilliped; e, Second maxilliped; f, Third maxilliped (inner face); g, Second pereopod; h, Third pereopod; i, Fourth pereopod; j, Fifth pereopod. numerous setae on outer surface (Fig. 2a). Maxillule with elongate lobe on coxo-basal endite, palp slender, unsegmented. Maxilla with elongate lobes on both coxal and basal endites, scaphoganthite with anterior lobe rounded (Fig. 2b, c). First maxilliped with short basal endite. Second maxilliped with exopod shorter than merus of endopod, dactyl short. Third max- illiped without exopod; ischium-merus sub- pediform, ischium with crista dentata on inner surface, with about 10 teeth; propodus longer than wide, distal margin emarginate, concave (Fig. 2d-f). Chelipeds dissimilar, unequal. Major cheliped strong, ischium elongate, widening distally, ventral margin finely serrated; me- rus broad, ventrally denticulate, bearing long setae, convex, without hooks or spines; car- pus subquadrate, longer than wide, ventral margin evenly convex, smooth; propodus with palm as long as carpus, ventral margin irregular, except on fixed finger, with long setae; cutting edge of fixed finger with 3—4 small teeth; dactylus slightly shorter than palm, with tufts of setae on dorsal margin, curved ventrally, tip hooking over outer surface of fixed finger, cutting edge with 3- 4 small teeth on proximal half (Fig. 1d). Ischium of minor cheliped narrow, mar- gins diverging distally; merus broad on dis- tal half, lacking ventral keel, serrations or hooks; carpus elongate, more than 2 times longer than wide, with long setae on ventral margin; dactylus longer than palm, both dactylar and fixed finger with cutting edges denticulate (Fig. le). Second pereopod chelate, fixed finger and VOLUME 108, NUMBER 1 | \ l \ d VN DN U) yy \ Y 7, Ng Y 4 Y yy . Y \ YH e b C Y a,b 105 SS S S SK d p+ 2mm 2mm 0.5mm Fig. 3. Sergio guaiqueri, new species, paratype male, IOV, cl 2.5 mm. a, First pleopod; b, Second pleopod; c, Third pleopod; d, Appendix interna of third pleopod. dactylus symmetrical, both cutting edges pectinate; carpus widening distally. Margins of chela and carpus setose; ventral margin of merus with row of evenly spaced long setae. Third pereopod with carpus widening anteriorly; propodus 2 times wider than long, outer face covered with numerous tufts of short setae; dactylus short, articulating on upper anterior margin. Fourth pereopod subchelate, with carpus and propodus broader distally; propodus and dactylus with setae, densely setose grooming apparatus on ventral margin of propodus. Fifth pereopod chelate, propodal finger short; dactylus elongate, apex rounded. Propodus and dac- tylus setose; carpus widening distally (Fig. 72-1). Telson short, nearly 2 times wider than long, subtrapezoidal, posterior margin di- vided by shallow median cleft into broadly rounded posterolateral lobes. Uropods lon- ger than telson; uropodal protopod bilobed, lacking spines; exopod broad, longer than endopod, upper exopodal plate shorter than lower plate, posterior margin flattened; en- dopod elongate, longer than wide, margins converging distally, apex rounded (Fig. 1g). First pleopod of male uniramous, 2-segmented, bearing a few setae, terminal segment hooked at tip (Fig. 3a). Second pleopod biramous, endopod and exopod of about the same length, blade-like exopod, with a few setae; endopod with a slender lobe and few setae. Appendix interna and appendix masculina absent (Fig. 3b). Ple- opods 3-5 alike, exopods and endopods fo- liaceous; endopods with small appendix in- terna embedded on inner margins; margins of endopods and exopods with plumose se- tae (Fig. 3c, d). Distribution.—Known so far only from north of Jose, Anzoategui State, Venezuela. Etymology.—The specific name honors an ancient indian tribe Guaiqueri from the northern Venezuelan coast, and the re- search vessel of the IOV. Remarks.— Sergio guaiqueri, new spe- cies, is most closely related to S. guara. The two can be separated by differences in the shape of the corneae, margins of carpus and 106 merus, and cutting edge of the dactyl of the major cheliped. In the new species, the cor- neae are larger, and more inflated than in S. guara; the interior margin of carpus of the major cheliped is smooth, and the lower margin of merus is convex and denticulate, whereas in S. guara the carpus of major cheliped has the proximal half of its interior margin serrated, and the merus has a strong serrated tooth near the proximal extremity of the lower margin. Additionally, the cut- ting edge of the dactyl of the major cheliped has only 3—4 small teeth in S. guaiqueri, whereas in S. guara the cutting edge has four strong irregular teeth. Acknowledgments The authors are indebted to Dr. R. Le- maitre and Dr. R. B. Manning, USNM, for their helpful advice concerning the status of the species and who were kind enough to examine the holotype, and offering numer- ous valuable suggestions to the manuscript. We are grateful to Dr. D. L. Felder, Uni- versity of Southwestern Louisiana, for his constructive comments on an earlier ver- sion of this manuscript. We especially wish to thank F. Arocha P., Rosenstiel School of Marine and Atmospheric Science, Univer- sity of Miami, for providing most of the necessary literature. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Literature Cited Biffar, T. A. 1970. Three new species of callianassid shrimp (Decapoda, Thalassinidea) from the western Atlantic. — Proceedings of the Biological Society of Washington 83:35—49. Blanco Rambla, J. P., & I. Limero Arana. 1994. New records and new species of ghost shrimps (Crus- tacea: Thalassinidea) from Venezuela.—Bulle- tin of Marine Science 55(1):16—29. Dana, J.D. 1852. Macroura. Conspectus Crustaceo- rum & conspectus of the Crustacea of the Ex- ploring Expedition under Capt. C. Wilkes, U.S.N.—Proceedings of the Academy of Nat- ural Sciences of Philadelphia 6:10-28. Manning, R. B. 1993. Two new species of Neocalli- chirus from the Caribbean Sea (Crustacea: De- capoda: Callianassidae).— Proceedings of the Biological Society of Washington 106:106—-114. —,&D.L. Felder. 1991. Revision of the Amer- ican Callianassidae (Crustacea: Decapoda: Thalassinidea).— Proceedings of the Biological Society of Washington 104:764-792. , & R. Lemaitre. 1994. Sergio a new genus of ghost shrimp from the Americas (Crustacea: De- capoda: Callianassidae).— Nauplius 1:39-43. Rodrigues, S. de A. 1971. Mud shrimps of the genus Callianassa Leach from the Brazilian coast (Crustacea, Decapoda). — Arquivos de Zoologia, Sao Paulo 20(3):191-223. Sakai, K. 1988. A new genus and five new species of Callianassidae (Crustacea: Decapoda: Thalas- sinidea) from northern Australia.— The Beagle, Records of the Northern Territory Museum of Arts and Sciences 5(1):51-69. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(1):107-116. 1995. Heteranthessius hoi, a new species (Copepoda: Pseudanthessiidae) from a sea-anemone in the Straits of Gibraltar, with remarks on the genus Pablo J. Lopez-Gonzalez and Mercedes Conradi (PJLG) Laboratorio de Biologia Marina, Departamento de Fisiologia y Biologia Animal. Fac. de Biologia. Universidad de Sevilla. Apdo 1095, 41080 Sevilla, Spain; (MC) Departamento de Biologia Animal, Vegetal y Ecologia, Fac. Ciencias del Mar. Universidad de Cadiz. Apdo. 40, 11510 Puerto Real, Cadiz, Spain Abstract. — An unknown species of Heteranthessius is described as H. hoi and is compared with the three previously known species: H. dubius (T. Scott 1903), H. scotti Bocquet et al., 1959 and H. furcatus Stock, 1971. This is the first time male and female of this genus have been found together. The diagnostic char- acters of this species are two claws on the terminal segment of the second antenna and third segment of leg 4 exopodite with the armature formula II, I, 5. The diagnosis of the genus is modified to incorporate these and other features. Resumen. —Se describe una nueva especie del género Heteranthessius, H. hoi, y se compara con las tres especies conocidas hasta ahora: H. dubius (T. Scott 1903), H. scotti Bocquet et al., 1959 y H. furcatus Stock, 1971. Esta es la primera vez que se ha encontrado el macho y la hembra de este género juntos. Las caracteristicas tipicas de la especie son: dos garfios terminales del ultimo segmento de la segunda antena y la armadura del tercer segmento del exopodito de la pata cuarta, II, I, 5. Se modifica la diagnosis del género in- corporando éstas y otras caracteristicas. In the course of research on copepod fau- na associated with marine invertebrates from the Straits of Gibraltar and nearby ar- eas, a female and male of copepod belonging to the genus Heteranthessius were found in the coelenteron of the sea-anemone Aipta- siogeton pellucidus (Hollard 1848). The ge- nus Heteranthessius was erected by T. Scott (1904) for Paranthessius dubius T. Scott, 1903. Paranthessius T. Scott was replaced by Heteranthessius since the former name had been preoccupied by Paranthessius Claus, 1889, another copepod crustacean. Three species have been recognized in Heteranthessius: H. dubius (T. Scott 1903), H. scotti Bocquet et al., 1959 and H. furcatus Stock, 1971. Heteranthessius dubius was de- scribed from a male dredged in British wa- ters. Heteranthessius scotti was found in washings of the calcareous alga Lithophyl- lum incrustans from Brittany. H. furcatus was collected in the branchial cavity of the ascidian Microcosmus sabatieri; this was the first time that the genus was found in Med- iterranean waters. Two males of an unde- termined Heteranthessius species were col- lected recently from the Irish coasts (Holmes & Gotto 1992). In this paper, Heteranthes- sius hoi, new species, is described and com- pared with the other species of the genus. The diagnosis of Heteranthessius is modi- fied to accommodate the new features shown by H. hoi. Material and Methods Aiptasiogeton pellucidus was collected on stones from the infralittoral zone (4 m deep). 108 The copepods were removed by dissection of the sea-anemone and preserved in 70% ethanol. Because the purpose of host dis- section was not to search for symbiotic co- pepods, the male specimen was slightly damaged. The specimens were stained with chlorazol black, dissected under a stero- microscope. Permanent mounts were made in lactophenol and sealed using entellan. All figures were drawn with the aid of a camera lucida. The letter after explanation of each figure refers to the scale at which it was drawn. Order Poecilostomatoida Thorell, 1859 Family Pseudanthessiidae Humes & Stock, 1972 Genus Heteranthessius T. Scott, 1904 Diagnosis (modified from Humes & Stock 1973).—Body of female transformed, pro- some swollen. Body of the male cyclopi- form, elongate. Urosome in female 5-segmented, in male 6-segmented. Caudal ramus with 5 or 6 setae. First antenna 7-segmented. Second antenna 4-segmented, with one or 2 claws. Mandible consisting of broad basal area and 2 unequal slender, re- curved lashes with denticulated edges. In females lashes may be partially fused or overlapping, so that it is difficult to distin- guish them. First maxilla with 2 terminal elements. Second maxilla with terminal ar- mature consisting of one long spine, ter- minally bifurcate in female, plus a short seta. Maxilliped in female non-prehensile, 3-segmented with rounded or pointed tip, in male prehensile, 4-segmented (that distal part of claw represents a fourth segment). Legs 1-4 with 3-segmented rami except leg 4 endopod which consists of a single small segment or knob with one or 2 terminal elements. Setae on all 4 legs short in female, long and plumose in male. Leg 4 exopodite with terminal segment having ITI, I, 5 or II, I, 5. Leg 5 without a free segment and rep- resented only by two setae. Found free, associated with ascidians or actinian coelenterates. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Type species: Heteranthessius dubius (T. Scott 1903). Heteranthessius hoi, new species Figs. 1-5 Type material. —1 2 and 1 6 from the sea- anemone Aiptasiogeton pellucidus (Hollard 1848) at Patricia, Cadiz (Spain), Feb. 1990. The holotype female and allotype male have been deposited in the Museo Nacional de Ciencias Naturales of Madrid, Spain (MNCN lot n° 20.04/334). Description. —Female: Body transformed (Fig. la, b); length (not including setae on caudal rami) 2 mm and greatest width 1 mm, based on one specimen in 70% etha- nol. Ratio of length to width of prosome 1.5:1. Ratio of length of prosome to that of urosome 3.8:1. Separation of pedigers weakly defined. Genital segment (Fig. Ic) wider than long, 258 x 387 wm. Genital areas located dorsolaterally on widest part of segment. Postgenital segments from an- terior to posterior 45, 58, and anal segment 142 um long. Caudal ramus about 3 times longer than wide. One outer anterolateral seta 96 um, dorsal seta 32 um, outermost terminal seta 96 um, intermost terminal seta 109 wm, and two long median terminal setae 167 wm (outer) and 251 wm (inner). All setae naked. Rostrum rounded posteroventrally. First antenna (Fig. 1d) about 408 wm long, lengths of 7 segments (measured along their posterior nonsetiferous margins) 48 (87 um along anterior margin), 132, 48, 45, 64, 42 and 29 um, respectively. Formula for ar- mature 4, 12, 2, 4, 4+1 aesthete, 2 and 6 setae. All setae naked. Second antenna (Fig. le) 4-segmented. First segment, 74 wm along its outer edge, 154 um along its inner edge; remaining seg- ments, 106, 42, 87, respectively. Formula for armature: 1, 1, 3, and 3 setae + 2 un- equal claws. All elements naked. Labrum (Fig. 1f), with two posteroventral lobes, with medial projection. Paragnaths smooth. Mandible (Figs. 1g, 5a) with concave side VOLUME 108, NUMBER 1 109 Fig. 1. Heteranthessius hoi, new species. Female. Holotype: a, dorsal (A); b, lateral (A); c, urosome dorsal (B); d, first antenna (C); e, second antenna (C); f, labrum, ventral (D); g, mandible (E). Scale bars: A, 1 mm; B, 400 um; C, 200 um; D, 100 um; E, 35 um. 110 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Heteranthessius hoi, new species. Female. Holotype: a, first maxilla (A); b, second maxilla (D); c, maxilliped (D); d, leg 1, anterior (B); e, leg 2, anterior (B); f, leg 3, anterior (B); g, leg 4, anterior (B); h, leg 4 endopodite, anterior (C). Scale bars: A, 35 wm; B, 200 um; C, 30 wm; D, 100 um. VOLUME 108, NUMBER 1 111 Table 1.—Comparison of the armature of legs 1-4 of the known females of Heteranthessius. (Roman numerals = spines; arabic numerals = setae; exp = exopodite; enp = endopodite.) H.. scotti H. furcatus H. hoi, Leg Bocquet et al., 1959 Stock, 1971 new species Ist exp: 1-0; I-0/I-1; I,1,3/1,1,4 1-0; I-1; III,1,4 I-0; I-1; ITI,1,4 enp: 0-0; 0-0; II,4 0-0; 0-1; II-4 O-1; O-1; II-4 2nd exp: I-0; 0-0; II,1,5/1I,1,4 1-0; I-1; III,1,5 I-0; O-1; II,1,5 enp 0-0; 0-0; III,3/1II,0 0-0; 0-2; III-3 O-1; 0-2, III-3 3rd exp: I-0; I-1; III,1,5 I-0; I-1; III,1,5 I-0; O-1; III,1,5 enp: 0-0; 0-2; II,2/1,0 0-0; 0-2; III,2 QO-1; 0-2; II-2 4th exp: I-0; 1-1; III,1,5 1-0; I-1; III,1,5 1-0; I-1; II,1,5 enp: I-0 II-0 I-0 produced into 2 spinulose lashes. First lash long; second smaller and apparently partly fused to long lash. First maxilla (Fig. 2a) unilobated and elongate with 2 terminal setae subequal in length. Second maxilla (Fig. 2b), 2-segmented, first segment unarmed, sec- ond trapezoidal, small with a strong ter- minally bifid spine (13.8 um) and one seta (6.3 wm). Maxilliped (Fig. 2c) 3-segmented. First and second segments unarmed, third small and pointed. Legs 1-4 (Figs. 2d.g) with 3-segmented rami, except leg 4 endopodite (Fig. 2h) which consists of single small segment with one seta shorter than segment. Formula for ar- mature as in Table 1. Setae of leg rami, basis and coxae small and naked. Leg 5, represented by 2 naked setae in other species of genus, not observed here. Male: Body cyclopiriform (Fig. 3a). Pe- diger | fused with cephalosome, remaining pedigers clearly defined. Length (excluding setae on caudal rami) 1.4 mm, greatest width 423 wm, based on one specimen in 70% ethanol. Ratio length to width of prosome 2:1. Ratio length of prosome to that uro- some 1.6:1. Leg 5 (Fig. 3c) small, 11 < 18 um, and bearing two setae. Genital segment (Fig. 3b) as long as wide. Four postgenital segments from anterior to posterior 74 x 143, 52 x 134, 35 x 95, and 69 X< 95 um. Caudal ramus resembling that of female, 122 x 52 um, length/width ratio 2.3:1. Se- tae plumose, except dorsal seta. Rostrum rounded posteroventrally. First antenna (Fig. 3e) similar to that of female, formula for armature 4, 11+2 aesthetes, 3, 3+1 aesthete, 2, 7+1 aesthete. Second antenna (Fig. 3f) similar to that of female, but ter- minal claws more slender and longer. La- brum (Fig. 3g) and paragnaths as in female. Mandible (Figs. 3h, 5b) more slender than that of female, with lashes well separated. First maxilla (Fig. 4a) smaller than that of female, with length of both setae similar to length of segment. Second maxilla (Fig. 4b) similar to female; second segment bearing a long (not bifid) spine and one seta. Maxilliped (Fig. 4c) 4-segmented; first, second and third segments unarmed; last segment transformed into long curved claw with one seta at basis. Inner surface of claw with spinules. Legs 1-4 (Fig. 4d—g) with 3-segment rami. Leg 4 endopodite reduced unisegmented with one long seta 3.8 times as long as seg- ment. Formula for armature as in Table 2. Setae of rami and coxae of legs plumose; those on basis naked. Leg 5 (Fig. 3c) with 2 naked setae, sub- equal in length. Leg 6 (Fig. 3d) consisting of two scarcely plumose setae 47 wm and 44.5 um. Spermatophore not observed. Sexual dimorphism. —The body 1s trans- formed with an inflated prosome in the fe- 112 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. Heteranthessius hoi, new species. Male. Allotype: a, dorsolateral (A); b, urosome, dorsal (B); c, leg 5, dorsal (C); d, leg 6, ventral (C); e, first antenna (D); f, second antenna (E); b, labrum, ventral (F); c, mandible (G). Scale bars: A, 300 um; B, 200 um; C, 30 um; D, 100 um; E, 100 um; F, 50 wm; G, 30 um. VOLUME 108, NUMBER 1 I) Fig. 4. Heteranthessius hoi, new species. Male. Allotype: a, second maxilla (B); b, first maxilla (A); c, maxilliped (B); d, leg 1, anterior (C); e, leg 2, anterior (C); f, leg 3, anterior (C); g, leg 4, anterior (C). Scale bars: A, 40 um; B, 100 um; C, 200 um. 114 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Tea a: _ Spare “aie i ‘ a. = ae Fig. 5. Heteranthessius hoi, new species: A, mandible of female; B, mandible of male. Notice lashes (white arrows) and basal portion (black arrows). Scale bars: 30 um. male, but cyclopiriform in male. The first antenna possesses only one aesthetasc in the female but five in the male. Terminal claws of the second antenna are more slender in the male than in the female. The lashes of the mandible are more divergent in the male than in the female. The elements of the first maxilla are proportionately longer in the male than in the female. The terminal spine of the second maxilla is bifurcated in the female and simple in the male. Sexual di- morphism in the maxilliped is as in other lichomolgoidean genera, 4-segmented and prehensile in the male and very simplified and reduced in the female. The armature of legs 1—4 is reduced in the female and free- swimming type in the male. The caudal rami are longer in the female than the male. Setae of the caudal rami are plumose in the male and naked in the female. Etymology. —This species is dedicated to Dr. Ju-Shey Ho (Long Beach, California) in recognition of his valuable contributions to the knowledge of parasitic and commensal copepods. Key to species of the genus Heteranthessius Known Females (Heteranthessius dubius is not included) 1. Terminal segment of second anten- na with 2 claws; third segment of leg 4 exopodite having armature for- mula TES ol. ee eee ees H. hoi Terminal segment of second anten- na with one claw; third segment of leg 4 exopodite having armature for- mula III, I, 5 2. Length/width ratio caudal ramus 3:1 or longer; third segment of maxil- liped pointed; single segment of leg 4 endopodite with 2 elements H. furcatus Length/width ratio of caudal ramus less than 2.3:1; third segment of maxilliped rounded; single segment VOLUME 108, NUMBER 1 of leg 4 endopodite with one ele- MCI hag ee Oa oe A ah earn 4 AZ, scotti Known Males (Heteranthessius scotti and H. furcatus are not included) 1. Terminal segment of second anten- na with 2 claws; third segment of leg 4 exopodite having armature for- men a ee a ce ese Ses Hi. hoi Terminal segment of second anten- na with one claw; third segment of leg 4 exopodite having armature for- ono) 0 00S) eo) ae re H. dubius Discussion With the discovery of Heteranthessius hoi, four species of the genus are now known. The host specificity at the generic level 1s not clear because hosts of Heteranthessius dubius and H. scotti are unknown, H. fur- catus was found in an ascidian and H. hoi in a sea-anemone. Heteranthessius are rare- ly found. H. dubius is known from the male, and H. scotti and H. furcatus from the fe- male. Only Heteranthessius hoi is known from both sexes. Two features of the genus Heteranthessius are not in a good agreement: the setation of the first antenna and the structure of the mandible. As to the formula of the first an- tenna, none of the described females of the genus (H. scotti, H. furcatus and H. hoi) has the same setation. The armature of H. scotti is 3,9, 2,4,4 + 1, 3, 6; that of H. furcatus 3,12, 4, 3,4 + 1, 3,6 + 1, and that of H. hoi 4, 12,2, 4,4 + 1, 2, 6. The three species have in common only the armature of the fifth segment, which bears 4 setae and | aesthete; H. furcatus has a second aesthete in the last segment. There is no concordance in the formula of the first antenna between the known males. The setation of H. dubius Peeoeiee oe be 138 1,2 15 | whereas that of H. hoi is 4, 11 + 2, 3,3 + 1,4 + 1, 2, 7+ 1. Such differences in the setation of the first antenna is rather rare 115 Table 2.—Comparison of the armature of legs 1-4 of the known males of Heteranthessius. (Roman nu- merals = spines; arabic numerals = setae; exp = exo- podite; enp = endopodite.) H. dubius H. hoi, Leg (T. Scott, 1903) new species Ist exp I-O; I-1; IfI,1,4 I-O; I-1; HI,1,1(?) enp: 0-I; ?;? I-0; 0-1; ? 2na exp: 2: 7:? I-O; I-1; TI,1,5 enp: 0-1; 0-2; III-3 O-1; 0-2; IlI-2 3rd ~exp: —_‘I-0; I-1; TII,1,5 I-0; O-1; ITI,1,5 enp: 0-1; 0-2; III-2 QO-1; 0-2; H-2 4th exp: I-0; I-1; IIU,1,5 I-0; I-1; I,1,5 enp: 0-0 0-1 for the genera Lichomolgoidea, they usually differ only in the number of aesthetes. These disimilarities in the armature of the first antenna (including differences between male and female) may indicate that Heteranthes- sius 18 highly a variable genus as reported for the genera Amarda Humes & Stock, 1972 and Indomolgus, Humes & Ho, 1966 (Humes & Stock 1973). The genus Paranthessius Scott, 1903 was erected for P. dubius, which has two lashes on the mandible. However, Heteranthessius scotti and H. furcatus were described with a single lash. The study of both sexes in H. hoi has allowed us to corroborate the exis- tence of two lashes in this structure (Fig. 5). As these lashes may be partially fused or overlapping in the female, the mandibles of H. scotti and H. furcatus should be re-ex- amined. Until then, the diagnosis of the su- perfamily Lichomolgidea Humes & Stock, 1972 must be slightly modified to accom- modate the mandible of the Heteranthessius species as follow: ‘““Mandible simple, with- out terminal elements, but often attenuate into a slender lash, rarely two lashes.” Het- eranthessius hoi may be easily distinguished from the remaining species of the genus by the presence of two claws on the terminal segment of the second antenna and the third segment of the leg 4 exopodite having II, I, 5. The female of Heteranthessius hoi differs from H. scotti and H. furcatus in the ar- 116 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON mature of legs 1-4 (see Table 1), particularly in the endopodal segments of legs 1-3 and terminal segment of leg 4 exopodite. The labrum of a Heteranthessius species is de- scribed and illustrated for the first time. Se- tae on the segments of the second antenna are larger in H. hoi than in the remaining species. Heteranthessius hoi is similar to H. scotti in the shape of the first maxilla. The second maxilla of H. hoi has a smaller basal segment than that in the remaining species. The maxilliped of H. hoi and H. furcatus are acuminate terminally, but rounded in H. scotti. The caudal ramus is equal to or more than 3:1 longer than wide in H. hoi and H. furcatus, and 2.2:1 in H. scotti. The male Heteranthessius hoi differs from H. dubius in the armature of legs (see Table 2). On inner margin of the third exopodite segment of the first leg, only one seta was observed in H. hoi. Because the right side of the copepod was damaged, the actual dif- ference in the armature of this segment is impossible to ascertain. The unisegmented leg 4 endopodite is slightly larger in H. hoi than in H. dubius. This segment also bears a long plumose seta which is lacking in H. dubius. The mandible is longer and slender in Heteranthessius hoi than H. dubius, and the two basal elements present in H. dubius are lacking in H. hoi. The first maxilla car- ries two moderately long elements, longer in H. hoi than in H. dubius. Acknowledgments We thank Dr. A. G. Humes (Woods Hole, Massachusetts) and Dr. J. H. Stock (Am- sterdam) for generous help with informa- tion and literature, and Dr. R. V. Gotto (Belfast, Northern Ireland) for revising the first draft of this manuscript and for valu- able suggestions. Literature Cited Bocquet, C., J. H. Stock, & F. Bernard. 1959. Co- pépodes parasites d’invertébrés des cOtes de France. IX. Description d’une nouvelle espéce remarquable de Lichomolgidae: Heteranthes- sius scotti n. sp. (Cyclopoida).— Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen. Amsterdam, series c, 62(2):111- 118. Claus, C. 1889. Ueber neue oder wenig bekannte halbparasitische Copepoden, insbesondere der Lichomolgiden-und-Ascomyzontiden-Gru- ppe.—Arbiten aus dem Zologischen Institute der Universitat Wien, 8(3):1—44. Hollard, M. H. 1848. Etudes dur l’organisation des actinies. — Thése présentées a la Faculté des Sci- ences de l’Université de Paris, 26 pp. Holmes, J. M. C., & R. V. Gotto. 1992. A list of the Poecilostomatoida (Crustacea: Copepoda) of Ireland.— Bulletin of the Irish Biogeographical Society 15:2-33. Humes, A. G., & J. H. Stock. 1973. A revision of the family Lichomolgidae Kossmann, 1877, cy- clopoid copepods mainly associated with the marine invertebrates. — Smithsonian Contribu- tions to Zoology 127:308 pp. Scott, T. 1903. On some new and rare Crustacea collected at various times in connection with the investigations of the Fisheries board for Scotland.— Twenty-first Annual Report of the Fishery Board for Scotland, 1902, 3:109-135. 1904. Notes on some rare and interesting marine Crustacea. — Twenty-second Annual Re- port of the Fishery Board for Scotland, 1903, 3:242-261. Stock, J. H. 1971. Découverte du genre Heteran- thessius (Copepoda) en Méditerranée: H. fur- catus n. sp.—Bulletin de la Société Zoologique de France 95:335-340. Thorell, T. 1859. Bidrag till kinnedomen om Krus- taceer som lefva i arter af slagtet Ascidia L.— Konglica Svenska Vetenskaps-Akademiens Handlingar, ny foljd, 3(2):1-84. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(1):117-124. 1995. Description of Amphiascoides atopus, a new species (Crustacea: Copepoda: Harpacticoida) from a mass culture system Guilherme R. Lotufo and John W. Fleeger Department of Zoology and Physiology, Louisiana State University, Baton Rouge, Louisiana 70803, U.S.A Abstract. —Amphiascoides atopus, a new species, is described from a mass culture system. Its natural habitats and distribution are unknown. This new species shares the possession of seven setae on the exopod 3 of leg 4 with ten other species in the genus. Amphiascoides atopus can be distinguished from congeneric species by the shape, relative length, and ornamentation of the setae on the leg 5 in both male and female. Possession of three setae grouped together on the proximal region of the second endopod segment of leg 2 distinguishes the males of this species from all other described males in this genus. A species of harpacticoid copepod has been successfully kept in a mass culture sys- tem in our laboratory at the Department of Zoology and Physiology, Louisiana State University, Baton Rouge, LA. Approxi- mately 300,000 individuals have been har- vested from 2.5 m2 culture surface area al- most every day for more than six months, with no sign of population decline. Speci- mens used to start this culture were ob- tained from a commercial bioassay labo- ratory, where it was identified as a fouling species in culture tanks of marine fish and invertebrates. It was identified as an un- described species of Amphiascoides. It was not possible to determine the source of the individuals present in the tanks as the origin of the various specimens was from many localities in the United States including oth- er laboratories. Therefore, the natural hab- itat and the distribution of this new taxon remain unknown. In the United States three species of Amphiascoides (A. lancisetiger Lang, 1965, A. petkovskii Lang, 1965, A. dimorphus Lang, 1965) have been described from the coast of California and were col- lected from tidal pools, shell-sand intertid- al, among algae and among stones (Lang 1965). On the Atlantic coast, the cosmo- politan A. debilis has been recorded from Scituate, MA living among algae by Rosen- field (1967). Specimens in the culture sys- tem display strong photophobia and have been successfully cultured in trays contain- ing small stones, suggesting that they might come from the habitats typically reported for Amphiascoides. The culture system has been kept at a temperature of 23-—26°C and 30%o salinity. Specimens examined came directly from our mass culture system. Hundreds of in- dividuals were drawn from the system using a Pasteur pipet and fixed with formalin 4%. Type material was sorted under a stereo- dissecting microscope. Whole specimens were examined in lactic acid and dissected parts mounted in Hoyers mounting medi- um. All figures were drawn using a camera lucida connected to a Zeiss microscope. De- tails were checked on an Olympus BX 50 microscope using phase contrast. Type ma- terial and an additional lot of specimens were deposited in the collections of the Na- tional Museum of Natural History, Smith- sonian Institution. Terminology follows that of Huys & Boxshall (1991). 118 Order Harpacticoida Sars, 1903 Family Diosaccidae Sars, 1906 Genus Amphiascoides Nicholls, 1941 Amphiascoides atopus, new species Figs. 1-3 Material. —Holotype female, ethanol preserved (USNM 259892). Allotype male, ethanol preserved (USNM 259893). Paratypes: two females dissected and pre- served on slides (USNM 259894); two males dissected and preserved on slides (USNM 259898); 10 females (USNM 259895) and 10 males (USNM 259896) ethanol pre- served. Additional lot with approximately one hundred adults and copepodites, etha- nol preserved (USNM 259897). All speci- mens were drawn from culture system col- lections. Description of female.—Length (tip of rostrum to end of caudal rami) of holotype, 841 wm; paratypes 780-1000 um (vn = 10). Body (Fig. 1a) moderately slender, slightly tapering towards the end. Rostrum (Fig. 1b) prominent, with pair of small sensory setae and tip finely serrate. Urosome (Fig. Ic), with ventro-lateral row of spinules on all segments; genital double somite with dorso- lateral sign of subdivision; genital field with attached spermatophore as shown in Fig. lc; third segment with ventral row of spi- nules; anal somite with distal border spi- nulate ventrally; operculum and pseudo- perculum lacking. Caudal rami (Fig. 1d, e) about twice wider than long and bearing short spinules on inner margin; setae II, IV and V well developed; seta II as long as seta VI; seta VII triarticulate; setae II, III, VI and VII naked; seta IV spinulose on outer margin and with few spinules on inner mar- gin, seta V sparsely spinulose on both mar- gins. Antennule (Fig. 1f). Eight-segmented. Ar- mature as follows: (Roman numeral = seg- ment; Arabic numeral = seta): I— 1; II— 10; III—5; IV—3 + aesthetasc; V—1; VI—4; VII—2; VIII—6 + aesthetasc. Antenna (Fig. 1g). Coxa short, asetose. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Allobasis partially divided by an incom- plete suture on outer face and with one inner seta. Second endopod segment with two strong spines and row of spinules laterally; distal edge with one spiniform seta, four geniculate setae and one slender seta con- fluent at base with outermost geniculate seta. Exopod 3-segmented; proximal segment with plumose seta, second segment very short and asetose, terminal segment with one lateral and two apical setae, all pinnate. Mandibular palp (Fig. lh). Broad basis with three pinnate and one naked seta on distal edge. Endopod elongate, 2-segmented and with one strong and pinnate basal seta on and two sub-apical setae on first segment and three apical setae on second segment. Exopod 2-segmented, first segment with one basal seta, second segment with three plu- mose apical setae. Maxillule (Fig. 11). Praecoxa with row of spinules medially and at base of coxa; ar- thrite with two surface setae and seven el- ements on inner margin, two proximal-most elements ornamented with spinules. Coxa partially fused to basis and lacking epipod- ite and bearing two setae. Basis with one spinular row, two sub-distal and three distal setae. Exopod with two setae, inner-most pinnate; endopod with three setae, the out- er-most pinnate. Maxilla (Fig. 1j). Syncoxa with spinule rows medially and distally and with three endites, first two bearing two setae and dis- tal endite bearing three setae. Allobasis with distal margin of endite bearing a fused claw and one seta. Endopod 3-segmented with one seta on first two segments and three setae on distal segment. Maxilliped (Fig. 1k). Well developed and prehensile. Syncoxa with spinule rows and three stout and pinnate setae. Basis with a row of long spinules and two naked setae on inner margin. Endopod 1-segmented with a terminal claw twice as long as the segment and two accessory setae. Leg 1 (Fig. 2a) with intercoxal sclerite wider than long, without ornamentation; VOLUME 108, NUMBER 1 119 IL/ Fig. 1. Amphiascoides atopus, Female: a. Habitus, dorsal; b. Rostrum; c. Urosome (first somite omitted), ventral; d. Caudal ramus, dorsal; e. Anal somite and caudal ramus, ventral; f. Antennule; g. Antenna; h. Mandibular palp; i. Maxillule; j. Maxilla; k. Maxilliped. Scale bars: a, c, e = 100 um; b, d, fk = 50 um. 120 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON YL 7 YL HE: p Uf > VQ ES me Ne Ss ye “guia Manat Ma SOS OS 50 SS: OX _ oO SSIS YX => SS SS Es eye \) SS anntan® a | \ $UNU LLU i Mla Meee) prremmrnn Fig. 2. Amphiascoides atopus, Female: a. Leg 1; b. Leg 2; c. Leg 3; d. Leg 4; e. Leg 5. Male: f. Urosome (first somite omitted). Scale bars = 100 um. coxa with four rows of spinules on anterior face; basis with rows of spinules on inner and distal margin and at base of inner spine; outer seta spiniform and as long as inner spine, both finely serrate; exopod and en- dopod 3-segmented; exopod with long spi- nules present on outer margin of proximal segment and both margins of second seg- VOLUME 108, NUMBER 1 Table 1.—Spine and seta formula for legs 1 to 4. 121 Exopod segment Endopod segment Coxa Basis Hes 2 3 jee 3 Leg 1 0-0 1-I I-0; I-0; I,1,2 O-1; O-1; 0,2,1 Leg 2 0-0 1-0 I-0; I-1; TI,1,1 0-1; O-1; 1,2,1 Leg 3 0-0 1-0 I-O; I-1; III,1,2 Was Malle Nw Leg 4 0-0 1-0 I-O; I-1; III,1,3 O-1; O-1; 1,2,1 ment; proximal and second segment lacking inner setae; terminal segment with two out- er spines and two apical setae; first segment of the endopod reaching beyond the distal margin of the exopod, with spinules on out- er and inner margins, and armed with inner seta plumose on inner margin of posterior end; second and third segments spinulose on outer margin; second segment with one plumose seta on inner margin; terminal seg- ment with outer apical seta serrate on outer margin, inner apical seta geniculate and slender inner seta. Legs 2 to 4 (Fig. 2b—d) with intercoxal sclerite about as long as wide and with spi- nules only on leg 2. Exopod 3-segmented; outer margin of all segments with a row of spinules; inner row of spinules present on all segments on leg 2, second segment on leg 3, and on first two segments on leg 4; distal margin spinulate on exopod | of all legs and exopod 2 leg 2. Endopod 3-segmented and with row of spinules on outer margin of all segments except endo- pod 3 leg 2; inner and distal margins spi- nulate only on endopod | and 2 leg 2. Spine and seta formula for legs 1 to 4 are shown in Table 1. Setation using system devised by Lang (1934) as follows: exopod endopod leg 1 0202022 eile e2A0 leg 2 O10) 2.3 eee Za leg 3 eet 2.3) ley ae leg 4 OeMe2 ZS ewe ten. Leg 5 (Fig. 2e) with baseoendopod and exopod distinct. Baseoendopod with inner expansion reaching about midpoint of ex- opod, ornamented with a row of spinules and armed with three inner and two apical setae; two innermost setae irregularly spi- nulate; outer seta naked. Exopod 1.5 times longer than wide, with row of spinules on inner and outer margins and armed with five setae, the innermost pinnate and the remaining naked. Leg 6 (Fig. 1c) represented by two setae, inner seta short and pinnate on inner margin, outer setae long and naked. Description of the male.—Length (tip of rostrum to end of caudal rami) of allotype 828 um; paratypes 724-882 um (n = 10). Body similar to the female. Urosome (Fig. 2f) with second and third somites not fused. Third and fourth segments with ventro-lat- eral row of spinules; anal somite with distal border spinulate ventrally. The male is identical to the female in all other respects. Antennule (Fig. 3a) 8-segmented, genic- ulation between sixth and seventh seg- ments. Armature as follows: (Roman nu- meral = segment; Arabic numeral = seta): I—1; Ii—8; I11—7; IV—7 + spine + aesth- etasc; V—1; VI—1 + 2 spines; VII—4; VIII—5 + aesthetasc. Leg 2 endopod (Fig. 3b) 2-segmented; en- dopod 1 with row of spinules on outer mar- gin and one short and pinnate seta on mid- point of inner margin. Endopod 2 produced to a strong mucruniform projection bearing three setae on inner margin, all grouped to- gether on the proximal region of the seg- ment, above a triangular lateral projection; proximal-most seta the shortest, sparsely spinulose on both margins; median seta the longest, 1.3 times longer than the segment and spinulate on inner margin; distal-most 122 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. Amphiascoides atopus, Male: a. Antennule; b. Leg 2; c. Leg 5. Scale bars = 50 ym. seta the most robust, spinulate on both mar- gins. Leg 5 (Fig. 3c) with confluent baseoen- dopods, not separated by intercoxal sclerite; inner expansion with row of spinules and armed with two irregularly spinulate spines, outermost one slightly longer; outer seta na- ked and very long. Exopod 1.3 times longer than wide and ornated with minute spinules and bearing five elements; innermost spine the longest, spinulate distally on inner mar- gin; second inner spine minutely pinnate distally on inner margin; middle setae na- ked and 2.5 times longer than the segment; two outer spines naked and equal in length. Leg 6 (Fig. 2f) represented by three naked setae set on a ventrolateral common plate. Distribution. —Unknown. Etymology. —The specific name is a Neo- Latin formation from Greek a-‘without? + topos- ‘place’, used as an adjective with the masculine form atopus. It is used as an al- lusion to the fact that the type material has not been collected from the species natural habitats, as they remain unknown. Remarks. —To the genus Amphiascoides VOLUME 108, NUMBER 1 belong twenty named species and Amphias- coides sp. Becker & Schriver, 1979, from which hitherto only the male has been de- scribed. Both males and females of A. ato- pus can be distinguished from the other species of the genus by the shape, relative length and ornamentation of the setae on the leg 5. Amphiascoides atopus shares with A. brevifurca (Czernivaski), A. neglectus (Nor- man & T. Scott), A. subdebilis (Willey), A. lancisetiger Lang, A. petkovskii Lang, A. di- morphus Lang, A. nichollsi, A. bulbiseta Pal- lares, A. koltuniand A. breviarticulatus Kunz, the leg 4 exopod three bearing 7 setae. The remaining species in this genus (A. debilis (Giesbrecht), A. dispar (T. Scott & A. Scott), A. limicolus (Brady), A. littoralis, A. nanus (Sars), A. nanoides (Sars), A. sterilis, A. go- likovi, and A. paradebilis Tschislenko, have this segment armed with 6 setae. This char- acter is not described for A. proximus, a poorly described species that is distin- guished from A. atopus and all other species in the genus by possessing the baseoendo- pod leg 5 of the female with only four setae. Among the species armed with seven se- tae on the leg 4 exopod 3, 4. brevifurca dif- fers from A. atopus by lacking inner setae on endopod 2 leg 1. Amphiascoides nichollsi is distinguished from A. atopus by having two projections on the tip of the rostrum, leg 1 endopod 1 1.3 times longer than the exopod. Amphiascoides neglectus differs from A. atopus in having the leg 1 endopod 1 1.4 times longer than the exopod and in the shape of the exopod leg 5, 2.6 times longer than wide. Amphiascoides subdebilis is distinguished from A. atopus in having the caudal seta VI shorter than seta III, leg 1 endopod 1 1.4 times longer than the ex- opod. Amphiascoides lancisetiger, A. pet- kovskii, and A. dimorphus, the three North American species described by Lang (1965), differ from A. atopus in having the man- dibular exopod 3-segmented and bearing five setae and endopod with seven (4. /ancise- tiger) or eight setae (A. petkovskii and A. 123 dimorphus), maxillule with four setae on the endopod and 11 (A. /ancisetiger) or 10 el- ements (A. petkovskii and A. dimorphus) on the arthrite, maxilla with 2-segmented en- dopod, and leg 1 endopod | extending far beyond exopod. Amphiascoides lancisetiger and A. petekovskii further differ from A. ato- pus in having three terminal setae on the distal segment of the antenna exopod. Am- phiascoides bulbiseta differs from the spe- cies here described in having the leg 1 en- Gopod | 1.5 times longer than the exopod, three terminal setae on the distal segment of the antenna exopod and caudal seta III modified. Amphiascoides breviarticulatus differs from A. atopus in having three ter- minal setae on the distal segment of the an- tenna exopod and inner seta of the leg 6 almost as long as the genital double seg- ment. Finally, Amphiascoides koltuni also differs from A. atopus in having the leg 1 endopod | 1.4 times longer than the exo- pod. The male of A. atopus is readily distin- guished from all described males of Am- phiascoides and from Amphiascoides sp. Becker & Schriver, 1979, by the shape and insertion of the setae on the endopod 2 leg 2. This segment is always modified in Am- phiascoides, but only in A. atopus are the three setae grouped together on the proxi- mal region of the segment. In all other spe- cies the three setae are inserted on different positions on the inner margin. Amphias- coides dispar bears only four setae on the leg 5 exopod in the male, differing in this respect from 4. atopus and all the other spe- cies in the genus. Males are unknown in A. proximus (T. Scott), A. sterilis (Monard), A. littoralis (T. Scott), A. nichollsi Lang, and A. golikovi Tschislenko. The leg 2 is not represented for A. koltuni Tschislenko, but it is apparent from the description that it is similar to A. subdebilis (Willey). Morphological data were obtained from Lang (1948) for all species represented in his publication. Data were complemented by Noodt (1955) for A. brevifurca; Bodin 124 (1964), Noodt (1955), and Pallares (1975) for A. subdebilis; Tschislenko (1967) for A. debilis; Mielke (1974) for A. nanus; Becker (1970) for A. dispar. Original descriptions were used as the sole source for the re- maining species cited. A mass culture system of A. atopus (adult dry weight of 5 wg) may generate enough copepods to be used in mariculture as a high nutritional value food source for a wide va- riety of larval fish and shellfish (Sun & Flee- ger, pers. comm.). The potential use of Am- phiascoides as an alternative food source in fish farming has already been suggested in the literature (Walker 1970, Kahan 1979, Shirgur 1989). Acknowledgments Specimens were obtained from the cul- ture system designed and maintained by Dr. Bin Sun. We are grateful to Jeff Tamplin for kindly inking the line drawings and to Dr. Janet Reid for her help with the bibliogra- phy. Support from Louisiana Sea Grant (Project Number R/A-35-PD) is gratefully acknowledged. G. Lotufo is financially sup- ported by CNPq (Brazilian Federal Gov- ernment). Literature Cited Becker, K. H. 1970. Beitrag zur Kenntnis der Co- pepoda Harpacticoidea sublitoraler Weichbdd- en in der Kieler Bucht.—Kieiler Meeresfor- chungen 26:56-73. — ., & G. Schriever. 1979. Eidonomie und Tax- onomie abyssaler Harpacticoidea (Crustacea, Copepoda). Teil III. 13 nue Tiefsee-Copepoda Harpacticoidea der Familien Canuellidae, Cer- viniidae, Tisbidae, Diosaccidae und Ameiri- dae.— Meteor ForschErgebn (Biology) 31:38-62. Bodin, P. 1964. Recherches sur la systématique et la distribution des Copépodes Harpacticoides des substrats meubles des environs de Marseille. — Recuieil des Travaux, Station Marine d’En- doume-Marseille 35:107-183. Huys, R., & G. A. Boxshall. 1991. Copepod evolu- tion. The Ray Society, London, 468 pp. Kahan, D. 1979. Vegetables as food for marine har- pacticoid copepods. — Aquaculture 16:345-350. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Kunz, H. 1983. WHarpacticoiden (Crustacea: Cope- poda) aus dem Litoral der Azoren.— Arquipé- lago 4:117—208. Lang, K. 1934. Marine Harpacticiden von der Camp- bell-Insel und einigen anderen siidlichen In- seln.— Lunds Universitets Arsskrift, N.F. (Avd. 2) 30(14):1-56. . 1948. Monographie der Harpacticiden: I. 1- 896; II. 897-1982. Hakan Ohlssons Booksellers, Lund. 1965. Copepoda Harpacticoidea from the Californian Pacific coast.—Kungliga Svenska Vetenskapsakademiens Handlingar, Fyarde Ser- ien 10(2):1—560. Mielke, W. 1974. Eulitorale Harpacticoidea (Cope- poda) von Spitzbergen. — Mikrofauna des Meer- esboden 37:161-—210. Nicholls, A. G. 1941. Littoral Copepoda from south Australia. I—Harpacticoida.—Records of the South Australian Museum (Adelaide) 6:38 1-427. Noodt, W. 1955. Marine Harpacticoiden (Crust. Cop.) aus dem Marmara Meer.— Revue de la Faculté des Sciences de l’Université d’ Istanbul, Série B, Sciences Naturelles 20(1—2):49-94. Pallares, R. 1975. Copépodos marinos de la Ria De- seado (Santa Cruz, Argentina). Contribucion sistematico-ecologica IV.— Physis 34A:213-227. Rosenfield, D.C. 1967. The external morphology of the developmental stages of some diosaccid har- pacticoid copepods (Crustacea) from Massachu- setts Bay. Unpublished Ph.D. dissertation, Bos- ton University, Boston, 307 pp. Shirgur,G. A. 1989. Phased fertilisation for culturing copepods. —Crustaceana 56:113-126. Tschislenko, L. L. 1967. Copepoda Harpacticoida of the Karelian coast of the White Sea. —Issledov- aniya Fauny Morei 7(15):48-196. [in Russian] . 1977. Harpacticids (Copepoda, Harpacticoi- dea) from sponges of Franz Josef Land. —Issle- dovaniya Fauny Morei 14(22): ““Biocoenoses of the shelf of Franz Josef Land and the fauna of adjacent waters”’:237-276. 1978. New species of copepod harpacticids (Copepoda, Harpacticoida) from the Possjet Bay of the Sea of Japan.—Trudy Zoologicheskogo Instituta, Akademiya Nauk SSSR, Leningrad 61: 161-192. [in Russian] Walker, J. 1979. Mechanisms of density-dependent population regulation in the marine copepod Amphiascoides sp. (Harpacticoida).— Marine Ecology Progress Series 1:209-221. Willey, A. 1935. Harpacticoid Copepoda from Ber- muda. Part II.—Annal and Magazine of Natural History Ser. 10. Vol. 15. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(1):125-129. 1995. Hydroids colonizing the carapaces of the ostracode Philomedes brenda from the Beaufort Sea, Arctic Ocean Louis S. Kornicker and Dale R. Calder (LSK) Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 29560, U.S.A.; (DRC) Department of Invertebrate Zoology, Royal Ontario Museum, Toronto, Ontario, Canada M5S 2C6 Abstract.—A hydroid in the superfamily Bougainvillioidea or Pandeoidea colonizing the carapaces of the mydocopid ostracode Philomedes brenda (Baird 1850) collected in the Beaufort Sea, Arctic Ocean, is described and illustrated. The Beaufort Sea forms that part of the Arctic Ocean lying north of Alaska and Can- ada, eastward of Point Barrow and west- ward of Banks Island and Prince Patrick Island. During a study of myodocopid os- tracodes collected in bottom sediments of the continental shelf and slope of the Beau- fort Sea from 1971 to 1976, an athecate hydroid was observed colonizing the outer surface of carapaces of the myodocopid os- tracode Philomedes brenda (Baird 1850) (Kornicker 1988: abstract, fig. 2a, b). The only previous report of Hydrozoa at- tached to the carapaces of myodocopid os- tracodes was that of Baker (1975:78, fig. 17b, c, J) who observed them attached to Eu- philomedes carcharodonta (Smith 1952), Scleroconcha trituberculata (Lucas 1931) and Eusarsiella ‘“‘tubipora’’ (Darby 1965) on the southern California continental shelf. Baker (1975:78) reported that the same spe- cles appeared to be present on all three os- tracode species and that both feeding and reproductive polyps were present, but it was not possible to identify the hydrozoan. The purpose of the present paper is to describe the morphology of the hydrozoan on the Arctic ostracodes, and to further identify it. For the latter purpose the junior author made a nematocyst slide preparation from one of the hydranths. The nematocyst categories are of types widespread in the superfamilies Bougainvillioidea and Pan- deoidea. Without more evidence concern- ing the life cycle of the hydroid it is not prudent to carry the identification to family or genus. Hydroids of the Arctic Seas of Russia have been monographed by Naumov (1969), and those of northern Canada by Calder (1970, 1972). The Beaufort Sea species seems dif- ferent from any mentioned in these works. Specimens were prepared for the Scan- ning Electron Microscope by critical point drying. All specimens, including those in alcohol, have been placed in collections of the National Museum of Natural History, Department of Invertebrate Zoology, and have been assigned USNM numbers. Bot- tom collections were made with a Smith- McIntyre Grab (SMG) covering 0.1 square meters. Class Hydrozoa Subclass Hydroida Order Leptolida Superfamily Bougainvillioidea or Pandeoidea Figs. 1-3 Material. — All hydroids are on carapaces of P. brenda that have been given USNM 126 Figs le Hydranth with tentacles and T-shaped segment of pseudohydrorhiza; b, Hydranth with tentacles, medusa bud, and segment of pseudohydrorhiza. (Scale bar 0.1 mm.) Hydroid on P. brenda USNM 193284B: a, numbers. USCGC_ Glacier, cruise WEBSEC71, 12 Sep 1971, station C7184, sample SMG-1017-11; 76°16'48’N, 150°18'30”W, 831 m: USNM 193284A, adult female, USNM 193284B, ovigerous female. Ice Station OCS-1, 29 Oct 1975, 71°19'06’N, 152°34'00”W, 59 m: sample SMG-1092-16, USNM 158419A, oviger- ous female; sample SMG-1091-12, USNM 158422, adult female. Ice Station OCS-1, sample SMG-1093-17, 30 Oct 1975, PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 71°21'36’N, 152°35’00”W, 102 m, USNM 158466, adult female. Ice Station OCS-?2, station PPB-100, sample SMG-1137-14, 19 May 1976, 71°26'30"N, 152°38'42”W, 99 m, USNM 158536, adult female. Ice Station OCS-6, station PP-100, 3 Nov 1976, 71°21'54"N, 152°33'24”"W, 99 m: sample SMG-1491-09, USNM 193282, adult fe- male; sample SMG-1494-10, USNM 193283, ovigerous female. Distribution. — Beaufort Sea, 59-831 m. Description (Figs. 1-3).— Colonies on car- apaces of adult and ovigerous females of P. brenda (carapace length 2.80-3.06 mm) (Figs. 2a, b, f, 3a, b, d). Pseudohydrorhiza filiform (stolon width about 0.07 mm; in- dividual stolon length about 0.5 mm or less) (Fig. 3d); polyp with maximum length about 0.75 mm; hydrocaulus unbranched; hy- dranths with filiform tentacles forming sin- gle whorl (length of longest hydranths and tentacles about 0.46 mm) (Figs. 1, 2c-e, 3a, b, e, f); larger hydranths with 4—6 tentacles (Fig. la), but most with 2—4 (Figs. 2c-e, 3e, f); stalks with wrinkled perisarc and either cylindrical (Fig. 3a—c) or with indentation at base (Fig. 3b), some with proximal flaring section (Fig. 1). Sparse oval processes (pos- sibly medusa buds or sporosacs, maximum diameter 0.17 mm) originating directly from pseudohydrorhiza (Fig. 1b). Nematocyst complement of a hydranth comprising desmonemes (4.1 um long X 2.8 um wide) and heterotrichous microbasic euryteles (6.5 um long X 2.8 wm wide). One specimen with a developing gonophore; de- velopment too early to determine whether it would become a medusa or a fixed gon- ophore. Discussion. — The ostracode P. brenda has pan-Arctic distribution (Kornicker 1982:4). Juveniles are not capable of swimming and are restricted to the bottom where they e1- ther crawl on the substrate or burrow into it. When they become adults, both sexes become capable of swimming and rise in the water to couple. After mating, the fe- male either rubs or breaks off its swimming VOLUME 108, NUMBER 1 127 Fig. 2. Hydroid on left valve of P. brenda USNM 158536 (length 3.06 mm): a, Pseudohydrorhiza and 3 hydranths (arrows), x 22.8; b, Detail of pseudohydrorhiza, from a, x 67; c, Detail of hydranth on lower right of a, 171; d, Detail of hydranth on upper right of a, x 336; e, Hydranth in d reoriented, x 336. Hydroid on right valve of USNM 158536: f, Pseudohydrorhiza and 4 hydranths (arrows), x 22.8. bristles and is again restricted to the bottom (Elofson 1969:165). The eight females hav- ing attached hydrozoans all have broken swimming bristles, indicating that they had mated and returned to the bottom. The presence of hydrozoans on adult females only, and not on juveniles, suggests that the hydroids may have become attached when 128 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. Hydroid on left valve of P. brenda USNM 158536: a, Detail from left end of Fig. 2f showing 2 hydranths (arrows), x 101; b, Detail of upper right of Fig. 2f showing hydranths (arrows), x 174; c, Detail showing surface crinkles of right hydranth shown in b, x 1448. Hydroid on right valve of P. brenda USNM 158466 (length about 3 mm): d, Detail of pseudohydrorhiza near anterior of valve, x 119; e, Hydranth with tentacles, x 326; f, Detail of tentacles in e, II > IV > I on right side but III = II > IV > I on left. Length varies from 23.6 to 29.5 mm (43.5-54.4% of ML). Oral surface with low, trabeculate protective membranes on both margins of all arms. Trabeculae short, narrow and attached to bases of suck- er Stalks. Suckers biserial, spherical in shape, of approximately uniform size for nearly en- tire length of all arms. Chitinous sucker rings of proximal suckers with 10-12 large, point- ed teeth on distal margins and several very small, rounded teeth on proximal margins. Small distal suckers, with about eight large teeth distally but no teeth proximally. Tentacles and clubs: Tentacles nearly twice as long as arms; length is 44.1 mm (81.4% of ML). Cross section is elliptical, almost triangular. Club only slightly ex- panded. Club very long; about 22.5 mm (ca. 51% of tentacle length). Aboral swimming keel originates approximately mid club and extends to distal end. Proximal half of club and stalk with separate, low median aboral keel. Low trabeculate protective mem- branes equally developed on both margins of clubs proximal of dactylus. Membranes most developed on long manus. Four series of suckers on full length of club; ventral and dorsal pairs of suckers on manus slightly separated by a sucker-free median area (Fig. 2c). Neither suckers or knobs on stalks. Suckers on manus about 350 um across on longest axis; oval, almost circular, with wide openings (Fig. 4) and not compressed like VOLUME 108, NUMBER 1 141 Fig. 4. Mid manal tentacular sucker, 54 mm ML, scale: 100 um. Raion F ZOKU 135% °»&#&«xfeL Fig. 5. Mid manal tentacular sucker, 108 mm ML, UMML 31.2369, scale: 100 um. 142 suckers in large P. adami. Suckers on carpus and dactylus more circular. Suckers on short carpus ca. 80% diameter of those on manus, suckers on moderately short dactylus ca. 60% diameter of those on manus. Periph- eral fringe of outer chitinous rings of suckers composed of tooth-like plates; larger, blunt, tooth-like structures present on medial plates with those near interior margins of suckers much larger. Three pronounced teeth on in- terior distal margins of inner chitinous ring. Buccal mass: Small; projects outward. Seven large lappets with well-developed connectives attached dorsally to arms I-II, ventrally to arms IIJ-IV. Buccal membrane prominent; translucent but firm. No buccal suckers. Beak and radula: Not extracted. Protrud- ing rostrums of upper and lower beaks dark- ly pigmented. Chromatophores: Present on all parts of specimen; mostly round or oval. Red-brown, mostly flared, vary in intensity from faint to vivid. Numerous and large along dorsal midline of mantle. Large, widely spaced on dorso-lateral and ventral surface of mantle. Form distinct wine-glass pattern on ventral surface. Dense on dorsal surfaces of fins and tail, and ventrally on outer halves of fins; absent on interior halves and ventral surface of tail. Few, mostly contracted, on ventral surface of funnel. Some widely scattered on ventral surface along dorsal margin of fun- nel groove, midline of head and anterior and posterior to eyes. Few large, irregularly spaced on dorsal surface of head posterior to eyes, and some on posterior and anterior center of head. Two or three indistinct hor- izontal rows along entire length of arms. Several very small, elongate on distal and proximal edges of suckers. Aboral row along full length of tentacular stalk, dense pattern aborally on club. On oral surface of club, scattered around the bases of the suckers stalks. Tentacular suckers and dermal cushions of 108 mm and 179 mm ML juveniles. — Tentacular suckers: Mostly oblong and lat- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON erally compressed with narrow openings at 108 mm ML, some partially open (Fig. 5). Laterally compressed with narrow openings at 179 mm ML, identical to the condition described for the adult by Voss (1956). Length from about 1000 to 1200 um at both 108 and 179 mm ML. Tooth-like structures on the infundibulum similar but more ro- bust than at 54 mm ML. Dermal cushions: Flat-topped and broad, not conical and pointed like in the specimen of 54 mm ML. Positioned relatively much closer, like in adult (Roper & Lu 1990). About 100 um in height and 300-400 um across at 108 mm ML; roughly circular or oblong, may have been slightly flattened during handling of specimen; vacuolated (Fig. 6). About 175-250 um in height and about 500 um across at 179 mm ML; rough- ly square or pentagonal; closer together than at 108 mm ML, often touching at the base; vacuolated (Fig. 7). Underlying mantle tis- sue more highly vacuolated than at 54 mm ML (Figs. 6b, 7b). Discussion There are notable differences in some morphological features of the SEFCAR specimen compared with the description and illustration of the later growth stages of P. adami in Voss (1956). The size of the spec- imen illustrated in Voss (1956) is given as 136 mm MLin the figure caption. However, this number is questionable: the mantle lengths of the holotype, paratypes, and ad- ditional material listed by Voss were all larger. Nevertheless, differences include the fins, which are shorter and more rounded in the small SEFCAR juvenile compared to larger specimens; the tail, which is longer and narrower in the juvenile; and the ten- tacular suckers, which are open in the small juvenile and not compressed like those of older individuals. Additionally, the devel- opment of the dermal cushions in juvenile P. adami with growth consists of a change from widely spaced cones to closely set pen- VOLUME 108, NUMBER 1 143 2aKU 2eas Fig. 6. Dermal cushions from mantle, 108 mm ML: (a) lateral view, scale: 100 um, and (b) cross section, scale: 50 wm. 144 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON b. igeau asag Fig. 7. Dermal cushions from mantle, 179 mm ML, UMML 31.465: (a) lateral view, scale: 500 um, and (b) cross section, scale: 100 um. VOLUME 108, NUMBER 1 tagons. The internal structure of the dermal cushions of my P. adami juveniles is similar to the vacuolated condition of dermal cush- ions for the adult described and illustrated by Roper & Lu (1990). There are also differences in the dermal cushions and tentacular suckers of the SEF- CAR specimen compared with the only de- scriptions of small Pholidoteuthis sp.; spec- imens of 14 mm and 32.8 mm ML in Clarke (1992). Clarke characterized the dermal structures of the specimen of 14mm MLas_ the same as those of adults. However, the dermal cushions of the SEFCAR P. adami are conical and pointed, so Clarke’s 14 mm ML specimen is probably a different species. Furthermore, Clarke stated that the dermal structures of a Pholidoteuthis sp. of 32.8 mm ML have small papillae around their mar- gins, similar to the papillose tubercles de- scribed by Pfeffer (1912) for the 30 mm ML, juvenile type of 7etronychoteuthis massyae. However, there are no marginal papillae on the dermal cushions of the SEFCAR spec- imen. The papillose tubercles of an imma- ture 7. massyae of 100 mm ML in Roper & Lu (1990) are similar to the dermal struc- tures of a mature Pholidoteuthis boschmai of 286 mm ML in Villanueva & Sanchez (1993), which is in accord with the conclu- sion of Nesis & Nikitina (1990) that T. mas- syae and P. boschmai are conspecific. Therefore, the similarities of the dermal structures of Clarke’s 32.8 mm ML Pholi- doteuthis sp. to the papillose tubercles of T. massyae imply that it is P. boschmai. Clarke (1992) described the tentacular suckers of his Pholidoteuthis sp. of 14 mm ML as being open and not compressed, and the suckers of his Pholidoteuthis sp. of 32.8 mm ML as compressed, the same as in adults. The tentacular suckers of the 54 mm ML SEFCAR specimen, however, are not compressed and are clearly different from those of my older museum juveniles, which have narrow openings like later growth stages of P. adami described by Voss (1956). The inconsistency in the size at which com- 145 pression of the tentacular suckers is present between the specimen at hand and Clarke’s specimens is further indication that his specimens are not P. adami. Additionally, the fins in an illustration of Clarke’s (1992) specimen of 14 mm ML are proportionally smaller than the SEFCAR juvenile, and the tail is not developed. These and the above incongruities suggest differ- entiation of Clarke’s specimens, at least one of which is likely P. boschmai, and my spec- imen at the generic level. This inference concurs with Toll’s (1982) finding that P. boschmai and P. adami may not be con- generic based on gladius morphology. How- ever, aS Roper and Lu (1989) concluded, the status of the type species, P. boschmai, must first be resolved before the status of P. adami can be revised. Acknowledgments I thank N. Voss for reviewing the manu- script and contributing much helpful in- sight, R. E. Young and an anonymous re- viewer for critique that substantially im- proved the text, C. McSweeny for the illus- trations, P. Blackwelder, M. Lynn, and N. Romer for their assistance with the SEM analysis, and M. McGowan for the suc- cessful capture of the specimen. This work was supported by National Oceanic and At- mospheric Administration (NOAA) Coop- erative Agreements #NA85-WCH-06134 and #NA90RAH00075 to the Cooperative Institute for Marine and Atmospheric Stud- ies (CIMAS), University of Miami. Re- search operations in the National Marine Sanctuary were conducted under National Marine Sanctuary Permits KLNMS and LKNMS-11-89. Literature Cited Arocha, F., L. Marcano, & R. Cipriani. 1991. Ceph- alopods trawled from Venezuelan waters by the R/V Fridtjof Nansen in 1988.— Bulletin of Ma- rine Science 49(1-2):23 1-234. Clarke, M.R. 1992. Family Pholidoteuthidae Adam, 146 1950. Pp. 168-170 in M. J. Sweeney, C. F. E. Roper, F. M. Mangold, M. R. Clarke, & S. v. Boletsky, eds., ““Larval’’ and juvenile cephalo- pods: a manual for their identification. —Smith- sonian Contributions to Zoology 513:1-282. Goldman, D. A. 1993. Distribution of cephalopod paralarvae across the Florida Current front in the Florida Keys: preliminary results. — Revista Biologia Tropical, Suplemento 41(1):31-34. —., & M. F. McGowan. 1991. Distribution and abundance of ommastrephid squid paralarvae off the Florida Keys in August 1989.— Bulletin of Marine Science 49(1-2):614—622. Lee, T. N., C. Rooth, E. Williams, M. McGowan, A. F. Szmant, & M. E. Clarke. 1992. Influence of Florida Current, gyres and wind-driven circu- lation on transport of larvae and recruitment in the Florida Keys coral reefs. —Continental Shelf Research 12(7/8):971—1002. Lipka, D. A. 1975. The systematics and zoogeogra- phy of cephalopods from the Gulf of Mexico. Unpublished Ph.D. dissertation, Texas A&M University, College Station, 351 pp. Nesis, K. N., & I. V. Nikitina. 1990. Revision of the squid family Lepidoteuthidae. — Zoologicheskii Zhurnal 69:38-49. Okutani, T. 1983. Mollusks. Pp. 189-354 in Takeda, M., & T. Okutani, eds., Crustaceans and mol- lusks trawled off Surinam and French Guiana. Japanese Marine Fishery Resource Research Center, Tokyo, 354 pp. Pfeffer, G. 1912. Die Cephalopoden der Plankton- Expedition.—Ergebnisse der Plankton-Expedi- tion der Humbolt-Stiftung 2:1-815, Atlas of 48 pls. Rathjen, W. F. 1981. Exploratory squid catches along the continental slope of the eastern United States. — Journal of Shellfish Research 1(2):153- 159. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Roper, C. F. E., & C. C. Lu. 1989. Systematic status of Lepidoteuthis, Pholidoteuthis, and Tetrony- choteuthis (Cephalopoda: Oegopsida). —Pro- ceedings of the Biological Society of Washington 102:805-807. —., & C.C. Lu. 1990. Comparative morphology and function of dermal structures in oceanic squids (Cephalopoda).—Smithsonian Contri- butions to Zoology 493:1—40. ———., M. J. Sweeney, & C. E. Nauen. 1984. FAO species catalogue. Volume 3. Cephalopods of the world. An annotated and illustrated cata- logue of species of interest to fisheries. -FAO Fisheries Synopsis 125(3):1—277. —, R.E. Young, & G. L. Voss. 1969. An illus- trated key to the families of the order Teuthoi- dea (Cephalopoda).—Smithsonian Contribu- tions to Zoology 13:1-32. Toll, R. B. 1982. The comparative morphology of the gladius in the Order Teuthoidea (Mollusca: Cephalopoda) in relation to systematics and phylogeny. Unpublished Ph.D. dissertation, University of Miami, Miami, 390 pp. Vecchione, M., & C. F.E. Roper. 1991. Cephalopods observed from submersibles in the western North Atlantic.— Bulletin of Marine Science 49(1 -2): 433-445. Villanueva, R., & P. Sanchez. 1993. Cephalopods of the Benguela Current off Namibia: new addi- tions and considerations of the genus Lycoteu- this. —Journal of Natural History 27:15—46. Voss, G. L. 1956. A review of the cephalopods of the Gulf of Mexico. — Bulletin of Marine Science of the Gulf and Caribbean 6(2):85-178. Wiebe, P. H., K. H. Burt, S. H. Boyd, & A. W. Morton. 1976. A multiple opening/closing net and en- vironmental sensing system for sampling zoo- plankton. —Journal of Marine Research 34:313- 326. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(1):147-152. 1995. A new species of the gorgonacean genus Narella (Anthozoa: Octocorallia) from Hawaiian waters Frederick M. Bayer Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, U.S.A. Abstract. — A new species of Narella collected in Hawaiian waters by the U.S. Bureau of Fisheries steamer Albatross in 1902, but overlooked when the col- lection of octocorals was originally studied, is now described and illustrated by scanning electron micrographs. The species differs from all others by the elaborate ornamentation of the sclerites of the polyps and the high, thin crests of the coenenchymal scales. During a review of gorgonacean corals of the genus Narella from Hawaiian waters in connection with the identification of spec- imens recently collected from guyots south of Hawaii, a fragmentary specimen ob- tained by the U.S. Bureau of Fisheries steamer A/batross during its Hawaiian cruise of 1901-1902 was found among the un- studied specimens in the Department of In- vertebrate Zoology, U.S. National Museum of Natural History. The bulk of the collec- tion of octocorals obtained around the Ha- waiian Islands by the A/batross during its cruise of 1901 and 1902 were reported upon by Prof. C. C. Nutting of the State Univer- sity of Iowa (Nutting 1908), but neither the present fragment nor the complete speci- men from which it must have come were mentioned by Nutting in his report. It can be referred without question to the genus Narella but its highly distinctive characters exclude it from any species of that genus heretofore described. Even though the specimen is only a frag- ment (fig. la) from a terminal branch of what must have been a much larger colony, it retains all diagnostic characters necessary for the establishment of a new species. The only feature that remains unknown is the form and branching pattern of the complete colony, a character that may be influenced by environmental factors and is not essen- tial for recognition of the species. Primnoidae Genus Narella Gray, 1870 Narella Gray, 1870:49.— Deichmann, 1936: 168. Stachyodes Studer [& Wright], 1887:49.— Wright & Studer, 1889:xlvii, 53.—Ku- kenthal, 1924:308 (references). Four species of Primnoidae referred to the genus Stachyodes (a junior synonym of Narella) were reported from waters around Hawaii by Nutting (1908): Stachyodes an- gularis Nutting, 1908; S. regularis “Wright & Studer, 1889”; S. bowersi Nutting, 1908; and S. dichotoma Versluys, 1906. The specimens described as the new spe- cies Stachyodes angularis by Nutting (1908) can be referred to the genus Calyptrophora Gray. The specimen Nutting reported as S. regularis (erroneously attributed to Wright & Studer) is an incomplete terminal branch of Narella dichotoma (Versluys, 1906), and the specimens reported as Stachyodes di- chotoma represent an entirely different new species of Narella, which will be described 148 elsewhere. Stachyodes bowersi remains a valid species now referred to Narella. The specimen here described cannot be referred to any species of Stachyodes (=Na- rella) previously reported from Hawaiian waters, to any species from Indonesian wa- ters obtained by the Siboga Expedition de- scribed by Versluys (1906), nor to any spe- cies from Japan described by Kinoshita (1907, 1908). Narella ornata, new species Figs. 1-3 Material examined. —Hawaii, Kauai: Hanamaulu warehouse S43°W, 8.1 miles, 550-409 fathoms (=1007-—748 m), bottom temperature 37.8°F, USFC str. Albatross sta. 4019, 21 Jun 1902. One small piece of branch with comatulid crinoid attached. Holotype, USNM 94617 (SEM 2352-2354). Diagnosis. — Narella with both buccal and basal scales distinctly separated adaxially; one pair of small adaxial buccal scales; ad- axial opercular scales not remarkably small- er than outer laterals; outer surface of oper- cular and body scales with radial sculpture developed as prominent, thin crests; corti- cal sclerites elongate, narrow plates and ir- regular rods, many with high, thin crests. Description. — The type specimen (Fig. 1a) is fragmentary so colonial form and manner of branching are unknown. The polyps (Fig. 2) are 3 mm long measured parallel with the branch, directed downward and arranged in whorls of three or four; 7—8 whorls occur in 3 cm of branch length. Members of basal (Fig. 1h) and buccal (Fig. 1g) scale pairs do not meet to form closed but unfused rings around the body of the polyp; medial (Fig. 1f) and buccal (Fig. 1g) scales similar in size; the closed opercular scales form a broad cone (Fig. 3, top) that projects beyond the margin of the buccal pair (Fig. 2, top); one pair of large, squarish adaxial buccal scales is pres- ent between the adaxial ends of the buccal scales and below the adaxial opercular scales PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON (Fig. 3, bottom). Tentacles with scattered, minute, elongate scales. The radial external sculpture (Fig. 1b) of the opercular (Fig. 1d) and body scales (Fig. 1f-h) is conspicuously developed as thin crests, giving the polyps an unusually ornate appearance. The inner surface of the body scales is covered with crowded, complex tu- bercles (Fig. 1c). The opercular scales (Fig. 1d) are of the usual triangular shape, with a strong, thin, apical keel on the inner sur- face, and are sculptured externally by a sys- tem of thin crests radiating from center of calcification (i.e., the “‘nucleus’’); the ad- axial opercular scales are somewhat smaller than the abaxial and lateral scales but are not unusually reduced in size. The sclerites of the cortex (Fig. le) are elongate scales and narrow, irregular rods, most having the outer surface ornamented with one or sev- eral prominent crests; the inner surface is covered by small, complex tubercles. Etymology. — Latin ornatus = decorated, beautiful, in allusion to the ornate sculp- turing of the sclerites. Adjective. Comparisons. —The most conspicuous feature of this species is the elaborate, cris- tate sculpture of the body scales and coen- enchymal sclerites, unlike any species of Narella heretofore recorded. Remarks.—The nature of this sample suggests that it was cut from a larger spec- imen of gorgonian in order to segregate the attached crinoid. As it is most unlikely that such a small part of a gorgonian branch with commensal crinoid would have been taken by the trawl, it is highly probable that a larger specimen from which it was cut did exist at one time. However, no trace of such specimen can now be found in the collec- tions of the National Museum of Natural History. The collection of octocorals ob- tained around the Hawaiian Islands by the Albatross during its cruise of 1901 and 1902 were entrusted by the Commission of Fish- eries to Prof. C. C. Nutting at the State Uni- versity of Iowa for study and report. Al- VOLUME 108, NUMBER 1 2354 10KY Boum Fig. 1. Narella ornata new species. USNM 94617. a, Holotype fragment; b, External sculpture of body scale; c, Internal sculpture of body scale; d, Opercular scales; e, Coenenchymal sclerites; f-h, Body scales. b-h, SEM 2354. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON RS ry ’ _- = x @ FS 4 + ae e, 7 ‘ we e a | * é ‘ , 2 eo 4 3% * * ; * o e ae PY ¥ * “4 * a? * i 7 +t * _ PA ~ . os * sa* = wm TF " large gland cell (scale bar = 25 um); 12, surface view of spermatangial sorus (scale bar = 10 um); 13, transection through a tetrasporangial thallus showing cruciate tetraspores in outer cortex (scale bar = 15 um); 14, surface view showing spermatia in periclinal rows above the spermatangial parent cell (scale bar = 5 um). Holotype. —DML-746 (Alg. Coll. US- 162777), including liquid-preserved speci- men and accompanying microscope slides; leg. C. Forsyth, B. L. Brooks, M. M. Littler & D.S. Littler, 20 Jul 1986 (Fig. 3). Type locality. — Attached to rock, ca. 8 m depth, Diamond Rock (14°26'60’N, 61°02'50”W), Martinique, French West In- dies, Lesser Antilles. Paratypes. — All from the type locality, at- tached to basalt rock, 12-15 m depth, Di- amond Rock, DML-17092 (Alg. Coll. US- 162662), 21 May 1989, leg. D. S. Littler, M. M. Littler, B. L. Brooks & S. A. Reed, 21 May 1989; and, DML-17223 @, 6, & ® (Alg. Coll. US-162775; MSM; MICH; SNU), leg. B. L. Brooks, S. A. Reed, D. S. Littler & M. M. Littler, 23 May 1989. Etymology. —Chrysymenia littleriana is named for our colleagues, Diane S. Littler and Mark M. Littler (National Museum of Natural History, Smithsonian Institution) in recognition of their phycological contri- butions and having collected this distinctive alga; in choosing the Latin adjectival,-iana, the specific name translates as “‘the Littler- ian Chrysymenia.” Chrysymenia nodulosa J. Norris & Ballantine, sp. nov. Figs. 2a, b, 4, 5, 7, 8a, b, 9 Latin description.—Thalli erecti, usque ad 30 cm alti, cinerascentes gelatinosi, stip- itibus brevibus, axibus ramisque juventute fere omino teretibus, usque ad 6 mm diam. leviter nodulosis aetate compressescentibus usque ad 12 mm latis, magis nodulosis, ramificatio oppositis aut alternatis, pler- umque plano singulari, ramosis lateralibus pinnatis usque ad 2 ordines. Axes et rami cavatae, medulla incolorata, 160 cellularum plerumque 3 stratis composita, cellulis intimis maximis, cellulis extrinsecus decrescentibus. Cellulae medullosae inti- mae irregulariter rectangulares, usque ad 300 um X 600 um. Fila rhizoidalia medullosa plerumque in partibus veteribus, usque ad 22 um diam., basaliter tumidis ex cellulis medullosis intimis, plerumque cellulis med- ullosis aliis, saepe distaliter minute papil- losis. Cellulae glandulosae abundantes ex intimis cellulis medullosis in cavitatem cen- tralem eminentibus, pyriformae, usque ad 35 um longae X 32 um diam., atrocoloratae, usque ad 25 in quoque cellula. Cortex bi- vel tristromaticus, cellulis parvis; pilis in partibus veteribus corticis adundantibus su- perficie in sicco corno. Tetrasporangia cruciata, elliptica, 28-36 um longa < 14-18 wm diam.; in strato cor- ticali parce dispersa. Spermatangia in soros superficiales dispositis. Spermatangia ro- tunda vel ovalia 2.2—3.6 wm lata in catenas anticlinales (1—)2—4(—5) per cellulas paren- tales spermatangiorum efferentia. Cystocar- pla prominentia hemirotunda, usque ad ca. 1 mm diam. ad superficiem thalli. Description. —Thalli erect, up to 30 cm tall above a short stipe, grayish, gelatinous, axes and branches when young almost en- tirely terete, to 6 mm diameter, and slightly nodulose, becoming compressed, to 12 mm broad, and more nodulose with age and size; branching mostly in a single plane with op- posite or alternate, pinnately branched lat- erals to 2 orders. Axes and branches internally hollow with a medulla generally of 3 layers of colorless cells which are largest internally and be- come smaller towards the surface. Inner- most medullary cells are irregularly rect- angular, up to 300 wm by 600 um (Fig. 7). Medullary filaments mostly in older por- tions, to 22 wm in diam. (Figs. 8a, b), in- flated at their origin from the innermost medullary cells (Fig. 8a) or occasionally from other medullary cells, and often with small papilla-like projections near distal ends (Fig. 8b). Gland cells (Figs. 7, 8a) are abundant PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON on innermost medullary cells facing into the central cavity, pyriform, to 35 wm long by 32 wm diam., darkly staining, and up to 25 per cell. Cortex composed of 2 or 3 layers of small cells (Fig. 7); hairs abundant in cor- tex of older portions of thallus, giving the surface a horny texture when dried. Tetrasporangia cruciate, oval, 28-36 (—38) um long by 14—18 (—20) um wide; sparingly scattered in the cortical layer (Fig. 7). Sper- matangia in sori on thallus surface. Sper- matangial parent cell produces an anticlinal chain of (1-)2—4(—5) spermatangia, spheri- cal to oval, 2.2—3.6 wm (Fig. 9). Cystocarps protruding, hemispherical, to about 1 mm diam. in surface view (Fig. 5). Holotype. —DLB-3108, cystocarpic (Alg. Coll. US-162770), leg. D. L. Ballantine, 5 May 1988 (Fig. 4). Type locality. —17 m depth, 1.5 km sea- ward of Media Luna Reef, La Parguera, Puerto Rico, Greater Antilles. Tsotypes. —DLB-3108, tetrasporangial (Alg. Coll. US-162772; MSM); and DLB- 3108, spermatangial (Alg. Coll. US-162771; MSM; and MICH). Paratypes.— All from Media Luna Reef, La Parguera, leg. D. L. Ballantine: DLB- 1837 (MSM), 24 Jan 1985; DLB-2128 (Alg. Coll. US-014407), 2 Nov 1985; DLB-3340 (MSM), 21 Feb 1989; and DLB-3375 (MSM), 26 Apr 1989. Etymology.—The specific epithet nodu- losa refers to the small, knobby swellings on the thallus surface, a characteristic unique to this new species of Chrysymenia. It is the diminutive of Latin nodosus (Stearn 1973), and means “‘full of little knobs.” Key to the Western Atlantic species of Chrysymenia i hallus fate 27 ee yD 1. Thallus terete to compressed ..... 4 2. Branches of the thallus broad (> 2.0 Cmi)fOlOSe Sb .s a. ees fee 3 2. Branches of the thallus narrow (<2.0 cm), irregularly branched to 4(—S) orders C. dickieana VOLUME 108, NUMBER 1 3. Thallus dichotomous to palmately branched, 2.0 to 2.5 cm broad ... 1 <0 6 6 GSAS RencentemmnnmedicertE C. agardhii 3. Thallus branching irregularly lobed, 225 Ci re C. planifrons 4. Thallus terete, occasionally com- pressed at and above the base .... 5 4. Thallus slightly to moderately com- pressed throughout 5. Branches highly constricted at the lDASS 6.50 C. enteromorpha 5. Branches not or barely constricted Att Une ako yr C. ventricosa 6. Branching dichotomous joe 008 oe C. halymenioides 6. Branching pinnate 7. Thalli with nodulose swellings (more pronounced in older and/or larger (inglli) oe. C. nodulosa 7. Thalli without nodulose swellings 506050 CRS te ae C. littleriana o © © © © © © © Discussion The genus Chrysymenia has close affini- ties to Botryocladia (J. Agardh) Kylin (1931: 17). These genera are separated by the veg- etative characters of solid axes and dia- phragms in Botryocladia and hollow axes without diaphragms in Chrysymenia. Bro- die & Guiry (1988) pointed out that the degree of stipe development in Botryocladia is variable. They speculated that the genus might be an artificial grouping of species, but they retained it with reservations. In our specimens, we observed the short, stiptate regions of the new species to be filled with rhizoidal filaments, in some places very densely. This suggests that the nature and origin of the cells in the stipe and axes may be a useful taxonomic character, i.e., the solid, parenchymatous medulla of the axes in Botryocladia vs. the rhizoidal filament- filled cavity of the stipe in Chrysymenia. Chrysymenia littleriana and C. nodulosa have some features in common, including the short stipe above a discoidal holdfast, compressed axes, and pinnate branching 161 (Figs. 1-4). These new species are also sim- ilar anatomically (Figs. 7, 11), with large medullary cells of C. /ittleriana generally in two layers and those of C. nodulosa mostly in three layers. Both species have two or three layers of cortical cells, although the outermost cortical cells in C. /ittleriana are elongate (Fig. 13). They both produce non- aggregated gland cells from inner medullary cells (Figs. 7, 8, 11), although they are much more abundantly produced in C. nodulosa. In C. nodulosa, the medullary filaments fre- quently possess very minute, papilla-like projections near their distal ends (Fig. 8b) and the inner medullary cells give rise to internal rhizoids that are inflated at their origin (Fig. 8a). Sometimes the internal rhi- zoids of C. littleriana are also similarly in- flated. Transections of both C. Jittleriana and C. nodulosa through the terete, stipitate basal region reveal cellular medullary and cortical regions similar to the distal portions of inflated thallus branches; however, the central medulla of this region is densely filled with branching, rhizoidal filaments. Reproductively the cruciately divided te- trasporangia are oval and cut off by an inner cell of the cortical layer in both new species (Figs. 7, 13); however, they are larger in C. nodulosa (28-36 um by 14-18 wm) than in C. littleriana (25 um by 16 wm). Female C. nodulosa and C. littleriana are readily rec- ognizable by their hemispherical and dome- shaped cystocarps that project well above the thallus surface (Figs. 5, 6). Lee (1969:figs. 1A, B, 2A, B) recognized two modes of spermatangial development in the Rhodymeniales: (1) the “separate type,’ where the parent cells are separate from each other, the parent cell and sper- matangia are surrounded by a common ge- latinous wall, and the spermatia produced are comparatively large; and (2) the “‘seriate type,” in which the parent cells originate from the same cortical cell in seriate rows and are pit connected to each other, the spermatangia and the parent cell each have an independent cell wall, and the spermatia 162 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Comparison of the new species of Chrysymenia with: A.) the morphologically similar species of form-group 1; and B.) the other known species of the tropical Western Atlantic. Branch Degree of Branching constric- No. cortical Habitat flattening pattern tions ~cell layers A. Form-group 1 C. enteromorpha (1, 2, 3) Deep, 15-90 m Terete Radial, common- A 1 ly whorled from distal end of axial segments C. grandis (4) Deep, 30 m Subcylindrical Simple or irregu- No to slightly larly branched compressed C. halymenioides (2, 5) Deep, 43 m Sub-terete to Dichotomous Slight at Several compressed most C. littleriana (3) 5-15 m Compressed Oppositely Slight at 3 pinnate most C. nodulosa (3) Deep, 17-24 m Compressed Alternate to op- _— Slight 2 to 3 positely pinnate C. ventricosa (Med.) (6) Shallow to 130 m_ Terete Alternate to op- No Several positely pinnate C. ventricosa (Carib.) (1, | Deep, to 90 m Compressed be- Irregularly alter- Slight Several 2) low, terete nate to opposite above C. wrightii (7, 8) Shallow Terete Radial, alternate, + 1 to 3 opposite or ir- regular B. Other tropical Western Atlantic species C. agardhii (1, 2) Shallow to deep, Flat Dichotomous to No 1 to 2 to 29 m palmately lobed C. planifrons (1, 2) Deep, 30 m Flat Irregularly lobed No 1 1) Borgesen 1920, 2) Taylor 1960, 3) This study, 4) Okamura 1933, 5) Harvey 1853, 6) Kuckuck 1912, 7) Lee 1978, 8) Ben Maiz et al. 1987. VOLUME 108, NUMBER 1 163 Table 1.—Extended. Pres- ence of med- No. ul- medullary lary cell Hol- layers Gland cells low ments Distribution Tetrasporangia Cystocarps Spermatangia 1 Scattered or =F Bermuda, Elliptical, Projecting small groups North Caroli- scattered, na, Florida, 28 x 39 um U:S.V.I., Bra- zil, P.R. + Be Japan Elliptical, 10— Small, roundish, 15 wm not prominent 4to5 =F + Bermuda, Flori- Spherical to Rounded, coni- da, Jamaica, elliptical cal Netherlands Antilles 2 to 3 Occasional, st Martinique, Elliptical, Conical, project- Average 2.4 um mostly soli- Guadeloupe scattered, ing to 1.2 mm diam., sperma- tary, on (12.5-)16 x diam. tangia in peri- medullary 25 wm clinal seriate se- cells ries 3 Abundant on + Puerto Rico Oval, scat- Hemispherical, Average 2.7 um medullary tered, 28-38 projecting to 1 diam., sperma- and subme- x 14-20 mm diam. tangia in anticli- dullary cells pm nal seriate series 2. to 3 Numerous, + Morocco, Medi- Elliptical, scattered terranean scattered, 20-26 x 15-19 wm 2 to 3 Occasional, + U.S.V.IL., Ber- Elliptical, Hemispherical, solitary; ob- muda, Jamai- scattered, projecting long ca, Venezuela 20 wm 3 to 5 Aggregated or + Japan, Korea, Spherical to Projecting, 850— 2.9 x 4 um, sper- solitary Mediterra- elliptical, 950 wm diam. matangia in anti- nean 38-42 x clinal seriate se- 48-53 um ries y, Occasional, + North Carolina, Spherical, 27 scattered Florida, Ber- wm muda, US.V.I. ito 2 Occasional, + Florida, Puerto scattered Rico, U.S.V.L., Netherlands Antilles 164 are comparatively small. The nature and development of spermatangia in Chrysy- menia have been little studied and are known for only three species (Table 1). Spermatangial production for both new spe- cies is the seriate type (Figs. 9, 10), similar to that reported by Lee (1978) for C. wrigh- tii, with the outer cortical cells cutting off up to 4 spermatangial parent cells. In C. nodulosa, the spermatangia are cut off se- rially from spermatangial parent cells in an- ticlinal rows (Fig. 9). In contrast, sperma- tangia are cut off periclinally in C. /ittleriana (Figs. 10, 14). As spermatangial thalli are discovered and described for other species of Chrysymenia, the different kinds (sepa- rate vs. seriate, and periclinal vs. anticlinal; Table 1) should be correlated to see if these characters have any taxonomic significance, perhaps for use in subgeneric groupings as has been proposed for Gracilaria by Ya- mamoto (1978). Chrysymenia nodulosa and C. littleriana are unique among Caribbean species of the genus (Table 1), and distinct from all the other species of morphological form-group 1 (as noted in the Introduction). Both new species lack the totally terete thallus with the highly constricted branches as seen in C. enteromorpha (Borgesen 1920). They are also easily separated from C. halymen- loides, which is dichotomously branched. Chrysymenia nodulosa and C. littleriana could be confused with C. ventricosa sensu Taylor (1960:460, see e.g., pl. 62:fig. 3). However, on close examination, the two new species are clearly different from it in several respects. Chrysymenia nodulosa differs from all known species of Chrysymenia (Table I) in having abundant, nodulose projections irregularly scattered across its thallus sur- face (Figs. 2a, b, 4). It also possesses more elongate and larger tetrasporangia, as well as having branches that are more constrict- ed at their origin than in specimens of C. ventricosa (MICH; US). Chrysymenia lit- tleriana branches to 3—4(—5) orders vs. 1-3 orders for C. ventricosa. The branches of C. littleriana are considerably more closely set PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON and have broadly obtuse branch apices, and the tetrasporangia are smaller and more elongate than those of C. ventricosa. Fur- thermore, the distal portions of the thalli are compressed in C. /ittleriana and C. no- dulosa, whereas they are terete in C. ven- tricosa. Of the non-western Atlantic species, Chrysymenia wrightii superficially resem- bles C. nodulosa; however, C. wrightii dif- fers in possessing terete axes and up to five layers of medullary cells (Lee 1978, Ben Maiz et al. 1987). These characters also serve to differentiate C. Jittleriana from C. wrigh- Bulbs Acknowledgments We thank S. Fredericq for the Latin de- scriptions and the photomicrographs of spermatangial preparations, D. S. Littler for the color photo of Chrysymenia littleriana, D. H. Nicolson for nomenclatural discus- sions, R. H. Sims for the type specimen pho- tos, M. J. Wynne (MICH) for a loan of spec- imens of C. ventricosa, and W. Fenical (Chief Scientist) for JN’s collaboration on the 1985 and 1986 Lesser Antilles expeditions of OR/V Cape Florida (University of Miam1) supported by National Science Foundation #CHE-86-20217 and in part by the Smith- sonian Institution’s Caribbean Coral Reef Ecosystem Program (CCRE contribution number 433), and the Smithsonian Marine Station at Link Port, Fort Pierce, Florida (SMSLP contribution number 366). For re- viewing the manuscript we appreciate the comments of N. Aponte, K. E. Bucher, and D. B. Lellinger. Literature Cited Abbott, I. A., & M. M. Littler. 1969. Some Rhody- meniales from Hawaii. — Phycologia 8:165-169. Agardh, J.G. 1842. Algae maris mediterranei et ad- riatici, observationes in diagnosin specierum et dispositioneum generum. Fortin, Masson, Par- is, xX + 164 pp. . 1851. Species genera et ordines algarum, seu descriptiones succinctae specierum, generum et ordinum, quibus algarum regnum constituitur. VOLUME 108, NUMBER 1 Species genera et ordines floridearum ... Vol. II, pt. 1. C. W. K. Gleerup, Lund, iii + xii + 1-351 pp. . 1876. Species genera et ordines algarum, seu descriptiones succinctae specierum, generum et ordinum, quibus algarum regnum constituitur. Epicrisis systematis floridearum. Vol. III, pt. 1. T. O. Weigel, Leipzig, 111 + viii + 1-724 pp. . 1892. Analecta algologica. Observationes de speciebus algarum minus cognitis earumque dispositione.— Lunds Universitets Arsskrift 28(6):1-182, pls. 1-3. Ben Maiz, N., C.-F. Boudouresque, R. Riouall, & M. Lauret. 1987. Flore algale de l’etang de Thau (France, Méditerranée): Sur la présence d’une Rhodymeniale d’origine japonaise, Chrysymen- ia wrightii (Rhodophyta). — Botanica Marina 30: 357-364. Borgesen, F. 1920. The marine algae of the Danish West Indies. Pt. III: Rhodophyceae (6), with addenda to the Chlorophyceae, Phaeophyceae and Rhodophyceae.— Dansk Botaniska Arkiv 3:369-504. Brodie, J., & M. D. Guiry. 1988. Life history and reproduction of Botryocladia ardreana sp. nov. (Rhodymeniales, Rhodophyta) from Portu- gal.— Phycologia 27:109-130. Cribb, A. B. 1983. Marine algae of the southern Great Barrier Reef—Rhodophyta. Australian Coral Reef Society, Handbook #2. W. Ferguson, Bris- bane, 175 pp., 71 pls. Fredericq, S., E. Serrao, & J. N. Norris. 1992. New records of marine red algae from the Azores. — Arquipélago, Life & Earth Sciences 10:1—-4. Grunow, A. 1889. Algae. Pp. 1-5 inK. M. Schumann & U. M. Hollrung, Der Flora von Kaiser Wil- helms Land. Beiheft zu den Nachtrichten Uber Kaiser Wilhelms Land und Bismarck Archipel. Asher, Berlin, 140 pp. Harvey, W. H. 1853. Nereis Boreali-Americana... Pt. II: Rhodospermae. [First Issue]. Smithson- ian Institution, Washington, D.C. & John Van Voorst, London, ii + 258 pp., pls. 13-36. [Third Issue, 1858 (‘1853’).—Smithsonian Contribu- tions to Knowledge 5(5):11 + 258 pp., pls. 13- 36.] Holmgren, P. K., N. H. Holmgren, & L. C. Barnett. 1990. Index Herbariorum. Part I: the Herbaria of the World. 8th Edition. New York Botanical Garden, Bronx, N.Y., 693 pp. (Regnum Veget- abile vol. 120). Howe, M.A. 1914. The marine algae of Peru.—Bul- letin of the Torrey Botanical Club 38:489-5 14. Kylin, H. 1931. Die Florideenordnung Rhodyme- niales.— Lunds Universitets Arsskrift, Ny Foeljd, Avdeling 2, 27(11):3—48, pls. 1-20. . 1956. Die Gattungen der Rhodophyceen. C. W. K. Gleerups, Lund, 673 pp. 165 Lee, I. K. 1969. On the male organs of Rhodyme- niales.— Korean Journal of Botany 12:143-150. . 1978. Studies on Rhodymeniales from Hok- kaido.—Journal of the Faculty of Science, Hok- kaido University, Series V (Botany), 11:1-194, 5 pls. Levring, T. 1974. The marine algae of the Archipel- ago of Madeira.— Boletim do Museu Municipal do Funchal 28(125):1-111. Norris, J. N., & K. E. Bucher. 1982. Marine algae of Belize. Pp. 167-223 in K. Ritzler & I. G. Mac- intyre, eds., The Atlantic Barrier Reef Ecosys- tem at Carrie Bow Cay, Belize, I: structure and communities.—Smithsonian Contributions to the Marine Sciences 12:1-539. Okamura, K. 1930. On the algae from the island Hatidyo.— Records of Oceanographic Works of Japan, Tokyo, 2:92-110, pls. 6-10. . 1933. Icones of Japanese Algae, Vol. VII, no. 1. Pp. 1-7 [English text] +1-7 [Japanese text], pls. 301-305. Kazamashobo, Tokyo. 1936. Nippon Kaiso Shi [Marine Algae of Japan]. Published by the Author, Tokyo, 12 + 6 + 964 + 11 pp. [Reissued in 1956 by Uchida Rokakuho Publishing House, Tokyo.] Stearn, W. T. 1973. Botanical Latin. 2nd Ed. David & Charles Ltd., Newton Abbot, xiv + 566 pp. Taylor, W. R. 1960. Marine algae of eastern tropical and subtropical coasts of the Americas. Uni- versity of Michigan Press, Ann Arbor, 870 pp. Tsuda, R. T., & I. A. Abbott. 1985. Collection, han- dling, preservation, and logistics. Pp. 67-86 in M. M. Littler & D. S. Littler, eds., Handbook of phycological methods, Vol. 4: ecological field methods: macroalgae. Cambridge University Press, Cambridge. Weber van-Bosse, A. 1928. Liste des algues du Si- boga, IV. Rhodophyceae: Gigartinales et Rho- dymeniales. Pp. 393-533, pls. 11-16 in M. We- ber, ed., Siboga Expeditie, Monographie #59d. E. J. Brill, Leiden. Wynne, M. J. 1986. A checklist of benthic marine algae of the tropical and subtropical western At- lantic.—Canadian Journal of Botany 64:2239- 2281. Yamada, Y. 1932. Notes on some Japanese algae, IlI.—Journal of the Faculty of Science, Hok- kaido Imperial University, Series V, 1(3):109- 123, pls. 21-25. —, & S. Segawa. 1953. On some new or note- worthy algae from Hachijo Island.— Records of Oceanographic Works, Japan, N.S., 1:100-114. Yamamoto, H. 1978. Systematic and anatomical study of the genus Gracilaria in Japan.—Mem- oirs of the Faculty of Fisheries, Hokkaido Uni- versity, 25(2):97-152, pls. 1-49. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(1):166. 1995. Applications published in the Bulletin of Zoological Nomenclature The following Applications were published on 20 December 1994 in Vol. 51, Part 4 of the Bulletin of Zoological Nomenclature. Comment or advice on these Appli- cations is invited for publication in the Bulletin and should be sent to the Executive Secretary, I.C.Z.N., % The Natural History Museum, Cromwell Road, London SW7 SBD, U.K. Case No. 2888 Valdivianemertes Stiasny-Wijnhoff, 1923 (Nemertea): proposed conserva- tion. 2908 Vejdovskyella Michaelsen, 1903 (Annelida, Oligochaeta): proposed prece- dence over Macrochaetina Bretscher, 1899. 2896 Scottia Brady & Norman, 1889 (Crustacea, Ostracoda): proposed designation of Scottia pseudobrowniana Kempf, 1971 as the type species. 2893 Temnorhynchus Hope, 1837 (Insecta, Coleoptera): proposed conservation 2865 BRACHYPTERINAE Erichson, 1845 (Insecta, Coleoptera) and BRACHYPTERINAE Zwick, 1973 (Insecta, Plecoptera): proposed removal of homonymy. Coproica Rondani, 1861 and Ischiolepta Lioy, 1864 (Insecta, Diptera): pro- posed conservation of usage by the designation of Limosina acutangula Zetterstedt, 1847 as the type species of Coproica Sphaerocera Latreille, 1804 and Borophaga Enderlein, 1924 (Insecta, Dip- tera): proposed conservation; Sphaerocera curvipes Latreille, 1805 and Phora flavimana Meigen, 1830: proposed conservation of the specific names. Scomber dentex Bloch & Schneider, 1801 (currently Caranx or Pseudocaranx dentex) and Caranx lugubris Poey, [1860] (Osteichthyes, Perciformes): proposed conservation of the specific names. Bagrus hoevenii Bleeker, 1846 (currently Hemibagrus hoevenii; Osteichthyes, Siluriformes): proposed designation of a neotype. Lycognathophis Boulenger, 1893 (Reptilia, Serpentes): proposed conserva- tion. Loris E. Geoffroy Saint-Hilaire, 1796 (Mammalia, Primates): proposed con- servation. VOLUME 108, NUMBER 1 167 The 122nd Annual Meeting of the Biological Society of Washington will be held on Friday, 12 May 1995, at 12:00 noon in the Waldo Schmitt Room, National Museum of Natural History, Washington, D.C. INFORMATION FOR CONTRIBUTORS Content.—The Proceedings of the Biological Society of Washington contains papers bearing on systematics in the biological sciences (botany, zoology, and paleontology), and notices of business transacted at meetings of the Society. Except at the direction of the Council, only man- uscripts by Society members will be accepted. Papers are published in English (except for Latin diagnoses/descriptions of plant taxa), with an Abstract in an alternate language when appropri- ate. Submission of manuscripts.—Submit three copies of each manuscript in the style of the Proceedings to the Editor, complete with tables, figure captions, and figures (plus originals of the illustrations). Mail directly to: Editor, Proceedings of the Biological Society of Washing- ton, National Museum of Natural History NHB-108, Smithsonian Institution, Washington, D.C. 20560. (Do not submit manuscripts to an associate editor). Presentation.—Clarity of presentation, and requirements of taxonomic and nomenclatural procedures necessitate reasonable consistency in the organization of papers. Telegraphic style is recommended for descriptions and diagnoses. The style for the Proceedings is described in “GUIDELINES FOR MANUSCRIPTS for Publications of the BIOLOGICAL SOCIETY OF WASHINGTON” a supplement to Volume 103, number 1, March 1990. Authors are encouraged to consult this article before manuscript preparation. Copies of the article are available from the editor or any associate editor. The establishment of new taxa must conform with the requirements of appropriate interna- tional codes of nomenclature. Decisions of the editor about style also are guided by the General Recommendations (Appendix E) of the International Code of Zoological Nomenclature. When appropriate, accounts of new taxa must cite a type specimen deposited in an institutional col- lection. Review.—One of the Society’s aims is to give its members an opportunity for prompt publi- cation of their shorter contributions. Manuscripts are reviewed by a board of Associate Editors and appropriate referees. Proofs.—Authors will receive first proofs and original manuscript for correction and ap- proval. Both must be returned within 48 hours to the Editor. Reprint orders are taken with re- turned proofs. Publication charges.—Authors are required to pay full costs of figures, tables, changes in proofs ($3.00 per change or revision), and reprints. Authors are also asked to assume costs of page-charges. The Society, on request, will subsidize a limited number of contributions per vol- ume. If subsidized manuscripts result in more than 12 printed pages, the additional pages must be financed by the author(s). Multiple authorship will not alter the 12 page limit (each author will be viewed as having used his/her 12 subsidized pages). Payment of full costs will facilitate speedy publication. Costs.—Printed pages @ $60.00, figures @ $10.00, tabular material @ $3.00 per printed inch per column. One ms. page = approximately 0.4 printed page. CONTENTS The karyotypes of two South American mouse opossums of the genus Thylamys (Marsupialia: Didelphidae), from the Andes, and eastern Paraguay R. Eduardo Palma Rediscovery and redescription of Sphenomorphus beyeri Taylor (Reptilia: Lacertilia: Scinci- dae) from the Zambales Mountains of Luzon, Philippines | Rafe M. Brown, John W. Ferner, and Rogelio V. Sison A new species of lygosomine lizard (Reptilia: Lacertilia: Scincidae; Sphenomorphus) from Mt. Isarog, Luzon Island, Philippines | Rafe M. Brown, John W. Ferner, and Luis A. Ruedas Revision of the South American freshwater fish genus Sternarchorhamphus Eigenmann, 1905 (Ostariophysi: Gymnotiformes: Apteronotidae), with notes on its relationships Ricardo Campos-da-Paz A new species of Neoperla (Insecta: Pleocoptera: Perlidae) from Mississippi Bill P. Stark Macrobrachium catonium, a new troglobitic shrimp from the Cayo District of Belize (Crus- tacea: Decapoda: Palaemonidae) H. H. Hobbs III and Horton H. Hobbs, Jr. Procambarus (Ortmannicus) nueces (Decapoda: Cambaridae), a new crayfish from the Nueces River Basin, Texas Horton H. Hobbs, Jr. and H. H. Hobbs III On a new Somanniathelphusa Bott, 1968, from Vietnam (Crustacea: Decapoda: Brachyura: Parathelphusidae) Peter K. L. Ng and Takeharu Kosuge A new species of Goreopaguras McLaughlin (Decapoda: Anomura: Paguridae) from the Pa- cific, and a comparison with its Atlantic counterpart Patsy A. McLaughlin and Janet Haig Epilobocera wetherbeei, anew species of freshwater crab (Decapoda: Brachyura: Pseudothel- phusidae) from Hispaniola Gilberto Rodriguez and Austin B. Williams Alpheus angulatus, a new species of snapping shrimp from the Gulf of Mexico and northwestern Atlantic, with a redescription of A. heterochaelis Say, 1818 (Decapoda: Caridea: Alpheidae) Matthew R. McClure A new species of freshwater crab of the genus Strengeriana from Colombia (Crustacea: Decapoda: Pseudothelphusidae) Martha R. Campos A new callianassid (Decapoda: Thalassinidea) from the southern Caribbean Sea Juan Pablo Blanco Rambla, Ildefonso Lifiero Arana, and Luis Beltran Lares M. Heteranthessius hoi, a new species (Copepoda: Pseudanthessiidae) from a sea-anemone in the Straits of Gibraltar, with remarks on the genus Pablo J. Lopez-Gonzalez and Mercedes Conradi Description of Amphiascoides atopus, a new species (Crustacea: Copepoda: Harpacticoida) from a mass culture system Guilherme R. Lotufo and John W. Fleeger Hydroids colonizing the carapaces of the ostracode Philomedes brenda from the Beaufort Sea, Arctic Ocean Louis S. Kornicker and Dale R. Calder Laonome albicingillum, a new fan worm species (Polychaeta: Sabellidae: Sabellinae) from Taiwan Hwey-Lian Hsieh A juvenile of the scaled squid, Pholidoteuthis adami Voss, 1956 (Cephalopoda: Oegopsida), from the Florida Keys David A. Goldman A new species of the gorgonacean genus Narella (Anthozoa: Octocorallia) from Hawaiian waters Frederick M. Bayer Two new species of the red alga Chrysymenia J. Agardh (Rhodymeniales: Rhodymeniaceae) from the tropical western Atlantic James N. Norris and David L. Ballantine _ International Commission of Zoological Nomenclature Notice 18 29 45 50 54 61 68 76 84 98 102 107 iy) W2S 130 136 147 5s 166 167 OLOGICAL SOCIETY WASHINGTON VOLUME 108 NUMBER 2 22 JUNE 1995 ISSN 0006-324X THE BIOLOGICAL SOCIETY OF WASHINGTON 1994-1995 Officers President: Janet W. Reid Secretary: Carole C. Baldwin President-elect: Stephen D. Cairns Treasurer: YT. Chad Walter Elected Council Robert J. Emry Susan L. Jewett Richard C. Froeschner Lynne R. Parenti Alfred L. Gardner F. Christian Thompson Custodian of Publications: Austin B. Williams PROCEEDINGS Editor: C. Brian Robbins Associate Editors Classical Languages: George C. Steyskal Invertebrates: Jon L. Norenburg Frank D. Ferrari Plants: David B. Lellinger Rafael Lemaitre Insects: Wayne N. Mathis Vertebrates: Thomas A. Munroe Membership in the Society is open to anyone who wishes to join. There are no prerequisites. Annual dues of $25.00 (for USA and non-USA addresses) include subscription to the Pro- ceedings of the Biological Society of Washington. Library subscriptions to the Proceedings are: $40.00 for USA and non-USA addresses. Non-USA members or subscribers may pay an ad- ditional $25.00 to receive the Proceedings by Air Mail. The Proceedings of the Biological Society of Washington (USPS 404-750) is issued quarterly. Back issues of the Proceedings and the Bulletin of the Biological Society of Washington (issued sporadically) are available. Correspondence dealing with membership and subscriptions should be sent to the Biological Society of Washington, P.O. Box 1897, Lawrence, Kansas 66044, U.S.A. Payment for membership is accepted in US dollars (cash or postal money order), checks on US banks, or MASTERCARD or VISA credit cards. Manuscripts, corrected proofs, editorial questions should be sent to the Editor, Biological Society of Washington, National Museum of Natural History, Smithsonian Institution, Wash- ington, D.C. 20560. Known office of publication: National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. Printed for the Society by Allen Press, Inc., Lawrence, Kansas 66044 Second class postage paid at Washington, D.C., and additional mailing office. POSTMASTER: Send address changes to PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON, P.O. Box 1897, Lawrence, Kansas 66044. This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(2):169-179. 1995. A new species of Raricirrus (Polychaeta: Ctenodrilidae) from wood collected in the Tongue of the Ocean, Virgin Islands Harlan K. Dean Department of Invertebrates, Museum of Comparative Zoology, 26 Oxford St., Cambridge, Massachusetts 02138, U.S.A. and John Hazen White School of Arts and Sciences, Johnson and Wales University, 6 Abbott Park Place, Providence, Rhode Island 02903, U.S.A. Abstract.— Raricirrus variabilis new species (Polychaeta: Ctenodrilidae) is described from the Tongue of the Ocean, St. Croix, Virgin Islands. This deep- sea species was collected from submerged wood at 4000 m and is an apparent organic-enrichment opportunist. Taxonomic differences between the genera Raricirrus and Raphidrilus are discussed. The life history characteristics of this species are examined. Woody plant material is rapidly broken down in the deep sea by bivalves belonging to the family Pholadidae (Mollusca), sub- family Xylophagainae. The activities of these pholads have been shown to provide a highly concentrated, tractable source of organic material that supports a community of associated organisms (Turner 1973, 1977, 1981). Many of these associated species sur- vive in the deep sea by specifically finding and exploiting organically enriched sites (Grassle & Morse-Porteous 1987, Smith & Hessler 1987, Desbruyéres & Laubier 1988). As part of a long-term study of such wood- associated communities by Dr. Ruth Tur- ner, specimens of an undescribed poly- chaete belonging to the family Ctenodrili- dae were recovered from pholad-riddled wood panels and stray wood (“wild wood’’) retrieved from the deep-sea floor in the Tongue of the Ocean. This paper describes Raricirrus variabilis, new species, and fur- ther elucidates the separation of two cten- odrilid genera Raricirris and Raphidrilus. Raricirrus variabilis, new species Material examined.—St. Croix, Virgin Islands: 17°56.63’N, 64°48.6'W, 4000 m. Holotype (MCZ 4008) from wood panel P-13; submerged 17 Dec 1978 (Alvin Dive 873), recovered 6 Dec 1980 (Alvin Dive 1079). Paratypes (MCZ 4009) 12 specimens from same panel and dates. Paratype (MCZ 4011) one specimen from wood panel P-2; submerged 17 Dec 1978 (Alvin Dive 873), recovered 13 Dec 1980 (Alvin Dive 1082). Paratypes (MCZ 4010) 6 specimens from wood panel P-12; submerged 17 Dec 1978 (Alvin Dive 873), recovered 13 Dec 1980 (Alvin Dive 1082). Paratypes (USNM 170552) 8 specimens from wood panel P-5 & P-13 wash material; submerged 17 Dec 1978 (Alvin Dive 873), recovered 6 Dec 1980 (Alvin Dive 1079). Paratypes (USNM 170553) 4 epitokous specimens from 6 Ft. plank, wild wood; recovered 20 Dec 1978 (Alvin Dive 876). Description. —Benthic form: Body elon- gate with widened posterior region; anterior and middle body segments longer than wide, posterior segments much wider than long (Fig. 1). Holotype 6.75 mm long, 0.43 mm maximum width (modified segments) with 27 setigers (Table 1, specimen S); paratypes 1.54-7.59 mm long, 0.25—0.61 mm wide, with 18-31 setigers (Table 1). Prostomium conical, eyespots lacking; nuchal organs dorsolateral, appearing as slits or round de- pressions with what appear to be cilia with- in. Prostomium with lateral oral folds; pro- 170 Figs. 1-2. Raricirrus variabilis, new species. 1. Dor- sal view of holotype (specimen S). B = branchiae, HB = heart body, MS = modified segments, N = nuchal organ, O = oocytes, OF = oral fold, S = enlarged spine, ST = reproductive stylet, SV = seminal vesicle. Scale bar = 1.0 mm. 2. Distal end of branchia (specimen MCZ 4011). Scale bar = 100 um. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON stomium, peristomium and first setiger con- tinuous dorsally, differentiated ventrally. Body of larger specimens brownish gray, prostomium brown posterodorsally, first two setigers and posterior body region gold brown across dorsum; smaller specimens uniformly pale tan to cream color. Branchial filaments greatly elongate (ap- proximately 60% of total body length) with club-like ends (Fig. 2), emerging postero- dorsal to notopodia. Branchial filaments fragile and easily lost or broken (incomplete branchiae on setigers 5, 7 and 8 in holotype), branchial scars on other segments not dis- cernible; anteriormost occurrence observed at setiger 2, posteriormost on last setiger anterior to the modified segments of an ep- itokous individual. Heart body a reddish brown convoluted tube (Fig. 1) extending for a variable (range shown in parentheses) number of segments (4-11) beginning on setiger 4 (4-6). Notosetae and neurosetae emerging di- rectly from body wall, lateral in anterior region, becoming ventrolateral elsewhere. Notosetae of three kinds: four to six capil- laries finely serrate distally, usually one (1- 3) much longer (ca. 73 body width) than oth- ers; short pectinate setae with narrow, wide- ly spaced teeth (Fig. 3A) beginning in an- terior half of body, grading into coarsely pectinate falcigers in the midbody region (Fig. 3B), and coarsely serrate setae (Fig. 3C) posterior to the modified setigers. Neuro- setae three to six pectinate falcigers (Fig. 3D) increasing in length posteriorly; 1-2 straight, coarsely serrate setae (similar to those of posterior notopodia) in the setigers posterior to the modified setigers (Fig. 4). A pair of modified segments present as first two segments of broadened posterior region in all but several smaller (<2.73 mm long) individuals (Table 1). Modified seg- ments distinct dorsally (Fig. 6), with one pair of large, curved notopodial spines emerging anterolaterally from second mod- ified segment (holotype with only a single spine on right side), fitting into deep grooves on posterior ventral surface of first modified VOLUME 108, NUMBER 2 171 Fig. 3. Setae of Raricirrus variabilis, new species. A-C. Notosetae: A. Short pectinate falciger with the bases of several capillary setae from anterior region. B. Coarsely pectinate falciger from midbody region. C. Coarsely serrate falciger from region posterior to the modified segments. D. Pectinate neuropodial falcigers from midbody region. Scale bars = 10 um. 172 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Figs. 4-7. Raricirrus variabilis, new species. 4. Coarsely serrate neuropodial falciger from region posterior to the modified segments. Scale bar = 50 um. 5. Posterior region of epitokous individual (specimen W), ventral view. GP = genital pore, SO = spherical organ. Scaie bar = 500 um. 6. Modified segments of epitokous individual (specimen W), dorsal view. Scale bar = 500 um. 7. Reproductive stylet (specimen R). Scale bar = 50 um. segment (Fig. 5). Three specimens (includ- ing holotype) with a pair of enlarged, straight spines in notopodia of 1—2 setigers anterior to modified segments. Median pore ob- served on ventral surface of second modi- fied segment (Fig. 5, GP), associated with a medial spherical area within first modified segment or segment immediately anterior (Fig. 5, SO). Spherical structure appeared empty or sometimes contained diffuse, coarse-grained material when examined in squash preparation. Pygidium rounded with dorsoterminal anus. Oocytes observed scattered in several an- terior setigers of several benthic specimens (numerous 10—25 wm diameter oocytes in VOLUME 108, NUMBER 2 setigers 5—9 of holotype); these setigers often thin-walled and delicate. Straight cuticular stylet (Fig. 7) with narrow lumen present from five to nine setigers (five in holotype) anterior to modified segments in several specimens, associated with a large sac (sem- inal vesicle) of what were revealed to be spermatozoa in squash preparation (Fig. 1). Presence and development of modified seg- ments apparently unrelated to presence of either observable oocytes or a stylet and seminal vesicle. Epitokes. —Specimens 6.67—20.64 mm long, 0.73-2.07 mm wide, 28-35 setigers (Table 1). Midbody region brown-gray, clearly differentiated from light cream col- ored and much narrower posterior setigers. Capillary setae more numerous in notopo- dial bundles (S—16) and with a greater num- ber (2-9) of elongate capillaries (ca. #4 body width) than in benthic specimens. Noto- podial falcigers lacking; straight, coarsely serrate setae present posterior to the mod- ified segments. Neurosetae similar to those of benthic individuals but with a greater number of falcigers (maximum of 8-10). Heart body extending through five to eight setigers starting at setigers 8-10. The body cavity of two epitokous specimens (Table 1, specimens U and X) contained numerous oocytes from setigers 6-8 to the first seg- ment anterior to the modified segments. Maximum diameter of these oocytes was approximately 140 um and each oocyte con- tained a round germinal vesicle in the nu- cleus, and was filled with coarse, yolky ma- terial. Other material.—While trends in mor- phological characters are apparent, Table 1 illustrates the great morphological variabil- ity expressed by this species. The anteri- ormost occurrence of the heart body (char- acter 4) in benthic forms was in setiger 4, usually extending through 9 segments to se- tiger 12 (from 4—6 to 7—14); in epitokes from 8-10 to 13-16. Modified segments (char- acter 5) were absent in some smaller spec- imens and present in the region of setigers WS 18—22 in larger individuals. The number of setigers posterior to the modified segments (character 6) was variable (2—9 in benthic forms, 9-11 in epitokes), but generally ranged from five to seven in larger benthic individuals. The variability in the distribution of setal types displayed in the benthic form of Rar- icirrus variabilis made the characterization of body regions based on setal characteris- tics impossible. The greatest number of cap- illary notosetae (character 7) occurred in the first or second setiger with from four to sev- en capillaries per setal bundle, from one to three of these capillaries much longer than the rest. Subsequent notopodia possessed from one to five capillaries with a single elongate capillary per setal bundle. The an- teriormost occurrence of notopodial pecti- nate falcigers (character 11) was usually at setigers 3-9, although the first occurrence of such setae was at setigers 10—13 in several specimens. Coarsely serrate notosetae (character 12) were usually present begin- ning immediately posterior to the modified segments and were absent in most of the smaller individuals (<3.20 mm long) ana- lyzed. Several specimens possessed serrate notosetae in the last setiger anterior to the modified segments. Similar coarsely serrate neurosetae (char- acter 15) were present in all but three of the smaller individuals and were commonly found only in the posterior setigers. Several specimens had coarsely serrate neurosetae in from one to five setigers anterior to the modified segments. Enlarged, curved spines of the modified segments (character 16) usu- ally occurred either as a single or double pair. Several of the specimens with well de- veloped modified segments either lacked enlarged spines or possessed only a single spine. Three of the four epitokous individuals examined were much larger than any of the other specimens and possessed a greater number of setigers than any of the benthic individuals examined (Table 1). The heart PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 174 AJASIEOO JO 9DUdLINDOO ISI = C] “UOTsa1 APOqprIw UI S1OBIO[e} [eIpodoimou jo Jaquinu = p ‘souids saANOnpoidol = 9| ‘eIpodoinosu UT SIOBIO[Vy 91 e1IOS [ ‘Sios1o[ey [etpodomou jo Joquinu WNnwIxeul = ¢[ “eIpodojou Ur sIdBIO[ Ry d1BLI9S A[ISIVOD JO 99U9LINDIO IsSIY = Z| ‘“eIpOdojoU UT sIOSIO[VJ 9}eUTID0d JO 9dUDLINIDO IsIy = [|] ‘APOQGpPrIu UI SoLIe][Ides Zuo] Jo 1oquinu — QI ‘SoLiel[Ides Zuo] jo Joquinu WNUWIxewW = 6 ‘s}UoWId9s APOQpIW UT soLie[[ideo Jo Joquinu = g ‘etpodojou Ur soLel[ides Jo JoquINU WNUWIxeul = / ‘S}UdSWIZIS POYIPOU! 0} IOLID}sod $198119S JO JoqUINU = 9g ‘S}USWIZEs PoYyIPOW Jo uOTedO] = ¢ “(SIOSTIOS JO 93ULI) UOTLDO] Apog jIe9Yy = p ‘(WW) YIpIM = ¢ ‘(WIW) YIZUS] = Z ‘sIJOBIVOS JO JaquInu = | sr 61 6—S Ol CC x GAs 9 TI-S cl TT Tc—0¢ vI-Ol LG |! v9 0C ce x ar i SC 11-6 IT Sc x 8S 6 SI-Cl 91 cl VC-EC SI-6 LOC IT LI St ial Loar Sc OI-8 Ol Sc x 8-4 8 AW 91 Ol VC-EC 91-6 911 91 vl ve A I ar CC 8-$ 8 CC x v~C v 8-9 8 6 61-81 Sls €L0 L9°9 8C M suoutoods snoyorrdy Gare 61 Ome g €¢ Ol I Cc tac L L Gan 06 CIV 19°0 6S°L 8C L [far © Ic a 9 Ic I I Cc eG 9 9 0c 6l GS €v'0 SLO LC S Ca C CC 9-S 9 €c 6 I € SV 9 L 1c-0¢ cI cs 0 Ivs 8C ua 0+ 0 CC Se 9 CC L l C t 9 S Tc—0¢ cIly be 0 LSY 9¢ fe) Gaur © CC 9-€ 9 CC 9 I (4 cme L L GEOG GWY 8V0 LSY 8C d yeom eae 0c SY g €C € I € Sv L L ONG Glsav, Iv0 vy 6¢ O ar | €c a 9 0c 9 I Lt b-t S 6 CCIE TI-S Ov 0 OL et Te N far | €C v S €C 6 I Cc eC v ¢ CCIC CIS 0£ 0 19°€ LC W fap €C v S €C 9 I Lt t=C S € CERN 1 9€°0 OS’ Sc I Pet 1 €c b-e 9 tc L I Cc ve 9) t CC IC cI4 0€°0 Ive SC aA x CC SY ¢ 1c el I € SY ¢ L 61-81 G\ a7 €v0 Of € LC f GAG EC v 9 €¢ 8 I Cc t S S CC-IC cI Iv'0 Oce L¢ I x vl v-t v x L I I (| v x x viv Sc 0 tL €¢ H Gar G CC SY ¢ vC S I Cc oma v € Tc—0¢ cI4 ce 0 LSC VC 3) x x v-t v x € I I cI 9 x x EmaV, ce 0 Osc 81 A x x b-t v x v I Lt 6 € x x OI-v €c 0 O€C Oc a bar i LI v-t S x € I Lt Cc v Cc €C-CC GWA? Lc0 5G GC SC d x vC WAS v x S I Cc I v x x II Of 0 81°C vC O x x Cc v x I I (6 I € x x 84 LeO0 sé6l 0c a x 81 Cc t Oc € Lt G I v x x LW Sc O bs'l 0c V susutoeds ory uag 91 SI vl el A U Ol 6 8 L 9 S v € 4 I [enpta JajovIeyD -Ipu] "§ uowioeds = odA}O[OH{ “UNSoId 1OU JO}OvVIeVYO = K “SoIdads MoU ‘SIJIQDIADA SNAMIIUDY JO WIOJ SNOYO}IdS oy} JO SUDUIIOOdS MOF PUL WIOJ SIYJUDQ DY} JO SUdUTIODdS QZ JOJ SIOJOVIeYO [eOIsO[OYdIOW oY} soZLIeWIUINS—-"] s[qQe LL VOLUME 108, NUMBER 2 body of the epitokous specimens did not occur until setiger 8—10, extending to setiger 13-16. Modified setigers were present at the same or slightly more posterior setigers than in the benthic form, but the number of se- tigers posterior to these segments was great- er in the epitokous specimens (9-1 2). Epitokous individuals possessed a greater number of capillary notosetae along with a greater number of elongate capillaries in the anterior and midbody setigers. Notopodial pectinate falcigers were absent in the epi- tokous specimens while serrate notosetae were present only in the segments posterior to the modified segments. Coarsely serrate neurosetae were found only posterior to the modified segments with the exception of one epitokous individual with such neurosetae present in the segment immediately anterior to the modified segments. Remarks. —Raricirrus variabilis differs from R. maculatus Hartman, 1961, and R. beryli Petersen & George, 1991, in the pos- session of a more anteriorly situated heart body, the presence of pectinate notopodial falcigers in the benthic stage, the presence of a seminal vesicle and reproductive stylet, and the apparent absence of an asexual re- productive mode. Raricirrus variabilis and R. maculatus share the presence of a mod- ified region with large curved spines in at least some specimens, although that of R. maculatus is a apparently single modified segment while that of R. variabilis is com- posed of two segments incompletely sepa- rated dorsally. Raricirrus variabilis differs from R. beryli in lacking the ciliated region anterior to the mouth opening and on the ventral surface of the first two setigers. Epitokous individuals of R. maculatus have not been encountered while a single immature male epitoke of R. beryli has been described (Petersen & George 1991). The epitoke of R. beryli was smaller (11 mm) than the benthic form, with elongate cap- illary setae approximately 150% of the body width. The epitokous individuals of R. var- labilis are (with one exception) much larger 7s) (14-20 mm) than the benthic form with the elongate capillary setae only 75% of the body width. Etymology. —The specific name refers to the variability in the distribution of setal types. Discussion The family Ctenodrilidae is now com- prised of the subfamilies Ctenodrilinae and Raphidrilinae, each with two described gen- era. The genera Ctenodrilus Claparéde, 1863, and Aphropharynx Wilfert, 1974, are united within the subfamily Ctenodrilinae by their lack of branchial filaments while the sub- family Raphidrilinae consists of the genera Raphidrilus Monticelli, 1910, and Raricir- rus Hartman, 1961, both possessing bran- chiae. Morphological differences between Raricirrus and Raphidrilus are slight, how- ever, and there has been some doubt as to whether their separation is justified (see dis- cussion in Petersen & George 1991). Based on the descriptions of Raphidrilus nemasoma, an analysis of the type material of Raricirrus maculatus and the analysis of specimens of Raricirrus beryli, Petersen & George (1991) retained separation of Ra- Dhidrilus and Raricirrus and identified sev- eral morphological characters that could be useful in differentiating these two genera. These characters include the position and extent of the heart body, the structure of the nuchal organs, the structure of the head re- gion, and the presence ofa region with mod- ified setae. Petersen & George (1991) had used the anteriormost occurrence of the heart body as a distinguishing character for the genera Raphidrilus and Raricirrus. That of Raphi- drilus nemasoma has been reported to first appear at setiger 4 while that of the two previously known species of Raricirrus does not occur anterior to setiger 8 or 9 (Peterson & George 1991). Unfortunately this gener- alization does not hold true for R. variabilis as the heart body begins at setiger 4 (4-6) in the benthic specimens taken from the 176 wood panels and does not appear until se- tigers 8—10 in the epitokous individuals from the wild wood. The heart body of poly- chaetes has been viewed as being of he- matopoietic function (Kennedy & Dales 1958) however, more recent work has in- dicated that this organ could also have a protective function. Vovelle et al. (1995) noted chemical differences in the heart bod- ies of specimens of Raricirrus beryli col- lected adjacent to and away from the North Sea oil platforms indicating a possible role in the detoxification of hydrocarbons in this polluted environment. While the anterior- most occurrence of the heart body seems to be a valid specific character and may pos- sibly be related to the ability of a species to cope with its environment, it is not useful in the differentiation of genera. Sokolov (1911) described the nuchal or- gans of Raphidrilus as often-closed slits with short cilia extending from the base of the pit while Petersen & George (1991) de- scribed those of Rarvicirrus as bare oval areas surrounded by fields of cilia. The majority of specimens of R. variabilis examined pos- sessed slit-like nuchal organs with no visible ciliation of the surrounding prostomial sur- face when viewed using scanning electron microscopy. Several specimens that were observed under the dissecting microscope had the rims of their nuchal organs widely expanded, exposing what appeared to be a uniformly ciliated surface within. The mor- phology of the nuchal organs of R. variabilis is more similar to that described for the genus Raphidrilus than for Raricirrus. How- ever, while the ultrastructure of nuchal or- gans has been utilized in the recognition of possible phylogenetic relationships among polychaete families (Purschke 1986), the use of their gross morphology as a taxonomic character at the generic level is question- able. Nuchal organs may be eversible (Whit- tle & Zahid 1974) with associated retractor muscles (Purschke 1986) and may vary in morphology at reproductive maturity, per- haps exhibiting sexual dimorphism PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON (Schlotzer-Schrehardt 1987, 1991). This variability in the morphology of nuchal or- gans within a species as well as a lack of knowledge regarding nuchal organ variabil- ity within a family or genus precludes their use (at the light microscope level) in differ- entiating between these two genera. Petersen & George (1991) also referred to the shape and arrangement of the head as perhaps being of possible taxonomic signif- icance. The prostomium, peristomium, and first setiger of Raricirrus beryli and R. ma- culatus are described by Petersen & George (1991) as forming a single “‘visual unit” with all these segments united along the dorsal surface. In the figures of Monticelli (1910) and Sokolov (1911) the prostomium and peristomium of Raphidrilus nemasoma ap- pear to be united dorsally while the first setigerous segment is clearly separated from these segments. The prostomium, peristo- mium and first setigerous segment of R. var- labilis are also united dorsally and this char- acter may be of utility in the separation of Raphidrilus and Raricirrus. Further analysis of material belonging to the genus Raphi- drilus would be required to substantiate the use of this morphological character in sep- aration of the two genera. Raricirrus variabilis is placed within the genus Raricirrus, rather than Raphidrilus, primarily due to its setal characteristics. Ra- phidrilus nemasoma is described as pos- sessing only capillary notosetae and neu- rosetae with the exception of several mod- ified segments possessing thick spines (Monticelli 1910, Sokolov 1911). Raricirrus variabilis, as well as the other two described species in this genus, possesses both serrate capillaries and coarsely serrate setae in their notopodia. R. variabilis differs from the oth- er species in the possession of short, coarse- ly serrate falcate spines (Fig. 3A) in the an- terior notopodia while R. maculatus and R. beryli both have long, natatory capillaries in their “dispersal” forms (Petersen & George 1991). The finely pectinate falcigers, grading to coarsely serrate forms posteri- VOLUME 108, NUMBER 2 orly, are characteristic of all three species in the genus. Based on the analysis of Raricirrus var- iabilis, the generic diagnosis for Raricirrus Hartman, 1961 emended by Petersen & George 1991, is further emended as follows: Raricirrus Hartman, 1961, emended. Type species. —Raricirrus maculatus Hartman, 1961, by monotypy and original description. Diagnosis. —Raphidrilinae with prosto- mium not obviously delimited from peri- stomium and setiger 1 dorsally; with or without ventral cilia on peristomium and first few segments; last 6—9 setigers shorter and wider than preceding ones, forming a distinct posterior region. Branchiae simple, filamentous, may be clublike distally. Heart body in variable number of anterior and middle segments. Notosetae serrate capil- laries and coarsely serrate forms posteriorly, some species also with short pectinate fal- cigers; neurosetae falcate and finely pecti- nate anteriorly, grading to coarsely serrate forms posteriorly; simple curved spines sometimes replacing most or all normal se- tae in | or 2 modified posterior segments of some species. With or without seminal vesicle and reproductive stylet. Remarks. — Raricirrus variabilis is an ex- ample of an opportunist in the deep sea which survives by finding and exploiting or- ganically enriched sites (Grassle & Morse- Porteous 1987). The two previously de- scribed species in the genus are also reported from environments with elevated levels of organic carbon being most positively cor- related with elevated levels of hydrocar- bons. Raricirrus maculatus is known from an area close to an industrial waste dis- charge site in fine sediments that are con- taminated with heavy metals and chlori- nated hydrocarbons (Hartman 1961, Peter- sen & George 1991). Raricirris beryli was first reported at low densities from a sewage sludge dumping ground and at very high densities in fine sediments with high hydro- carbon concentrations from northern North 177 Sea oilfields (Moore 1991, Petersen & George 1991). Raricirris beryli has also been collected by A. Norrevang in shallow (5 m) waters in Skopun Harbor in the Faroes, a harbor without any apparent elevated hy- drocarbon input (Petersen 1994 pers. comm.). Based on strong association with high concentrations of certain components of the aromatic fraction of hydrocarbons, Moore (1991) has characterized R. beryli as a polychaete indicator species for sediments containing pollution levels of hydrocar- bons. The epitokes and/or larvae of R. var- labilis are most likely attracted to sites in the deep sea with high concentrations of organic material such as decaying wood. The largest eggs observed in female epi- tokes (Table 1, specimen U) of Raricirrus variabilis were approximately 140 um in di- ameter. While any determination of mode of development based on egg size should be done with great caution (Bridges 1993), it is probable that this species has either direct development ora short lecithotrophic larval mode of development. Assuming that these eggs approximate egg size at maturity, their diameter compares favorably with those of the unknown species of Raphidrilus (Ra- phidrilus nemasoma (sic)) studied by Qian & Chia (1989). By following the develop- ment of newly fertilized eggs, Qian & Chia found that the eggs of Raphidrilus sp. de- veloped directly into free-crawling pre- adults. Similarly the eggs of R. variabilis may produce preadults through direct de- velopment subsequent to deep-sea dispersal by the epitokous body form. Bridges (1993), for example, has reported a lecithotrophic mode of development for a morph of Stre- blospio benedicti with eggs 100-200 um in diameter and it is also quite possible, based upon egg diameter, that R. variabilis may also have a lecithotrophic mode of devel- opment rather than direct development. Additionally, dispersal may also be accom- plished by some type of asexual dispersal form similar to that reported for R. beryli by Petersen & George (1991). 178 Two body forms were recognized in Rar- icirrus variabilis, a ““normal” benthic form and a form similar to what Petersen & George (1991) identified as an epitokous phase for R. beryli. They considered this body form as an epitoke because it was sex- ually mature with the body usually packed with gametes. The epitokous form of R. var- iabilis was loosely organized with much free movement of the numerous large eggs with- in the interior of the worms. This may have been an artifact of poor fixation or may be indicative of the transient nature of this life stage with the individuals reduced to little more than egg containers. The epitoke of R. variabilis differs from the one immature male epitoke of R. beryli described by Pe- tersen & George (1991) in that its capillaries are only 0.75 times the body width while those of R. beryli were about 1.5 times body width. Petersen & George (1991) also described a third body form they referred to as a “‘dis- persal’’ form which had capillaries 2-4 times the body width but with no obvious ga- metes. Petersen & George (1991) hypothe- sized that this dispersal form may disperse, settle and then develop into spawning epi- tokes. There were no specimens of R. var- iabilis recovered, from the wood panels or the piece of wild wood, that were analogous to the dispersal phase recognized for R. ber- yli. Many specimens of the two shallow-water species, R. beryli and R. maculata, exam- ined by Petersen & George (1991) displayed evidence of regeneration subsequent to frag- mentation. Both of these species seem to rely on asexual reproduction as a means of exploiting their high organic, high hydro- carbon content, environment. Following colonization of a patch of suitable habitat by some type of dispersal form, asexual re- production would provide a means of rapid population growth in order to better utilize the available resource (Schroeder & Her- mans 1975). There was no evidence of an asexual re- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON productive stage for R. variabilis (although the possibility exists that temporally restric- tive sampling could have missed such an asexual stage if it were a seasonal event). This species seems to have adapted a dif- ferent reproductive strategy in the exploi- tation of its patchily distributed resource. Many specimens of the benthic form of R. variabilis contained both developing eggs and sperm and were evidently hermaph- roditic. Hermaphroditism has often been associated with patchy, unpredictable en- vironments (Petraitis 1991) and may allow R. variablis to enhance its exploitation of widely scattered patches of organic material on the deep-sea floor. The benthic form could function as both genders or as the gender which would maximize reproductive success under existing environmental con- ditions similar to the strategy proposed for Capitella capitata by Petraitis (1991). Once established on a patch of suitable organic resource, the benthic form of R. var- iabilis may engage in sexual reproduction, maximizing its reproductive success through its hermaphrodism. The resultant free-liv- ing larvae would increase the population size at the home site although at a reduced rate when compared to the exploitation of a comparable resource in shallower waters. Some of the benthic individuals could de- velop into the epitokous form, either as a routine percentage of the population as seen in Dedocaceria caulleryi Oersted (Gibson & Clark 1976) or perhaps in response to de- teriorating resource, as a means of coloniz- ing suitable habitat elsewhere. Acknowledgments These polychaetes were collected and dis- sected from wood panels as part of R. D. Turner’s studies of deep-sea wood-boring communities, a project supported by the Of- fice of Naval Research, Contract no. N0O0014-84-0258 with Harvard University. The SEM pictures were taken with an ABT 55 SEM generously provided by R. Prezant VOLUME 108, NUMBER 2 of the Department of Biology, Indiana Uni- versity of Pennsylvania. An early draft of this manuscript was reviewed and im- proved by E. Ruff of Ruff Systematics, So- lana Beach, California. The manuscript was greatly improved by the suggestions and supporting materials provided by M. Pe- tersen (Zoological Museum, University of Copenhagen) and J. Blake (ENSR Consult- ing and Engineering, Woods Hole). Literature Cited Bridges, T.S. 1993. Reproductive investment in four developmental morphs of Streblospio (Poly- chaeta: Spionidae).— Biological Bulletin 184: 144-152. Desbruyéres, D., & L. Laubier. 1988. Exploitation d’une source de matiére organique concentrée dans l’océan profond: intervention d’une an- nélide polychéte nouvelle.—Comptes Rendus de l’Académie des Sciences, Paris 30(III): 329- 335. Gibson, P. H., & R. B. Clark. 1976. Reproduction of Dodecaceria caulleryi (Polychaeta: Cirratu- lidae).—Journal of the Marine Biological As- sociation of the United Kingdom 56:649-674. Grassle, J. F., & L. S. Morse-Porteous. 1987. Mac- rofaunal colonization of disturbed deep-sea en- vironments and the structure of deep-sea ben- thic communities. — Deep-Sea Research 34(1 2): 1911-1950. Hartman, O. 1961. Polychaetous annelids from Cal- ifornia. — Allan Hancock Pacific Expeditions 25: 1-226. Kennedy, G. Y., & R. P. Dales. 1958. The function of the heart-body in polychaetes.—Journal of the Marine Biological Association of the United Kingdom 37:15-31. Monticelli, F.S. 1910. Raphidrilus nemasoma Mon- tic. Nuovo Ctenodrilide del Golfo di Napoli (Revisione de’Ctenodrilidi).— Archivio Zoolo- gico 4(4):401—436, plates 12-13. Moore, D. C. 1991. Raricirrus beryli Petersen & George (Ctenodrilidae): a new polychaete in- dicator species for hydrocarbon-polluted sedi- ments. Pp. 477-486 in M. E. Petersen & J. B. Kirkegaard, eds., Systematics, biology and mor- phology of world Polychaeta. Proceedings of the 2nd International Polychaete Conference, Co- penhagen 1986. Ophelia Supplement 5. Petraitis, P. S. 1991. The effects of sex ratio and density on the expression of gender in the poly- chaete Capitella capitata. —Evolutionary Ecol- ogy 5:393-404. 179 Petersen, M. E., & J. D. George. 1991. A new species of Raricirrus from northern Europe, with notes on its biology and a discussion of the affinities of the genus (Polychaeta: Ctenodrilidae). Pp. 185-208 in M. E. Petersen & J. B. Kirkegaard, eds., Systematics, biology and morphology of world Polychaeta. Proceedings of the 2nd In- ternational Polychaete Conference, Copenhagen 1986. Ophelia Supplement 5. Purschke, G. 1986. Ultrastructure of the nuchal organ in the interstitial polychaete Stygocapitella sub- terranea (Parergodrilidae).— Zoologica Scripta 15:13-20. Qian, P.-Y., & F.-S. Chia. 1989. Sexual reproduction and larval development of Raphidrilus nema- soma Monticelli, 1910 (Polychaeta: Ctenodril- idae).— Canadian Journal of Zoology 67:2345- 2351. Schlotzer-Schrehardt, U. 1987. Ultrastructural in- vestigation of the nuchal organs of Pygospio ele- gans (Polychaeta). IJ. Adult nuchal and dorsal organs.— Zoomorphology 107:i169-179. 1991. Nuchal and dorsal organs in Pygospio elegans. Pp. 633-640 in M. E. Petersen & J. B. Kirkegaard, eds., Systematics, biology and mor- phology of world Polychaeta. Proceedings of the 2nd International Polychaete Conference, Co- penhagen 1986. Ophelia Supplement 5. Schroeder, P. C., & C. O. Hermans. 1975. Chapter 1. Annelida: Polychaeta. Pp. 1-213 in A. C. Giese & J. S. Pearse, eds., Reproduction of marine invertebrates. Volume III. Annelids and Echiu- rans. Academic Press, New York. Smith, C. R., & R. R. Hessler. 1987. Colonization and succession in deep-sea ecosystems. — Trends in Ecology and Evolution 2:359-363. Sokolov, I. 1911. Uber eine neue Ctenodrilusart und ihre Vermehrung. — Zeitschrift fur wissenschaf- tliche Zoologie, Leipzig 97(3):546-603. Turner, R. D. 1973. Wood-boring bivalves: oppor- tunistic species in the deep sea.—Science 180: 1377-1379. 1977. Wood, mollusks, and deep-sea food chains.— Bulletin of the American Malacologi- cal Union for 1976:13-19. 1981. ‘‘Wood Islands” and the ‘Thermal Vents” as centers of diverse communities in the deep sea. — Soviet Journal of Marine Biology 7(1): 1-9. Vovelle, J., M. E. Petersen, M. Grasset, & P. Beaunier. 1995. Iron bioaccumulation in the heart body of Raricirrus beryli Petersen & George (Poly- chaeta: Ctenodrilidae).— Proceedings of the 4th International Polychaete Conference (in press). Whittle, A. C., & Z. R. Zahid. 1974. Fine structure of nuchal organs in some errant polychaetous annelids. — Journal of Morphology 144:167-184. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(2):180-200. 1995. Six copepodid stages of Ridgewayia klausruetzleri, a new species of copepod crustacean (Ridgewayiidae: Calanoida) from the barrier reef in Belize, with comments on appendage development Frank D. Ferrari Department of Invertebrate Zoology (MRC 534), National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, U.S.A. Abstract. —Ridgewayia klausruetzleri from the barrier reef in Belize differs from the five other Atlantic species of Ridgewayia in the morphology of the female genital complex and fifth leg of both females and males. Developmental patterns of setation suggest that there are six enditic lobes associated with the syncoxa and basis of maxilla 2, and that the ramal segments are exopodal. The maxilliped has a distomedial lobe on its basis and five endopodal segments, three of which are added proximally from the penultimate segment during the copepodid phase of development. The first through fourth legs exhibit the common, and presumed ancestral, pattern of segmental development. Left and right endopods of the male fifth leg are one-segmented, a condition resulting from developmental convergence for the two rami. The male fifth leg also exhibits setal loss during development of the right endopod and both exopods. Ridgewayia Thompson & Scott, 1903 along with three other genera of epibenthic, pseudocyclopoidean calanoids, Brattstrom- ia Fosshagen (in Fosshagen & Iliffe 1991), Exumella Fosshagen 1970, and Placocalan- us Fosshagen, 1970 comprise the copepod family Ridgwaylidae of Wilson (1958). Spe- cies of Ridgewayia are found around and in crevices and caves in coral reef habitats. Of the nine nominal species of Ridgewayia, five are found in the Atlantic Ocean, R. marki (Esterly, 191la), R. gracilis Wilson, 1958, R. shoemakeri Wilson, 1958, R. wilsonae Fosshagen, 1970, and R. fosshageni Humes & Smith, 1974. Ridgewayia typica Thomp- son & Scott, 1903, R. canalis (Gurney, 1927), R. krishnaswamyi Ummerkutty, 1963, and R. flemingeri Othman & Green- wood, 1988, are Indo-Pacific. On 17 July 1989 specimens of a new spe- cies of Ridgewayia were collected from a solitary swarm of copepods off the northern cay of Tobacco Range in Belize. Tobacco Range (16°54’N, 88°05’W) is a group of four cays about 1 km from the lagoonal side of the barrier reef of Belize. The cays surround a shallow grass flat. The copepod swarm was collected in 3-6 m of water in an area of peat blocks which are fractured and slumped from the northwest shore of the northern cay (MacIntyre et al. 1989). No swarms were observed in June 1988, or subsequently in May 1992 and June 1993. The first descrip- tion of the copepodid phase of development for a species of Ridgewayia is presented here from the six copepodid stages in the swarm. Methods The copepods were fixed with 4% form- aldehyde in 35% sea water and preserved in 0.5% propylene phenoxytol/4.5% pro- pylene glycol/95.0% fresh water. In the lab- oratory, specimens were cleared in steps through 50.0% lactic acid/50.0% fresh water to 100% lactic acid and stained by adding VOLUME 108, NUMBER 2 a solution of chlorazol black E dissolved in 70.0% ethanol/30.0% fresh water, or treated with a 15% solution of KOH in water and stained. Prosome and urosome are designated Pr and Ur. First through sixth copepodid stages are designated CI to CVI; CVI is the adult. Thoracic and abdominal somites are num- bered according to their appearance during development as interpreted from data of dulsemann (1991). The first and oldest tho- racic somite bears the maxilliped and is fused with the cephalon. The youngest is the sev- enth; among calanoids it is the only thoracic somite without an appendage. In adult cal- anoids the seventh is the first somite of the urosome, and in adult females it is fused to the second abdominal somite to form the genital complex. The first and oldest ab- dominal somite is the most posterior; it bears the caudal rami. The youngest is immedi- ately anterior to the oldest, and the remain- ing abdominal somites increase in age, and decrease in numerical designation, anteri- orly. Cephalic appendages are abbreviated Al = antennule; A2 = antenna; Mn = mandi- ble; Mx1 = maxillule; Mx2 = maxilla. Ap- pendages on thoracic somites are Mxp = maxilliped (thoracopod 1); Pl-5 = swim- ming legs (thoracopods 2-6). The caudal ra- mus is CR. Designations of appendage seg- ments generally follow Huys & Boxshall (1991) except for Mx2 and Mxp; exopod = Re; endopod = Ri; medial lobe of a segment = li, lateral lobe = le. Terminal segments of Mx2 are exopodal. Mxp has at most five endopodal segments. Ramal segments on the thoracopods (Mxp and P1—5) are numbered by their appear- ance during development (Hulsemann 1991, Ferrari & Ambler 1992, and here) and not proximal-to-distal as is the usual case for copepod descriptions. On the Mxp the distal segment is the first endopodal segment, and the second endopodal segment is immedi- ately proximal to the first. The third en- dopodal segment is immediately distal to 181 Table 1.—Setation of the maxilliped of Ridgewayia klausruetzleri for stages CI-CVI; setation of syncoxa and basis (columns 1-6) is complete at CII while se- tation of endopodal segments 2—5 is not complete until CVI. Lobes of the syncoxa (sl—s4), the basis (b) and its distomedial lobe (1), and the endopodal segments (nI—n5 numbered by developmental age) are arranged from left, proximally, to right, distally. a = segment not formed. sl s2 s3 s4 b l n3 n4 n5 n2 nil CI Ongar DO DD ele daa. LY 4 CII [bea Ate ee eee Ser a deh Cll Le eS eae te DT) awe aes ed = eg CIV ley sini ARES ass ye Rl yie2, anew np 2s. oh CV Ngee ee Aaa Si Sue? OS) Iu, Dae Fa CVI Wn De ee SiS. OE eee Bi sh Bg the basis. The fourth endopodal segment is immediately distal to the third. The fifth endopodal segment is the middle segment. On an Mxp with a 5-segmented endopod, the second and first segments are more dis- tal and the third and fourth segments are more proximal (Table 1 and Figs. 3E, 9G, 11F). On P1-5, the distal segment of a ra- mus is the first segment. The second seg- ment is immediately distal to the basipod. If present, the third segment is immediately proximal to the distal (or first) segment. For a 3-segmented ramus, the proximal segment is the second segment, the middle segment is the third segment, and the distal segment is the first segment (Figs. 4F, 7B, 9J). The number of setae recorded for the segments follows this same scheme. Armament elements of appendages here are termed setae regardless of their position or degree of rigidity. Examples of the po- sition and morphology of setae are shown in the illustrations. Two setae and one aesthetasc on a segment of Al are desig- nated 2 + 1. Setules are epicuticular exten- sions of a seta; denticles are epicuticular ex- tensions of an appendage segment; spinules are epicuticular extensions of a somite. Von Vaupel Klein’s organ (Ferrari & Steinberg 1993) on P1 (the appendage of thoracic so- mite 2) consists of the curved basipodal seta, 182 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Ridgewayia klausruetzleri n. sp., CVI female: A, animal left lateral; B, genital complex, right lateral; C, genital complex, ventral; D, rostrum right lateral; E, rostrum ventral; F, CR dorsal. CVI male: G, Th 6-7 and abdominal somites. Line 1 = 0.1 mm for A; line 2 = 0.1 mm for G; line 3 = 0.1 mm for B—F. and distolateral corner of the second en- dopodal segment. Many setae on P1-5 of Ridgewayia narrow abruptly distally; set- ules usually are present only on the distal section. Ridgewayia klausruetzleri, new species Figs. 1-11 Material.—CI—528 specimens; CII—505 specimens; CIII—403 specimens; CIV — 319 specimens; CV—636 specimens; CVI— 3687 specimens from Tobacco Range (16°54'N, 88°05’W) in Belize are deposited in the National Museum of Natural History, Smithsonian Institution (USNM). An un- dissected female holotype (USNM 268277), an undissected male allotype (USNM 268278), and one lot of undissected para- types, 10 females and 10 males (USNM 268279) are in vials of glycerin. The re- VOLUME 108, NUMBER 2 maining specimens (CI—524, CII—497, Clli— 394, CIV—279, CV—596, CVI— 3663) comprise a lot of undissected para- types (USNM 268280) in 0.5% propylene phenoxytol/4.5% propylene glycol/95.0% fresh water. CVI female.—Length range of 15 speci- mens 0.84—0.90mm (mean 0.87); average Pr length/Ur length = 3.1. Pr (Fig. 1A): 6 segments; Ist a complex of 5 cephalic somites plus thoracic somite 1; thoracic somites 2—6 are simple and ar- ticulated. Ur (Fig. 1A): 4 segments; Ist a genital complex of thoracic somite 7 and abdom- inal somite 2 (Fig. 1B); genital complex slightly asymmetrical, as viewed ventrally (Fig. 1C); several sensilla near left postero- lateral corner. Copulatory and oviducal openings separate. Abdominal somites 3, 4, 1 articulated; somite 1 small. Rostrum (Fig. 1D, E): a short thick plate. Al (Fig. 2A-C): 26 articulated segments Teese 22+ 1 2+ 1,2 + 1, - ae i, 2 ar 1, Bap iD ar il Ye ae I A ae » =P il, 2 ap 1, Bae I 2 ae ly Bae leg es eins?) | 12.4 + 2 setae + aesthet- ascs; 13th through 22nd segments with row of small denticles. A2 (Fig. 4A): coxa with | seta; basis with 2 setae. Re 8-segmented with 1, 1, 1, 1, 1, 1, 1, 4 setae. Ri 2-segmented with 1, 15 (7 terminal, 8 subterminal) setae. Mn (Fig. 4B, C): coxa more heavily scle- rotized medially; basis with 4 setae. Re 4-segmented with 1, 1, 1, 3 setae. Ri l-segmented with 15 setae (11 terminal, 4 subterminal). Mx! (Fig. 3A, B): le with 9 setae. Re 1-segmented with 11 setae. Basis with 5 se- tae; Ri with sets of 4 and 4 medial, and 7 terminal setae. Li 2 and 3 with 5 and 4 setae. Li 1 with 9 apical, 1 anterior setae and 4 posterior setae; denticles on posterior sur- face. Mx2 (Fig. 3C): li 1-4 of coxa with 5, 3, 3, 3 setae; li 5 and 6 of basis with 4 and 3 setae. Re indistinctly segmented with 7 se- tae. 183 Mxp (Fig. 3D, E): syncoxa with 4 lobes of 1, 2, 4, 3 setae; basis with 5 setae (2 on a distal medial lobe). Ri 5-segmented with 4, 4, 4, 4, 3 setae. Three areas of denticles on coxa and a longitudinal row of denticles on basis. P1 (Fig. 5A, B): coxa with medial seta; basis with curved, medial seta. Re 3-segmented with 7, 2, 2 setae. All segments with row of denticles toward distal edge; segment 2 with longer denticles at distolat- eral corner; distolateral margin of segment 3 with a finger-like process with denticles along outer margin and an attenuate pro- cess. Ri 3-segmented with 6, 1, 2 setae. Von Vaupel Klein’s organ includes setules and seta of basis, and denticles and 2 pores on an attenuate distolateral edge of segment 2. P2 (Fig. 4D): coxa with medial seta and lateral area of denticles; basis with medial area of denticles. Re 3-segmented with 8, 2, 2 setae; segments | and 3 with row of den- ticles toward distal edge. Ri 3-segmented with 8, 1, 2 setae. P3 (Fig. 4E): coxa with medial seta and lateral denticles; basis unarmed. Re 3-segmented with 9, 2, 2 setae; posterior face of segments | and 3 with denticles to- ward distal edge. Ri 3-segmented with 8, 1, 2 setae; posterior face of segment 3 with denticles toward distal edge. P4 (Fig. 4F): coxa with medial seta, and lateral and anterior denticles; basis with lat- eral seta. Re 3-segmented with 9, 2, 2 setae; posterior face of segments | and 3 with den- ticles toward distal edge. Ri 3-segmented with 7, 1, 2 setae. P5 (Fig. 5C): coxa unarmed; basis with lateral seta and posterior denticles. Re 3-segmented with 8, 1, 2 setae; segment 1 with denticles on anterior face. Ri 2-segmented with 7, O setae. CR (Fig. 1F): 5 terminal setae of differing lengths, and a small dorsal seta. CVI male. —Differs from CVI female as follows: length range of 15 specimens 0.77- 0.82mm (mean 0.79); average Pr length/Ur length = 3.0. 184 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Ridgewayia klausruetzleri n. sp., CV1 female: A, Al articulated segments 1-9; B, Al articulated segments 10—20; C, Al articulated segments 21-26. CVI male: D, right Al articulated segments 1-10; E, right Al articulated segments 11-18; F, right Al articulated segments 19-21. Line = 0.1 mm. VOLUME 108, NUMBER 2 185 \ x N \ ge [ \| \\\\ ; / Fig. 3. Ridgewayia klausruetzleri n. sp., CVI female: A, Mx1 anterior; B, lil of Mx1 posterior; C, Mx2 posterior; D, syncoxa and basis of Mxp posterior; E, endopod of Mxp posterior (numbers to the right indicate the appearance of endopodal segments during development). Line = 0.1 mm. 186 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON { Fig. 4. Ridgewayia klausruetzleri n. sp., CVI female: A, A2; B, coxa of Mn; C, palp of Mn; D, P2 posterior, exopod detached; E, P3 posterior, exopod detached; F, P4 posterior, exopod detached (numbers to the right indicate the appearance of exopodal segments during development). Line = 0.1 mm. VOLUME 108, NUMBER 2 187 ~ = Pays NS t t t t r y N \ y r Fig. 5. Ridgewayia klausruetzleri n. sp., CVI female: A, P1 anterior, exopod detached; B, basis, endopod 2 and 3 of Pl medial; C, P5 posterior. CVI male: D, right PS and coxa of left P5 anterior; E, right exopod | of P5 posterior; F, left basis, exopod and endopod of P5 anterior; G, left exopod | and 3 of PS anterior. Line = 0.1 mm. 188 Ur (Fig. 1G): 5 somites; thoracic somite 7 with left lateral genital aperture. Abdom- inal somites 2, 3, 4, 1 simple and articulat- ed; somite | small. Right Al (Fig. 2D—-F): 21 articulated seg- MEMES Withs2 sealee ora lee eel PSAs se ar iL ealce Oe) cs le S) seg) Dap Nee ar Ne 2p Is 4, 2, 3 ae 1, 2s De 1, 2, 3 + 3 setae + aesthetascs; segments | 1- 16 with row of denticles and segment 17 with 3 rows. Partly complete arthrodial membrane in segment 9; 2 partly complete ventromedial arthrodial membranes in seg- ment 11; segment 17 probably of 3 fused segments based on 3 rows of denticles. Ge- niculation between 17th and 18th articu- lated segments. P5 (Fig. 5D-G): right coxa unarmed; ba- sis with lateral seta, and long, medial sen- silla proximally and short medial denticles distally. Re 2-segmented with 2, 1 setae; 3 ridged pads distally on segment 1. Ri l-segmented with midlateral seta and den- ticles. Left coxa unarmed; basis with lateral seta. Re 3-segmented with 4, 1, 1 setae. Rel hook- like, articulating at medial corner with 3rd segment; its proximal seta with 3 bends and an extension of its hyaline membrane on convex surface of the 3rd bend; 2nd and 3rd seta simple; 4th seta with 1 bend and an extension of its hyaline membrane on con- cave surface of the bend. Re2 outer, spine- like seta reaching to distal edge of Re3. Re3 outer, spine-like seta not reaching beyond setae of Rel. Ri 1-segmented, unarmed. CV female. — Differs from CVI female as follows: length range of 27 specimens 0.73- 0.80mm (mean 0.76); average Pr length/Ur length = 3.0. Ur (Fig. 6A, B): 4 segments; thoracic so- mite 7 and abdominal somites 2, 3, 1 ar- ticulated. Arthrodial membrane between Th7 and Ab2 not as pronounced as those between other somites. Copulatory opening apparently forms at articulation of Th7 and Ab2 (Fig. 6B). Al (Fig. 6C): articulating segments 4-14 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON wil 2 ap iw, Zar il, A Bap il, 2,2 ce ll, 2 2, 2, 2 + 1 setae + aesthetascs. A2 (Fig. 6D): Ri terminal segment with 14 setae (7 terminal, 7 subterminal). Mx! (Fig. 6E): Re with 10 setae. Ri with medial sets of 3 and 4 setae, and 6 terminal setae. Mxp (Fig. 6G): Ri 5-segmented with 4, 3, 3, 3, 2 setae. P1: Re2 distal margin simple; Re3 (Fig. 6F) distal margin with 2 simple extensions of segment. Ri2 without denticles or pores and elongation of distolateral margin less pronounced. P5 (Fig. 6H): coxa unarmed; basis with lateral seta. Re 2-segmented with 9, 1, setae. Ri 2-segmented with 6, O setae. CV male. — Differs from CV female as fol- lows: length range of 13 specimens 0.68- 0.74mm (mean 0.71); average Pr length/Ur length = 3.0. Ur (Fig. 61): pronounced arthrodial mem- brane between Th7 and Ab2. P5 (Fig. 6J): right Re 2-segmented with 7, 1 setae. Ri 2-segmented with 4, O setae. Left Re 2-segmented with 7, 1 setae. Ri an unarmed segment. CIV female. — Differs from CV female as follows: length range of 27 specimens 0.64— 0.69mm (mean 0.66); average Pr length/Ur length = 3.1. Ur (Fig. 7A): 3 segments; thoracic somite 7 and abdominal somites 2, 1 articulated. Al (Fig. 8A, B): 25 segments with 1 + 1, 2 tl bed ee ae eal 2 ele legal WD, =P My Bs ar A 4 ae le, 2, 232 ar il, I, I, 2,2 + 1, 2, 4 + 2 setae a, aesthetascs: denticle rows on 13th through 21st seg- ments. A2 (Fig. 8C): Ri terminal segment with 13 setae (7 terminal, 6 subterminal). Mn (Fig. 8D): Ri with 3 and 8 setae on segments 1 and 2. Mx! (Fig. 8E): basis with 4 setae. Re with 8 setae. Ri with medial sets of 3 and 3 setae, and 5 terminal setae. Mxp (Fig. 8F): Ri 5-segmented with 4, 2, Dl setae: VOLUME 108, NUMBER 2 189 Fig. 6. Ridgewayia klausruetzleri n. sp., CV female: A, body right lateral; B, Th7 and abdominal somites; C, Al articulated segments 4-14; D, endopod 2 of A2; E, exopod and endopod of Mx1; F, Pl exopod 3; G, distomedial lobe of basis and endopod of Mxp; H, P5 posterior. CV male: I, Th7 and abdominal somites lateral; J, P5 anterior. Line 1 = 0.1 mm for A; line 2 = 0.1 mm for B, I; line 3 = 0.1 mm for C—H. 190 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ( Fig. 7. Ridgewayia klausruetzleri n. sp., CIV female: A, body right lateral; B, exopod and endopod of P2 (numbers to the right indicate the appearance of exopodal segments during development); C, exopod and endopod of P3; D, exopod and endopod of P4; E, basis, exopod and endopod of P5. CIV male: F, basis, left exopod and endopod, and right endopod of PS. Setae which were broken and not studied are designated with a wavy-line cutoff. Line 1 = 0.1 mm for A; line 2 = 0.1 mm for B—F. P1 (Fig. 8G): Re 2-segmented with 8, 2 P4 (Fig. 7D): Re 2-segmented with 9, 1 setae. Ri 2-segmented with 8, 1 setae. setae; segment 2 with medial denticles. Ri P2 (Fig. 7B): Re 2-segmented with 9, 2 2-segmented with 7, | setae. setae. Ri 2-segmented with 8, | setae. PS (Fig. 7E): basis unarmed. Re P3 (Fig. 7C): Re 2-segmented with 9, 2 1-segmented with 7 setae. Ri 1-segmented setae. Ri 2-segmented with 8, 1 setae. with 4 seta. VOLUME 108, NUMBER 2 191 Fig. 8. Ridgewayia klausruetzleri n. sp., CIV female: A, Al articulated segments 1-17; B, Al articulated segments 18-25; C, endopod 2 of A2; D, endopod of Mn; E, basis, exopod and endopod of Mx1; F, distomedial lobe of basis and endopod of Mxp; G, exopod and endopod of P1. Line = 0.1 mm. 192 CIV male. —Differs from CIV female as follows: length range of 13 specimens 0.62-— 0.65mm (mean 0.64); average Pr length/Ur length = 3.0. P5 (Fig. 7F): right basis unarmed. Re l-segmented with 7 setae. Ri 1-segmented with 4 setae. Left basis unarmed. Re 1-segmented with 7 setae. Ri 1-segmented without setae. CIII.—Differs from CIV female as fol- lows: length range of 25 specimens 0.50- 0.58mm (mean 0.54; average Pr length/UR length = 3.4. Ur (Fig. 9A): 2 segments; thoracic somite 7 and abdominal somite | articulated. Al (Fig. 9B, C): 24 articulated segments with le ese lll tle Ons Ow tle Os eee eee lal lel Der A lly + 1, 2,4 + 2 setae + aesthetascs. A2 (Fig. 9D): Ri terminal segment with 11 setae (6 terminal, 5 subterminal). Mn (Fig. 9E): Ri2 with 7 apical setae. Mx! (Fig. 9F): le with 8 setae. Ri 2-segmented; Ril with medial groups of 2 and 3 setae. Lil with 7 apical setae, 1 re- duced in length. Mxp (Fig. 9G): Ri 4-segmented with 4, 2, 1, 1 setae. P1 (Fig. 9H): proximal 3 setae of Ril and seta of Ri2 reduced in size. P2: proximal, medial seta of Rel and me- dial seta of Re2 reduced in size. Proximal, medial 2 setae of Ril and medial seta of Ri2 reduced in size. P3 (Fig. 91): Re 2-segmented with 7, 1 setae. Ri 2-segmented with 7, | setae. P4 (Fig. 9J): coxa and basis unarmed. Re l-segmented with 7 setae. Ri 1-segmented with 6 setae. P5 (Fig. 9K): ventrally-directed bud with 2 medial and 3 apical setae. CIT. — Differs from CIII as follows: length range of 25 specimens 0.44—0.47mm (mean 0.46); average Pr length/Ur length = 2.9. Pr (Fig. 10A): 5 segments; Ist a complex of 5 cephalic somites plus thoracic somite 1; thoracic somites 2—5 simple and articu- lated. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Ur (Fig. 10A): 2 segments; thoracic so- mite 6, with lateral lobes, and abdominal somite 1 articulated. Al (Fig. 10B, C): 17 articulated segments witht? 42.0 Ve Oe tl Osa O sale + 1,1,1,1,2+ 1, 2,4 + 2 setae + aesthet- ascs; denticles on segments 12 and 13. A2 (Fig. 10D): Ri terminal segment with 10 (6 terminal, 4 subterminal) setae. Mxl1 (Fig. 10E): le with 5 setae. Re 1-segmented with 7 setae. Basis with 3 setae. Ril with medial sets of 2 and 2 setae. Lil with 2 posterior setae; li2 with 4 setae. Mx2 (Fig. 10F): li 5 and 6 of basis with 3 and 1 setae. Re indistinctly segmented with 6 setae. Mxp (Fig. 10G): Ri 3-segmented with 4, 1, 1 setae. P1 (Fig. 10H): Re 2-segmented with 8, 1 setae. Ri 2-segmented with 7, 1 setae. P2 (Fig. 101): Re 2-segmented with 7, 1 setae. Ri 2-segmented with 7, 1 setae. P3 (Fig. 10J): coxa unarmed. Re l-segmented with 7 setae. Ri 1-segmented with 6 setae. P4 (Fig. 10K): ventrally-directed bud with 2 medial, 1 lateral, and 2 apical setae. CI. — Differs from CII as follows: length range of 25 specimens 0.36—0.40mm (mean 0.38); average Pr length/Ur length = 2.6. Pr (Fig. 1 1A): 4 segments; articulation be- tween thoracic somites 3 and 4 not as dis- tinct as between other segments. Ur (Fig. 11A): 2 segments; thoracic so- mite 5, with lateral lobes, and abdominal somite | articulated. Rostrum absent Al (Fig. 11B): 10 articulated segments Wan Sao JU aes Weil ae JO, Wo, Yar i. 2, 4 + 2 setae + aesthetascs; denticles absent. A2 (Fig. 11C): Ri terminal segment with 8 (5 terminal, 3 subterminal) setae. Mn (Fig. 11D): Ri2 with 5 setae. Mx! (Fig. 11E): le with 4 setae. Ri with medial groups of 2 and | setae. Li2 with 2 setae. Mxp (Fig. 11F): syncoxa with 4 lobes of O, 1, 2, 2 setae; basis with 3 (1 on a distal VOLUME 108, NUMBER 2 193 Fig. 9. Ridgewayia klausruetzleri n. sp., CIII: A, body left lateral; B, Al free segments 1-15; C, Al free segments 16-24; D, endopod 2 of A2; E, endopod 2 of Mn; F, Mx1; G, distomedial lobe of basis and endopod of Mxp (numbers to the right indicate the appearance of endopodal segments during development); H, endopod of P1; I, exopod and endopod of P3; J, exopod and endopod of P4 (number to the left indicates the appearance of exopodal segment during development); K, P5. Setae which were broken and not studied are designated with a wavy-line cutoff. Line 1 = 0.1 mm for A; line 2 = 0.1 mm for B—-K. 194 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 10. Ridgewayia klausruetzleri n. sp., Cll: A, body right lateral; B, Al articulated segments 1-10; C, Al articulated segments 11-17; D, endopod 2 of A2; E, Mx1; F, distal lobes of basis and exopod of Mx2; G, basis and endopod of Mxp; H, exopod and endopod of P1; I, exopod and endopod of P2; J, P3; K, P4. Setae which were broken and not studied are designated with a wavy-line cutoff. Line 1 = 0.1 mm for A; line 2 = 0.1 mm for B-K. VOLUME 108, NUMBER 2 NOS) Fig. 11. Ridgewayia klausruetzleri n. sp., Cl: A, animal right lateral; B, Al; C, endopod 2 of A2; D, endopod 2 of Mn; E, Mx1; F, Mxp (numbers to the right indicate the appearance of endopodal segments during devel- opment); G, Pl; H, P2; I, P3. Setae which were broken and not studied are designated with a wavy-line cutoff. Line 1 = 0.1 mm for A; line 2 = 0.1 mm for B-I. 196 medial lobe) setae. Ri 2-segmented with 4, 1 setae. P1 (Fig. 11G): coxa and basis unarmed. Re 1l-segmented with 8 setae. Ri 1-segmented with 7 setae. P2 (Fig. 11H): coxa and basis unarmed. Re l-segmented with 7 setae. Ri l-segmented with 6 setae. P3 (Fig. 111): ventrally-directed bud with 2 medial and 3 apical setae. Remarks.—The six Atlantic species of Ridgewayia can be divided into two groups. In the first group, of R. klausruetzleri, R. marki, R. shoemakeri, and R. fosshageni, the endopod of the male left PS is a rela- tively simple, unarmed segment. In the sec- ond group, of R. wilsonae and R. gracilis, this segment is divided into several finger- like extensions. Females of the second group can be separated from each other by the number of setae on the endopod of the fe- male P5, six on R. wilsonae or seven on R. gracilis. Among males in the first group, the ex- ternal spine-like seta on the third (middle) exopodal segment of the left P5 of R. shoe- makeri and R. fosshageni extends beyond the setal elements of the first (distal) seg- ment; in R. klausruetzleri and R. marki, as redescribed by Yeatman (1969), this seta does not extend beyond the smallest of these elements. An examination of specimens of R. marki deposited in the National Museum of Natural History by Dr. Harry Yeatman (USNM 190873) shows that R. marki can be separated from R. klausruetzleri by the following characters. Ridgewayia klaus- ruetzleri is smaller (length range of adult females 0.84-0.90mm, males 0.77- 0.82mm) than R. marki (females 0.98- 1.05mm, males 0.90—1.03mm). The copu- latory pore is located more anteriorly on the female genital complex of R. klausruetzleri (compare Figs. 1B, C with 12A, B). The distomedial corner of the second (proximal) endopodal segment of the female P5 is not as pronounced as that of R. marki (compare Figs. 5C with 12C). The external, spine-like PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON seta of the third (middle) exopodal segment of R. klausruetzleri reaches slightly beyond the second seta of the first (distal) segment, while in R. marki this spine reaches to the edge of the distal spine-like seta of the first segment. The right basis of the male P5 of R. klausruetzleri has a set of long sensilla proximally and a set of short denticles dis- tally; males of R. marki (compare Figs. 5F and 12D) havea pad-like thickening distally but no proximal sensilla. The right endo- podal seta is located on the distal half of the outer margin in R. klausruetzleri but on the proximal half in R. marki. The external, spine-like seta on the second (proximal) seg- ment of the left exopod of R. klausruetzleri reaches to the end of the well-developed outer tip of the third (middle) segment; in R. marki this spine reaches beyond the poorly developed tip (Fig. 12D). Ecological notes. —No other calanoid co- pepods were collected in the swarm of co- pepodids of R. klausruetzleri; no nauplii were present. The copepods appeared col- orless in the water, but after capture all stages appeared red in direct sunlight. Examina- tion of live specimens with a dissecting mi- croscope indicated that pigment was con- centrated at margins of body somites. This color pattern is not apparent in preserved specimens today. Number (and percentage) of each copepodid stage of R. klausruetzleri in the swarm were as follows: CI— 528 (8%); CII— 505 (8%); CIII— 403 (7%); CIV—319 (5%); CV—636 (11%); CVI—3687 (61%). This aggregation like other calanoid swarms (Kimoto et al. 1988) includes CI, and con- trasts with swarms of the cyclopoid copepod Dioithona oculata (Farran, 1913) which sel- dom include this stage (Ambler et al. 1991). Percentage of females in the last three stages was CIV— 68%; CV—68%; CVI— 74%; fe- males and males of the CIV and CV cannot be determined without dissection so per- centages for those stages were estimated from a subsample of 40 specimens. Identities and homologies.—Ferrari & Steinberg (1993) have noted for Scopalatum VOLUME 108, NUMBER 2 7 Fig. 12. Ridgewayia marki (Esterly, 1911a), CVI female: A, genital complex ventral; B, genital complex lateral; C, PS posterior, exopod detached. CVI male: D, P5 posterior; E, left exopod 1 and 3 of P5 anterior. Line 1 = 0.1 mm. 198 vorax (Esterly, 1911b) that the setation of the distal part of maxilla 2 is complete by CII. This early completion of setation is similar to the setal development of the ex- opods of antenna 2 and mandible, but it is unlike any known endopodal pattern for an- terior copepod appendages. Thus the distal part of maxilla 2 is considered an exopod. The setation of R. klausruetzleri develops similarly. The sixth lobe of maxilla 2 is con- sidered a lobe of the basis and not an en- dopodal segment (Huys & Boxshall 1991). It is not an endopodal segment because its setation is complete at CII; it is not an ex- opodal segment because, like the other en- ditic lobes, it bears more than | seta. New endopodal segments of the maxil- liped (segments 3-5) are formed within the penultimate segment (Figs. 10G & 11F); the antepenultimate segments of CII, CIII, and CIV are the new ones. The distomedial lobe of the basis, which is antepenultimate at CI, adds its last seta at CII. This early comple- tion of setation proceeds identically to that of the syncoxal lobes and non-ramal seg- ments of other appendages, and suggests that this lobe is part of the basis (Von Vaupel Klein 1982) and not a sixth endopodal seg- ment (Huys & Boxshall 1991). Each new endopodal segment of the max- illiped initially possesses a single seta (Table 1), but setae also are added to non-terminal segments (including the new segments) at CIV, CV, and CVI so that there are two kinds of setae on the new segments: one seta formed as the segment is formed and several setae added after the segment has formed. On the remaining thoracopods, the inner seta of the second (proximal) exopodal seg- ment of thoracopods 2—5 (swimming legs 1-4) is the only seta added after segmen- tation. Segmentation of Pl—4 appears to follow the common pattern of development which is presumed to be ancestral for copepods (Ferrari 1988). Pl and P2 develop similarly. Although the morphology of the last nau- pliar stage is unknown, PI and P2 are re- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON organized appendages with 1-segmented rami at CI; these rami gain a second segment at CII, after which there is no further seg- mentation until CV. P3 develops serially and similarly, but one stage out of register with Pl and P2. It is a bud at CI, is reor- ganized at CII, and adds its second segment at CIII. P4 develops serially, but one stage out of register with P3; it is a bud at CII, is reorganized at CIII, and adds its second seg- ment at CIV. Pl-4 add their third ramal segments simultaneously during the molt to CY. The number of outer setae on the third (middle) exopodal segment of P1 in species of Ridgewayia has been interpreted as one or three elements (Wilson 1958, Humes & Smith 1974, Yeatman 1969). Most cope- pods have one external seta on the third (middle) exopodal segment. Ridgewayia klausruetzleri has one seta (an articulating element whose thick base narrows abruptly; this morphology is identical to that of the other setae on the swimming leg segments); the two structures medial to the seta do not articulate and instead are complex attenu- ations of the segment. This interpretation is clearer at CV because generally there is no change in the number of setae from CV to CVI. In CV R. klausruetzleri there is one abruptly narrowing seta and two simple at- tenuations of the segment (Fig. 6F); the lat- ter two presumably will become the com- plex attenuations of CVI. The left and right endopods of the male P5 are one-segmented, but they develop by different patterns. At CIV the left endopod is an unarmed segment; neither segments nor setae are added during later develop- ment. At CIV the right endopod is one-seg- mented with four setae. A second, unarmed segment is added proximally at CV and there is no change in setation. At CVI the endo- pod is again one-segmented and just one seta is present. The development of the oth- er three setae and the arthrodial membrane between the two segments present at CV presumably is repressed. VOLUME 108, NUMBER 2 Setae also are lost during the exopodal development of male P5. Both the left and right exopods are two-segmented with sev- en and one setae at CV. At CVI five setae are lost from the first (distal) segment of the right exopod. Two are lost from the first (distal) segment of the left exopod as one seta becomes associated with the new, third (middle) segment. Etymology. —This species honors Dr. Klaus Ruetzler for his continued support of, and contributions to coral reef research. Acknowledgments Dr. Mark Grygier, Silver Spring, Mary- land, and five reviewers improved this ar- ticle. This is contribution number 440 of the Caribbean Coral Reef Ecosystems Pro- gram. Literature Cited Ambler, J., F. Ferrari, & J. Fornshell. 1991. Popu- lation structure and swarm formation of a cy- clopoid copepod, Dioithona oculata, near man- grove cays.—Journal of Plankton Research 13: IDS aNA 2: Esterly, C. 191la. Calanoid Copepoda from the Ber- muda Islands.— Proceedings of the American Academy of Arts and Sciences 47:219-226 +4 plts. . 1911b. Third report on the Copepoda of the San Diego region.—University of California Publications in Zoology 6:313-352. Farran, G. 1913. Plankton from Christmas Island, Indian Ocean. II. On Copepoda of the genera Oithona and Paroithona.—Proceedings of the Zoological Society of London, 1913:181-193. Ferrari, F. 1988. Developmental patterns in numbers of ramal segments of copepod post-maxillipedal legs. —Crustaceana 54:256-293. —, & J. Ambler. 1992. Nauplii and copepodids of the cyclopoid copepod Dioithona oculata (Oi- thonidae) from a mangrove cay in Belize. — Pro- ceedings of the Biological Society of Washington 105:275-298. —., & D. Steinberg. 1993. Scopalatum vorax (Es- terly, 1911) and Scolecithricella lobophora Park, 1970 calanoid copepods (Scolecitrichidae) as- sociated with a pelagic tunicate in Monterey Bay.— Proceedings of the Biological Society of Washington 106:467-489. Fosshagen, A. 1970. Marine biological investigations 199 in the Bahamas 15. Ridgewayia (Copepoda, Cal- anoida) and two new genera of calanoids from the Bahamas. —Sarsia 44:25-58. —, & T. Iliffe. 1991. A new genus of calanoid copepod from an anchialine cave in Belize. — Bulletin of the Plankton Society of Japan, Spe- cial Volume, pp. 339-346. Gurney, R. 1927. Report on the Crustacea: —Cope- poda (littoral and semi-parasitic). Zoological re- sults of the Cambridge expedition to the Suez Canal, 1924, no. 35.—Transactions of the Zoo- logical Society of London 22:451-577. Hulsemann, K. 1991. The copepodid stages of Dre- panopus forcipatus Giesbrecht, with notes on the genus and a comparison to other members of the family Clausocalanidae (Copepoda Calan- oida). — Helgolander Meeresuntersuchungen 45: 199-224. Humes, A., & W. Smith. 1974. Ridgewayia fosshag- enin. sp. (Copepoda; Calanoida) associated with an actiniarian in Panama, vith [sic] observations on the nature of the association.— Caribbean Journal of Science 14:125-139. Huys, R., & G. Boxshall. 1991. Copepod Evolution, The Ray Society, London, 648 pp. Kimoto, K., J. Nakashima, & Y. Morioka. 1988. Di- rect observations of copepod swarm in a small inlet of Kyushu, Japan.— Bulletin of the Seikai Regional Fisheries Research Laboratory 66:41- 58. MacIntyre, I., M. Littler, & D. Littler. 1989. Sub- merged fractured peat, Tobacco Range, Belize: biological and geological studies of a unique ma- rine habitat. Abstract, p. 97 from Colloque Biol- ogie et Geologie des Recife Coralliens, 1989 An- nual Meeting of the International Society for Reef Studies, 180 pp. Othman, B., & J. Greenwood. 1988. A new species of Ridgewayia (Copepoda, Calanoida) from the Gulf of Carpentaria.— Memoirs of the Queens- land Museum 25:465-469. Thompson, I., & A. Scott. 1903. Report on the Co- pepoda collected by Professor Herdman, at Cey- lon, in 1902.—Ceylon Pearl Oyster Fisheries, Supplemental Report 7:227-307. Ummerkutty, A. 1963. Studies on Indian copepods— 7. On two calanoid copepods, Ridgewayila typica Thompson & Scott and R. Krishnaswamyi N. sp.[sic].— Bulletin of the Department of Marine Biology and Oceanography, University of Ker- ala 1:15-28. Vaupel Klein, J. von. 1982. A taxonomic review of the genus Euchirella Giesbrecht, 1888 (Cope- poda, Calanoida). II. The type-species, Euchi- rella messinensis (Claus, 1863). A. The female of f. typica. —Zoologische Verhandelingen, Lei- den, 198:1-131 + 23 plts. 200 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Wilson, M. 1958. A review of the copepod genus Yeatman, H. 1969. A redescription of copepod, Ridgewayia (Calanoida) with descriptions of new Ridgewayia marki, with description of an un- species from the Dry Tortugas, Florida. —Pro- usual specimen.—Journal of the Tennessee ceedings of the United States National Museum Academy of Science 44:710. 108:137-179. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(2):201-206. 1995. Sinoniscus cavernicolus, a new genus and species of terrestrial isopod crustacean from a cave in China (Styloniscidae: Oniscidea) George A. Schultz 15 Smith St., Hampton, New Jersey 08827-9531, U.S.A. Abstract. —This is the first Oniscidea or terrestrial isopod crustacean recorded from a cave in the Peoples Republic of China. The cave where the new species of a new genus of Styloniscinae of Styloniscidae was collected is located near Guilin in Guangxi Province. The species is blind, pigmentless and well adapted to live in a cave. Kwon & Taiti (1993) in their review and comprehensive account of 49 species of Oniscidea from southern China included no species of Styloniscidae (or of related Tri- choniscidae) and no oniscidean specifically from a cave. This is the first record of a troglobitic oniscidean from China. The new genus is in Styloniscinae of Styloniscidae. The specialized oniscidean can be added to Kwon & Taiti’s 23 “‘endemic species”’ from China (pg. 80). Kwon & Taiti described two blind Philosciidae, neither from a cave— Papuaphiloscia granulata and P. arcangelii. Sinoniscus, new genus Diagnosis. — Blind. Pigmentless. Frontal line distinct; supra-antennal line medially rounded (Fig. 1B). Three flagellar articles; middle article almost two times as long as proximal article; tiny apical article tipped with several long setae. Pereopods without marked sexual dimorphism. Pleon narrow- er than pereon with all neopleurons laterally appressed. Endopod of pleopod | with strong muscular attachment (Fig. 1E). Female with exopods of pleopods of segment | and of segment 2 of pleon firmly attached in single scale-like structure fitting across width of each segment (Figs. 3A,B). Derivation and gender of name. —“*Sino-” is a prefix for Chinese to which “-oniscus” is attached referring to an oniscidean from China. Masculine. Type species.—Sinoniscus cavernicolus, new species. Type by original designation. Affinities. —The new genus is in Stylon- iscidae because it has conspicuous, strong muscles connected to upper part of endopod of male pleopod | (Cf 1E and Vandel 1952: 11, fig. 6). The genital apophysis, although quite narrow, has a tiny apical cone or pro- jection. The left mandible has 2 penicillate setae and the right has only one. The char- acters above are those on which Vandel (1952:95) defined Styloniscidae. However, Vandel stated that Styloniscidae had a “‘téte de type trichoniscien,” or a head of the tri- choniscid type. Vandel (1960:137) stated that in trichoniscids “‘la ligne frontale n’est pas differenciée (disparition probablement du a une regréssion).”” The cephalon of the type of the new genus clearly has a well de- fined frontal line (Fig. 1B). The presence might represent the primitive state and per- haps it is regressed in contemporary forms, however, more knowledge about the shape of the cephalon of already known species must be obtained since most species of Sty- loniscidea (and of related Trichoniscidae) are described only on differences in mor- phology of mouth parts and male pleopods. 202 The new genus is in Styloniscinae of Sty- loniscidae because the dorsum is smooth and all neopleurons are appressed laterally on the pleon. All other characters of the new genus are more closely related to species of Styloniscus as described by Vandel (1952) who reviewed many species of that genus. Remarks.—Schmalfuss (1989:21), after presenting a cladogram of family and higher than family clads of Oniscidea, stated that he could not find any characters which could be used to keep Styloniscoidea and Tri- choniscoidea distinct. The superfamilies are separated essentially on the same characters as are the nominate families so species of Styloniscidae are closely related to those in Trichoniscidae. Most genera of trichonis- cids with species adapted to cave life are in Vandel’s (1960) Premiére Division (now Tribe) of Trichoniscinae. However, Sinon- iscus (without reference to the muscles of the male pleopods or to the frontal line on the cephalon) fits most closely into Vandel’s “Deuxieme Division’ with exception that the elongate shaft (tige) projecting from the en- dopod of male pleopod 1 is not plumose as stated in Vandel’s definition of the Deux- iéme Division (pp. 138, 151). Sinoniscus 18 not related to the blind, monotypic Thailandoniscus Dalens (1989), a cave adapted styloniscid inhabiting water in caves in Thailand, as comparisons of the cephalons and male pleopods 2, among oth- er things, immediately suggest. The new species described here was found on the mud floor of the cave. Dalens (1989:6) stated that the difference between Trichoniscidae and Styloniscidae is “‘d’ordre quantitatif et non qualitatif’ and the families are distin- guished solely “sur l'appareil male.’ Tri- choniscidae contains about 86 valid genera and Styloniscidae contains about 11 valid genera. Five genera of Trichoniscidae based on only blind species were discussed by Schultz (1994). The two families must be reevaluated to determine if they can be merged into a monophyletic unit. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Sinoniscus cavernicolus, new species Figs. 1-3 Description.— About two and one half times longer than broad, dorsum smooth. Anterolateral lobes large (Fig. 1B). Antenna 1 with article 3 longest and tipped with sev- en aesthetacs, apical one longest. Antenna 2 relatively short with long seta on distal peduncular segment. Right mandible with well developed molar; one seta and lacinia mobilis tipped with tiny teeth between mo- lar and incisor. Incisor with two large, strong outer teeth with inner smaller teeth. Left mandible with well developed molar with two setae on lacinia mobilis; incisor with few teeth. Exopod of maxilla 1 with four outer teeth, one medial tooth and three in- ner short unnotched teeth; inner margin with tiny, plumose seta (Fig. 10). Endopod of maxilla | tipped with three compound sen- sory setae, apical one knobed. Maxilla 2 narrows apically with rounded sensory edge, fringe of setae on medial margin. Maxilliped with palp of two segments; first short and broad; second with rounded outer margin ending in point, inner or medial margin straight with several paired setae and one large seta with accompanying shorter seta near apex. Endite narrows apically with large apical spine medial to short, setose apical segment. Exopod of maxilliped about one third length of maxilliped proper with rounded apex and setae filled indentation on Outer margin. All pereopods about same length and with simple setae on inner margins; all with scale- like setae on outer margin of propodus and carpus and scale of setae apically on outer margin of carpus and merus; hair-like setae distally on propodus and proximally on dactylus of each pereopod; each dactylus with single claw and brush-like dactylar or- gan. Male pereopod I with short dactylus, propodus with two inner setae and scale- like setae on outer edge; carpus with one especially long seta with three other long VOLUME 108, NUMBER 2 203 Fig. 1. Sinoniscus cavernicolus, new genus, new species: A, Holotype male 6.5 mm long; B, Frontal view cephalon; C, Antenna 1; D, Antenna 2 with detail of flagellum; E-I, Male pleopods 1-5 respectively; J, Uropod; K, Pleotelson; L, Left mandible; M, Right mandible; N, Maxilla 2; O, Maxilla 1. 204 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Sinoniscus cavernicolus: A, Maxilliped with detail of endite; B—E, Male pereopods I, II, VI and VII respectively; F-I, Female pereopods I, II, VI and VII respectively. VOLUME 108, NUMBER 2 A 205 BOCK Fig. 3. Sinoniscus cavernicolus, allotype female: A, B, Pleopods underside segments 1 and 2 respectively; C— E, Exopod pleopods 3-5 respectively. setae On inner margin; few setae on inner margin of merus and ischium. Male pereo- pod II with three long setae on inner margin of propodus; three, one very long, on inner margin of carpus; other segments with few setae on inner margins. Male pereopod VI with three long setae on inner margin; four on carpus; few on other segments. Male pe- reopod VII with three long setae on inner margin; carpus with six long setae, two on outer apical margin; few on other segments. Narrow pleon with neopleurons not de- veloped. Exopod of male pleopod | simple with pointed posterolateral border; endo- pod small tipped apically with long, non- plumose shaft, shaft about twice as long as basal segment (Fig. 1E). Exopod of male pleopod 2 small, about one and one half times as wide as long; endopod elongate with short proximal segment and second elon- gate segment about four times as long as proximal segment; second segment narrows apically, apex rounded (Fig. 1F). Male ple- opods 3 to 5 simple, with marginal setae and one seta on face of exopod. Pleotelson, short, with rounded posterior border, no marginal scales or setae. Uropod with con- ical rami each extending beyond posterior point of pleotelson, both tipped with two long setae. Pereopods of female much like those of male (Figs. 2F—I). Pleopods 1 in single elon- gate structure medially split into two lobes. Pleopods 2 with exopods and endopods nar- row, elongate and medially located. Exo- pods of pleopods 3 to 5 of female much like those of male (Figs. 3A—E). Materials examined. —Nine specimens (3 males and 6 nongravid females). Measurements.— Males 6.5 to 6.8 mm long; females 6.8 to 7.2 mm long. Type locality.—A cave at Taiping Yau, Lin Chuan County (just north of Guilin; Reed Flute Cave, a tourist cave, is nearby), Guangxi Province, Peoples Republic of China. Collected 1 Aug 1993 by D. A. Hub- bard. Ecology.—The species was taken from highly organic mud sediment on the floor of the cave. Distribution. —Known only from the type locality. Derivation of name.—The Latin caverna means “‘cave”’ and -co/us means “dwelling in”’ so cavernicolus refers to its cave dwell- ing life style. Deposition of types. —Holotype male, al- lotype female and paratypes (1 male and 3 females), Institute of Zoology, Academia Sinica, No. 940925; paratypes, 1 male 2 fe- males. National Museum of Natural His- tory, USNM 267281, Washington, D.C. Acknowledgment The author would like to thank Dr. John R. Holsinger of Old Dominion University, Norfolk, Virginia, for sending him the Onis- 206 cidea collected in the Peoples Republic of China. Literature Cited Dalens, H. 1989. Sur un nouveau genre d’oniscoide ‘‘aquatique”’ provenant du sud-est Asiatique: Thailandoniscus annae.—Spixiana 12(1):1-6. Kwon, D. H., & S. Taiti. 1993. Terrestrial Isopoda (Crustacea) from southern China, Macao and Hong Kong.—Stuttgarter Beitrage zur Natur- kunde, Serie A (Biologie) 490:1-83. Schmalfuss, H. 1989. Phylogenetics in Oniscidea.— PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Monitore Zoologico Italiano (Nuovo Serie) Monografia 4:3—27. Schultz, G. A. 1994. Typhlotricholigioides and Mex- iconiscus from Mexico and Cylindroniscus from North America (Crustacea: Oniscidea: Trichon- iscidae).— Journal of Crustacean Biology 14(4): 763-770. Vandel, A. 1952. Les trichoniscides (Crustaces—Is- opodes) de l’hémisphére austral.— Mémoires du Muséum National d’Histoire Naturelle, No- vuelle Série, Série A, Zoologie 6:1-116. 1960. Isopodes terrestres (Premiére Par- tie). Faune de France 64:1-416. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(2):207-211. 1995. Sphaerolana karenae, a new species of hypogean isopod crustacean from Nuevo Leon, Mexico Gabino A. Rodriguez-Almaraz and Thomas E. Bowman (GAR-A) Facultad de Ciencias Biologicas, Universidad Autonoma de Nuevo Leon, Ap. postal 105-F, Ciudad Universitaria, San Nicolas de los Garza, Nuevo Leon, Mexico; (TEB) Department of Invertebrate Zoology, NHB-163, Smithsonian Institution, Washington, D.C. 20560, U.S.A. Abstract. —Sphaerolana karenae is described from two localities in Nuevo Leon, Mexico. It is the third known species of Sphaerolana Cole & Minckley, 1970 and differs from the two previously known species in having three rather than two pleonites visible dorsally, in having the telson pointed rather than rounded, and in having only one rather than two retinacula on the endite of the maxilliped. Resumen. —Sphaerolana karenae es descrita de dos localidades de Nuevo Léon, México. Esta especie es la tercera conocida para Sphaerolana Cole & Minckley, 1970 y difiere de las dos especies previamente conocidas en poseer tres pleonitos dorsalmente visibles en vez de dos, el telson termina en punta mas que redondeada, y ademas, de tener solamente un retinaculum y no dos sobre el endito del maxilipedio. In 1970 Cole & Minckley described the very distinctive troglobitic isopod genus Sphaerolana, from the Cuatro Ciénegas ba- sin of central Coahuila state, Mexico. The two species included, C. interstitialis (type species) and C. affinis, were characterized by their ability to roll into a ball, pleons with only 2 free pleonites, and uropods with tiny rami inserted in a lateral notch of the protopod. We describe herein a third spe- cies from the neighboring state to the south- east, Nuevo Léon. Sphaerolana karenae, new species Figs. 1—23 Sphaerolana affinis. —Reddell, 1981:87-88 [Misidentification provided Reddell by T. E. Bowman.] Material. — Holotype 6, UANL (Univer- sidad Autonoma de Nuevo. Leon) CCRICIOO!, and 3 paratypes, UNALCC- RICOO2, from small spring adjacent to Pi- lon River, Montemorelos, Nuevo Leon, Mexico, about 25°27’N, 99°48’W, 20 Jun 1988 and 17 May 1990, leg. M. Valdéz-Mar- roquin and C. Rodriguez-Almaraz.— 3 paratypes on wood in mine below Cueva de la Boca, Santiago, Nuevo Léon, about 25°27'N, 100°06'W, 22 May 1971, leg. S. Peck, deposited in National Museum of Natural History, USNM 139138. Description. —Length of holotype 7 mm, of largest paratype from USNM 139138 9.7 mm. Head almost 2 x as wide as long, with- out rostrum; frontal lamina about as wide as long; clypeus 4.2 x as wide as long. Pleon with 3 free pleonites; pleonite | longest; pleonite 3 very short and inconspicuous. Pleotelson about as wide as long, expanded laterally anteriorly, ending in rounded point. Antenna-1l peduncle 3-segmented; flagel- lum 9—13-segmented; single esthete on 5th through subapical segments. Antenna-2 pe- 208 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Figs. 1-11. Sphaerolana karenae, 6 from Montemorelos (except 2'): 1, Habitus, dorsal; 2, Antenna 1; 2’, Antenna 1, é from Santiago; 3, Antenna 2; 4, Telson; 5, Right Mandible; 6, Maxilla 1; 7, Maxilla 2; 8, Maxilleped; 9, Pereopod 1; 10, Pereopod 2; 11, Pereopod 3. VOLUME 108, NUMBER 2 209 Figs. 12-16. 1, 6. 15-16, Sphaerolana karenae from Santiago: 15, Left uropod, dorsal; 16, Rami of left uropod, dorsal. duncle 5-segmented, segment 5 longest; fla- gellum 12—15-segmented, reaching pereon- ite 4 when reflexed. Right mandible spine- row process with 11 spines; molar with 15 curved spines; palp segment 2 with 9 pec- tinate lateral setae and setules along entire medial margin; segment 3 with 11 pectinate setae on lateral margin, distal 2 stouter than others. Maxilla-1 proximal endite with 3 long plumose setae and 2 shorter naked se- tae; distal endite with 11 stout dentate spines. Maxilla-2 proximal endite medially pubescent, with 5 long plumose setae and 4 shorter setae; distal endite bilobed, inner lobe with 6 setae, outer lobe with 3 setae. Maxilliped endite with single retinaculum, 4 stout plumose setae, and 4 setules; palp segments 3 and 4 produced medially into weak lobes, outer margin of segment 4 with a stout seta, segment 5 with 9 long setae and numerous setules. Pereopods 1-3 with longer dactyls than those of pereopods 4—7. Pereopod 1 dactyl 12-14, Sphaerolana karenae from Montemorelos: 12 pleopod 3, 6; 13, Pleopod 2, 6; 14, Pleopod with a few short setae on anterior and pos- terior margins. Merus, carpus and propus with 3,3,2 stout toothed spines in pereopod 2; 2,3,2 spines in pereopod 3. Pleopod 1 protopod with 3 retinacula; ex- opod elongate, armed with 18 terminal and subterminal plumose marginal setae and row of setules on medial margin. Pleopod 2 pro- topod with 4 retinacula, 2 plumose setae, and 4 setules; exopod with 15 terminal and subterminal plumose marginal setae; ap- pendix masculina reaching beyond endo- pod to tip of exopod. Pleopods 3-5 rami unarmed; exopods with partial transverse sutures. Pleopod 3 protopod with 4 setae and 5 setules; rami subequal. Uropod pro- topod with straight medial margin and broadly rounded apex; rami inserted in notch near apex. Etymology. —Named for the first author’s daughter Ana Karen. Comparisons.—Table 1 compares taxo- nomic characters of the three known species 210 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Figs. 17-24. 17-18, Sphaerolana karenae from Santiago: 17, Buccal area; 18, Right pereopod 1, dactyl. 19- 24, Sphaerolana karenae from Montemorelos: 19, Right pereopod 4; 20, Right pereopod 5; 21, Right pereopod 6; 22, Right pereopod 7; 23, Right pleopod 4; 24, Right pleopod 5. VOLUME 108, NUMBER 2 Table 1.—Comparison of species of Sphaerolana. Ziel interstitialis affinis karenae No. of pleonites demarcated dorsally Dy Y, 2 Telson apex broadly rounded broadly rounded rounded point Pleotelson L/W 0.90-0.95 1.1-1.2 0.96 Up protopod: length to notch —— 0.74 0.88 0.85 length to apex A-1 flagellar segs. 10-17 10 8-12 A-1 reaches pereonite 3 pereonite 4 pereonite 4 A-2 flagellar segs. 15 16 12 Md lacinia teeth 9-16 v 11 Md molar teeth 10-14 & 15 Mxp retinacula ) 2) 1 Body length (mm) up to 17.1 up to 22.1 up to 9.7 of Sphaerolana. Sphaerolana karenae is easily distinguished from its two congeners by its three free pleonites, pointed pleotel- son, and single retinaculum on the maxil- liped endite. Cole & Minckley’s 1970 diagnosis of Sphaerolana must be emended by changing the third sentence to read, “‘Pleonites 1—2 or 1—3 present, visible dorsally; no evidence of either pleonites 3-5 or 4—5”. Acknowledgments Part of this work was carried out in Wash- ington by the first author during his tenure of a Short-term Visitor Award from the Smithsonian Institution’s Office of Fellow- ships and Grants. We give thanks to Pro- fessor Ernesto Campos for his general com- ments and a critical review of the manu- script. Literature Cited Cole, G. A., & W. L. Minckley. 1970. Sphaerolana, a new genus of cirolanid isopod from northern Mexico, with description of two new species. — Southwestern Naturalist 15(1):71-81. Reddell, J.R. 1981. Areview of the cavernicole fauna of Mexico, Guatemala and Belize.— Bulletin of the Texas Memorial Museum 27:1-327. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(2):212-219. 1995. Natatolana nukumbutho, a new species (Crustacea: Isopoda: Cirolanidae) from deep water off Suva, Fiji Niel L. Bruce and Jorgen Olesen Zoologisk Museum, University of Copenhagen, Universitetsparken 15, DK 2100, Kebenhavn, Denmark Abstract. — Natatolana nukumbutho, a new species, is described from spec- imens taken by baited trap at a depth of 470 m, off Suva, Viti Livu, Fiji. The species is characterized by having acute posterolateral margins on pleonites 3 and 4, entire sutures on all coxae, the appendix masculina slender and shorter than the endopod, and by the spine details of the pleotelson and uropods. The marine isopod fauna of Fiji is poorly known, with only a few records in the recent literature (e.g., Bruce 1984, 1995a; Bruce & Harrison-Nelson 1988, Kensley 1979). This present contribution adds one more species to the few already known from the area. Specimens are deposited at the Zoological Museum, University of Copenhagen (ZMUC), and one pair of paratypes at the Australian Museum, Sydney, Australia. Ab- breviations: BL—body length; CPS—cir- cumplumose setae; PMS—plumose mar- ginal setae; PS—plumose setae. Taxonomy Family Cirolanidae Dana Genus Natatolana Bruce, 1981 Remarks.—The cirolanid isopod genus Natatolana is one of the largest of the family with over 60 named species, and numerous undescribed species. It has been recorded from all the world oceans, from shallow wa- ter, rarely intertidal, to depths in excess of 2000 meters (Bruce 1986, Brusca et al. 1995). There are, however, no records of the genus from the central South Pacific, and very little collected material exists from this region. The description of the new spe- cies below establishes the presence of the genus in the central South Pacific Ocean. Natatolana nukumbutho, new species Figs. 1-4 Material examined.—Holotype: 6 (12.3 mm), off Lauthala Bay, Suva, Fiji, 18°13’S, 178°29’E, 23 Jan 1981, 470 m, in baited trap, coll. Institute of Marine Resources, University of the South Pacific (ZMUC CRU-580). Paratypes: 2 66 (11.4, 13.2 [dissected] mm), 11 2 (non-ovig 12.2, 13.8, 14.0, 14.0, 14.2, 14.5, 15.5, 16.0, 16.5, 16.8, 17.0 mm), same data as holotype (ZMUC CRU-581, 582; 1 6, 1 2 AM P43423). Description. —Male: Body about 2.8 times as long as greatest width; maximum width at pereonites 4—6. Cephalon with submar- ginal anterior suture, entire anterior and dorsal interocular sutures present, dorsal suture medially weak. Eyes large, rectan- gular, 1.8 times as long as wide; ocelli dis- tinct, 12-13 horizontally, 8-9 vertically. Pereonite 1 with 2 fine lateral sutures, dorsal suture not extending to posterior margin of pereonite; coxae all with entire carina, pos- terolateral angles of coxae 6 and 7 acute. Pleon 14.3% BL, pleonites 2-4 each with lateral longitudinal suture; pleonite | lateral margins not produced, those of 3 and 4 pro- duced and acute, dorsal and ventral margins of pleonite 4 both convex. Pleotelson slight- ly longer than pleon, about 16.7% BL, 80% VOLUME 108, NUMBER 2 213 Fig. 1. Natatolana nukumbutho, new species. A-E holotype, remainder 6 paratype 13.2 mm. A, dorsal view; B, lateral view; C, cephalon, dorsal view; D, frons: E, detail of anterior cephalic sutures; F, pleonites, lateral view; G, antennule; H, antenna; I, left mandible; J, maxillule. Scale 2.0 mm. 214 as long as greatest width, shield shaped, with indistinct narrow and shallow anterodorsal depression which lacks a median notch; posterior margin 12 spines set among PMS, with single seta anterior to anterior-most spine; apex not produced. Antennule peduncle article 2 about half (0.57) as long as 1, article 3 longest, articles 1 and 2 combined 1.3 times as long as article 3; article 2 with prominent stout broad based pappose seta at ventrodistal margin, nearly as long as and extending near to distal mar- gin of article 3; article 3 with 1 plumose sensory and 3 simple setae at anterodistal angle; flagellum slightly shorter (0.9) than peduncle, with 12 articles, extending to pos- terior margin of eye, proximal article lon- gest. Antenna peduncle articles 1 and 2 short, their combined length slightly less (0.88) than article 3, article 5 1.6 times as long as article 4, posterodistal angle with 2 densely plumose sensory setae; flagellum with 21 articles, extending to pereonite 4. Frontal lamina 3.4 times as long as basal width, ventrally flat, lateral margins weakly concave, anterior margin rounded, slightly wider than posterior margin. Mandible spine row with 15 spines; molar process with lat- eral fold, with about 28 spines along ante- rior margin. Maxillule as for others of the genus. Maxilla lateral lobe with 5 setae, middle lobe with about 17 in 2 ranks, 10 ventral and 7 dorsal, medial lobe with about 13 CP and distally 3 simple spines. Max- illiped palp articles 2—5 with both margins setose, those of lateral margins being longer than those of mediai margins; endite with 2 coupling hooks and 6 long CPS. Pereopods 1-3 without conspicuously elongate spines on posterior margins. Pe- reopod | basis with cluster of about 8 long slender simple setae at posterodistal angle, anterior margin 10 widely spaced long slen- der simple setae; ischium anterodistal mar- gin with about 15 long slender simple setae, distomedial lateral margin with row of 7 simple setae, posterior margin with 9 stiff simple setae; merus anterodistal lobe with PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON prominent apical spine and lateral row of about 14 long setae medially with about 5 setae, posterior margin sinuate, with 10 spines (2 of which are medial); carpus with 1 small and 1 large acute spines at postero- distal angle; propodus about 3 times as long as wide, with 4 acute spines on palm be- coming progressively larger distally, prom- inent unornamented blunt spine opposing dactylus, anterodistal angle with cluster of 3 stiff setae, further medial cluster of setae at posterodistal margin; dactylus slender, about 0.75 as long as propodus and about 5.2 times as long as basal width. Pereopod 2 with more numerous, larger and more ro- bust spines than pereopod 1 except for pro- podal palm which has only 2 small acute spines; ischium anterodistal margin without spines, posterodistal margin with 3 acute spines; merus anterodistal margin with 7 stiff setae and 6 stout spines, distally with further 3 long slender simple setae; carpus posterodistal margin with 2 long acute and 4 short acute spines. Pereopods 5-7 similar in morphology, pereopod 4 intermediate between 3 and 5. Pereopod 7 basis broad, about half as wide as long, anterior margin proximal three-quarters straight, with con- tinuous PS, posterior margin nearly straight, without setae, medial carina with PS along entire length, distal angle with about 20 long PS and laterally a further 6 shorter PS; is- chium posterior margin with 5 evenly spaced single acute spines and about 16 long PS, spines at distal angle, anterodistal angle with 2 larger acute and 4 small biserrate spines, mediodistal margin with 2 acute spines; merus posterior margin with 2 short acute spines and 5 slender finely serrate setae, an- terodistal angle with 2 spines and 2 setae, posterodistal angle with cluster of about 9 spines and 5 slender finely serrate setae; car- pus posterior margin with 2 short acute spines and a large cluster of about 4 long and short spines and 3 finely serrate setae at distal angle, anterodistal angle with clus- ter of about 10 long acute spines several of which are finely biserrate; propodus poste- PA) VOLUME 108, NUMBER 2 Fig. 2. Natatolana nukumbutho, new species. Figures of 6 paratype 13.2 mm, except C, holotype. A, maxilla; B, maxilliped; C, pleotelson posterior margin; D, pereopod 1; E, pereopod 2. 216 Vy Zs A }) 4 A un kA i <—SS SS aera (\ INN Nt SS Qs SSS —/ ) SS SSS SSSSSS . See HE Ress Fig 1. enlarged. Mx2 with proximal endite composed of 2 straplike lobes, distal endite subtriangular, elongate and flanked laterally by slightly curved straplike palp; scaphognathite great- ly expanded, anterior subovate lobe with densely setose margin bearing longest setae along distomesial sector, stout posterior subtriangular lobe fringed on distomesial margin with long strong somewhat tangled and stiffened setae. Mxpl with prominent 2-lobed endite; palp rather stout, bearing terminal and sub- terminal tufts of setae; exopod symmetri- cally rounded and setose distally; large epi- pod foliaceous, bilobed. P1 rather stout, reaching (if extended) be- yond antennal peduncle by at least length of fingers but normally folded on itself at mero-carpal joint at right angle or diagonal to axis of leg, with chela and carpus oriented toward midline. Outline of chela shaped like “bird-head with bent beak”’ (flamingo-like); palm inflated and smooth. Fingers curved and closing without gape; flexor surfaces 231 —————— SS arta —— = ~~. = o on Ke a, a; oe ote RQ b Opaepele loihi, new genus and species. Female holotype: a, lateral view, scale = 5 mm; 5, rostrum concave; extensor surfaces of each finger convex, opposed edges uniformly offset, each armed with row of almost uniform, minute, erect, closely set setae; acute tip of each fin- ger slightly spooned by curved setal row, that of dactylus with elongate teeth slanting distad and curving around its distal end; line of curved sensory setae mesial to each cut- ting edge; dactyl broadened at base for ar- ticulation with palm but uniformly nar- rowed throughout most of length to about ’; width of fixed finger. Carpus inflated, ir- regularly funnel-shaped, extensor surface bent almost at right angle near tapered prox- imal end articulating with merus, distal margin irregularly flared, cupped to con- form with shape of palm in flexed or ex- tended position, strong tooth on flexor sur- face with patch of setae on its external sur- face. Merus club-shaped but somewhat flat- tened, submarginally concave along flexor surface for reception of flexed carpus. P2 slightly more slender than other legs, extending to near distal end of antennal pe- 232 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Opaepele loihi, new genus and species. Female holotype: a, anterior region (dorsal view); b, telson and uropods (dorsal view); c, right first pereopod. Paratype female: spine between coxae of fifth pereopods, d, ventral view, e, oblique lateral view, left side. Paratype male: appendix masculina, f lateral, g, mesial. Scales: 1 (a—b); 2 (c); 3 (d-e) = 1.0 mm; 4 (~g) = 0.1 mm. duncle; chela and carpus about equal in length; fingers, slightly longer than palm, similar in size and shape, opposed edges without gape, each pectinate with single row of short spines directed obliquely distad and increasing slightly in size to end in notice- ably stronger terminal spine crossing op- posite member when closed. P3-—5 similar in structure but increasing in relative length from p3 to p5, p3 reaching well beyond tip of antennal peduncle by full length of propodus and dactylus; dactylus of each essentially equal in length and armed with 6 strong curved spines on flexor sur- face, distalmost strongest; propodi increas- ing in length from p3 to pS, that of p5 twice length that of p3, that of p4 intermediate in length, each armed with setae scattered along flexor surface, mainly on distal '2 and most dense on distal 3, prominent slender setae along extensor surface most prominent dis- tally and in dense distal tuft; respective length of carpi, meri and ischia on p3, p4, p5 respectively similar in length, carpi dis- 233 VOLUME 108, NUMBER 2 rr ZZ VA / A A A XZ |. tow Fig. 3. Opaepele loihi, new genus and species, female paratype: a, right mandible; 5, right first maxilla; c, same, tip of palp, internal view; d, right second maxilla; e, right first maxilliped (ventral view); f, right second maxilliped; g, third maxilliped, mesial oblique view. Scales: 1 (d—g); 2 (a—c) = 1 mm. 234 tally extended in dorsal “‘stop”’ overhanging base of propodi to prevent hyperextension. Posterior thoracic somite (XIV) with por- tion of sternite posterior to base of fifth cox- ae smooth, slightly inclined anteriorly to- ward base of fifth coxae and shallowly con- cave mesially, that portion of sternite lying posteromesial to each fifth coxa drawn me- sially from each side into flattened, weakly calcified median process, tapered, distally subtruncate, and bearing small acute spur anteriorly near base; process often obscured by appressed fifth coxae. Pls 2—5 with appendix interna on mesial ramus slightly increasing in size from an- terior to posterior in series, well developed coupling hooks only on appendix of pl 5. Small appendix masulina on mesial ramus of pl 2 in males bearing 4 terminal setae. Uropods with both rami elongate oval, subequal in length, exceeding distal end of telson, outer ramus wider than inner ramus, with sinuous diaresis and movable spine be- neath rather broad, asymmetrically acu- minate distolateral tooth. Measurements in mm.—Holotype 2, cl 11.5, scl 10.1, al 21.8; allotype 4, cl 13.6, scl 11.9, al 21.8; paratype 2 251450, cl 13.1, scl 11.3, al 23.0; paratype 6 251449, cl 7.9, scl 6.7, al 13.8. Paratypes 2, nm = 26, scl 9.1-12.0, ¥ = 10.3. Paratypes ¢ that can be measured, n = 4 scl 6.8-9.3, ¥ = 6.5. Color.—Color when collected, intensely orange (astaxanthin pigment). All preserved specimens examined basically white but variously mottled with deciduous orange to yellowish deposit, densest and most intense on segments of abdomen and tail fan, an- tennular and antennal peduncles, and prox- imal segments of their flagella, and as bands on pereopods, especially on carpi and ad- jacent portions of meri and propodi. This deposit is an accumulation of the iron ox- yhydroxide particles that cover the summit of the seamount. Some of the particles are formed inorganically; others originate from the mat created by the growth and metab- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON olism of iron-oxidizing bacteria (Karl et al. 1989, color figs.). Etymology. —The specific name, a noun in apposition, refers to the seamount from which the species was collected. Remarks. —Holthuis (1993) gave an am- ple discussion of the family makeup of the infraorder Caridea, with particular atten- tion to recent treatments of family realign- ments within the group. Among the latter was recognition by Christoffersen (1989) of a dichotomy among the Bresiliidae that in his view justified elevation of one of these clusters to family status, the Alvinocaridi- dae, which appear to be associated only with hydrothermal vents, cold brine and hydro- carbon seeps (Segonzac et al. 1993). The latter authors, in their exhaustive exami- nation of trophic behavior, accepted this subdivision, acknowledging that many fea- tures of the group are similar to those of the genus Bresilia but noting absence of exo- podites on thoracic appendages posterior to the maxillipeds and reduced eyestalks more or less attached to each other and adjacent regions. However, appealing, the action leaves the thus diminished Bresiliidae as a less unified group than is the family in its undivided state. Since neither of these choices is entirely satisfactory, we accept the more conservative approach of Holthuis (1993), following Chace (1992), in retaining the Bresiliidae as a unified though some- what disparate assemblage. Among species of bresiliids associated with hydrothermal vents and cold brine or hydrocarbon seeps are some shared mor- phological characters, but other features cut across these seemingly neat clusters in con- trasting manners. Opaepele loihi is imme- diately recognizable as one of these bresi- liids by virtue of the characteristic chelipeds shared by all members of the group. Within the group this species seems to occupy an intermediate position between those having a carinate carapace with well developed ros- trum that reaches at least to mid-length of the antennal peduncle and antennal scale VOLUME 108, NUMBER 2 (including Alvinocaris lusca Williams & Chace, 1982, and A. markensis, A. muri- cola, A. stactophila Williams, 1988), and those in which the carapace 1s not carinate and the rostrum is absent (including Rim- icaris exoculata Williams & Rona, 1986), or greatly reduced in length (including Chor- ocaris chacei (Williams & Rona, 1986) and C. vandoverae Martin & Hessler, 1990). Moreover, the rostral teeth of O. /oihi are much smaller and less numerous than in species that have a long rostrum. Viewed dorsally, the lateral rostral margins of O. loihi are triangular in outline and merge im- perceptibly with the orbital margins, but the rostrum of C. vandoverae, broadly oval in dorsal view, has margins that join each or- bital margin at the site of an orbital prom- inence. The anterior margins of the carapace in the sharply rostrate species bear antennal and pterygostomian spines, as does O. loihi; this set of species shares spinose margins on the posterolateral corners of the fourth and fifth abdominal segments. Those species with short rostrum or no rostrum lack these spines. It thus appears that there are two groups of species that have contrastingly spinose or non-spinose sectors on both car- apace and abdomen, but this neat cleavage is compromised by other morphological features. Alvinocaris lusca has a telson with some- times sinuous terminal margin. This margin is convex in other species of the genus, though general breadth of the telson is much greater in R. exoculata, C. chacei and C. vandoverae. The scaphognathite of Mx2 has a greatly expanded or anteriorly subtruncate anterior lobe in Alvinocaris, Rimicaris and Choro- caris species, whereas this element is more rounded and relatively narrower in O. loihi. The relatively broad dactyls of P3-5 in Chorocaris species bear both marginal and plantar rows of setae on the flexor surface whereas those dactyls of the sharp spined Alvinocaris species have only a single row 235 of spines on the flexor surface; between these two extremes, R. exoculata exhibits only marginal rows of spines on the flexor surface of its comparable rather broad dactyls. P12-5 of Rimicaris and Chorocaris spe- cies bear obvious coupling hooks on the ap- pendix internae of the inner ramus; A/v- nocaris species apparently do not bear such hooks, and Opaepele loihi has them well developed only on the appendix internae of ps. A feature possessed by all species dis- cussed above that has not been emphasized in published descriptions is the process on thoracic sternite 5 (somite XIV) lodged be- tween the coxae of p5. This process 1s much more prominent in O. /oihi than among oth- er genera and species in the family. There is considerable variety in development of the process, and to our knowledge only in O. loihi does it bear a small acute spur an- teriorly near its base. The process is present on both males and females. Its function is unknown. Bresiliid shrimps associated with hydro- thermal vents, brine and hydrocarbon seeps are known from depths as great as 3660 m (Martin & Hessler 1990). Only two of these species are reported from depths less than 1000 m, A. stactophila from hydrocarbon seeps at 534 m in the Gulf of Mexico (Wil- liams 1988), and the present species at 980 m. A key to the 12 currently accepted genera of the family Bresiliidae, modified to ac- commodate Opaepele new genus, 1s adapted from the key given by Holthuis (1993). 1. Eyes with distinct dorsal spine on peduncle, overreaching base of cornea. Dactyl of first pereopod longer than fixed finger and with single large tooth on basal part of Cutting edge’... wise sees: Encantada — Eyes without spine on peduncle. Dactyl of first pereopod, if longer than fixed finger, without single large tooth on cutting edge. ..... 2 236 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON . Exopods on first two pairs of per- eopods at most. Telson with three or more dorsolateral spines. First pereopod with ischium and merus distinct Exopods on all five pairs of per- eopods. Telson with three or fewer dorsolateral spines. First pereopod with ischium and merus fused .. 8 First two pairs of pereopods with exopods None of pereopods with exopods Articulation of palm and carpus of first pereopod normal, proximal end of palm articulated with distal end of carpus. Pleurobranch at base of fifth pereopod reduced. Arthro- branchsjabsent\...5 2... 4a) Bresilia Carpus of first pereopod articulat- ing with chela at middle of prop- odus below base of dactyl, so that larger posterior part of propodus projects beyond this articulation. Pleurobranchs associated with all five pereopods. Arthrobranchs present at base of all pereopods, but that of pereopod 5 reduced Agostocaris ee © © © © © © © © © © © © © © ew ew ew ew ee . Carapace with well developed ros- trum bearing dorsal teeth and at least 1 ventral tooth (may be min- ute). Pterygostomian spine pres- ent. Eyes on separate movable stalks variously fused mesially .. 6 Carapace with rostrum absent or visible as depressed angle of fron- tal margin, without any teeth. Pter- ygostomian spine absent. Eyes re- duced, fused mesially . Carapace with compressed ros- trum, dorsal and ventral teeth clearly evident. Eyes on separate movable stalks narrowly fused me- Sially at base ............ Alvinocaris Carapace with rostrum broadly tri- angular in dorsal view, dorsal teeth and ventral tooth minute (some- times absent). Eyes on separate 10. movable stalks rather broadly fused mesially Opaepele Carapace inflated, rostrum absent, frontal margin of carapace con- cave. Stylocerite, scaphocerite and antennular peduncle fitting tightly and forming an operculiform structure. Third maxilliped with 4 long and 2 short articles. .. Rimicaris Carapace not inflated, rostrum present, broadly rounded anteri- orly. Antennules and antennae not forming an operculum. Third maxilliped with 3 long and 2 short articles ya.% 3 .ccteon bathe Chorocaris Antennal scale narrowing distally, blade not overreaching distolateral tooth. Mandible without deep di- vision between incisor and molar processes Antennal scale broad distally, blade Overreaching distolateral tooth. Mandible with moderate to deep division between incisor and mo- lar processes 10 . Mandible with palp. Last three pereopods pseudochelate, 1.e., dac- tyl opposing distal spine of prop- odus, forming structure somewhat resembling a chela. Carapace with- out pterygostomian spine Bo Set Rouen diy SOs ae NES Pseudocheles Mandible without palp. Last three pereopods normal, not pseudoche- late, no long distal spines on prop- odus. Carapace with pterygosto- mian spine present ........ Kirnasia Rostrum reaching to end of anten- nular peduncle. Third abdominal somite forming gibbous cap over base of fourth somite. Third max- illiped with terminal article obliquely truncate distally. Dactyl of first pereopod not semicircular Wass leaner onead «auth, te bei suc cxian Sow enla Lucaya Rostrum not reaching to end of an- tennular peduncle. Third abdom- inal somite not forming gibbous cap over base of fourth somite. VOLUME 108, NUMBER 2 Terminal article of third maxilli- ped lanceolate. Dactyl of first pe- reopod semicircular ........... 11 11. Carapace with supraorbital and branchiostegal spines present. Exo- pods on pereopods 1 to 3. Telson with 5 lateral and 5 distal spines sa coke 66 0 es nee eee Tridiscias — Carapace without supraorbital and branchiostegal spines. Exopods on all pereopods. Telson with 2 or 3 lateral and 3 posterior spines .. Discias Acknowledgments We gratefully acknowledge the assistance of the Captain and crew of the R/V Kila, and DSRV Pisces V operations team, and the staff of the Hawaiian Undersea Research Laboratory. Craig Moyer collected the shrimps, and we appreciate his and David Karl’s participation in studies at Loihi Sea- mount. This research was funded in part by a grant from the National Oceanic and At- mospheric Administration, project # R/OM- 8, which is sponsored by the University of Hawaii Sea Grant College Program SOEST, under Institutional Grant No. NA89AA-D-SG063 from the NOAA Office of Sea Grant and by the NOAA-National Undersea Research Program, Department of Commerce. Our views expressed herein do not necessarily reflect the view of NOAA or any of its sub-agencies. We are indebted to Keiko Hiratsuka Moore for rendering the excellent illustrations, and to B. B. Collette, F. A. Chace, Jr., and R. B. Manning for critical review of the manuscript. This manuscript is Sea Grant publications UNI- HI-SEAGRANT-JC-94-21 and contribu- tion No. 3870 from the University of Ha- wail School of Ocean and Earth Science and Technology. Literature Cited Chace, F. A., Jr. 1992. On the classification of the Caridea (Decapoda).—Crustaceana 63:70-80. Christoffersen, M. L. 1989. Phylogenetic relation- ships between Oplophoridae, Atyidae, Pasi- Dow phaeidae, Alvinocarididae Fam. N.., Bresiliidae, Psalidopodidae and Disciadidae (Crustacea Caridea Atyoidea).—Boletim de Zoologia. Universidade de Sao Paulo 10:273-281. Edmond, J. M., A. C. Campbell, M. R. Palmer, K. K. Falkner, & T. S. Bowers. 1987. Chemistry of low temperature vent fluids from Loihi and Lar- son’s Seamounts.—EOS Transactions Ameri- can Geophysical Union 68:1553-1554. Grassle, J. F. 1986. The ecology of deep-sea hydro- thermal vent communities.— Advances in Ma- rine Biology 23:301-—362. Holthuis, L. B. 1993. The recent genera of the cari- dean and stenopodidean shrimps (Crustacea, Decapoda); with an appendix on the order Am- phionidacea. [ed., C. H. J. M. Fransen & C. van Achterberg], Nationaal Natuurhistorisch Mu- seum, Leiden, Netherlands, 328 pp. Karl, D. M., G. M. McMurtry, A. Malahoff, & M. O. Garcia. 1988. Loihi Seamount, Hawaii: a mid- plate volcano with a distinctive hydrothermal system.— Nature 335:532-535. ——., A. M. Brittain, & B. D. Tilbrook. 1989. Hy- drothermal and microbial processes at Loihi Seamount, a mid-plate hot-spot volcano.— Deep-Sea Research 36:1655-1673. Martin, J. W., & R. R. Hessler. 1990. Chorocaris vandoverae, a new genus and species of hydro- thermal vent shrimp (Crustacea, Decapoda, Bresiliidae) from the Western Pacific. —Contri- butions in Science, Natural History Museum of Los Angeles County, No. 417:1-11. Sedwick, P. N.,G. M. McMurtry, & J. D. Macdougall. 1992. Chemistry of hydrothermal solutions from Pele’s vents, Loihi Seamount, Hawaii. — Geochimica et Cosmochimica Acta 56:3643- 3667. Segonzac, M., M. de Saint Laurent, & B. Casanova. 1993. L’énigme du comportement trophique des crevettes Alvinocarididae des sites hydro- thermaux de la dorsale médio-atlantique. — Ca- hiers de Biologie Marine 34:535-571. (English Transl. IFREMER, 1994. The enigma of the trophic behaviour of alvinocaridid shrimps from hydrothermal vent sites on the Mid-Atlantic Ridge, pp. 1-20.) Williams, A. B. 1988. New marine decapod crusta- ceans from waters influenced by hydrothermal discharge, brine, and hydrocarbon seepage. — Fishery Bulletin, U.S. 86:263-287. , & F. A. Chace, Jr. 1982. A new caridean shrimp of the family Bresiliidae from thermal vents of the Galapagos Rift.—Journal of Crus- tacean Biology 2:136-147. , & P. A. Rona. 1986. Two new caridean shrimps (Bresiliidae) from a hydrothermal field on the Mid-Atlantic Ridge.—Journal of Crus- tacean Biology 6:446-462. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(2):238-246. 1995. Remarks on the taxonomy of Sudanonautes chavanesii (A. Milne-Edwards, 1886) (Brachyura: Potamoidea: Potamonautidae) from Central Africa Neil Cumberlidge Department of Biology, Northern Michigan University, Marquette, Michigan 49855, U.S.A. Abstract.—The fresh-water crab Sudanonautes chavanesii (A. Milne-Ed- wards, 1886) is redescribed from new material which is compared to previous descriptions of the male type from Gabon. The gonopods and mandibles of a male from Cameroon are described and illustrated for the first time. New records of specimens from several museum collections are presented. The spe- cies 1s recognized by a combination of characters of the carapace, chelipeds, mandibles, and gonopod 1. The range of S. chavanesii has been found to be restricted to an area from south Cameroon to Gabon; the species does not occur in West Africa (from Senegal to Nigeria), as previously thought. Sudanonautes chavanesii (A. Milne-Ed- wards, 1886), a fresh-water crab from Cen- tral Africa, has been recognized as a valid species by a number of authors (A. Milne- Edwards 1887, Rathbun 1905, Balss 1929, Chace 1942, Capart 1954). However, more recent authors (Bott 1955, 1959; Monod 1977) have considered this taxon to be a subspecies of Sudanonautes (Sudanon- autes) africanus (A. Milne-Edwards, 1869). Data presented here indicate that S. chay- anesii is indeed a valid species. Further- more, no characters were found to support the inclusion of S. chavanesii in any sub- genus, as has been suggested by other work- ers (Parathelphusa H. Milne-Edwards, 1854 (A. Milne-Edwards 1887, Rathbun 1905), Potamonautes MacLeay, 1838 (Ortmann 1897), or Sudanonautes Bott, 1955 (Bott 1955, 1959; Monod 1977, 1980)). Bott’s (1955) description of S. (S.) afri- canus chavanesii consisted of a one-line di- agnosis apparently made without reference to type material, and gonopod 1 was not described. Published figures of gonopod 1 of the type specimen of this species (Capart 1954) indicate a short, straight terminal seg- ment. A series of adult males from Came- roon examined in the present study indicate a different structure for gonopod I: the ter- minal segment is long, slim, and curves out- ward. It is likely that gonopod | of the type specimen from Gabon is unusual. Findings are presented here based upon the structure of the entire adult male gonopods of S. chay- anesii from Cameroon, together with other characters of the mandibles, carapace and chelipeds. Four dimensions of the carapace were re- corded using digital calipers as follows: car- apace length (CL) measured along median line, carapace width (CW) at widest point, measured beneath the large tooth on the anterolateral margin, carapace height (CH) maximum height of cephalothorax, and front width (FW) width of front measured along anterior margin. Carapace propor- tions were calculated according to carapace length. These data were pooled and used for descriptions of growth. Statistical compar- isons between species were made between sexually mature adults only. The distribu- tion of S. chavanesii described here is based on data from direct examination of speci- VOLUME 108, NUMBER 2 mens, since literature records are not reli- able. The following abbreviations are used: MNHN = Muséum national d’Histoire na- turelle, Paris; USNM = National Museum of National History, Smithsonian Institu- tion, Washington, DC, U.S.A.; NHM = Natural History Museum, London, U.K.; RCM = Royal Congo Museum, Tervuren, Belgium; IFAN = Institute Fondamental d’Afrique Noire, Dakar, Senegal; NMU = Northern Michigan University, U.S.A., = male, 2 = female. Sudanonautes chavanesii (A. Milne-Edwards, 1886) (Figs. 1-3) Thelphusa chavanesii A. Milne-Edwards, 1886:150. Parathelphusa chavanesii.—A. Milne-Ed- wards, 1887:145-146, pl. 7, fig. 3a, b.— Rathbun, 1900:285. Potamon (Potamonautes) chavanesi. —Ort- mann, 1897:305, 309. Potamon (Parathelphusa) chavanesii. — Rathbun, 1905:232, pl. 11, fig. 1.—Balss, 1929:127.—Balss, 1936:166. Potamon chavanesi.—Chace, 1942:209.— Capart, 1954:829-830, figs. 7, 11. Sudanonautes (Sudanonautes) africanus chavanesii.— Bott, 1955:299, fig. 97.— Bott, 1959:1005.— Monod, 1977:1217 (not figs. 103—107).— Monod, 1980:384. Material. —Gabon: Adult 6, type (CW 54, CL 39 mm), MNHN BS5079, vicinity of Franceville, on the river Alima, Mission de Brazza. Adult 6 (CW 54 mm) cotype, USNM 30034, La de Franceville, Mission de Braz- za. Cameroon: One °?, MNHN BS5081, Ohana, Besa, Nyong. One 6 (CW 46 mm), MNHN BS5077, Mt. M. Banbarto, river Noun. One 6, NHM 1936.2.27.1-—3, Batouri District. Two specimens, (RCM) 51.554, 1 ? mature (CW 70 mm), 1 6 (CW 48 mm), Soari, 26 Sep 1964, Thijs van den Auden- aerde. RCM 53.282, 1 6(CW 43 mm), Be- labo, 16 May 1970, Thijs van den Auden- aerde. RCM 53.290, 1 2 adult, Ebogo, 24 May 1970, Thijs van den Audenaerde. Three 2o9 adult 6 (CW 72.7, 69.6, 59.1 mm), RCM 53.291, Ebogo, 21 May 1970, Thijs van den Audenaerde. Type locality. — Vicinity of Franceville, on the river Alima, Gabon. Diagnosis. —Epibranchial tooth large, pointed outward, set back behind mid-point of postfrontal crest, positioned in line with widest point of carapace; distance between epibranchial tooth and intermediate tooth twice the distance between intermediate and exo-orbital teeth. Postfrontal crest spanning entire carapace, crest curving sharply back- ward before meeting epibranchial tooth; distinct notch in crest behind exo-orbital tooth; anterolateral margin smooth poste- rior to epibranchial tooth. Semi-circular, urogastric, cardiac, transverse branchial grooves very deep. Exo-orbital, intermedi- ate teeth large, sharp, pointed forward; epi- branchial tooth large, pointed outward. Vertical suture on flank forming y-shaped depression beneath intermediate tooth (Fig. 2b). Carapace very flat (CH/CL = 0.41). Mandibular palp 2-segmented; terminal segment single, undivided, with small hard, hair-fringed flap at junction between seg- ments (Fig. 2a—b). Terminal segment of gon- opod | thin and needle-like, subterminal segment gonopod | slim (Fig. 2d—f). Dac- tylus of major chelipid arched, forming oval interspace when closed; 2—3 large fused teeth in proximal regions of both fingers of the cheliped, rest of cutting edges lined with rows of rounded teeth (Fig. le-—f). Redescription. —Carapace (Fig. la, b) ovoid, widest in anterior third (CW/CL 1.37), very flat (CH/CL 0.42), semi-circular; urogastric, cardiac, transverse branchial grooves very deep, regions covered with raised circular blisters; cervical grooves present but weak. Branchial regions of car- apace with fields of faint raised short lines. Front deeply bilobed, anterior margin in- dented, curving down, relatively narrow, less than 3 carapace width (FW/CW = 0.30). Postfrontal crest smooth, spanning entire carapace, straight part consisting of fused epigastric, postorbital crests, then curving 240 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON e ee ae 16 Abts et Fe EMER ae ht ET TER ee _ A ad ° SAP, We MN: Sudanonautes chavanesii(A. Milne-Edwards, 1886), adult male (CW 72.7 mm) from Ebogo, Cameroon Fig. 1. (RCM 53.291). a, carapace, dorsal aspect (left epibranchial tooth broken); b, carapace, frontal aspect; c, sternum; d, abdomen; e, right cheliped, frontal view; f, left cheliped, frontal view; g, carpus, and merus of right cheliped, superior view; h, carpus, and merus of right cheliped, ventral view; 1, left periopod 2. Scale bar equals 20 mm (a—d), and 16 mm (e-1). VOLUME 108, NUMBER 2 sharply backward behind intermediate teeth to meet anterolateral margins at the epi- branchial teeth. Distinct notches in crest be- hind exo-orbital teeth; anterolateral margin smooth posterior to epibranchial tooth. Exo- orbital, intermediate teeth large, sharp, pointed forward. Epibranchial tooth large, pointed outward, set back behind mid-point of postfrontal crest, positioned in line with widest point of carapace; distance between epibranchial tooth and intermediate tooth twice the distance between intermediate and exo-orbital teeth. Flanks mostly smooth, with faint gran- ules in suborbital regions. Each flank with 2 sutures, | longitudinal, | vertical, dividing flank into 3 parts (Fig. 1b). Longitudinal suture dividing suborbital, subhepatic regions from pterygostomial region, begin- ning medially at lower margin of orbit, curving backward across flank. Short ver- tical suture beneath intermediate tooth di- viding suborbital region from subhepatic re- gion, suture forming y-shaped depression beneath intermediate tooth (Fig. 1b); stem of y-shape meeting longitudinal suture. First transverse groove on sternum (between sternal segments 2 and 3) complete; second groove (between sternal segments 3 and 4) consisting of 2 small notches at sides of ster- num (Fig. Ic). Third maxillipeds filling en- tire oral field, except for transversely oval efferent respiratory openings at superior lat- eral corners; long flagellum on exopod of third maxilliped (Fig. 2c); ishium of third maxilliped smooth, with clear vertical groove (Fig. 2c). Mandibular palp 2-segmented; terminal segment single, un- divided, with small hard, hair-fringed flap at junction between segments (Fig. 2a—b). Segments 1—6 of abdomen four sided, last segment triangular, sides indented, rounded at distal margin (Fig. 1d); segment 3 broad- est, segments 4—7 tapering inwards (Fig. 1d). Chelipeds (Fig. l1e—h) unequal, right lon- ger, higher than left. Dactylus of right che- liped long, arched; palm of propodus swol- len; proximal region of fingers of digits of 241 right cheliped each with 2-3 large fused teeth, forming oval interspace when closed; rest of cutting edges of fingers lined with smaller cheliped small to very small, fingers forming narrow interspace when shut. In- ferior margins of merus with rows of small teeth, cluster of granules surrounding larger pointed tooth at distal end. Inner margin of carpus of cheliped with 2 large, slender, pointed teeth, second half size of first (Fig. 1g—h). Dactylus of left cheliped not arched, otherwise similar to right, but smaller in all respects (Fig. 1f). Walking legs 2—5 (pereio- pods = P) slender (Fig. 11), P4 longest, P5 shortest. Posterior margin of propodus of P2-5 smooth or slightly serrated, dactyli of P2-5 tapering to point, each bearing 4 rows of downward-pointing sharp bristles; dac- tylus of P5 shortest of the 4 legs (Fig. 11). Gonopod 1 with very slender terminal segment, long (14 as long as penultimate seg- ment), curving strongly outward along en- tire length, tapering to pointed tip, longi- tudinal groove visible from caudal and su- perior views (Fig. 2d—e), not visible from cephalic view (Fig. 2f). Subterminal seg- ment of gonopod | very slim (Fig. 2d, f), with raised flap extending halfway across segment, forming roof of chamber for gon- opod 2; subterminal segment beneath flap forming lower floor of chamber for gonopod 2 (Fig. 2f). Gonopod 2 (Fig. 2g) shorter than gonopod | (reaching only to junction be- tween last 2 segments of gonopod 1). Ter- minal segment of gonopod 2 extremely short, only ’,, as long as subterminal segment, sides folded inwards to form spoon-shape, tip rounded. Subterminal segment gonopod 2 widest at base, then tapering sharply in- ward, forming long, thin, pointed, upright process which supports short terminal seg- ment; rounded collar at junction between terminal segment and subterminal segment. Adult female.—Chelipeds same propor- tions as male of same size, right cheliped enlarged, measuring longer and higher than the left cheliped. Mature female abdomen very wide reaching coxae of pereiopods 2- 242 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig.2. Sudanonautes chavanesii(A. Milne-Edwards, 1886), adult male (CW 72.7 mm) from Ebogo, Cameroon (RCM 53.291). a, left mandible, anterior view; b, left mandible superior view; c, left third maxilliped (flagellum broken in this specimen); d, left gonopod 1, caudal view; e, terminal segment of left gonopod 1, superior view; f, left gonopod 1, cephalic view; g, left gonopod 2, caudal view. Scale bar equals 3 mm (a, b, d, e, f, g), and 7 mm (c). VOLUME 108, NUMBER 2 5. Segments of female abdomen rectangular, wider than long, distal segments longest and widest; abdomen widest at groove separat- ing segments 4, 5; segment 6, telson together forming near semicircle. Growth (Fig. 3a, b).—Sexual maturity judged by development of female abdomen: abdomen of mature females overlapping bases of coxae of walking legs, pleopods broad, hair-fringed. Pubertal molt, from pu- bertal stage to sexual maturity, occurring between CW 35-45 mm. Largest known specimen (male from Cameroon) CW 81 mm. Dimensions of carapace varying with age (Fig. 3a). Relative width of carapace (CW/CL), relative height of carapace (CH/ CL), and relative width of the frontal mar- gin (FW/CL) not changing with age: pro- portions of adults remaining constant with age; proportions of adults not significantly different (P > 0.05) from those of juvenile and pubescent animals (Fig. 3b). Right and left chelipeds of juveniles (CW <35 mm) even sized, palms not inflated, cutting edges leaving no gap when closed. Distribution. —Coastal rain forest from south Cameroon to Gabon. It is likely that S. chavanesii is also present in Equatorial Guinea. Bott (1955) reported S. chavanesii from a wider region of Central Africa (from Cameroon to Zaire). However, there are no confirmed records of this species in the Cen- tral African Republic, the Republic of the Congo, or in Zaire. Bott (1959) and Monod (1977, 1980) extended the range of S. chav- anesii further to include West Africa, from Guinea all the way to Zaire. However, there is reason to doubt these records. Bott (1959) reported S. chavanesii from Guinea and Mali. Monod (1977) illustrated these same specimens, from ravin de Sokonafing, near Bamako, Mali (Monod 1977, figs. 103, 104, 106, 107), and from Fenaria, Guinea (Mo- nod 1977, fig. 105). This same material, de- posited in IFAN, from Dabola, chutes de Tinkasso, Guinea (Kindia, 5 Apr 1954, A. Villiers), and from Mali (Ravin de Sokon- afing, Bamako, May 1945), were all iden- 243 tified here as Liberonautes latidactylus (De Man, 1903). There are, therefore, no con- firmed records of S. chavanesii in any coun- try in West Africa (i.e., from Senegal to Ni- geria). Remarks.—Bott (1955) established the genus Sudanonautes (type species by orig- inal designation, Thelphusa africana A. Milne-Edwards, 1869), and recognized two subgenera, Sudanonautes s.s., and Convex- onautes Bott, 1955 (type species by original designation, Thelphusa aubryi H. Milne- Edwards, 1853). Sudanonautes chavanesii was one of three subspecies of Sudanon- autes (Sudanonautes) africanus, viz. S. (S.) a. africanus (A. Milne-Edwards, 1869), S. (S.) a. chaperi (A. Milne-Edwards, 1887), and S. (S.) a. chavanesii (A. Milne-Edwards, 1886) recognized by Bott (1955, 1959) and Monod (1977, 1980). Cumberlidge (1985) subsequently referred S. (S.) a. chaperi to the genus Liberonautes Bott, 1955. Data presented here indicate that Sudanonautes africanus chavanesii (A. Milne-Edwards, 1886) should also be regarded as a valid species. Identification of S. chavanesii depends on considering the characters of gonopod | of S. chavanesii (Fig. 2d-f) in conjunction with other characters of the carapace (Fig. 1a—b), sternum (Fig. 1c), chelipeds (Fig. le—h), and mandibles (Fig. 2a—b). Sudanonautes chay- anesil is most likely to be confused with other large species occurring in the rain for- ests of Central Africa such as S. africanus, S. faradjensis (Rathbun, 1921), and S. au- bryi (A. Milne-Edwards, 1853), all of which possess a gonopod | with a similar shaped terminal segment: slim, tapering, and curv- ing outward. Sudanonautes chavanesii can be distin- guished from S. africanus by the position of the postfrontal crest and the shape of the epibranchial teeth. In S. chavanesii the lat- eral ends of the postfrontal crest curve sharply backward before meeting the epi- branchial teeth which are set back posterior to the mid groove of the crest (Fig. 1a), PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 244 “SOSSEID OZIS UIOMI10q (FO) < qd) UONLIOIIOO JUvOyTUSIS ou 9}eoIpul (6 = fp 18 MA ‘ET = JP 1® HO ‘MO) Sonyea 4 (JD) 9z1s Apog 0} poredutos “(TD/MA) WIPIM JUO’ pue “(TD/HO) 14sI0y sdedeIed “(TD/MD) WIPIM s0edvivo Jo suopodold datjeol ‘q ‘Sosse[d oZIs UsIMI19q (1000 > d) UONeaLI09 jUROyTUBIS ATYSIY & d}eOIpUT (6 = /p 18 MA ‘ET =/P 18 HO ‘AAD) Sonyea 4 (JD) 9zIs Apoqg 0} poreduros (MJ ‘HO ‘MO) 20edeied oY} JO SUOISUSUIP “e “(O88 ‘SplemMp_-sUIII “Y) MSaUDADYI SajnDUOUDpNG Jo susutIdeds pI JO suostIedWI0D = “¢ “SI{ (wu) y}3uaT] adedeie5 (uu) y}3ua7] aoedeiey 9 0S OF 0€ 02 09 Os OF 0€ 02 2 BRR 0 fora 13é000%0+ @'0= TOMS TOL000 + €€°T = TO/MOD uorwodorg adedeiez (WW) UOIsUaWIG adedeIeZ I60=1 TOIO+SL's =Md S60=1 1Oh0+8T'0- =HD 660=1 TOUT + ELT - VOLUME 108, NUMBER 2 whereas in S. africanus the lateral ends of the crest curve forward to meet the epi- branchial teeth in line with the mid groove of the crest. In S. chavanesii the epibran- chial tooth is large, sharp, and pointed out- ward, and the distance between the epi- branchial tooth and intermediate tooth is twice the distance between the intermediate and exo-orbital teeth (Fig. la, b). In S. af- ricanus the epibranchial tooth is small, blunt, pointed forward, and positioned in line with the mid groove of the crest, so that the dis- tance between the epibranchial tooth and intermediate tooth is the same as the dis- tance between the intermediate and exo-or- bital teeth. Finally, in S. chavanesii the ver- tical suture on the flank forms a y-shaped depression beneath the intermediate tooth (Fig. 1b), whereas in S. africanus the vertical suture on the flank is simple. Sudanonautes chavanesii can be distin- guished from S. faradjensis by examination of the anterolateral margin behind the epi- branchial tooth: that of S. chavanesii is smooth, whereas in S. faradjensis there is a row of sharp teeth. In addition, the dactylus of the major cheliped of S. chavanesii is arched, forming a single oval interspace when closed, whereas that of S. faradjensis is straight, not arched, and forms two in- terspaces when closed. Sudanonautes chavanesii can be distin- guished from S. aubryi by examination of the carapace: that of S. chavanesii is signif- icantly flatter (P < 0.001) than that of S. aubryi (CH/CL S. chavanesii = 0.41, S. au- bryi = 0.52), and has patches of raised blis- ters, while that of S. aubryi is completely smooth. The postfrontal crest of S. chava- nesii meets the anterolateral margin at the epibranchial tooth, whereas that of S. aubryi meets the anterolateral margin behind the epibranchial tooth. In addition, the dactylus of the major cheliped of S. chavenesii is arched, with large, fused teeth (Fig. le); these features are lacking in S. aubryi. Finally, there is a small flap on the man- dibular palp at the junction between the two 245 segments (Fig. 2a—b) of S. chavanesii which is not present in S. africanus, S. faradjensis and S. aubryi. Acknowledgments The following people are thanked for hosting visits: Paul Clark (NHM); Dr. R. Roy (IFAN); Drs. D. Guinot and J. Forest (MNHN); and Dr. R. Manning (USNM). Dr. R. Joqué (RCM) is thanked for his kind hospitality and for loaning material. I am especially grateful to artist Anne C. Mich- elsen (NMU), for all of the illustrations used in this paper. Part of this work was sup- ported by a Faculty Grant from NMU. Two anonymous referees and Dr. Rafael Lemai- tre, Associate Editor of the Proceedings, are thanked for their helpful suggestions to im- prove the manuscript. Literature Cited Balss, H. 1929. Crustacea V. Potamonidae. Jn Th. Monod, ed., Contribution a l’étude de la faune du Cameroun, mission Monod 1925-1926 part 2, Faune Colonies Frangaises 3:115—129. 1936. Beitrage zur Kenntnis der Potamidae (Susswasserkrabben) des Kongogebeites. — Re- vue du Zoologie et Botanie d’Afrique 28:65- 204. Bott, R. 1955. Die Sitisswasserkrabben von Afrika (Crust., Decap.) und ihre Stammesgeschichte. — Annales du Musée Royal du Congo Belge, (Ter- vuren, Belgique) C. Zoologie 1 (3, 3):209-352. . 1959. Potamoniden aus West-Afrika.— Bul- letin de l'Institut Fondamental D’Afrique Noire, Serie A 21 (3):994—1008. Capart, A. 1954. Révision des types des espéces de Potamonidae de |’Afrique Tropicale conservés au Muséum d’ Histoire naturelle de Paris. — Vol- ume Jubilaire Victor Van Strallen, Director de l'Institut Royale des Sciences Naturelles de Bel- gique, 1925-1934, II:819-847. Chace, F. A. 1942. Scientific results of a fourth ex- pedition to forested areas in eastern Africa, III: Decapod Crustacea.— Bulletin of the Museum of Comparative Zoology, Harvard College 91(3): 185-233. Cumberlidge, N. 1985. Redescription of Liberon- autes chaperi (A. Milne-Edwards, 1887) n. comb., a fresh-water crab from Ivory Coast (Brachyura, Potamonautidae).— Canadian Journal of Zoology 63:2704—2707. 246 De Man, J.-G. 1903. On Potamon (Potamonautes) latidactylum, a new fresh-water crab from Up- per Guinea. — Proceedings of the Zoological So- ciety of London 1:41-47. Macleay, W. S. 1838. Illustrations of the zoology of South Africa; being a portion of the objects of natural history chiefly collected during an ex- pedition into the interior of South Africa, under the direction of Dr. Andrew Smith, in the years 1834, 1835, and 1836; Fitted out by “The Cape of Good Hope Association for exploring Central Africa.” in A. Smith, Illustrations of the Zoology of South Africa: consisting chiefly of figures and descriptions of the objects of natural history col- lected during an expedition into the interior of South Africa, in the years 1834, 1835, and 1836: fitted out by “The Cape of Good Hope Asso- ciation for Exploring Central Africa.’’, (Inver- tebrates). 75 pp., 4 pl., London. Milne-Edwards, A. 1869. Révision du genre Thel- phusa et description de quelques especes nou- velles faisant partie de la collection du Mu- séum.— Nouvelles Archives du Museum d’His- toire naturelle, Paris 5:161-191. 1886. La description de quelques Crustacés du genre Thelphusa recueillis par M. de Brazza dans les régions du Congo.— Bulletin de la So- ciété Philomathique de Paris, Série 7, 10:148- 151. 1887. Observations sur les crabes des eaux douces de Il’Afrique. — Annales des Sciences Na- turelles, Zoologie, Paris (7)4:121-149. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Milne-Edwards, H. 1853. Observations sur les affin- itiés zoologiques et al classification naturelle des Crustacés.—Annales des Sciences Naturelles, Zoologie, Paris (3)20:163—228. 1854. Notes sur quelques Crustacés nou- veaux Ou peu connus conserves dans la collec- tion du Museum d’Historie Naturelle.—Ar- chives du Muséum d’Histoire naturelle, Paris 7:145-192, pl. 9-16. Monod, T. 1977. Sur quelques crustacés Décapodes africaines (Sahel, Soudan).— Bulletin de Mu- séum national d’Histoire naturelle, Paris 3, 500: 1201-1236. 1980. Décapodes. in Flore et faune aqua- tiques de l’Afrique Sahelo-Soudanienne, 1:369- 389. Ed. J.-R. Durand and C. Léveque, OR- STOM, I.D T. 44, Paris. Ortmann, A. E. 1897. Carcinologische Studien— Zoologische Jahrbucher, Abteilung fur Syste- matik, Geographie und Biologie der Thiere 10: 258-372. Rathbun, M. J. 1900. The decapod crustaceans of West Africa. — Proceedings of the U.S. National Museum 22:221-316. 1905. Les crabes d’Eau Douce (Potamoni- dae).— Nouvelles Archives du Muséum d’His- toire naturelle (Paris) (4) 7:159-322. . 1921. The Brachyuran crabs collected by the American Museum Congo expedition 1909- 1915.—Bulletin of the American Museum of Natural History 43:379-474. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(2):247-253. 1995. First zoea of Dissodactylus glasselli Rioja and new range and host records for species of Dissodactylus (Brachyura: Pinnotheridae), with a discussion of host-symbiont biogeography Gerhard Pohle and Fernando Marques (GP) Atlantic Reference Centre, Huntsman Marine Science Centre, Brandy Cove, St. Andrews, New Brunswick, EOG 2X0, Canada; (FM) University of Toronto, Department of Zoology, 25 Harbord Street, Toronto, Ontario, M5S 1A1, Canada Abstract. — Dissodactylus glasselli Rioja, is a small pinnotherid crab living as an ectosymbiont on sand dollars of the American Pacific coast. The first zoea of D. glasselli is described in detail and compared to that of other species within the Dissodactylus complex. Morphometrically the larva differs from those of the sympatric species D. nitidus Smith, D. lockingtoni Glassell, and D. xantusi Glassell, in the relative length of carapace spines. The zoea of D. glasselli most closely resembles that of D. mellitae (Rathbun) from the Atlantic, the two being considered trans-isthmian geminate species. A geographic range extension of Dissodactylus lockingtoni is reported beyond the Gulf of California to Costa Rica, where it occurs on Mellita kanakoffi Durham, not reported previously as a host species. The range of Dissodactylus glasselli is extended southward from El Salvador to Costa Rica and it is reported for the first time on the hosts M. kanakoffiand Encope wetmorei Clark. Dissodactylus mellitae was found on the new host Encope aberans Martens in the Gulf of Mexico. Members of the Dissodactylus complex, comprising the genera Dissodactylus Smith, 1870 and Clypeasterophilus Campos & Grif- fith, 1990, are known as symbionts of echi- noids in tropical and subtropical regions of the Americas (Schmitt et al. 1973). Disso- dactylus is represented by four Atlantic and five Pacific species, and Clypeasterophilus by three and one species, respectively. Lar- val development is known for five Atlantic species (Pohle & Telford 1981b, 1983; Pohle 1984, Marques & Pohle 1995a, 1995b), and for Pacific species, larvae of D. lockingtoni Glassell, 1935, D. xantusi Glassell, 1936 and D. nitidus Smith, 1870 have been described (Pohle 1989, 1994). Dissodactylus glasselli overlaps geographically with the latter three sympatric species in the southern parts of the Gulf of California. The first zoea of D. glasselli Rioja, 1944 is described in this pa- per and compared to that of other species within the Dissodactylus complex. New sand dollar hosts are reported for Dissodactylus glasselli, D. lockingtoni and D. mellitae (Rathbun, 1900), and the occurrence of D. lockingtoni has been established outside the Gulf of California. Materiais and Methods During June to July 1992, first zoeae of Dissodactylus glasselli were obtained from a number of females collected along the Pa- cific coast of Costa Rica. Individual rearing techniques used to culture larvae, although successful for other species of Dissodactylus and Clypeasterophilus (Pohle & Telford 1981b, Pohle 1984, 1989; Marques & Pohle 1995a, 1995b), were not successful in this case. None of the larvae developed beyond 248 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.— Dimensions (mm) of zoea | structures of Dissodactylus glasselli Rioja and other selected species. Spine length Antennal Carapace Species Rostral Dorsal Lateral Rostrodorsal length length Dissodactylus 0.35 +0.02 0.20+0.02 0.20+0.01 0.88 +0.04 0.10+0.01 0.37 + 0.01 glasselli (0.32-0.37) (0.17-0.22) (0.17-0.22) (0.81-0.94) (0.09-0.10) (0.36—-0.39) D. lockingtoni' 0.26+0.01 0.16+0.01 0.14+0.01 0.71 +0.03 0O.10+0.01 0.35 + 0.01 D. nitidus? 0245-10102) 012-5 01019) (O12 = 0:01 0:68 =. 0102, OO OO 1036e 20:01 D. xantusi' O26): O10” OAD 25 0:01) 013) O10 10165: == 0:02) _ OL08F=2 O10 ea ONs 7a OL0II D. mellitae’ 0.30 +0.02 0.21+001 0.16+0.02 0.71+0.01 O.11 0.01 0.32 + 0.01 D. mellitae? 0:30)== 0102) 0:24-0:01 10:20 = O10), 202872270102) 2205102001036 O05 Note: Values are given as the mean + standard deviation, with range in parentheses for D. glasselli. ' From Pohle (1994); 2 from Pohle (1989); * from Marques & Pohle (1995b). the first zoea. Selection of the most lively larvae, addition of antibiotics and the use of alternate food organisms, such as oyster larvae and fertilized sand dollar eggs, did not change the outcome. Ten specimens were measured and used for morphological description. The descrip- tion of setae follows Pohle & Telford (1981a), but here includes only analysis by light microscopy, using an Olympus BH-2 microscope with Nomarski Differential In- terference Contrast and camera lucida. Measurements follow Pohle & Telford (1981a). Specimens of the first zoeal stage have been deposited at the National Mu- seum of Natural History, Smithsonian In- stitution, Washington, D.C. Results Eggs of Dissodactylus glasselli were in- cubated by crabs for 11-13 days at 32°C. Larval morphometrics are given in Table 1. First Zoea (Fig. 1) Carapace (Fig. 1A).— With long rostral, dorsal and lateral spines. Thickened pos- terior and ventral margin lacking setae. Eyes sessile. Two simple setae flanking dorsal spine. Chromatophores on each side of car- apace posterior to eyes, on ventrolateral margin and base of antennules; single chro- matophores located posterior to base of dor- sal spine, frontally between eyes and dor- sally on gut; abdominal somites 1-5 with paired melanophores; single chromato- phore on labrum, mandibles and basipod- ites of first maxillipeds. Abdomen (Fig. 1B).—Five somites and telson. Somite | naked, somites 2 and 3 with pair of dorsolateral spines; somites 2—5 with pair of simple setae dorsally. Telson (Fig. 1C).— Bifurcated, with prox- imal minute furcal spine. Furcal shafts spi- nulous, except tips. Furcal arch with 3 plu- modenticulate setae on either side of deep median depression. Denticulettes present in clusters on ventral and dorsal surface. Antennule (Fig. 1D).—Unsegmented, smooth, conical. Terminally with 1 short and 2 long aesthetascs. Antenna (Fig. 1E).—Elongate, unira- mous. Tapered protopodite with 2 rows of spinules distally. Maxillule (Fig. 1F).—Coxal endite bear- ing 4 graded plumodenticulate setae and proximal microtrichia. Basal endite with proximal microtrichia, 3 terminal plumo- denticulate cuspidate setae and 2 subter- minal plumodenticulate setae. Two-seg- mented endopodite with 4 terminal plu- modenticulate setae. Maxilla (Fig. 1G).—Coxal endite single- lobed, inflated, with semicircle of 4 plumose setae flanked by single proboscate (sensu Pohle & Telford 1981a) seta; scattered mi- crotrichia. Basal endite with 4 plumoden- ticulate setae on either side of slight inden- VOLUME 108, NUMBER 2 tation; microtrichia present. Endopodite unsegmented, with 3 terminal plumoden- ticulate setae and marginal microtrichia. Scaphognathite with 4 densely plumose se- tae marginally, tapering to sharp terminal process bearing microtrichia. Maxilliped 1 (Fig. 1H).—Coxopodite with a developing seta. Basipodite with 10 plu- modenticulate setae arranged in four groups of 2, 2, 3, 3 proximally to distally. Five segmented endopodite with 2, 2, 1, 2, 4 + 1 setae proximally to distally; all plumo- denticulate except single simple seta on seg- ments 1, 2, and 5. Exopodite with 4 long natatory plumose setae. Maxilliped 2 (Fig. 11).—Coxopodite na- ked. Basipodite with 4 plumodenticulate se- tae. Endopodite 2-segmented, first segment naked, distal segment with 4—5 plumoden- ticulate setae. Exopodite with 4 long nata- tory plumose setae. Maxilliped 3.—Not discernible. Pereiopods.—May be present as minute buds. Pleopods. — Absent. Range Extension and New Hosts for Species of Dissodactylus Dissodactylus lockingtoni Glassell, 1935 Dissodactylus lockingtoni Glassell, 1935: 100, pl. 27, figs. 5-8, text-fig. 68 (type locality, Punta Penasco, Sonora, Mexi- co).—Griffith 1987a: 401, 413, 419, figs. 8B, 9B, 11B, 13B, 14B; 1987b:2292—2310. D. smithi Rioja, 1944: 149, figs. 1-6, 11-15 (type locality, Playa San Benito, Chiapas, 50 km from Tapachula, Mexico). Material examined. — Puntarenas Beach, Puntarenas, Costa Rica, 9°56'N, 84°48'W; 2 Jul 1992, free-diving 0.5—1 m, sand bot- tom, | male cw 3.6 mm on Mellita kanakoffi Durham, 1961.—Puntarenas Beach, Pun- tarenas, Costa Rica, 9°56'N, 84°48'W; 18 Jul 1992, low intertidal, sand bottom, 3 fe- males cw 5.4, 5.1 and 4.8 mm on M. kan- akoffi. —Puntarenas Beach, Puntarenas, Costa Rica, 9°56'N, 84°48'W; 7 Aug 1992, 249 intertidal, low tide 0 m, sand bottom, 4 fe- males cw 4.6, 5.0, 5.1 and 5.3 mm on M™. kanakoffi. Previous range records. —Punta Penasco, Sonora, Mexico; “San Felipe, Gulf of Cal- ifornia and Punta Penasco... undoubtedly ranges throughout the Gulf of California’ (Glassell 1935). Campo et pescador, North of San Felipe (3 1°04'N, 114°50’W) (Campos et al. 1992). La Choya Bay, Punta Penasco, Sonora, Mexico (Pohle 1994). Previous host record. —Encope californica Verrill, 1870; E. grandis Agassiz, 1841; E. micropora Agassiz, 1841; Mellita longifissa Michelin, 1858 (Glassell 1935). Remarks.—The host Encope californica Verril listed by Glassell (1935) is presently considered a junior synonym of FE. micro- pora (cf. Brusca 1980) but both are probably separate species (M. Telford, in litt.). Dissodactylus glasselli Rioja, 1944 D. glasselli Rioja, 1944: 150, fig. 7-10, 16- 21 (type locality, Playa San Benito, near Tapachula, Chiapas, Mexico).—Griffith 1987a: 413, 420, figs. 8H, 9H, 11H, 13H; 1987b: 2292-2310. Material examined. —Manuel Antonio National Park, Manuel Antonio, Costa Rica, 9°06’N, 84°11'W; 20 June 1992, SCUBA diving 1.5—3 m, sand bottom, 3 males cw 3.6, 3.6 and 3.8 mm, 3 females cw 3.3, 4.2 and 4.2 mm on Mellita kanakoffi and En- cope wetmorei Clark, 1946.—Brasilito Beach, Guanacastes, Costa Rica, 10°25’N, 85°47'W; 22 Jun 1992, SCUBA diving 1-3 m, sand bottom, 4 males cw 2.4, 3.2, 3.9 and 4.0 mm, 2 females cw 3.4 and 4.6 mm on Encope micropora.—Brasilito Beach, Guanacastes, Costa Rica, 10°25'N, 85°47'W;; 13 Jul 1992, SCUBA diving 1-3 m, sand bottom, 3 females cw 3.0, 3.8 and 5.0 mm on Encope micropora. Previous range records.—Playa San Be- nito, near Tapachula, Chiapas, Mexico (Glassell 1935). Puerto el Triunfo, El Sal- vador (Griffith 1987a). 250 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Ko A aba PU PIT D vy Aaah \ Kt YA fp WY Fig. 1. First zoea of Dissodactylus glasselli Rioja, 1944. A, lateral view; B, dorsal view of abdomen; C, telson; D, antennule; E, antenna; F, maxillule; G, maxilla; H, maxilliped 1; I, maxilliped 2. VOLUME 108, NUMBER 2 Previous host record.—Mellita longifissa Michelin (Rioja 1944). Remarks.—The specimens collected at Brasilito Beach were sharing the same host with Dissodactylus nitidus Smith, 1870. Dissodactylus mellitae (Rathbun, 1900) Echinophilus mellitae Rathbun, 1900: 590 (type locality, Pensacola, Florida). D. mellitae Rathbun, 1901: 22.—Griffith 1987a: 413, 420, figs. 81, 9I, 111, 13], 14G; 1987b; 2292-2310. Material examined. —East beach, Desoto Fort, St. Petersburg, Florida, Gulf of Mex- ico; 20 Aug 1992, free-diving 2—3.5 m, sand bottom, 2 females cw 4.0 and 4.2 mm on Encope aberans Martens, 1867. Previous host records. —Echinarachnius parma (Lamarck, 1816), Mellita quinquies- perforata (Leske, 1778) (cf. Rathbun 1901, Telford 1982, Bell & Stancyk 1983, Bell 1984, 1988); Mellita tenuis Clark, 1940 (Marques & Pohle 1995b); Encope mich- elini Agassiz, 1841; and Clypeaster subde- pressus (Gray, 1825) (cf. Williams et al. 1968). Comparison of D. glasselli Larvae with those of Other Species of the Dissodactylus Complex Meristic characteristics of the first zoea of Dissodactylus glasselli are identical to other species of this genus. There are, how- ever, some morphometric differences be- tween D. glasselli and the three other sym- patric Pacific species, D. nitidus Smith, 1870, D. lockingtoni Glassell, 1935 and D. xantusi Glassell, 1936 (Pohle 1989, 1994). While the first zoea of the latter three are almost indistinguishable, Dissodactylus glassellidif- fers by significantly longer (P < 0.01) car- apace spines (Table 1). Carapace length, however, overlaps with the other Pacific species. The closely related Clypeasterophi- lus ususfructus (Griffith, 1987a), which has recently been removed from Dissodactylus (Campos & Griffith 1990), is also known 251 from near the mouth of the Gulf of Cali- fornia (Hendrickx 1990). Although larvae of C. ususfructus are unknown, it is expected that the zoeae will differ from Dissodactylus by the absence of dorsolateral spines on ab- dominal somite 3, as known for Clypeas- terophilus stebbingi (Rathbun, 1918) and C. rugatus (Bouvier, 1917) (cf. Marques & Pohle 1995a, Pohle 1984). Larvae of D. glasselli resemble most closely those of D. mellitae from the Atlan- tic (Marques & Pohle 1995b), D. glasselli differing only by the relatively longer rostral and lateral spines (Table 1). The similarity between these two species supports the con- clusion by Grifhth (1987b), based on an analysis of adult synapomorphies, that these are twin, or geminate, species. Trans-isth- mian pairs of closely related species have been described for many taxa, including about 45% of extant decapods of Panama (Abele 1976). Under vicariant biogeograph- ic theory (Rosen 1975), these species evolved by allopatric speciation, following the clos- ing of the Panama seaway. The relatively recent separation accounts for the similarity of eastern Pacific and Caribbean species, such as that seen between Dissodactylus mellitae and D. glasselli. Host-symbiont Biogeography Dissodactylus lockingtoni has never be- fore been reported outside the Gulf of Cal- ifornia and thus its occurrence in Costa Rica represents about a 20° southward extension. The hosts Mellita longifissa and Encope mi- cropora extend further south to Panama and Chile, respectively (Ghiold 1988). Thus, the geographic range of Dissodactylus locking- toni may also extend beyond Costa Rica. Dissodactylus glasselli was only known from its type locality in Mexico until Grif- fith (1987a) discovered a specimen amongst alot in the Smithsonian collection identified as D. lockingtoni from El Salvador. The rec- ord in Costa Rica represents a 3° southward extension. The hosts Mellita longifissa and M. kanakoffi both extend further south to MS)22 at least Panama (Harold & Telford 1990), indicating that D. glassel/li may also be found there. In terms of host specificity, Dissodactylus mellitae is the most generalist species within the Dissodactylus complex, inhabiting echi- noids from three distinct clypeasteroid fam- ilies. Other Atlantic species of Dissodacty- lus, except for D. primitivus and D. schmitti which live on heart urchins (Grifhtth 1987a), can be found on mellitid sand dollars and two species of Clypeaster. In contrast, the Pacific species of this genus have never been found on any of the Clypeaster species oc- curring in the area. In contrast to Dissodactylus, Atlantic and Pacific species of Clypeasterophilus are found exclusively on Clypeaster (Griffith 1987a, Hendrickx 1990). The following ev- idence suggests that species of C/ypeaster- ophilus are also more host dependent than those of Dissodactylus: 1) an analysis of gut contents of Clypeasterophilus rugatus and three species of Dissodactylus by Telford (1982) showed that only C. rugatus fed ex- clusively on host tissue, whereas D. primi- tivus took no more than about half of its food from the spatangoid host; 2) larvae of Clypeasterophilus (Pohle 1984, Marques & Pohle 1995a) could not complete larval de- velopment in absence of a host, whereas Dissodactylus primitivus could (Pohle & Telford 1983). Acknowledgments This work was supported through re- search grant A2313 to G. Pohle, from the Natural Sciences and Engineering Research Council, Canada. Dr. A. Dittel is thanked for arranging the use of facilities at the Cen- tro de Investigacion de Ciencias del Mar (CIMAR), Universidad Nacional, Costa Rica, and for providing advice and equip- ment. We also thank technical support per- sonnel at CIMAR. Dr. M. Telford from the Department of Zoology at the University of Toronto kindly identified the echinoid hosts. Drs. M. Telford and H. Griffith also re- viewed the manuscript. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Literature Cited Abele, L.G. 1976. Comparative species composition and relative abundance of decapod crustaceans in marine habitats of Panama. — Marine Biology 38:263-278. Agassiz, L. 1841. Des Scutelles. Monographies des Echinodermes Vivans et Fossiles. Monograph 2:1-151. Bell, J. L. 1984. Changing residence: dynamics of the symbiotic relationship between Dissodactylus mellitae Rathbun (Pinnotheridae) and Mellita quinquiesperforata (Leske) (Echinodermata). — Journal of Experimental Marine Biology and Ecology 82:101-115. 1988. Distribution and abundance of Dis- sodactylus mellitae Rathbun (Pinnotheridae) on Mellita quinquiesperforata (Leske) (Echinoder- mata).—Journal of Experimental Marine Biol- ogy and Ecology 117:93-114. —, & S. E. Stancyk. 1983. Population dynamics and reproduction of Dissodactylus mellitae (Brachyura: Pinnotheridae) on its sand dollar host Mellita quinquiesperforata (Echinoderma- ta).— Marine Ecology Progress Series 13:141- 149. Bouvier, M. E.-L. 1917. Gonoplacidés et Pinno- theridés nouveaux receuillis au cours des cam- pagnes ameéricaines du ‘‘Hassler’’ et du ‘Blake’. — Bulletin du Muséum national d’His- toire naturelle, Paris 23:391-398. Brusca, R.C. 1980. Common intertidal invertebrates of the Gulf of California. The University of Ar- izona Press, Tucson, 513 pp. Campos, E., & H. Griffith. 1990. Clypeasterophilus, a new genus to receive the small-palped species of the Dissodactylus complex (Brachyura: Pin- notheridae). — Journal of Crustacean Biology 10: 550-553. —., A. R. de Campos, & J. Ramirez. 1992. Re- marks on distribution and hosts for symbiotic crustaceans of the Mexican Pacific (Decapoda and Isopoda).—Proceedings of the Biological Society of Washington 105:753-759. Clark, A. H. 1946. Echinoderms from the Pearl Is- lands, Bay of Panama, with a revision of the Pacific species of the genus Encope.—Smith- sonian Miscellaneous Contributions 106 (5):1- 11, 4 pits. Clark, H. L. 1940. Revision of the keyhole urchins (Mellita).— Proceedings of the United States National Museum 89:435-444. Durham, J. W. 1961. The echinoid Mellita in the Pacific coast Cenozoic.—Los Angeles County Museum Contributions in Science 48:1-12. Ghiold, J. 1988. Species distributions of irregular echinoids. — Biological Oceanography 6:79-162. Glassell, S. A. 1935. New or little known crabs from the Pacific coast of northern Mexico.—Trans- VOLUME 108, NUMBER 2 actions of the San Diego Society of Natural His- tory 8:91-106. 1936. New porcellanids and pinnotherids from tropical North American waters. —Trans- actions of the San Diego Society of Natural His- tory 8:277-304. Gray, J. E. 1825. An attempt to divide the Echinida, or sea eggs, into natural families.—Annals of Philosophy 26:423—431. Griffith, H. 1987a. Taxonomy of the genus Disso- dactylus (Crustacea: Brachyura: Pinnotheridae) with descriptions of three new species. — Bulle- tin of Marine Science 40:397—422. 1987b. Phylogenetic relationships in the ge- nus Dissodactylus Smith, 1870 (Crustacea: Brachyura: Pinnotheridae).— Canadian Journal of Zoology 65:2292-—2310. Harold, A. S., & M. Telford. 1990. Systematics, phy- logeny and biogeography of the genus Mellita (Echinoidea: Clypeasteroida).— Journal of Nat- ural History 24:987-1026. Hendrickx, M. E. 1990. Range extension and host record for Dissodactylus ususfructus Griffith, 1987 (Crustacea: Brachyura: Pinnotheridae). — Proceedings of the Biological Society of Wash- ington 103:106—107. Lamarck, J. B. 1816. Histoire naturelle des animaux sans vertébres. III. Verdiére, Paris, 130 pp. Leske, N. G. 1778. Additamenta ad Jacob Theodori Klein naturalem dispositionem echinoderma- tum et lucubratiunculam de aculeis echinorum marinorum. Upsala, 216 pp. Marques, F., & G. Pohle. 1995a. The complete larval development of Clypeasterophilus stebbingi (Decapoda: Brachyura: Pinnotheridae) and a comparison with other species within the Dis- sodactylus complex.— Bulletin of Marine Sci- ence (in press). SSS, 0 1995b. Laboratory-reared larval stages of Dissodactylus mellitae (Decapoda: Brachyura: Pinnotheridae) and developmental patterns within the Dissodactylus complex.— Canadian Journal of Zoology (in press). Martens, E. von. 1867. Uber ostasiatische Echino- dermen (Fortsetzung).—Archiv ftir Naturfor- schung. Jahrgang X XXIII (1):106-119, 3 plts. Michelin, M. H. 1858. Du genre Mellita Famille des Clypeasteroides.— Revue et Magasin de Zool- ogie 8:2—7. Pohle,G. 1984. Larval development of Dissodactylus rugatus Bouvier, 1917 (= D. calmani Rathbun, 1918) (Brachyura: Pinnotheridae) reared under laboratory conditions.—Journal of Crustacean Biology 4:572-588. 1989. Development and morphology of lab- oratory-reared larvae of Dissodactylus nitidus Smith, 1840 (Decapoda: Brachyura), with a dis- cussion of phylogenetic aspects in the Pinno- 253 theridae. — Journal of Crustacean Biology 9:278- 296. 1994. Dissodactylus xantusi and D. locking- toni (Decapoda: Brachyura: Pinnotheridae): lar- val development of two sympatric Pacific spe- cies. — Canadian Journal of Zoology 72:575-590. ——., & M. Telford. 198la. Morphology and clas- sification of decapod crustacean larval setae: a scanning electron microscope study of Disso- dactylus crinitichelis Moreira, 1901 (Brachyura: Pinnotheridae).— Bulletin of Marine Science 31: 736-752. ,& . 1981b. The larval development of Dissodactylus crinitichelis Moreira, 1901 (Brachyura: Pinnotheridae) in laboratory cul- ture.— Bulletin of Marine Science 31:753-773. Ss, 1983. The larval development of Dissodactylus primitivus Bouvier 1917 (Brachy- ura: Pinnotheridae) reared in the laboratory. — Bulletin of Marine Science 31:753-773. Rathbun, M. J. 1900. Synopsis of North American invertebrates. XI. The catometopous or grap- soid crabs of North America.— American Nat- uralist 34:583-591. 1901. The Brachyura and Macrura of Porto Rico.— Bulletin of the United States Fish Com- mission 20:1—127. . 1918. The grapsoid crabs of America. — Unit- ed States National Museum Bulletin 97:1-461, 161 pits. Rioja, E. 1944. Estudios carcinologicos XVI. Obser- vaciones sobre algunas especies de cangrejos del genero Dissodactylus Smith (Braquiros, Pinno- teridos) de las costas Mexcicanas del Pacifico. — Anales del Instituto de Biologia de México 15: 147-160. Rosen, D. E. 1975. A vicariance model of Caribbean biogeography.—Systematic Zoology 24:431- 464. Schmitt, W. L., J.C. McCain, & E.S. Davidson. 1973. Decapoda I. Brachyura I. Family Piinnotheri- dae. Jn H. E. Gruner & L. B. Holthuis, eds., Crustaceorum Catalogus III. W. Junk B. V., Den Haag, The Netherlands, 160 pp. Smith, S.I. 1870. Notes on American Crustacea. No. 1. Ocypodoidea.—Transactions of the Con- necticut Academy of Arts and Science 2:113- 176, plts 2—S. Telford, M. 1982. Echinoderm spine structure, feed- ing and host relationships of four species of Dis- sodactylus (Brachyura: Pinnotheridae).— Bulle- tin of Marine Science 32:584—594. Vermill, A. E. 1870. Description of echinoderms and corals from the Gulf of California.— American Journal of Science 49:93-100. Williams, A. B., I. R. McClosky, & I. E. Gray. 1968. New records of brachyuran decapod crustaceans from the continental shelf off North Carolina, U.S.A.—Crustaceana 15:41-66. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(2):254-265. 1995. A new crab species of the genus Pseudorhombila H. Milne-Edwards, 1837 (Crustacea: Decapoda: Goneplacidae) Ana Rosa Vazquez-Bader and Adolfo Gracia Instituto de Ciencias del Mar y Limnologia, Universidad Nacional Autonoma de México, Apdo. Postal 70-305, México D. F., 04510, México, Abstract.—A new crab species of Pseudorhombila was found in the south- western Gulf of Mexico. A description of the new species, Pseudorhombila ometlanti, is given and is compared with other known species in the genus from the western Atlantic: P. octodentata, P. quadridentata and P. guinotae. Among western Atlantic species of the genus, P. ometlanti can be distinguished by carapace texture, dentition and sexual characters. Restimen. —Una nueva especie del genero Pseudorhombila se recolecto en el suroeste del Golfo de México. Se presenta la descripcion de Pseudorhombila ometlanti, especie nueva, y se compara con otras especies descritas para el Atlantico oeste: P. octodentata, P. quadridentata y P. guinotae. Entre estas especies, P. ometlanti se distingue por la textura y denticion del caparazon, y la morfologia de los pleopodos en machos y gonoporos en las hembras. Epibenthic macroinvertebrates from shrimp grounds along the southwestern Gulf of Mexico were studied during a three year program. Several goneplacid crab species of the genus Pseudorhombila H. Milne-Ed- wards, 1837, were found among the deca- pod crustaceans collected. As Guinot (1969) has stated, this is a genus in need of study. Species of this genus are typical catometo- pian crabs with a large part of the eighth sternite visible at the level of the second abdominal segment and its juncture with the seventh sternite. The sternal position of the male genital openings is another diag- nostic characteristic. Pseudorhombila crabs are a common component of the macroinvertebrate fauna of the southwestern Gulf of Mexico (Vaz- quez Bader & Gracia G. 1994). While study- ing our collections, we found that some of the specimens did not represent any of the species of this genus in the western Atlantic, all of which occur in the Gulf of Mexico (P. octodentata Rathbun, 1906; P. quadriden- tata (Latreille, 1828) and P. guinotae Her- nandez-Aguilera, 1982), and actually rep- resent an undescribed species. A complete description of this new spe- cies of the genus Pseudorhombila is given and its affinities with other known species of the genus are discussed. Materials and Methods The specimens used in this study were collected during the study ““Monitoreo de las Fases de Pre-reclutamiento de las Es- pecies Estuarino-dependientes de Impor- tancia Comercial frente a Laguna de Ter- minos” (MOPEED) conducted in Banco de Campeche, Campeche, Mexico; and during the cruises by the R/V Oregon IT in the southern Caribbean (Panama, Colombia). In the Gulf of Mexico samples were taken quarterly on board the R/V Justo Sierra of Universidad Nacional Autonoma de Meéx- ico, using a semicommercial otter trawl. The material examined is listed as fol- lows: location (MOPEED 1 and 3 or R/V Oregon IT station number), latitude, longi- VOLUME 108, NUMBER 2 DS) Fig. 1. 26.5 mm (SMIOM 4065); b, Pseudorhombila quadridentata (Latreille), male, carapace length 25.2 mm (USNM 171616); c, Pseudorhombila guinotae Hernandez-Aguilera, male holotype, carapace length 29.0 mm (SMIOM 000304); d, Pseudorhombila octodentata Rathbun, male, carapace length 24.1 mm (photograph by D. Guinot). tude, depth, date, number, sex, and muse- um number. All specimens have been de- posited in the Coleccion de la Secretaria de Marina, Mexico, D. F., México (SMIOM) and in the National Museum of Natural History, Smithsonian Institution, Washing- ton, DC (USNM). Other abbreviations used are, p, pereiopod, and pl, pleopod. Pseudorhombila ometlanti, new species Figs. la, 2, 3a, 4a—b, 5a, 6a, 7a, Table | Material examined. —Southwestern Gulf of Mexico: MOPEED3 sta C-2 (Banco de Campeche, Campeche, Mexico) 19°30.96'N, 91°50.09'W; 45 m, 14 Sep 1992, 1 6 holo- type, SMIOM 4065; MOPEED 1 sta R-1 (vicinity of Laguna de Carmen y Machona, Tabasco, Mexico), 18°38.43'N, 93°46.68'W; Carapace, dorsal surface. a, Pseudorhombila ometlanti, new species, male holotype. Carapace length 79 m, 18 Feb 1992, 1 6, 1 2 SMIOM 4066. Southern Caribbean: R/V Oregon II sta 10259 off Barranquilla, Colombia, 10°59'N, 75°17'W; 72 m, 12 Feb 1968, 2 8 USNM 268442; sta 11234, Gulf of Uraba, Colom- bia 8°49'N, 76°53’W, 50 m, 11 Feb 1970, 1 6 USNM 268443. Description. —Carapace much broader than long, coarse granulation visible at na- ked eye, more evident on and adjacent to margins. Regions partially marked. Cara- pace broadest at second anterolateral tooth. Frontorbital width more than '2 carapace width. Front convex divided by distinct me- dian notch into 2 rectangular lobes. Supraorbital margin bissected by 2 su- tures with granulation continued to outer orbital tooth (first anterolateral); suborbital margin mesially with triangular tooth visi- ble in dorsal view. Anterolateral margin 256 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Pseudorhombila ometlanti, new species, male holotype (SMIOM 4065). a, ventral surface of anterior region; b, sternum, postero-ventral view. VOLUME 108, NUMBER 2 convex, projecting 2 teeth; first tooth gran- ulated, rounded; second tooth triangular, granulate, directed anteriorly, acute and dis- tinctly separated from first. Posterolateral margin converging posteriorly. Epistome broadest at mid-length, with thick lip medially, finely granulate. Sternum granulate, first segments and abdominal de- pression with long soft setae, episternal su- tures 4, 5, and 6 marked, 7 poorly defined; abdominal depression broad and shallow. In male, posterior sternal segments sepa- rated from preceding segments by genital groove. In female, gonopores large, circular, with margin slightly elevated posteriorly. Abdomen in both sexes broad, punctate. In male, segment | widest and longest of all, segment 2 narrow; large portion of eighth sternite visible between second and third abdominal segment, segment 3 with lateral angles not acute and not reaching coxa of fifth pereiopod; segments 3—S partially fused, sutures remaining conspicuous. Abdomen of female widest at segment 1; segment 6 longer than any of preceding ones, last seg- ment spear-like. Eyestalks very broad, anterior surface with coarse granulation, with long soft setae dor- sally and anteriorly; cornea dilated. Anntenules prominent, folding trans- versely, basal article thickest, with longi- tudinal crest granulate; second article elon- gate, subcylindrical; third article nearly equal in length to second, with long marginal setae distally. Basal anntenal article broad, closing or- bital hiatus between frontal lobe and sub- orbital tooth, basally with operculum of ex- cretory pore occupying part of closed suture at anterolateral corner of epistome; follow- ing articles free, elongate, subcylindrical, and decreasing in length. Third maxilliped very granulated, pilose, ischium with medial longitudinal furrow, merus densely granulated with depression at either side of raised median area, dis- tomesial margin terminating in angular lobe, distolateral corner rounded, granulated. 25) Chelipeds (p1) slightly shorter than first walking leg (p2), slightly unequal in female. Movable finger or major cheliped with lob- iform proximal tooth. Fixed finger with 2 longitudinal ridges on outer surface. Palm with inner surface smooth; outer surface smooth to punctate; inferior surface with granulated crest. Carpus with strong, tri- angular tooth on inner angle, tip directed anteriorly, concentration of coarse granules below tooth and on superior and inferior margin. Merus with external surface very granulated and with stiff bristles, inner sur- face smooth, superior margin ending in sharp distally-directed tooth. Dactyls of walkings legs 1—4, each with corneous tip and with longitudinal rows of setae. Propodi of legs 1-3 granulate, elongate and with long setae, propodi of p4 with shorter setae; su- perior surface of meri of pereiopods with depression delimited by small granules; in- ferior surface with strong granulation (less evident on p4). Carpi with superior margin densely granulated, (more so in p1-3), in- ferior margin with coarse granules. Male gonopod (pll) elongate, apex ter- minating in hump continued in large lobe with approximately 19 spines. Anterome- sial margin with approximately 34 small, spaced spinules, distal end with 19 large, recurved spines; proximal boss on poster- omesial surface with row of about 8 bristles. Male pl2 elongate, narrow, sigmoid, ter- minal process curved, with small spinules and distally strap-like. Size. —Carapace length and width of male holotype 26.5 and 35.2 mm, respectively; males 27.8 and 38.0 mm (SMIOM 4066), 24.0 and 31.8 mm (USNM 263442), 21.7 and 27.9 mm (USNM 268442), 25.0 by 33.8 mm (USNM 368443); female 27.4 and 37.0 mm (SMIOM 4066). Color.—Not known in fresh specimens. In alcohol, the specimens are salmon-pink. Known range and habitat. — Western At- lantic from the southwestern Gulf of Mex- ico (Banco de Campeche, Campeche, Mex- ico) to southern Caribbean (Golfo de Uraba, 258 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. Male abdomen and telson. a, Pseudorhombila ometlanti, new species, male holotype (SMIOM 4065), abdomen and telson; b, Pseudorhombila quadridentata (Latreille) (USNM 171616); c. Pseudorhombila guinotae Hernandez-Aguilera, holotype (SMIOM 000304); d, Pseudorhombila octodentata Rathbun, holotype (USNM 32690). Scales = 1 mm. VOLUME 108, NUMBER 2 Me)s) Fig. 4. Male first gonopod. a—b. Pseudorhombila ometlanti, new species, holotype (SMIOM 4065): a. left mesial surface; b, enlarged tip of same posterior surface. c-d, Pseudorhombila quadridentata (Latreille) (USNM 171616); c, right mesial surface; d, enlarged tip of same posterior surface. e-f, Pseudorhombila guinotae Her- nandez-Aguilera holotype (SMIOM 000304); e, left mesial surface; f, enlarged tip of same posterior surface. g— h, Pseudorhombila octodentata Rathbun, holotype (USNM 32690); g, left posterior surface; h, enlarged tip of same mesial surface [g—h, after Guinot (1969)]. Scales = 1 mm. 260 and off Barranquilla); from 23 to 72 m depth. During the MOPEED study, P. ometlanti was collected on substrates composed most- ly of mud. Etymology.—The specific name is from the nahuatl language, ome meaning two, and tlanti meaning tooth, and is a noun in ap- position. The name is given in reference to the two anterolateral teeth of the carapace. Remarks. —Pseudorhombila ometlanti is the fourth species of the genus Pseudorhom- bila known to occur in the western Atlantic. The new species can be distinguished from the other species on the basis of carapace texture and dentition, and shape of male and female abdomen. Also, the forms of the first and second pleopods in the male, and the female gonopores, are distinctive (Table 1). The carapace granulation of P. ometlanti is Stronger than in P. quadridentata (Fig. 1b) and P. guinotae (Fig. 1c), but it is less strong than in P. octodentata (Fig. 1d). With re- spect to carapace shape and dentition, P. ometlantiis similar to P. quadridentata. Both have two anterolateral teeth, whereas the other two species have three and four teeth. However, Hernandez-Aguilera (1982), re- ported that P. guinotae can present two an- terolateral teeth. Abdomen shape is another character that distinguishes the new species. The abdomen in the other three species is narrower than in P. ometlanti, with seg- ments 3—5 fused and undistinguishable su- tures (Fig. 3b, c, d). The holotype male of P. ometlanti has a broader abdomen than the other three species, and is fitted in a shallower and broader abdominal depres- sion. Although in male paratypes the ab- domen is not so broad as in the holotype, it can be easily distinguished from the ab- domen of the other three species. In all the specimens examined of P. ometlanti, the su- tures of the abdominal segments are pres- ent; only the suture of segment 4 is inter- rupted in a small portion of the median line. With respect to pll, the four species of Pseudorhombila show distinct differences in PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON the shape of lobe apex and number of spines (Figs. 4c—d, e-f, g—h; Table 1). The P12 of the three previously known species is slight- ly shorter than in P. ometlanti, not recurved, and the terminal processes are distally ta- pered; whereas in the new species the ter- minal process is strap-like (Fig. 5b, c, d). Although the female abdomen of all four species have all segments free, the last seg- ment differs in shape (Fig. 6b, c). The gon- opores in the four species also differ. Pseu- dorhombila guinotae and P. octodentata ex- hibit oval-shaped gonopores, whereas in P. quadridentata and P. ometlanti the gono- pores are nearly circular (Fig. 7b, c, d). The following key separates the species of Pseudorhombila known in the western At- lantic. Key to the Species of Pseudorhombila 1. Anterolateral margin of carapace with 2) teeth 30 mm and total lengths oc- casionally >110 mm. Frontal margin of carapace (Figs. la, 5a, b) with three anterior prominences, each acute or terminated by short spine, median of which is slightly more produced on tri- angular base to form rostrum, lateral of which overlie inner margins of antennal pe- duncles; low, obtuse tooth on anterolateral margin just lateral to outer margin of an- tennal peduncle. Rostrum extending less than 45 visible length of eyestalks in dorsal view. Carapace lacking rostral carina, with distinct linea thalassinica, and with a de- fined dorsal oval marked posteriorly by deep transverse cardiac furrow, furrow extending anteroventrally to either side above linea thalassinica as shallow sinuous groove de- marcating posterior half of dorsal oval. Shallow cervical groove originating imme- diately below linea thalassinica on anterior half of branchiostegite and curved antero- ventrally, intersecting raised sinuous ridge in anterior third of branchiostegite; portion of ridge anterior to intersection positioned ventrolaterally to rounded hepatic boss and usually sculpted by weak crenelations on upper surface. Subantennular region of ep- istome with dense tuft of long setae brack- eting those of antennular peduncle. Eyestalks flattened, weakly concave dor- sally, length equal to or slightly less than 2 times basal width, in dorsal view tips reach- ing to or nearly to distal end of basal an- tennal article; mesial surfaces of eyestalks 269 closely opposed over proximal portion of midline, divergent terminally; weakly sin- uous anterolateral margin arching from broad base of eyestalk to anterior taper, joining mesial margin anteriorly in variable terminus, ending as spinule, tubercle or acute corner; rarely with subterminal spinule or tubercle in addition to terminal one. Cornea dark, area of pigmentation sexually dimor- phic; in both sexes, rounded corneal surface is small, subterminal, anterolateral in po- sition, larger relative to eyestalk and more bulbous in immature than in adult animals; in adult males, pigmented area large, often diffuse, almost broad as stalk, often occu- pying much of distal third to *% of eyestalk; in adult females and juveniles, pigmented area smaller, more restricted to area of rounded corneal surface. Antennular peduncle shorter and heavier than antennal peduncle, terminal article slightly longer than penultimate and reach- ing to or beyond midlength of terminal ar- ticle on antennal peduncle; penultimate and terminal articles of peduncle with ven- tromesial and ventrolateral rows of long se- tae, those of ventrolateral row longer, dens- er, and continued onto ventral ramus of fla- gellum; rami of flagellum subequal in length, 4—5 times length of terminal article of pe- duncle; dorsal ramus with sparse tufts of long setae, distal half with heavier articles bearing dense tufts of short ventral setae, articles comprising tapered tip with dense line of short ventral aesthetascs. Antennal peduncle with penultimate article subequal to or slightly longer than terminal article; basal article with dorsolateral carina above laterally produced excretory pore; second article with deep, diagonal ventral suture, distolaterally with single tuft of setae, small rounded vestige of dorsal scale articulated at joint with third article; third article elon- gate, narrower than second, slightly longer than combined lengths of first two, proxi- molaterally with partially fused condylar process articulated to distolateral extreme of second article; fourth article narrower than 270 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON b,d,f,h,j oc e, i a g Fig. 1. Sergio mericeae, new species, type materials from Fort Pierce Inlet, Florida; a—c, f, male holotype (cl 29.8 mm), USNM 268645; d, e, female paratype (cl 26.5 mm), USLZ 3544. a, Carapace front, eyestalks, basal antennal segments, dorsal view, setae not shown; b, Major cheliped of mature male, carpus and chela, external surface; c, Major cheliped of mature male, ischium and merus, external surface; d, Major cheliped of mature female, carpus and chela, external surface; e, Major cheliped of mature female, ischium and merus, external surface; f, Telson, uropods and part of sixth abdominal segment, dorsal view. Sergio guassutinga (Rodrigues) from Aracaju, Brazil, female (cl 23.5 mm), USNM 268644; g, Carapace front, eyestalks, basal antennal segments, dorsal view, setae not shown; h, Major cheliped of mature female, carpus and chela, external surface; 1, Major cheliped of mature female, ischium and merus, external surface; j, Telson, uropods and part of sixth abdominal segment, dorsal view. Scale lines indicate 5 mm. VOLUME 108, NUMBER 2 third; flagellum 2-2.5 times length of an- tennular flagellum. Mandibles (Fig. 2a) with large, 3-segmented palp, elongated third article of palp slightly tapered and terminally round- ed, long setae on external and distal surfaces of second article and on proximal extensor surface of third, field of short setae on most of extensor surface and terminal end of third article; incisor process with well-defined, terminally corneous teeth on cutting mar- gin, teeth largest, less coalesced on proximal two thirds of cutting margin, internal sur- face with lip giving rise to molar process proximal to incisor teeth, molar process with 4—6 small marginal teeth; paragnath (not fig- ured) rounded, scaliform, poorly sclero- tized, set against and below molar process. First maxilla (Fig. 2b) with endopodal palp long, narrow, terminal article deflected proximally at articulation; proximal endite densely setose on sinuous margin, terminal lobe additionally with field of large, termi- nally bifurcate setae; distal endite elongate, proximally narrow, broadening terminally where armed with short stiff bristles and field of short setae, some with bifurcate tips; exopodite low, truncate and setose. Second maxilla (Fig. 2c) with endopod narrowed distally, first and second endites each lon- gitudinally subdivided, internal surface of first endite fused to broad, rounded, mar- ginally setose plate; exopod forming large, broad, scaphognathite. First maxilliped (Fig. 2d) with proximal endite narrowly pro- duced, dentiform, marginally setose; distal endite robust, subquadrate, mesial half of external surface and margins heavily setose, longest setae terminal, internal surface con- cave; exopod ovoid, divided by transverse suture marking notch on mesial margin, longest setae in field on external surface and mesial margin proximal to notch; epipod large, broad, anterior end tapered to narrow terminus. Second maxilliped (Fig. 2e) with long, narrow endopod; endopodal merus length exceeding 4 times width, flexor mar- gin with dense fringe of long, close-set setae; 271 carpus short; propodus slightly arcuate, heaviest distally, length almost 3 times width; dactylus short, about % length of propodus, with terminal brush of stiff bris- tles; exopod phylloform, nearly as long as endopodal merus, marginally fringed by long setae; epipod small, sutures subdividing into 2 proximal lobes and minute trianguliform terminal lobe. Third maxilliped (Figs. 2f, 3a) without exopod; endopod with long, dense setation on mesial margin, terminal 3 articles also with long setation on extensor margins; length of endopodal merus-ischi- um about 2 times width; ischium subquad- rate, slightly longer than broad, diagonal length in adults about 1.6 times width at suture with merus, proximomesial margin forming subacute corner, internal surface with poorly defined, longitudinally oriented elevation bearing curved row of small den- ticles; merus subtriangular, distinctly broader than long; carpus heavy and sub- triangular, with setose lobe on flexor mar- gin, internal surface with dense field of fine setae in distal third; propodus large, subquadrate in adults, height often exceed- ing length, internal surface with narrow me- dian field of fine, dense setae, opposable margin slightly emarginate; dactylus nar- row, arcuate, in adults distinctly shorter than height of propodus, terminally with small brush of stiff bristles. Branchial formula includes exopods and epipods as described for first and second maxillipeds above; branchiae limited to sin- gle rudimentary arthrobranch on second maxilliped, pair of arthrobranchs on third maxilliped, and pair of arthrobranchs on each of the first through fourth pereopods. First pereopods of two sides forming dis- similar chelipeds, major cheliped heavy, massive in adults of both sexes (Fig. 1b-e), much less altered from minor cheliped in juveniles than in adults (Fig. 5c—f). Major cheliped of adults strongly calcified; ischi- um slender, superior margin weakly sinu- ous, inferior (flexor) margin with row of minute denticles; merus about twice as long P22 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON SN NS RNS NS a SA f a,b,c,d,e Fig. 2. Sergio mericeae, new species, type materials from Fort Pierce Inlet, Florida; ae, female paratype (cl 26.5 mm), USLZ 3544; f, male holotype (cl 29.8 mm), USNM 268645, right mouthparts. a, Right mandible, external surface; b, First maxilla, external surface; c, Second maxilla, external surface; d, First maxilliped, external surface; e, Second maxilliped, external surface, rudimentary arthrobranch not shown; f, Third maxilliped, internal surface. Scale lines indicate 5 mm. as high, superior margin weakly sinuous, sometimes with few minute tubercles prox- imally, inferior margin more strongly ar- cuate, proximally with strong projecting of- ten bifurcate process of 2—3 fused spines, margin distal to process with short unarmed region beyond which margin is lined by small inferodistally directed teeth; carpus broad, relatively shorter and higher in adults than in immature specimens, in large adults length about %, of height, superior margin nearly straight, forming thin unserrated keel slightly overhanging internal surface, prox- imoinferior margin regularly rounded in outline, serrations of keel most evident on internal surface; propodus heavy, length (in- cluding fixed finger) in large adults about 1.7 times height, height greatest basally, su- perior margin of palm forming unserrated keel, especially in proximal half, inferior VOLUME 108, NUMBER 2 ——— TN SS SE a => > aN = 285 11 Figs. 11-15. Reynoldsia brevitarsis, 6: 11. Tibia III, anterior view (from Albuquerque’s notes, no scale); 12. Abdomen, lateral view; 13. Sternite V, dorsal view; 14. Cercal plate and surstyli, dorsal view; 15. Cercal plate, surstyli and phallic complex, lateral view. 286 Genitalia. —Cercal plate large (Fig. 14); distiphallus large (Fig. 15). Material examined. — Argentina: 1 6, Bar- iloche, paratype no. 49875, Rio Negro, Nov 1926, R. & E. Shannon. Reynoldsia pteropleuralis Malloch Figs. 16-18 Reynoldsia pteropleuralis Malloch, 1934:232 [key], 237-238 [description 42], 238 [Fig. 41b: hind tibia, 41c: forceps and penis]; Pont, 1972:36 [catalog], Carvalho et al., 1993:132. Holotype. —°. Argentina. Lake Nahul Huapi (BMNH, London). Diagnosis. —Frons brown, silvery polli- nose; antenna with flagellomere 1 about 2.5 length of pedicel; acrostichal presutural bristles not differentiated; brown vittae at acrostichal and intralar surfaces, the 2 lat- eral ones reaching apex of scutellum; halter yellow with knob black; anepimeron with short cilia at disc; hind femur at postero- ventral surface with a comb-like row of bris- tles at apical 3rd; hind tibia at anteroventral surface with 2 bristles at middle 3rd; ab- domen grayish pollinose with lateral quad- rate dark spots in all tergites. Sternite VII with a row of small spines (Figs. 16-17). Spermatheca as in Fig. 18. Differs from R. pectinata as follows: Coloration. —Parafacial, parafrontal, face, lunule, and genae brown, silvery pollinose. Halter brownish yellow with knob dark brown. Legs dark brown with apical half of femora and entire tibiae yellow. Abdomen grayish pollinose as thorax, with lateral quadrate dark spots in all tergites. Female.—Body length 9.0 mm; wing length 7.5 mm. Head. —Eyes separated, at anterior ocelli level, by a space about 0.38 of head width. Antenna with flagellomere 1 about 2.5 times length of pedicel. Thorax. —Acrostichal presutural not dif- ferentiated, humerus with 3 bristles. Ane- pimeron with a few fine hairs at middle. Mid PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON femur at anterior surface with a row of bris- tles at basal half and one bristle inserted at apical 3rd, ventral surface with sparse long bristles at basal half, posterior surface with a complete row of bristles and with 2 preap- ical bristles. Mid tibia at anterodorsal sur- face with 3 long bristles at middle 3rd, pos- terior surface with 2 bristles at middle 3rd. Hind femur at anterodorsal and anteroven- tral surfaces with a row of long and sparse bristles. Hind tibia at anteroventral surface with 2 bristles at middle 3rd, anterodorsal and posterodorsal surfaces with 3 bristles at basal half, the anterior ones weaker. Abdomen. —Tergites I-IV with devel- oped lateral bristles at disc and apex. Genitalia. —Epiproct with a reentrance on basal margin; sternite VII with a row of small spines (Figs. 16-17). Material examined. — Argentina: 1 °, San Martin de los Andes, 1900 m, 28 Dec 1989. S. A. Marshall. Comments.—Malloch (1934: 238, Fig. 41c) noted the peculiar male hypopygium, a character that should distinguish this spe- cies from its congeners. The modification of male phallic complex, however, is found in all studied species. Reynoldsia rufoapicata Malloch, 1934 Figs. 19-26 Reynoldsia rufoapicata Malloch, 1934:231 [key], 239 [description 42], 238 [Fig. 41d: hind tibia]; Pont, 1972:36 [catalog], Car- valho et al., 1993:133. Holotype. —é. Argentina. Bariloche (USNM, Washington). Diagnosis. —Frons.dark, with whitish-gray or brownish-gray dust; antenna with flagel- lomere | about 2.0 length of pedicel; acros- tichals presutural bristles differentiated; brown vittae at acrostichals and intralar sur- faces, the 2 lateral ones reaching apex of scutellum; halter yellow with knob black; hind femur at posteroventral surface with- out a comb-like row of bristles at apical 3rd; hind tibia at anteroventral surface with 2 VOLUME 108, NUMBER 2 AIRE Z Maura” PPTL C CT Ls rane WLU O*} een Figs. 16-21. Reynoldsia pteropleuralis, 9: 16. Ovipositor, dorsal view; 17. Ovipositor, ventral view; 18. Spermatheca. Reynoldsia rufoapicata, 6 and °: 19. Sternite V, dorsal view; 20. Cercal plate and surstyli, dorsal view; 21. Cercal plate, surstyli and phallic complex, lateral view. 287 288 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Awl WAZ PSG ro Wy " ’ ’ ween mys tie iy yyy MOU ye he tt une lp flue 4, yu’ pi lind > n te tel nest Www Q'} 1 1, i TLL eZ Wee al WU Q'} revel Yr! *. af Sigg aly Figs. 22-26. Reynoldsia rufoapicata, 6 and 9: 22. Phallic complex, lateral view; 23. Proboscis, 2; 24. Ovi- positor, dorsal view 25. Ovipositor, ventral view; 26. Spermatheca. VOLUME 108, NUMBER 2 bristles at middle 3rd; abdomen black, lat- eral spots more visible in female. Differs from R. pectinata as follows: Coloration. —Parafacial, parafrons, face, lunule and genae brown, gray pollinose. Legs dark brown with apical 3rd of femora and entire tibiae yellow. Abdomen black, lateral spots more visible in female. Male. —Body length 8.0 mm; wing length 7.5 mm. Head. —Eyes separated, at anterior ocelli level, by a space about 0.40 of head width. Frontal row with 8 long pairs of bristles. Antenna with flagellomere 1 about 2.0—2.1 times length of pedicel. Thorax. —Mid femur at anterior surface with a row of bristles at basal 73 and one at apical 3rd, anteroventral surface with a row of bristles at basal half, posterior surface with 2 preapical bristles. Mid tibia at an- terodorsal surface with 3 long bristles at middle 3rd, posterior surface with 2 bristles at middle 3rd. Hind femur at anterior sur- face with 3 bristles at middle 3rd, antero- dorsal surface with a complete row of bris- tles, anteroventral surface with a row of bristles at basal 73 and one bristle at apical 3rd. Hind tibia at anterodorsal and pos- terodorsal surfaces with 3 bristles at middle 3rd, anteroventral surface with 2 bristles at middle 3rd. Abdomen. —Sternite V with anterior con- cavity deep (Fig. 19). Genitalia. —Cercal plate straight; surstyli longer than cercal plate (Figs. 20, 21, and 22). Female.—Body length 7.0 mm; wing length 8.0 mm. Similar to male. Genitalia. —Epiproct completely divided; sternite VII with row of short spines in each side (Figs. 24—25). Spermatheca as in Fig. 26. Material examined. — Argentina: 6 4, 3 8, San Martin de los Andes, slope of Cerro Chapelco, 1.500 m, 28 Nov 1989. S. A. Marshall (MNRJ). Comments. — Malloch (1934) stated that 289 this species is very similar to R. pectoralis, differing mainly in the lack of pteropleural hairs, coloration on apex of femora and the apical bristling of hind tibia. Reynoldsia scutellata Malloch Reynoldsia scutellata Malloch, 1934:232 [key], 234—235 [description 2]; Pont, 1972: 36 [catalog], Carvalho et al., 1993:133. Holotype. —é. Argentina, Puerto Blest (BMNH, London). Diagnosis. —Frons brown, brown dusted; antenna with flagellomere 1 about 3.0 times length of pedicel; acrostichals presutural bristles differentiated; brown vittae at ac- rostichals and intralar surfaces, none of them reaching scutellum, which is uniformly brownish-gray; halter brownish-yellow; tro- chanters yellow; hind femur at posteroven- tral surface without a comb-like row of bris- tles at apical 3rd; hind tibia at anteroventral surface with 1 bristle at middle 3rd; abdo- men grayish with lateral dark-brown sub- triangular spots in all tergites. Differs from R. pectinata as follows: Coloration. —Parafrontal, frons and lu- nule brownish yellow. Parafacial and gena silvery pollinose. Gena brown at vibrissal level. Mesonotum with the lateral brown vittae not continuing to scutellum, which is entirely brownish gray. Halter brownish- yellow. Trochanters yellow. Abdomen gray, tergites III and IV with very faint lateral brown spots. Female.—Body length 7.0 mm; wing length 6.5 mm. Head.—Eyes separated by a space about 0.55 of head width, at anterior ocelli level. Thorax.—Humerus with 2 bristles, the outer one longer. Fore femur at postero- dorsal and posteroventral surfaces with a row of bristles. Fore tibia at posterior sur- face with a median bristle. Mid femur at anterior surface with 4 sparsed bristles at basal half, ventral surface with 3 bristles at 290 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Figs. 27-31. Reynoldsia trochanterata, 2: 27. Ovipositor, dorsal view; 28. Ovipositor, ventral view; 29. Spermatheca; 30. Egg; 31. Egg, lateral view. VOLUME 108, NUMBER 2 basal 3rd, the last one longer, one median bristle and a row of short bristles at apical 3rd. Mid tibia at anterodorsal surface with 2 bristles at middle 3rd, apical bristles at anteroventral and posteroventral surfaces long. Hind femur at anterodorsal, anterior, anteroventral, ventral and posteroventral surfaces with a row of bristles, the anterior one irregular. Hind tibia at anteroventral surface with 3 bristles at middle 3rd, the basal one weaker, anterodorsal surface with 4 bristles, posterodorsal surface with 5 bris- tles, the 2nd and the 4th longer, dorsal and posterodorsal surfaces with a preapical bris- tle, anterior surface with 3 apical bristles and ventral surface with one. Genitalia. —Ovipositor not so long as in the other species; tergites larger (Figs. 27- 28). Spermatheca as in Fig. 29. Egg as in Figs. 30 and 31. Material examined.—Chile: 1 °, Cerros de Nahuelbuta, Angol 650 m, 12 Oct 1931, D. S. Bullock (MNRJ). Comments. —Malloch’s (1934) descrip- tions of R. scutellata and R. trochanterata are very similar. Both species have yellow trochanters and a uniformly colored scutel- lum, which are diagnostic characters to Malloch (1934). A more detailed analysis of the types could clarify if they are really synonyms. 291 Acknowledgments I am grateful to Dr. Stephen Marshall (University of Guelph, Ontario) for the do- nation of specimens to the Museu Nacional, Rio de Janeiro, and to Dr. Raymond Gagné (Systematic Entomology Laboratory, De- partment of Agriculture, Washington, D.C.) for suggestions and comments. Literature Cited Carvalho, C. J. B. de. 1989. Classificacao de Mus- cidae (Diptera): uma proposta através de analise cladistica.— Revista Brasileira de Zoologia 6(4): 627-648. , M. S. Couri, A. Pont, D. Pamplona, & S. M. Lopes. 1993. Part II. Muscidae. Jn A catalogue of the Fanniidae and Muscidae of the Neotrop- ical Region. C. J. B. Carvalho, ed., Sociedade Brasileira de Entomologia: 1-201. Cour, M. S., & S. M. Lopes. 1985. Neotropical gen- era of Coenosiinae—nomenclatural notes and key to identification (Diptera, Muscidae). — Re- vista Brasileira de Biologia 45(4):589—595. Hennig, W. 1965. Vorarbeiten zu einem phyloge- netischen System der Muscidae (Diptera: Cy- clorrapha).—Stuttgart Beitrage Naturkunde 141: 100 pp., 53 figs. Malloch, J. R. 1934. Muscidae. Jn Diptera of Pata- gonia and South Chile VII (2):171-346, figs. 22— 60. London. Pont, A. C. 1972. A catalogue of the Diptera of the Americas south of the United States.— Museu de Zoologia, Universidade de Sao Paulo 97:111 Dp. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(2):292-297. 1995. Rediagnosis of the brittlestar genus Ophiosyzygus and notes on its type species O. disacanthus (Echinodermata: Ophiuroidea: Ophiomyxidae) based on the type specimens from Japanese waters and new material from the Gulf of Mexico Richard L. Turner and Robyn M. Heyman (RLT) Department of Biological Sciences, Florida Institute of Technology, 150 West University Boulevard, Melbourne, Florida 32901, U.S.A.; (RMH) Florida Department of Environmental Protection, Florida Marine Research Institute, 100 Eighth Avenue S.E., St. Petersburg, Florida 33701, U.S.A. (Present address) Georgia Natural Heritage Program, Wildlife Resources Division, Georgia Department of Natural Resources, 2117 U.S. Highway 278 S.E., Social Circle, Georgia 30279, U.S.A. Abstract.—Two specimens of the ophiomyxid brittlestar Ophiosyzygus dis- acanthus H. L. Clark, 1911 represent the first new record of the species since the original description. The new specimens are from the Gulf of Mexico and are conspecific with the type material from Japan. The generic diagnosis is emended to include radial shields and dorsal arm plates. Evidence is given that these types of ossicle have been overlooked frequently in ophiomyxids. In 1911, H. L. Clark described a new monotypic genus of ophiomyxid brittlestar, Ophiosyzygus, based on two specimens from U.S. Fisheries Commission Steamer A/ba- tross stations off the southwestern coast of Japan. Among other characters, the genus was diagnosed as lacking radial shields and dorsal arm plates. Two specimens of the type species, O. disacanthus H. L. Clark, 1911, were recently collected from the Gulf of Mexico off the southwestern coast of Florida, U.S.A., and they differ little from the type material. Our discovery of radial shields and dorsal arm plates in the Gulf of Mexico specimens and, subsequently, in Clark’s North Pacific specimens prompts us here to emend the generic diagnosis. Ex- amination of some museum material and a review of literature on the Ophiomyxidae indicate to us that small radial shields and thin dorsal arm plates have been overlooked often in this family. Family Ophiomyxidae Ljungman, 1867 Subfamily Ophiomyxinae Ljungman, 1867 Genus Ophiosyzygus H. L. Clark, 1911 Ophiosyzygus. —Clark, 1911:275-276 (di- agnosed), 279 (compared to Ophiolepto- plax).—Clark, 1915:172 (listed).—Fell, 1960:14 (key).—Spencer & Wright, 1966: 89 (listed in systematic review). Diagnosis (emended).— Disc covered with skin in which are embedded, at least near margin, numerous minute delicate scales, on which are borne larger, presumably cal- careous granules. Radial shields small. Ten- tacle scales lacking. Arm spines few, with upper ones successively united to each other by a broad, thin, horizontal membrane. Dorsal arm plates thin, simple, multiper- forate. Teeth a cluster ofa few, minute, rough spinelets at apex of jaw. Oral papillae small, few, and unlike teeth. VOLUME 108, NUMBER 2 Type species. — Ophiosyzygus disacanthus H. L. Clark, 1911, by original designation. The genus is monotypic. Etymology. —Given by Clark (1911), in reference to the webbing of the upper arm spines by horizontal and vertical webs of skin in dried and alcoholic specimens. Hen- dler & Miller (1991) found that the webbing of arm spines on dried Ophiogeron supinus is an artifact from desiccation of the gelat- inous tissues of the arm. They were unable to induce swimming in live O. supinus by methods that worked successfully on four other species. These authors joined Mor- tensen (1932) in refuting Clark’s (1911) ini- tial implication and his (Clark 1941) later prediction that the webbed spines of ophio- myxids are used for swimming. Based on the findings of Hendler & Miller (1991), arms of Ophiosyzygus disacanthus in life proba- bly are fleshy, not webbed. Remarks. —Clark’s (1911) diagnosis of the genus included the absence of radial shields and dorsal arm plates. We found both kinds of plate in the material listed below for O. disacanthus. Clark (1911, fig. 139a) illus- trated small triangular radial shields (Fig. 1B) on O. disacanthus, but he described them as the “expanded ends of genital plates.” Dissection of the new material from the Gulf of Mexico and re-examination of the types revealed that they indeed are radial shields, each articulating with a paired slender gen- ital plate and genital scale (Fig. 1A). In fact, Clark’s (1911) figure includes the ends of the genital plates below the misidentified radial shields (Fig. 1B). The dorsal arm plates of type and new specimens are simple mul- tiperforate plates (Fig. 2) that are difficult to see in situ and similar to those of Ophio- dera serpentaria (Mortensen 1933a, figs. 1, 2) and of Ophioscolex inermis (Mortensen 1933b, fig. 36b). Clark (1911) overlooked them, although he reported similar dorsal arm plates in Ophioleptoplax megapora. Paterson (1985:18) described the Ophio- myxidae generally as having “‘rudimentary and inconspicuous radial shields.”’ Equally 293 Or more rudimentary and inconspicuous in some ophiomyxids are the dorsal arm plates. Byrne & Hendler (1988) concluded that the Ophiomyxidae have a reduced skeleton in which the function of some parts has been largely assumed by mutable collagenous tis- sue. The loss of basic elements of the ophiu- roid skeleton in some ophiomyxids might, therefore, not be surprising. Numerous gen- era have been diagnosed as lacking radial shields and dorsal arm plates. Subsequent examination by us and other authors has, however, confirmed the presence of radial shields or dorsal arm plates in some of these genera. Clark (1911) mistook radial shields for the distal ends of genital plates in his new genera Ophioleptoplax and Ophiocynodus as he did in Ophiosyzygus. Our examination of type material (Ophioleptoplax megapora, USNM 25619; Ophiocynodus corynetes, USNM 25607) revealed his error. Koehler (1914) described Ophioleptoplax atlantica from a specimen that only had “traces [of its disc] left on the upper face of the arms.” Koehler did not indicate the presence of ra- dial shields; but we have found them at- tached to the genital plates and scales in the holotype (USNM 32304) as they appear un- labelled in Koehler’s (1914) photograph (pl. 15, fig. 6) of the specimen. Ophioscolex was included by Fell (1960) among the ophio- myxid genera lacking radial shields, but Ly- man (1882) and Paterson (1985) described them to be present, although inconspicuous, in the type species, O. g/acialis. Verrill (1899) described as “‘rudimentary”’ the radial shields that were overlooked by Lyman (1875) in his original description of Ophio- scolex stimpsonii. Mortensen (1933b:314) alluded to the po- tential for oversight of dorsal arm plates when he wrote about Ophioscolex inermis: ‘“‘At first glance one would rather say that dorsal plates are lacking; but... they are so delicate as to let the vertebrae shine through .... Dorsal arm plates in many ophio- myxid species are described as small, thin, 294 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A ipenle Ophiosyzygus disacanthus. A. Paratype, oblique view of arm base, with part of ventral interradial integument removed to reveal basal five lateral arm plates with spines and the articulation of the radial shield (rs) with the genital scale (gs) and genital plate (gb). Calcareous granules (g) are shown among the dorsal scales of the disc. B. Aboral view of part of the disc and two arms (after Clark 1911, fig. 139a; probably based on the Paratype). Arrows indicate the exposed ends of the genital plates below the radial shields (rs). multiperforate, and hyaline (Lyman 1878, 1880, 1882, 1883; Clark 1911; Koehler 1914; Matsumoto 1915; Mortensen 1933a; Fell 1960). In many cases, the plates were overlooked in earlier or later studies; and we found similar delicate plates in Ophio- cynodus corynetes, which Clark (1911) di- agnosed as lacking dorsal arm plates. Ly- man’s (1883) statements about the absence of dorsal arm plates in Ophiobyrsa hystricis, Ophiobyrsa rudis, and Ophiobrachion un- cinatus were qualified by his reference to paired nodules or spiny tubercles over each arm joint. The purported absence of dorsal arm plates should be re-evaluated in several other ophiomyxid genera, particularly in light of the recent discovery by Hotchkiss (1993) of dorsal arm plates in the Paleozoic Oegophiurida. It may be assumed that the small radial shields and delicate dorsal arm plates, as well as the apparent absence of either type of plate, in ophiomyxids are apomorphic reductions or losses of typically robust plates found in many other ophiuroids. Byrne & Hendler (1988; see above) gave a plausible explanation for skeletal reduction in ophio- myxids. In view of the high potential for homoplasy in reductive character states, speculation about the phylogenetic infor- mativeness of these characters is, however, best deferred until a more complete revision of the Ophiomyxidae can be accomplished. Ophiosyzygus disacanthus H. L. Clark, 1911 Figs. 1-2 Ophiosyzygus disacanthus Clark, 1911:18 (station list), 276-277 (described, part), fig. 139.—Clark, 1915:172 (material).— Fell, 1960:14 (listed). —Spencer & Wright, 1966:89 (listed).—Downey, 1969:183 (listing of type material, part). Non Ophiosyzygus disacanthus. — Clark, 1911:277 (material, part).—Downey, 1969:183 (listing of type material, part). Material examined. —Holotype: USNM 25671, 12 mm disc diam, Albatross, Sta 4934, 16 Aug 1906, 30°58'20’N, 130°32'00’E, 103-152 fm, rocky, off Sata Misaki Light, Eastern Sea, Japan.—Para- type: MCZ 3277, 8 mm disc diam, Alba- tross, Sta 4936, 16 Aug 1906, 30°54’'40’N, 130°37'30”E, 103 fm, stones, off Kagoshima Gulf, Japan. Non-type material: USNM E44353, 1 dry specimen, 5 mm disc diam, U.S. Bureau of Land Management, Southwest Florida Shelf VOLUME 108, NUMBER 2 Fig. 2. Ophiosyzygus disacanthus. LACM 81-218.1. A large dorsal arm plate and five small plates of the ventral interradius of the disc. Ecosystems Study, Year 2, Cruise II (BLM 321-II), Sta 35, 26 Jul 1981, 25°44.8'N, 84°21.0'W, 159 m, thin sand and rubble over rock, triangle dredge, southeastern Gulf of Mexico.—LACM 81-218.1, 1 alcoholic specimen, disc damaged, BLM 321-II, Sta 36, 3 Aug 1981, 25°16.8'N, 83°57.4'W, 127 m, deep sand with some thin sand veneer over rock. Diagnosis. — With characters of the genus. Description.—The two specimens from the eastern Gulf of Mexico (southwestern Florida shelf) are not a new species, for they differ from the type material only in minor ways. Arm spines are thornier, many bear- ing two rows of two or three subapical teeth on opposite sides of the shaft. Fewer arm spines on the type specimens are thorny, and they bear fewer teeth (one or two per row). There are generally three oral papillae per row on the jaws, but the number is high- ly variable within specimens (Japan speci- mens, 2-4; Florida, 2-5). The integument also includes white, opaque, irregular gran- ules (Fig. 1) that reach and extend above the surface and appear like icebergs. The granules look imperforate, lacking the typ- ical stereomal-stromal organization of echi- noderm ossicles. The granules probably are not calcitic ossicles, for they disaggregate in sodium hypochlorite and do not effervesce in weak acid. Granules of the two specimens from Florida are smaller (84 + 21.9 um, n = 9: 124 + 32.8 wm, n = 10) than those of the holotype (232 + 67.8 um, n = 34) and paratype (221 + 53.8 um, n = 21). Consid- ering the poor condition of the types and new specimens, more rigorous comparisons must await the availability of better mate- rial. Flat, multiperforate plates 111 + 22 wm (n = 24) in diameter are embedded deep in 296 the skin of the ventral interradius of the disc (Fig. 2). They are visible in the type speci- mens without treatment with sodium hy- pochlorite and were probably overlooked by Clark (1911), who wrote that the ventral interradius is naked. Five measurable, prolately spheroidal eggs out of seven removed from an ovary of one specimen (LACM 81-218.1) ranged in greatest diameter from 100 um to 140 um. Ossicles of the gonadal wall are irregular perforate plates. Type locality. —Off the southern coast of Kyushu, Japan. Distribution. —Japan; eastern Gulf of Mexico. Habitat.—Type specimens were taken from rocky and stony substrata at 188-278 m in the Pacific Ocean. Stations from which specimens were collected in the eastern Gulf of Mexico at 127-159 m were described by Woodward-Clyde Consultants & Continen- tal Shelf Associates, Inc. (1985). At both stations in the Gulf of Mexico, the bottom was gently sloping, rarely with outcrops of rock or other relief, and mostly covered with a veneer of sand over hard substratum (Sta- tion 35) or with deep sand (Station 36). The most diverse benthic taxa in descending or- der at Station 35 were cnidarians, echino- derms, sponges, and crustaceans; and at Sta- tion 36, crustaceans, cnidarians, echino- derms, and sponges. A green encrusting alga dominated the epibenthic cover at Station 35. At Station 36, cover was dominated by sponges in summer and by crinoids in win- ter. Etymology. —Given by Clark (1911), in reference to the paired arm spines. Remarks. —In his original description of O. disacanthus, Clark (1911) was uncertain of the identity of one paratype (USNM 26217), which lacked a disc. The specimen clearly is not O. disacanthus. Although dis- tal arm spines are glassy and terete, proxi- mal arm spines are flattened, robust, and obtuse, and they do not appear glassy. Most arm segments bear three arm spines, rarely PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON four, and only a few proximal segments bear the two that are the typical number for O. disacanthus. Upper arm spines are not webbed. Much of the oral structure is ob- scured by mud, but the jaws can be seen to carry a vertical row of 3-6 teeth (one jaw has none), an apical cluster of tooth papillae, and 4—8 spiniform oral papillae. Acknowledgments The material for this study was provided by K. Spring and was collected during the Southwest Florida Shelf Ecosystems Study, Year 2, under contract 14-12-0001-29144 (formerly AA851-CT1-45) to Continental Shelf Associates, Inc., from the U.S. De- partment of the Interior, Minerals Manage- ment Service. We thank K. Fauchald, Smithsonian Institution, National Museum of Natural History, who approved our dis- tribution of samples between repositories, and R. M. Woollacott, Harvard University, Museum of Comparative Zoology, and D. L. Pawson, C. Ahearn, and M. E. Downey, National Museum of Natural History, for the loan of material in their care. Helpful comments on the manuscript were given by J. H. Dearborn, University of Maine, G. Hendler, Natural History Museum of Los Angeles County, and P. M. Mikkelsen, Del- aware Museum of Natural History. We ac- knowledge with gratitude the collaboration with and encouragement from J. E. Miller (formerly of Harbor Branch Oceanographic Institution) earlier in this study. Literature Cited Byrne, M., & G. Hendler. 1988. Arm structures of the ophiomyxid brittlestars (Echinodermata: Ophiuroidea: Ophiomyxidae). Pp. 687-695 in R. D. Burke, P. V. Mladenov, P. Lambert, & R. L. Parsley, eds., Echinoderm biology. A. A. Balkema, Rotterdam, 818 pp. Clark, H. L. 1911. North Pacific ophiurans in the collection of the United States National Muse- um.— United States National Museum Bulletin 75:1-302. . 1915. Catalogue of Recent ophiurans: based on the collection of the Museum of Comparative VOLUME 108, NUMBER 2 Zoology.—Memoirs of the Museum of Com- parative Zoology at Harvard College 25:165- 376. 1941. Reports on the scientific results of the Atlantis Expeditions to the West Indies, under the joint auspices of the University of Havana and Harvard University. The echinoderms (oth- er than holothurians).—Memorias de la Socie- dad Cubana de Historia Natural “Felipe Poey” 15(1):1-154. Downey, M. E. 1969. Catalog of Recent ophiuroid type specimens in major collections in the Unit- ed States.—United States National Museum Bulletin 293:1-239. Fell, H. B. 1960. Synoptic keys to the genera of Ophiuroidea.— Zoology Publications from Vic- toria University of Wellington 26:1—44. Hendler, G., & J. E. Miller. 1991. Swimming ophiu- roids—real and imagined. Pp. 179-190 in T. Yanagisawa, I. Yasumasu, C. Oguro, N. Suzuki, & T. Motokawa, eds., Biology of Echinoder- mata. A. A. Balkema, Rotterdam, 590 pp. Hotchkiss, F.H.C. 1993. Anew Devonian ophiuroid (Echinodermata: Oegophiurida) from New York State and its bearing on the origin of ophiuroid upper arm plates.— Proceedings of the Biologi- cal Society of Washington 106:63-84. Koehler, R. 1914. A contribution to the study of ophiurans of the United States National Mu- seum.— United States National Museum Bul- letin 84:1-173. Ljungman, A. 1867. Ophiuroidea viventia huc usque cognita.—Ofversigt af Kongliga Vetenskaps- Akademiens Forhandlingar, 1866 9:303-336. Lyman, T. 1875. Ophiuridae and Astrophytidae, in- cluding those dredged by the late Dr. William Stimpson. —Illustrated Catalogue of the Muse- um of Comparative Zoology at Harvard College 8(2):1-34. 1878. Ophiuridae and Astrophytidae of the exploring voyage of H.M.S. “‘Challenger’’, under Prof. Sir Wyville Thomson, F.R.S.— Bulletin of the Museum of Comparative Zoology at Har- vard College, in Cambridge 5:65-168. . 1880. A structural feature hitherto unknown among Echinodermata found in deep sea 297 ophiurans.— Anniversary Memoirs of the Bos- ton Society of Natural History 6:1-12. 1882. Report on the Ophiuroidea dredged by H.M.S. Challenger during the years 1873- 76.—Report on the Scientific Results of the Voyage of H.M.S. Challenger during the Years 1873-76, Zoology 5:1-383. . 1883. Report on the Ophiuroidea.— Bulletin of the Museum of Comparative Zoology at Har- vard College, in Cambridge 10:227-287. Matsumoto, H. 1915. A new classification of the Ophiuroidea: with descriptions of new genera and species.— Proceedings ‘of the Academy of Natural Sciences of Philadelphia 67:43-92. Mortensen, T. 1932. On an extraordinary ophiurid, Ophiocanops fugiens Koehler, with remarks on Astrogymnotes, Ophiopteron, and on an albino Ophiocoma. — Videnskabelige Meddelelser fra Dansk naturhistorisk Forening 1 Kobenhavn 93: 1-22. 1933a. Ophiuroidea.—The Danish Ingolf- Expedition 4(8):1-121. . 1933b. Echinoderms of South Africa (Aster- oidea and Ophiuroidea).— Videnskabelige Med- delelser fra Dansk naturhistorisk Forening i Ko- benhavn 93:215—400. Paterson, G. L. J. 1985. The deep-sea Ophiuroidea of the North Atlantic Ocean.— Bulletin of the British Museum (Natural History), Zoology 49: 1-162. Spencer, W. K., & C. W. Wright. 1966. Asterozoans. Pp. 4-107 in R. C. Moore, ed., Treatise on in- vertebrate paleontology, part U, Echinodermata 3, Asterozoa-Echinozoa. Geological Society of America, Boulder, Colorado, and University of Kansas Press, Lawrence, 695 pp. Verrill, A. E. 1899. Report on the Ophiuroidea col- lected by the Bahama Expedition in 1893.— Bulletin from the Laboratories of Natural His- tory of the State University of Iowa 5(1):1-86. Woodward-Clyde Consultants & Continental Shelf As- sociates, Inc. 1985. Southwest Florida Shelf Ecosystems Study — Year 2, vol. 2. United States Department of the Interior, Minerals Manage- ment Service, Metairie, Louisiana, irregular pagination. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(2):298-301. 1995. Designation of a lectotype for Crocodilus siamensis Schneider, 1801 (Reptilia: Crocodylia) Charles A. Ross, Gregory C. Mayer, and Roger Bour (CAR) Department of Vertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, U.S.A.; (GCM) Department of Biological Sciences, University of Wisconsin—Parkside, Kenosha, Wisconsin 53141, U.S.A.; (RB) Laboratoire de Zoologie (Reptiles et Amphibiens), Muséum National d’Histoire Naturelle, 57 rue Cuvier, 75005 Paris, France Abstract. —Schneider (1801) named Crocodilus siamensis on the basis of three specimens from Siam that were described and figured (but not named) in an earlier publication (Anon. 1688). None of the three specimens are known to have been preserved. Cuvier (1807) considered the syntypical series composite and proposed the replacement name Crocodilus galeatus for one of the included species. Cuvier’s (1807) replacement name is assessed and rejected as unjus- tified. In the interest of nomenclatural stability, a lectotype of C. siamensis Schneider is designated. The type locality of C. siamensis is shown to be Louvo, Siam (= Ayutthaya, Thailand). Résumé. —Schneider (1801) a créé ’espéce nominale Crocodilus siamensis a partir de trois spécimens du Siam décrits et figurés (sans étre nommeés) dans un travail publié antérieurement (Anon., 1688). Aucun de ces spécimens n’a ete conserve. Cuvier (1807) a considéré que la série-type de Crocodilus siamensis était composite, et a propose en conséquence un nom de remplacement, Croc- odilus galeatus. La validite de ce nom de remplacement est évaluée, et il est rejeté car étant injustifieé. Dans l’interét de la stabilité de la nomenclature, un lectotype de C. siamensis Schneider est désigné. La localité-type de C. siamensis est done fixée 4 Louvo, Siam (= Ayutthaya, Thailande). Crocodilus siamensis Schneider, 1801, is based on three specimens originally de- scribed from Siam (= Thailand) by “les Péres Jésuites Francois” in a report to the Aca- démie Royale des Sciences, published in 1688. This report, seemingly authored anonymously, is sometimes attributed to Thomas Gouye, and it is so registered in the library of the Muséum National d’Histoire Naturelle, Paris. Pages 1-47 and plates I- III in this report comprise a “Description anatomique de trois crocodiles. Avec les Réflexions de Monsieur du Vernay [sic, = Duverney], de l’Académie Royale des Sci- ences’’. This section is a description of three crocodiles from Louvo, the former capital of the Siamese Kingdom, that measured (in French feet; 1 ft. = 324 mm) 10’ 814”, 9’ 4", and 1’ 414”. The greater part of the descrip- tion is of the largest of these specimens, which also is illustrated in the five figures of plate I, showing the head, anterior snout, and whole body. Plate II shows a ventral dissection and various internal organs of this same specimen; other internal structures and a gecko foot are illustrated in plate III. Schneider (1801:158) explicitly based Crocodilus siamensis on the text description VOLUME 108, NUMBER 2 and plates I and III in the 1688 report. Al- though the ‘“‘Description anatomique de trois crocodiles ...’’ appeared in several other forms (see below), only the original 1688 version has the pagination given by Schnei- der. The date, 1737, given by Schneider ap- parently is an error. Cuvier (1807) discussed the three croco- diles reported by the French missionaries but used a later version of their ““Descrip- tion anatomique de trois crocodiles ...” found on pages 253-280 and plates 64-66 of Perrault (1733). The text is identical to that in the original (1688) version used by Schneider (1801), but the plates are slightly modified, viz., the figure of a gecko foot originally present on plate III is lacking from plate 66. Besides these editions, other re- prints appeared in French, English and Ger- man. Neither Schneider (1801) nor Cuvier (1807) examined the specimens described and figured by the French Jesuits. Schneider (1801) and Cuvier (1807) based their de- scriptions solely on the text and plates con- tained in the essentially identical 1688 and 1733 versions of the “Description anato- mique de trois crocodiles’’, respectively. Al- though Schneider (1801) considered the three specimens to be of the same species, Cuvier (1807:52) argued that the third and smallest specimen described by the mis- sionaries actually represented another spe- cies, because it lacked the median cranial crests and small eyes of the larger speci- mens, described and prominently illustrat- ed in plate 64 (= plate I). There are no rec- ords of any of the specimens being sent to a museum, and we suspect that none were preserved. Although Cuvier (1807:52) recognized flaws in the original figures, commenting on the faulty rendition of the feet and tail, ev- idently he believed the other details to be correct. Because he considered this material to represent two species, and thus that two species of crocodile occurred in Siam, he rejected Schneider’s (1801) name, C. sia- 299 mensis, aS inappropriate, and introduced the replacement name C. galeatus for the larger specimens and allocated the third and smallest specimen to C. biporcatus Cuvier, 1807 (= C. porosus Schneider, 1801). Gray (1862) was the first author to allo- cate correctly a museum specimen (illus- trated by Gunther, 1864) to C. siamensis. We now know that two specimens of C. siamensis from Java, variously misidenti- fied as Crocodilus raninus Miller & Schle- gel, 1844, Crocodilus rhombifer Cuvier, 1807, and Crocodilus vulgaris Cuvier, 1807 (= C. niloticus Laurenti, 1768), had previ- ously reached Europe (Ross 1992). Gray (1862, 1867, 1872) briefly described the British Museum (Natural History) speci- men, collected in ““Cambogia’”’ (= Kampu- chea) by M. Mouhot (BMNH 61.4.12.22), and, in comparing its head with that figured in the description (plate I) used by Schnei- der (1801), suggested that (1862:270, 1867: 144, 1872:13) “‘the two keels which are pres- ent in that specimen [i.e., the specimen in plate I] are either an individual peculiarity, or perhaps a character that developed itself as the animal approached old age.” Crocodylus siamensis remained poorly known until the 20th century, when detailed descriptions and analyses were published by Smith (1919) and Miller (1923). These au- thors and our examination of museum spec- imens (American Museum of Natural His- tory 28358, 49231; Museum Zoologicum Bogoriense 15; Rijksmuseum van Natuur- likke Historie, Leiden, 7939 and one skull with no number; Natur-Museum Senck- enberg 8090) clearly indicate that the orig- inal figures used by Schneider (1801) and Cuvier (1807) to formulate their concepts of the species exaggerated the size of the distinctive longitudinal interorbital ridge (which is poorly developed or absent in young animals), depicted a greatly exagger- ated crest at the posterior end of the cranial table, and incorrectly showed the eyes to be much smaller than in other crocodile spe- cies. Cuvier’s (1807) belief that the third 300 and smallest specimen referred to C. sia- mensis by Schneider (1801) was probably C. biporcatus (= C. porosus Schneider) thus was unfounded, albeit possibly correct. As no type specimens of C. siamensis are known to exist, it is possible Cuvier’s (1807) view that the type series included two spe- cies might be correct. Because of this con- fusion, and in the interest of nomenclatural stability, we believe that a lectotype should be designated, and, hereby designate as lec- totype of Crocodilus siamensis Schneider, 1801, the specimen whose head is figured on plate I of the 1688 “Description anato- mique de trois crocodiles...” (Internation- al Commission on Zoological Nomencla- ture [ICZN] 1985, Art. 72c(v) and 74c). This specimen was the largest of the three indi- viduals and the subject of the greater part of the text and other figures. The figure of the head of this specimen was reproduced by Cuvier (1807, pl. 2, fig. 9) and by Tie- demann et al. (1817, pl. 11, fig. 2), who also reproduced, with modifications based on Cuvier’s comments (1807:52), other parts of plate I of the 1688 ‘““Description anato- mique de trois crocodiles ...”’ (pl. 11, figs. 1, 3). Following provisions of the Interna- tional Code of Zoological Nomenclature (ICZN 1985: Art. 72e), this specimen also becomes the lectotype of Crocodilus galea- tus Cuvier, 1807, which is a junior objective synonym of C. siamensis Schneider, 1801. Few authors subsequent to Schneider (1801) and Cuvier (1807) apparently have examined the original version(s) of the Jesu- its’ descriptions and accompanying plates. Neither Schneider (1801) nor Cuvier (1807) provided precise locality data for the ma- terial described by the missionaries to Siam, and invariably the type locality has been given simply as “Siam” (e.g., Smith 1931; Mertens 1943; Wermuth 1953; Wermuth & Mertens 1961, 1977; King & Burke 1989). Upon examining the original text (Anon. 1688), we were surprised to discover that the Jesuits’ gave a precise locality for their three specimens. Accordingly, and follow- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ing Recommendation 72H of the Interna- tional Code of Zoological Nomenclature (ICZN 1985), the type locality of C. sia- mensis Schneider can be stated fully as the ““Menam qui baigne le pied des remparts de Louvo’’, Siam (= the Chao Phraya River, at Ayutthaya, Ayutthaya Changwat, Thai- land). Acknowledgments We would like to thank W. R. Heyer, R. W. McDiarmid, and G. R. Zug for their comments on drafts of this paper and for providing research facilities to CAR and GCM, and M. Hoogmoed, K. Klemmer, G. Foley, and R. G. Zweifel for the loan of specimens. Literature Cited Anonymous. 1688. Observations physiques et ma- thématiques pour servir a |’Histoire naturelle et a la Perfection de l’Astronomie et de la Géo- graphie: envoyées de Siam a l’Académie Royale des Sciences a Paris, par les Péres Jésuites Fran- ¢ois qui vont a la Chine en qualité de Mathé- maticiens du Roy. Avec les réflexions de Mes- sieurs de l’Académie et quelques notes du P. Goiiye, de la Compagnie de Jésus. Paris, chez la Veuve d’Edme Martin, Jean Boudot et Es- tienne Martin, vi + 278, i-1i, pls. 1-3. Cuvier, G. 1807. Sur les différentes espéces de croco- diles vivans et sur leurs caractéres distinctifs. — Annales du Muséum d’Histoire Naturelle 10:8— 66, pls. 1-2. Gray, J. E. 1862. A synopsis of the species of croc- odiles.—Annals and Magazine of Natural His- tory 3rd Ser., 10(58):265-274. . 1867. Synopsis of the species of recent Croc- odilians or Emydosaurians, chiefly founded on the specimens in the British Museum and the Royal College of Surgeons. — Transactions of the Zoological Society of London 6(4):125-169, pls. 31-34. . 1872. Catalogue of shield reptiles in the col- lection of the British Museum. Pt. 2. Emydo- saurians, Rhynchocephalia, and Amphisbaeni- ans. London, vi + 41 pp. Ginther, A. C. L. G. 1864. The reptiles of British India. London, xxvii + 452 pp., pls. i-xxvi. International Commission on Zoological Nomencla- ture. 1985. International code of zoological nomenclature, 3rd ed. London, xx + 338 pp. VOLUME 108, NUMBER 2 King, F. W., & R. L. Burke. 1989. Crocodilian, tu- atara, and turtle species of the world. A taxo- nomic and geographic reference. Association of Systematics Collections, Washington, D.C., xxii + 216 pp. Laurenti, J. N. 1768. Specimen Medicum, exhibens Synopsin Reptilium. Vienna. 214 pp., 5 pls. Mertens, R. 1943. Die rezenten Krokodile des Natur- Museums Senckenberg.—Senckenbergiana 26(4):252—312. Miller, L. 1923. Crocodilus siamensis Schneid. und + Crocodilus ossifragus Dubois.—Palaeontolo- gia Hungarica 1:119-122. Miiller, S., & H. Schlegel. 1844. Over de Krokodillen van den Indischen Archipel. 28 pp., pls. 1-3 in C. J. Temminck. 1839-1844. Verhandelingen Over de natuurlijke geschiedenis der Neder- landsche overzeesche bezittingen, door de leden der Natuurkundige Commissie in Indié en an- dere Schrijers. Leiden, 259 pp., pls. 1-70. Perrault, Claude. 1733. Mémoires pour servir a1’ His- toire naturelle des Animaux. Seconde partie (n° 18-34). Compagnie des Libraires, Paris, 294 pp., pls. 35-67. Ross, C. A. 1992. Designation ofa lectotype for Croc- odilus raninus S. Miller and Schlegel (Reptilia: 301 Crocodylidae), the Borneo crocodile. — Proceed- ings of the Biological Society of Washington 105(2):400-402. Schneider, J. G. 1801. Historiae Amphibiorum na- turalis et literariae. Fasciculus Secundus. Jena, vi + 374 pp., pls. I-II. Smith, M. A. 1919. Crocodilus siamensis. —Journal of the Natural History Society of Siam 3(3):217- 222, pl. 6. 1931. The fauna of British India, including Ceylon and Burma. Reptilia and Amphibia. Vol. 1. Loricata, Testudines. London, xxviii + 185 pp., pls. 1-2. Tiedemann, F., M. Oppel, & J. Liboschitz. 1817. Na- turgeschichte der Amphibien. Erstes Heft. Gat- tung. Krokodil. Heidelberg, 88 pp., 15 pls. Wermuth, H. 1953. Systematik der rezenten Kro- kodile.— Mitteilungen aus dem Zoologischen Museum, Berlin 29(2):375—514. —, & R. Mertens. 1961. Schildkréten, Kroko- dile, Briickenechsen. Gustav Fischer Verlag, Jena, xxvi + 422 pp. ——, & 1977. Liste der rezenten amphi- bien and reptilien. Testudines, Crocodylia, Rhynchocephalia. Das Tierreich 100, xxvii + 174 pp. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(2):302-318. 1995. Description of a new large-bodied species of Apomys Mearns, 1905 (Mammalia: Rodentia: Muridae) from Mindoro Island, Philippines Luis A. Ruedas Department of Biology, University of Puerto Rico, Cayey University College, Avenida Antonio R. Barcelo, Cayey, Puerto Rico 00736 Abstract.—Apomys gracilirostris, a new species in the Philippine endemic murid genus Apomys, is described from the Mount Halcon highlands on Min- doro Island. The new species is compared with its larger-bodied congeners: A. abrae, A. datae, A. sacobianus, and the two nominal subspecies of A. insignis, A. i. insignis and A. i. bardus. Twenty-eight cranial, mandibular, and dental measurements were taken on the 83 specimens representing these taxa and the new species. All species examined formed discrete groups in a principal com- ponent analysis; there were no apparent differences between the two nominal subspecies of A. insignis. The new species differs from congeners in its com- bination of large body, very long tail, dark pelage, gracile cranium with a long rostrum, and highly compressed upper incisors; the most diagnostic features of the species are the extremely fine upper and lower incisors (with no enamel on the lower), and the degree of molar reduction and simplification. In 1905, Edgar A. Mearns proposed the generic appellation Apomys for represen- tatives of a new species of rodent he had collected on Mount Apo, in southern Min- danao Island, Philippines (Mearns 1905). Different investigators historically have shown considerable flux with respect to the taxonomic placement of the genus (review by Musser 1982). However, since the ex- haustive review and diagnosis of Apomys by Musser (1982), there has been no doubt as to the validity of these organisms con- stituting a genus level category. As currently defined (Musser 1982, Musser & Heaney 1992), Apomys is restricted to the Philip- pine Archipelago (excluding the Sunda Shelf island of Palawan and associated peripheral islands). Eight species are now included in the ge- nus: A. hylocoetes and A. insignis from Min- danao Island, A. Jittoralis from Mindanao, Leyte, Biliran, and Bohol Islands (Heaney etal. 1989, Musser & Heaney 1992, Rickart et al. 1993), A. microdon from southern Lu- zon and Catanduanes Islands, A. musculus from Luzon and Mindoro Islands, and A. datae and A. abrae from Luzon Island. An additional species, currently undescribed, recently has been reported from a number of islands in the central Visayas group of the Philippine Archipelago (Musser 1982, Musser & Heaney 1992; L. R. Heaney, pers. comm.). During a biodiversity inventory survey carried out in 1992 on some components of the major Pleistocene island masses (sensu Heaney 1986), one of the localities that was visited was the Mount Halcon massiff, in the northern part of the island of Mindoro. Mount Halcon, reported to reach ca. 2500 m in elevation, had been explored by Edgar A. Mearns in 1906 (Merrill 1907a); how- ever, only 17 of the 496 mammals brought back by him were from Mindoro. Most of his collection from the Mt. Halcon area con- sisted of Crocidura; all told, eight nominal VOLUME 108, NUMBER 2 taxa in four orders were collected by him, including the holotypes of five nominal taxa. The object of the current exploration of the Mount Halcon area was to increase sample sizes of some of these taxa in order to arrive at a better assessment of geographic and nongeographic variation, as well as to de- termine whether additional species, which Mearns may have missed, also existed in the area. One mammal species found between 1255 m and 1900 m elevation on the north slope of Mount Halcon was a medium-sized ro- dent, similar to Apomys in having a slender rostrum, as opposed to the short, high, blunt rostrum of Rattus. The lower incisors were more procumbent than those of Rattus, and lacked pigment. The general gestalt, mostly related to how gracile this rodent was, there- fore pointed, barring the relatively large size, to its pertenence in the genus Apomys. Sub- sequent examination of cranial morpholo- gy, and conversations with colleagues, con- firmed this suspicion. Accordingly, individ- uals collected in. that taxon are described below as a new species of Apomys from the northern highlands of Mindoro. Methods Except for specimens of the new species (at the Cincinnati Museum of Natural His- tory), and those of Apomys sacobianus (Na- tional Museum of Natural History, USNM), all specimens examined are from the Field Museum of Natural History (FMNHBH). Ter- minology referent to molar cusps follows Miller (1912) for upper molars, and Musser and Heaney (1992) for lower molars. Cra- nial and mandibular measurements were taken with a digital caliper to the nearest 0.01 mm. Twenty-eight cranial, mandibu- lar, and dental measurements, listed below, were taken; the 23 cranial measurements were described and illustrated in Musser (1979); mandibular measurements are de- fined below. Measurements (abbreviations in parentheses) included: breadth of brain- 303 case (BB), breadth across incisive foramina (BIF), breadth across incisor tips (BIT), breadth of palatal bridge at first, and at third molars (BM1, BM3), breadth of mesoptery- goid fossa (BMF), breadth of rostrum (BR), breadth of zygomatic plate (BZP), height, and length, of auditory bulla (HB, LB), height of braincase (HBC), interorbital breadth (IB), incisive foramina to Ml (IFM1; modified to be straightline shortest distance, rather than anteroposterior dis- tance), length of diastema (LD), length of incisive foramina (LIF), alveolar length of maxillary toothrow (LMT), length of nasals (LN), length of palatal bridge (LPB), length of rostrum (LR), occipitonasal length (ONL), palatal length (PL), postpalatal length (PPL), and zygomatic breadth (ZB). Five addition- al measurements not used by Musser (1979) were taken: length of dentary (LMAND), the greatest length of dentary from upper incisive alveolar rim to mandibular con- dyle; length of mandibular toothrow (LMT), alveolar length of mandibular toothrow; height of mandible (HMC), greatest height of mandible; height of mandibular ramus (HMR), lowest height of ramus in mandib- ular incisive diastema; and thickness of mandible (TM), measured at thickest point below m2. Statistical analyses were performed on an 80486 equipped 33 MHz microcomputer using the Statistical Analysis System soft- ware, version 6.03 (SAS Institute 1988a, 1988b), generally following Ruedas & Bick- ham (1992). Univariate statistics, including mean, standard error, moment statistics (skewness, g,; and kurtosis, g,) and tests of normality were performed invoking the ‘normal’ option of the UNIVARIATE procedure, which tests for normality using the Shapiro-Wilk statistic, W, and provides the associated probability value. A principal component analysis was carried out to de- termine whether any group separation oc- curred using the measurements taken. Such a posteriori grouping methods are preferred herein over a priori grouping methods (mul- 304 Table 1.— Values for the first through third eigen- vectors from the principal component analysis carried out on the correlation matrix of the cranial and man- dibular morphometric data of specimens of Apomys. Refer to text for character abbreviations. The first four principal components together account for 91.5% of the variation (68.5%, 13.3%, 5.5%, and 2.2%, respec- tively). Results of the principal component analysis are graphically summarized in Fig. 1. Component 1 2 3 BB 0.2092 =023 7/2 —0.0489 BIF 0.1498 0.3402 0.0601 BIT 0.1244 0.2985 0.4763 BM1 0.1684 —0.0566 (0,707) BM3 0.1942 (0) 2A —0.1403 BMF 0.0993 0.3870 —0.0400 BR 0.2279 —0.0040 0.2225 BZP 0.2509 0.1233 0.0151 HB 0.1408 0.3926 0.0275 HBC 0.2454 —0.0082 —0.0001 IB 0.0362 (LI S2z 0.3550 IFM1 0.1443 —0.3582 0.0222 LB 0.2012 0.0836 —0.2576 LD 0.2465 =O} TNS) —0.0298 LIF 0.2353 0.1279 —0.2145 LM13 0.2081 0.2802 —0.1150 LN 0.2610 —0.0246 0.0379 LPB 0.2370 =O) SZ 0.1331 LR 0.2453 —0.1326 0.1170 ONL 0.2653 —0.0583 0.0164 PL 0.2651 0.0501 —0.0369 PPL 0.2576 —0.0188 0.0451 ZB 0.1841 = (0-211 335) 0.2830 tiple range tests, canonical discriminant analysis) because there is no prior hypoth- esis as to the putative identity of the spec- imens being examined. Specimens Examined A. abrae: Philippines: Luzon Island, Ilo- cos Norte Prov.; Mount Simminublan, 4300-4350 ft. FMNH nos. 92752-92755, males; 92756-92762, females. Abra Prov.; Massisiat, 3500 ft. FMNH 62749, female. Mountain Prov.; Mt. Data, 5300-8000 ft. FMNH 62700, 62724, 62726, 62728, 62738-62740, males; 62719-62723, fe- males. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A. datae: Philippines: Luzon Island, Mountain Prov.; Mount Data, 5300-8000 ft. FMNH 62695, 62696, 62699, 62720, 62741, males; 62706, 62709, 62711, 62712, 62725, 62727, 62731, 62733-62735, 62742, 62744, females. Mountain Prov.; no specific locality. FMNH 62761, female. A. insignis bardus: Philippines: Mindanao Isl., Zamboanga del Norte Prov.; Grand Malindang Mountains, Dapitan Peak, 6700- 7450 ft. FMNH 87568 (young), 87571- 87573, males, 87567, 87569, 87570, fe- males. A. 1. insignis: Philippines: Mindanao Isl., Bukidnon Prov.; Malaybalay, Mount Ki- tanglad, 4200-5000 ft. FMNH 92806, 92808-92812, 92815, 92816, 92818, 92819, 92821, males; 92807, 92813, 92814, 92817, 92820, 92822, females. Davao Prov.; East slope of Mount McKinley, 3500-7100 ft. FMNH 56267, 56270, 56276-56278, males; 56266, 56268, 56269, 56272-56275, 56282, 56285, 56288, females. A. sacobianus: Philippines: Luzon Isl., Pampanga Prov.; Angeles, Clark Air Base, Sacobia River. USNM 304352, 557717. Apomys Nn. sp.: see below, holotype and paratypes. Results Descriptive statistics for species of Apo- mys examined in this study are listed in Appendix 1. Results of the principal com- ponent analysis (Fig. 1) show considerable, non-overlapping variation among different nominal taxa within Apomys. The first four principal components account for 91.5% of the variation (Table 1). Principal compo- nent 1 accounts for 68% of the variation and separates A. insignis from remaining Apomys. Variation is spread fairly evenly over all characters in principal component 1, but characters with heavy loadings in- clude occipitonasal length, palatal length, length of nasals, and postpalatal length. Since A. insignis is smaller in all measurements than the Apomys considered herein, prin- cipal component | appears to be constituted VOLUME 108, NUMBER 2 3 insignis Principal Component II en) <9 305 sacobianus abrae f] datae © gracilirostris 0 5 Principal Component | Fig. 1. primarily on the basis of size. Principal component 2 accounts for 13% of the vari- ation and separates the Mount Halcon tax- on from remaining Apomys. This principal component shows heavy loadings on height of bulla, breadth of mesopterygoid fossa, straightline distance between incisive fora- men and M1, breadth of incisive foramina, breadth of incisor tips, and length of max- illary toothrow. These are measurements in which the Mount Halcon taxon is quite dis- tinct (either larger or smaller) from remain- ing Apomys examined herein; as a conse- quence, it also appears to be constituted on the basis of magnitude of linear measure- ments. Inspection of descriptive statistics (Ap- pendix), of results of principal component analysis, (Table 1 and Fig. 1), and of several discrete characters, clearly identifies the specimens of Apomys from the Mindoro highlands as a new species. The species is described below, being named Plot of principal component scores (PC 1 vs. PC 2) for six species of Apomys examined in this study. Apomys gracilirostris, new species Figs. 2—5 Holotype. —National Museum of the Philippines (NMP) No. 3482 (fide Pedro C. Gonzales; also Cincinnati Museum of Nat- ural History [CMNH] No. 650, and field collection number NMP/CMNH 1136), an adult male snared by Pedro Bangol, a Man- gyan native from Lantuyan (= Dulangan Dos); the specimen will be permanently housed at the National Museum of the Phil- ippines. The provenance is: Philippines: Mindoro Island; Mindoro Occidental Prov- ince; Municipality of San Teodoro, North Ridge approach to Mount Halcon, ca. 1580 m; ca. 13°16’48”N, 121°59'19’E; this area is known to the Mangyan tribespeople as ‘“‘HanglG.”’ Date of collection is 12 June 1992. Specimen preserved as a dried skin and skull, in good condition. Paratypes and localities. — Besides the ho- lotype, an additional 15 specimens are 306 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Dorsal and ventral views of Apomys gracilirostris CMNH 646 (left in each pair) and 649 (right). Specimen 646 is homogeneously colored dorsally and ventrally, while 649 is paler (buffy) in ventral coloration. Length of head and body and tail length (in millimeters) are 147 and 175 (CMNH 646) and 146 and 179 (CMNH 649). known. Four, snared by Pedro Bangol at the same locality as the holotype, but at ele- vations ranging between 1675 an 1731 m, are prepared as dried skins with a full skel- eton extracted from the left side, body skel- eton from the right: CMNH 646 (NMP/ CMNH 1119), an adult female (right zy- gomatic arch broken), 7 Jun 1992; CMNH 647 (NMP/CMNH 1126), young adult male snared 9 Jun 1992; CMNH 648 (NMP 3480; NMP/CMNH 1131), adult male, 10 Jun 1992; and CMNH 649 (NMP 3481; NMP/ CMNH 1132), old adult male, 11 Jun 1993. Twelve specimens are fluid preserved, with no skeletal material extracted to date. These are CMNH 642 (NMP/CMNH 1069), a scrotal male collected 28 May 1992, and CMNH 643 (NMP 3479; NMP/CMNH 1100), an adult female collected 1 Jun 1992, from the Dulangan River Valley (Philip- pines: Mindoro Isl., Mindoro Oriental Province; Municipality of San Teodoro; VOLUME 108, NUMBER 2 307 Table 2.—Standard external measurements of the holotype of Apomys gracilirostris, CMNH 650, and uni- variate statistics of male and female specimens of A. gracilirostris. There were no significant differences between males and females (P = 0.05), however, males displayed non-normally distributed measurements in total length (P = 0.0039) and mass (P = 0.0005). Sample size is four females and 13 males (including holotype) for all measurements except mass, twelve males. All measurements expressed in millimeters except mass, in grams. Females Males Holotype Character (CMNH 650) Mean + SD Range Mean + SD Range Total length 294 S30G2 se 23211 283-330 314.1 + 30.98 288-398 Tail length 157 NOU se NAO) 147-175 159.0 + 14.50 135-185 Length of hind foot 39 Boo) 28 So 1/0) 33-41 38.8 + 3.05 34-45 Length of ear 21 19.2 + 1.26 18-21 18.6 + 2.72 14-22 Mass 80 soe) as LY) 7/ 75-115 87.4 + 18.28 71-140 North approach to Mt. Halcon, Dulangan River Valley, ca. 1255 m, approx. 13°17’27’N, 121°59'32”W); CMNH 644 (NMP/CMNH 1108), a scrotal male col- lected 4 Jun 1992, and CMNH 645 (NMP/ CMNH 1118), also a scrotal male, collected 7 Jun 1992 from the same locality as the holotype. Seven specimens, CMNH 634, 637 and 639-641 (inclusive), all males, and 635 and 636, females, are from an elevation of ca. 1900 m, an area known to the Man- gyan tribespeople as Patok-tok, just above the type locality. All paratypes are in ex- cellent condition except as noted. The fol- lowing specimens will be permanently housed at the National Museum of the Phil- ippines: CMNH 634 (NMP 3475), 635 (3476), 637 (3477), 641 (3478), 643 (3479), 648 (3480), 649 (3481), and the holotype, CMNH 650, NMP 3482. Diagnosis. —Apomys gracilirostris is dis- tinguished from its congeners by the follow- ing combination of traits: with respect to remaining Apomys, upper incisors are asul- cate and extremely compressed and thin, breadth of incisor tips averaging only 1.56 + 0.053, versus extremes in the remaining large bodied Apomys of 1.66 + 0.158 (A .i. insignis males from Davao Province) to 2.23 + 0.178 (A. datae males); lower incisors un- pigmented and elongate. Lower edge of in- cisive alveolus beginning (in the vertical plane) just in front of m1, whereas the body of the ramus between the first molar and incisor in other species of Apomys is much longer. The small width of the incisors be- comes even more apparent as a ratio against occipitonasal length; in A. gracilirostris this is 0.039 + 0.002, versus extremes in re- maining large bodied Apomys of 0.050 (A. abrae females from Mountain Province) to 0.061 (a single A. insignis bardus female). The skull is gracile, with a long rostrum. The combination of large body (combined head and body length, 143.6 + 3.97; range 137-147), very long tail (length, 168.2 + 9.47; range 157-179), dark pelage, gracile cranium, long rostrum (length, 14.11 + 0.425; range 13.86—-14.86), compressed and thin upper incisors (breadth at incisor tips, 1.56 + 0.054; range, 1.51—1.62) and elon- gate, unpigmented lower incisors are sufh- cient to diagnose the species. Description. — All specimens adults, pel- age dark brown on dried skins (Fig. 2), body relatively large for an Apomys (external measurements in Table 2), with a long, slen- der snout, and long tail and feet. Holotype and paratypes similar in dorsal and lateral coloration. Fur silky and soft; proximal part of dorsal hairs light grey, turning pale brown distally; with numerous black guard hairs. Overall color dark brown, although freshly captured specimens with a dark blue/green sheen; some specimens darker along mid- dorsal region. Color somewhat paler later- ally. Ventral color variable. One specimen (646; Fig. 2) not differing between dorsal 308 and ventral appearance with respect to tone; 647 and 646 are marginally paler, 647 with silver-tipped grey hairs, 648 with grey hairs tipped with pale brown; 649 and 650 buffy ventrally, (particularly the former; see Fig. 2) with grey hairs tipped to a greater (649) or lesser (650) extent with brown. Tail generally uniformally dark, however, sometimes paler on ventral surface, some individuals displaying a white extremity, varying from 2 mm in length (637) to 6 mm (640) or 10 mm (639). Scale row counts 14/ cm at both base and midsection of tail. Mystacial vibrissae present; anteriormost ones dark grey, dorsad and caudad ones black, with some fading to grey distally. Ears small relative to size of head; dark brown on inner surface, slightly paler on outer sur- face. Hindfeet relatively long (females: mean, 36.5 mm, SE 3.70, range, 33-41; males, 38.8 SE 3.05, range 34—45), and dark dor- sally. Plantar surface also dark, with paler tubercules forming a contrast. Claws rela- tively long; foreclaws ca. 3 mm in length; hindfoot claws ca. 4 mm long. Vertebral counts identical except for number of caudal vertebrae: atlas, axis, 5 innominate cervical, 13 thoracic, six ab- dominal, and three pelvic vertebrae (cran- iad weakly fused to two, strongly fused cau- dad vertebrae). Specimen 648 with 32 cau- dal vertebrae, while 646 and 647 with 35. Premaxillae projecting approximately 2.5 mm beyond anterior edge of incisive alve- oli; top of nasals projecting marginally be- yond this. Nasals widest at their anterior extremities, curving somewhat downward to meet premaxillae; nasals narrowing somewhat caudally, narrowest at fronto-na- sal suture. Nasals and lower surfaces of pre- maxillae, maxillae, and palatine almost on same plane, with little tapering towards ros- trum when viewed laterally. Incisive foram- ina long and narrow (4.9 + 0.102), typically end about 1.3 mm anterior to first molar. Zygomatic arches and zygomatic plates slender and delicate. As with other mem- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON bers of genus, edges of interorbital and post- orbital areas rounded, without the beading characteristic of Rattus. Braincase smooth and round, the only protuberances those as- sociated with occipital bone. Caudalmost end of cranium formed by an external oc- cipital protuberance at caudalmost end of interparietal that meets squamous part of occipital (or supraoccipital bone), causing occipital bone to project beyond occipital condyles. Paired protuberances present where caudal ends of squamosal bones meet supraoccipital; without lambdoidal ridge. Mastoid somewhat inflated. Squamoso- mastoid foramen present, but not as well developed (as a fenestra) as in A. insignis, for example; present only on one side in the holotype (650), 648, and 649 (extremely re- duced); on both sides 646 and 647. Large postglenoid fossa separating squamosal from periotic bones. Foramina of alisphenoid region essen- tially as described for Apomys (Musser 1982), with slight modifications. Lateral strut of alisphenoid bone slender, much more so than illustrated for A. insignis (AMNH 207571) by Musser (1982). No separation between anterior opening of al- isphenoid canal and coalesced masticatory- buccinator foramina, so that all three are consolidated. Masticatory nerve emerging from posterior end of this consolidated fo- ramen, leaving deep furrow in alisphenoid bone, particularly in region proximal to its point of emergence. From a ventral per- spective, auditory bulla separated from al- isphenoid bone by middle lacerate foramen, which is only thinly separated from the postglenoid vacuity. Sphenopterygoid va- cuity variable in conformation, being single (separating alisphenoid from pterygoid), or consisting of the same, with an additional opening at caudal edge of pterygoid. A vary- ing number of sphenopalatine foramina of varying conformation present. Minute dor- sal palatine foramen present in orbital re- gion, above roots of M2. Sphenopalatine foramen, just anterior to dorsal palatine fo- VOLUME 108, NUMBER 2 309 Fig. 3. Dorsal, ventral, and lateral views of cranium and mandibles of the holotype of Apomys gracilirostris (CMNH 650). Occipitonasal length 39.4 mm, zygomatic breadth, 17.6 mm, length of dentary, 20.3 mm. ramen, likewise inconspicuous (in some cases closed); where present, only evident as a depression in maxilla. Sphenoid bone with relatively small optic canal separated by a strut from the much larger orbital fis- sure. Dentaries (Fig. 3) with angular process projecting beyond plane of mandibular con- dyle. Lower edge of incisive alveolus begin- ning (in the vertical plane) just anterior to ml. With a large protuberance (capsular process sensu Musser & Heaney 1992) forming around end of root of lower incisor, on labial side of dentary, just anterior and ventral to coronoid process, forming narrow shelf that narrows posteriorly into mandib- ular condyle. Upper incisors asulcate, extremely com- pressed and thin, their combined width av- eraging 1.56 + 0.053. Enamel pale orange. Lower incisors extremely elongate and slen- der, also asulcate, and lacking pigment on their enamel surfaces. Molars (Fig. 5) typical of Apomys (refer to fig. 15 in Musser 1982), with a few no- table differences. All cusps occupying rela- 310 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 4. Ventral view of pterygoid and mesopterygoid regions in the holotype of Apomys gracilirostris; bottom left is the bony eustachian tube, anterior is to the top. tively greater area of crown than in other species of A4pomys. Anterior labial cleft sep- arating the hypothesized front two laminae of m1 not present; as a consequence, antero- lingual or antero-labial cusp not evident. Anterior lamina consisting simply ofa large, crescent-moon-shaped ring of enamel sur- rounding a lacuna of dentine, presumably formed from fusion of anteroconid, proto- conid, and metaconid. All three laminae of M1 relatively broad as well, occupying entire occlusal surface. M1 with a small distinction still evident be- tween cusps t3 and t1/t2, which are contin- uous. Cusps t8 and t9 occupying almost en- tire caudad half of M1. No evidence of a VOLUME 108, NUMBER 2 Siel igs: toothrow, 6.3 mm, length of lower molar toothrow, 6.4 mm. posterior cingulum. M2 lacking cusp tl (vide A. musculus), and t3 reduced and tightly compressed between t5 and t6 (continuous- ly fused with t4) and t9 of M1. Cusp t8 large, ovoid, and comprising entire caudad half of M2. M3 with only the slightest trace of t3 existing (more obviously present in other Apomys), making this essentially a unicus- Upper (left) and lower molar toothrow of the holotype of Apomys gracilirostris; length of upper molar pid molar. Specimen 649 differs from the other A. gracilirostris specimens in having only two upper molars in right maxilla (full complement in left) and two molars in right dentary (full complement in the left). One female (CMNH 646) had 3 embryos, measuring 2 mm crown-rump length; two of these were on the right horn of the uterus; 312 636 displayed a swollen uterus with no pla- cental scars. Two of five males from the area surrounding 1900 m were scrotal; males from lower elevations (646-650) all dis- played scrotal testes. Testes measurements were 6 X 4 mm (647 and 648) and 7 x 4 mm (649); no measurements were obtained on testes of 650. Distribution. —Apomys gracilirostris is known only from the type locality and gen- eral vicinity, with an elevational range of ca. 1250-1950 m; it may occur throughout the highlands of Mindoro, particularly northern Mindoro, where there still remain large tracts of relatively undisturbed mon- tane and midmontane forests. Etymology. —gracilis + rostrum —a com- bination of the latin words describing the particularly long and slender rostrum of this species of Apomys (with respect to other Apomys, and most rodents). Ecology. —Starting at ca. 1250 m on the north slope of Mount Halcon, the canopy height of the forest (in the general category of Montane Forest) averaged 14-16 m in sheltered areas, with some emergents to 20 m; more exposed areas displayed a canopy height of 6-8 m with emergents to 11 or 12 m. The canopy generally was uneven, al- lowing considerable light to penetrate in places, although the forest immediately ad- jacent to the Dulangan River had a very dense canopy. All areas had heavy under- growth up to 2 m high. Tree ferns (Cyathea), birds’ nest fern (Asplenium), climbing bam- boo, palms, and pandans (particularly Pan- danus) were very common. Also indicative of the extremely wet nature of the area was the heavy growth of moss on trees. The most dominant trees (typically com- posing at least 25% of plot surveys) were Tristania sp. (Myrtaceae), Lithocarpus sp. (Fagaceae) and Leptospermum flavescens (Myrtaceae). Important trees (constituting 5—15% of plot surveys) were Decaspermum paniculatum and Tristaniopsis sp. (Myrta- ceae), Praravinia sp. and Psychotria sp. (Ru- biaceae), Ficus sp. (Moraceae), Pinanga sp. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON (Arecaceae), Elaeocarpus sp. (Elaeocarpa- ceae) and Adinandra sp. (Theaceae). Pri- mary understory trees were Saurauia sp. (Actinidiaceae), Syzygium sp. (Myrtaceae), Astronia sp. (Melastomataceae), and Sym- plocos sp. (Symplocaceae). Numerous herbs and shrubs also abounded. Forest compo- sition in this general area of Mount Halcon also was detailed by Ames (1907), Broth- erus (1907), and Merrill (1907b). From 1600 m to 1950 m, numerous areas were Overgrown with bamboo, with the only tree species present being Agathis philippi- nensis (“‘almaciga’’). In areas of closed for- est, canopy heights ranged from 7-10 m in most areas, with the tallest trees (in shel- tered areas) growing up to ca. 20 m. The canopy was considerably more closed (up to 80% closed) at this elevation that at the lower elevations. Pandans and gingers were SO COmmon in some areas as to make pas- sage quite difficult. Leaf litter was approx- imately 6-8 cm thick. Numerous epiphytic ferns and orchids were present, although none occurred on the almaciga trees. The gracile skull, long rostrum, reduction in relative size of upper incisors, and ex- treme elongation of lower incisors of A. gra- cilirostris all point to a diet of soft-bodied invertebrates, although additional, freshly sacrificed specimens will need to be exam- ined to confirm this hypothesis (stomach contents were not readily recognizable). The long tail, and relatively long hind feet with well-developed claws point to a scansorial lifestyle. Much more ecological data, per- haps from mark and recapture studies, still need to be gathered on this little known spe- cies. Comparisons. —The most conspicuous feature of the cranium in 4. gracilirostris (Fig. 3) is the long and slender rostrum; the length in the five crania examined was 14.11 + 0.425 (range: 13.86—14.86). Only A. datae males exceed A. gracilirostris in this respect; for 10 females examined, 13.42 + 0.702 (12.5-15.0); 7 males, 14.14 + 0.480 (13.30- 14.78); means in other species ranged to a VOLUME 108, NUMBER 2 low of 10.45 + 0.292 for male A. insignis from Davao Province. The ratio of rostrum length to skull length is surpassed only by male A. datae and female A. abrae from Mountain Province, Luzon; for 4. gracili- rostris, 0.354 + 0.005 (0.348—0.360), versus extremes of 0.329 (female A. insignis from Mount Kitanglad, Bukidnon Province, Mindanao) to 0.356 (for 7 male A. datae from Davao City Province, Mindanao Is- land); A. abrae females from Mountain Province displayed a mean of 0.355 + 0.021. The interorbital region, averaging 6.0 + 0.233 mm, is relatively the narrowest among Apomys species: the ratio of interorbital width to occipitonasal length is 0.150, ver- sus extremes in other Apomys of 0.151 (fe- male A. abrae from Ilocos Norte, Luzon) to 0.167 (female A. insignis from Mt. McKin- ley, Davao, Mindanao Isl. [this locality may currently be more easy to locate as Mt. Tal- omo, within the boundary of Davao City Province]). The conformation of the carotid arterial patterns in A. gracilirostris clearly unite this new species in a group with A. datae and A. sacobianus. The presence of a stapedial fo- ramen in the auditory bulla, together with a relatively large carotid canal for the in- ternal carotid artery, and a prominent fur- row between the middle lacerate foramen and the foramen ovale (Fig. 4) for the in- ternal maxillary artery, also is a trait shared by A. gracilirostris, A. datae and A. saco- bianus. It should be noted, however, that this pattern is hypothesized to constitute the ancestral arterial pattern among muroid ro- dents (Hill 1935, Musser 1982, Musser & Heaney 1992), and as such should not be interpreted as expressing phylogenetic afhin- ity between these Apomys species. The min- ute dorsal palatine foramen present in or- bital region, above roots of M2, distinguish- es the new species from 4. insignis, which has an easily visible foramen above the junction between M1 and M2. A more ex- plicit statement of hypothesized relation- ships among these three species Apomys, 313 and among all the members of the genus, will have to await results of an upcoming revision. The dentaries (Fig. 3) display distinct dif- ferences between A. gracilirostris and other Apomys having to do with the relative length of the angular process and mandibular con- dyle. When viewed from the side, most Apo- mys (e.g., A. datae, A. insignis, and A. mus- culus, as illustrated in fig. 8 of Musser 1982: 24) display a mandibular condyle that pro- jects somewhat beyond the vertical plane of the angular process. In contrast, in A. gra- cilirostris, it is the angular process that pro- jects beyond the plane of the mandibular condyle. Another characteristic distinguish- ing this species from all other Apomys is that the lower edge of the incisive alveolus begins (in the vertical plane) just in front of m1, whereas the body of the ramus between the first molar and incisor in other species of Apomys is much longer. A large protu- berance (capsular process sensu Musser & Heaney 1992) forms around the end of the root of the lower incisor, on the labial side of the dentary, just anterior and ventral to the coronoid process, forming a narrow shelf that fades back into the mandibular con- dyle. The teeth of A. gracilirostris are the most diagnostic feature separating the species from its congeners: in fact, the upper and lower incisors are sufficient to diagnose the species, in a museum or in the field. Among Apomys, A. gracilirostris likewise displays an increasing degree of molar reduction and simplification of crowns, epitomized by the reduction in number of molars found in A. gracilirostris specimen 649. In fact, upon first examination, the extreme reduction of the molars together with the lack of a pig- mented enamel on the lower incisors, led me to believe that A. gracilirostris consti- tuted a new genus, which it still may, once stricter and more careful analyses are car- ried out of Apomys and potential sister taxa. Outgroup taxa that must be included in such an analysis are the native Philippine shrew 314 rats (Crunomys, Chrotomys, and Rhynco- mys). Inclusion of these taxa will enable testing one of the three hypotheses of ge- neric relationships of Apomys proposed by Musser & Heaney (1992): that Apomys may be more related to the native Philippine shrew rats than to other Philippine endem- ics or genera from other archipelagos. The somewhat inflated mastoid bone of A. gra- cilirostris is reminiscent of a condition not- ed in Tarsomys apoensis (Musser & Heaney 1992). It was impossible to determine the homology between the principal cusp com- prising the almost unicuspid M3 in Apomys and the same structure in other Murinae; Musser and Heaney (1992) hypothesized that the Apomys M3 cusp may be homol- ogous with t5 of other murines. The intent of this work was not to assess the specific status of the two nominal taxa currently included in A. insignis, A. 1. insig- nis and A. 1. bardus. However, it is clear that the lack of morphological differences be- tween these two taxa lends support to the hypothesis that they clearly do not consti- tute separate species (Musser 1982), and probably not even separate subspecies. Discussion Finding a new species of Apomys in the Philippines is not as surprising as the fact that it was found in an area that already had been explored by naturalists. Edgar A. Mearns spent some time in the general area, but further east than the new species was collected (Merrill 1907a). This highlights the fact that as good as the turn of the century collectors were, there still are areas in the Philippines that need more thorough study in order to arrive at a fuller understanding of their existing biodiversity. Itis even more critical today than before to document the biodiversity in the Philippines, as continu- ing loss of forest cover leaves ever dimin- ishing pristine habitats available for wild- life. We observed some limited scale logging (with chain-saws and water buffalo) going PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON on up to an elevation of 950 m in the Mount Halcon area, in spite of the fact that the entire area is purportedly protected from exploitation by all save the native Mangyan tribal peoples. These forests harbor a rich endemic fauna, including Apomys gracili- rostris, the rare endemic Anonymomys min- dorensis (pers. obv., Musser 1981), very high population levels of Chrotomys mindorensis (pers. obv., Musser et al. 1982), and other, as yet undescribed species. The ineluctable conclusion with respect to the conservation of the Mount Halcon Highlands is that the remaining areas of forest must at all costs be protected. Acknowledgments I thank most of all Pedro Bangol, the Mangyan tribesman who is responsible for catching the specimens described herein: he had a lot to teach. I thank the Protected Areas and Wildlife Bureau of the Philippine Department of the Environment and Nat- ural Resources for facilitating permits that allowed us to carry out this work. Within these organizations, I especially thank A. Alcala, A. Ballesfin, J. Caleda, C. Catibog- Sinha, C. Custodio, D. Ganapin, and S. Penafiel. I thank A.-M. and R. Tate, and the Miranda family for extending welcome hospitality during overly long stays in Ma- nila. Assistance in the field was provided by the following CMNH volunteers and un- compensated staff, and NMP staff: R. M. Brown, J. Barcelona, D. T. Busemeyer, R. E. Fernandez, J. W. Ferner, A. Ippolito, and D. D. Keller. Specimens examined in this work kindly were made available by L. R. Heaney and W. T. Stanley, FMNH. Speci- mens of A. sacobianus (USNM) were mea- sured by R. Garcia Perea through the good graces of A. L. Gardner. J. C. Brown assisted with measuring specimens at FMNH. L. R. Heaney, T. A. Munroe, G. G. Musser, and E. A. Rickart reviewed, and greatly im- proved, preliminary drafts of this manu- script. G. G. Musser kindly diagnosed these VOLUME 108, NUMBER 2 creatures as a new species. This work was made possible in part by a grant from the John D. and Catherine T. MacArthur Foun- dation to P. C. Gonzales (NMP) and R. S. Kennedy (CMNH). This paper constitutes a contribution of the joint National Muse- um of the Philippines/Cincinnati Museum of Natural History Biodiversity Inventory Project. Electron micrographs courtesy of F. Ezra and D. Jacobs, Miami Valley Labo- ratories of Procter & Gamble. Literature Cited Ames, O. 1907. Orchidacee halconenses: an enu- meration of the orchids collected on or near Mount Halcon, Mindoro, chiefly by Elmer D. Merrill.— Philippine Journal of Science, Series C (Botany) 2:311-337. Brotherus, V. F. 1907. Musci halconenses.—Philip- pine Journal of Science, Series C (Botany) 2:339- 343. Heaney, L. R. 1986. Biogeography of mammals in SE Asia: estimates of rates of colonization, ex- tinction and speciation.— Biological Journal of the Linnean Society 28:127-165. , P. D. Heideman, E. A. Rickart, R. B. Utzur- rum, & J. S. H. Klompen. 1989. Elevational zonation of mammals in the central Philip- pines. —Journal of Tropical Ecology 5:259-280. Hill, J. E. 1935. The cranial foramina in rodents.— Journal of Mammalogy 16:121-129. Mearns, E. A. 1905. Descriptions of new genera and species of mammals from the Philippine Is- lands.— Proceedings of the United States Na- tional Museum 28(1402):425-460. Merrill, E. D. 1907a. The ascent of Mount Halcon, Mindoro. — Philippine Journal of Science, Series A (General Science) 2:179-203. 1907b. The flora of Mount Halcon, Min- doro.— Philippine Journal of Science, Series C (Botany) 2:251-309. 315 Miller, G. S., Jr. 1912. Catalogue of the mammals of Western Europe (Europe exclusive of Russia) in the collection of the British Museum. London, British Museum (Natural History), xv + 1019 pp. Musser, G. G. 1979. Results of the Archbold expe- ditions. No. 102. The species of Chiropodomys, arboreal mice of Indochina and the Malay Ar- chipelago.— Bulletin of the American Museum of Natural History 162(6):377-445. 1981. Results of Archbold expeditions. No. 105. Notes on systematics of indo-malayan mu- rid rodents, and descriptions of new genera and species from Ceylon, Sulawesi, and the Philip- pines.— Bulletin of the American Museum of Natural History, 168(3):225-334. 1982. Results of the Archbold expeditions. No. 108. The definition of Apomys, a native rat of the Philippine Islands.— American Museum Novitates 2746:1-43. ——, & L. R. Heaney. 1992. Philippine rodents: definitions of Tarsomys and Limnomys plus a preliminary assessment of phylogenetic patterns among native Philippine murines (Murinae, Muridae).— Bulletin of the American Museum of Natural History 211:1-138. , L. K. Gordon, & H. Sommer. 1982. Species- limits in the Philippine murid, Chrotomys. — Journal of Mammalogy 63:514—521. Rickart, E. A., L. R. Heaney, P. D. Heideman, & R. C. B. Utzurrum. 1993. The distribution and ecology of mammals on Leyte, Biliran, and Mar- ipipi Islands, Philippines.—Fieldiana (Zoolo- gy), New Series No. 72:1-62. Ruedas, L. A., & J. W. Bickham. 1992. Morpholog- ical differentiation between Rhogeessa minutilla and R. tumida (Mammalia: Chiroptera: Ves- pertilionidae).— Proceedings of the Biological Society of Washington 105:403-409. SAS Institute Inc. 1988a. SAS/STAT user’s guide, release 6.03 edition. Cary, North Carolina, SAS Institute Inc., 1028 pp. 1988b. SAS procedures guide, release 6.03 edition. Cary, North Carolina, SAS Institute Inc., 441 pp. 316 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Appendix 1.—Means (plus or minus one standard deviation; sample size in parentheses) and ranges for 28 cranial and mandibular measurements in 12 populations of Apomys examined. Superscript ‘a’ indicates holotype of A. gracilirostris. Taxon: A. abrae (2) A. abrae (2) A. abrae (6) A. abrae (2) Character (Abra Prov.) (Ilocos Norte Prov.) (Ilocos Norte Prov.) (Mountain Prov.) BB — 14.71 + 0.52(3) 14.2-15.2 13.79 + 0.60 (3) 13.1-14.3 14.86 + 0.18 (2) 14.7-15.0 BIF i NOOO) 20-9 OI OOS) QO UO EOI) 182.0 BIT = ES 2 EO) ES BONG) NGENO WS O03) 1aEN.e BM1 19m 9320-5027 Gr 2.9236 2.70/21 01361G) una 322.9) a 3 8i-e 0108) Oma aore BM3 = 4.23 = 0.247) 3.8-4.5. 3.98 = O.25/G)in 3f=4-2) 4.25) OOO) mea =a) BMF = 227 = 02200) 20226 27420500) 62423 238 +0202) ES BR at 5.93 + 0.23(7) 5.5-6.2 5.61 +0.21(3) 5.4-5.8 6.32+0.05(2) 6.3-6.4 BZP 21 °° 2842 016@) 26230 275 = 0073). 27-28 92:67 0032) momo HB 45 461 +0.37(3) 4.2-4.9 4.30 + 0.02 (2) ae Mure O21 @) A225 HBC =) 1060 014) 10425107 10.21 = 0: 7114) 9721-3) 10.70 10.221 OC) enor s10rs IB ue 5.42 +0.16(7) 5.2-5.7 5.30 +0.13(3) 5.2-5.4 5.40 + 0.05 (2) = IFM1 18) 169 £0167) 15-19 1:49 = 0.236) 12=7 > 1627-7 0.25, Cease LB 5.0 4.96 + 0.26(2) 4.8-5.1 4.75 +40.11(2) 47-48 4.48+0.11(@2) 4.4-4.6 LD 8.8 8.54+0.30(7) 8.0-8.9 8.66 +0.60(3) 8.0-9.2 8.06 +0.52(2) 7.7-8.4 LIF 45, 473-015) 4.6449 4.932023) 47-5) 4 61 = 0.401 Oa an9 LM13 6.5 6.64+0.24(7) 6.4-7.0 6.86+0.55(4) 6.1-7.4 7.00 + 0.05 (2) = LN 13.3. 13.05 + 0.55(7) 12.3-13.6 13.39 + 0.54 (3) 12.8-13.8 12.96 + 1.06 (2) 12.2-13.7 LPB 84 97:96 + 0:08 (7) 7-8-8.) | 97.46 = 0.02) 724275 1.40 = Ono Ons ates LR 11.9 11.96 + 0.39(7) 11.4-12.5 11.82 + 0.63 (3) 11.2-12.5 12.54 + 1.24(2) 11.7-13.4 ONL 35.1 35.36 + 0.95 (4) 34.3-36.3 34.98 + 1.00(3) 33.9-35.9 35.24 + 1.42 (2) 34.2-36.2 PL 18.0 17.52 + 0.29(7) 17.1-17.9 16.64 + 0.51 (3) 16.2-17.2 17.30 + 1.02(2) 16.6-18.0 PPL 9.8 10.76 + 0.60(3) 10.1-11.3 11.51 + 1.82(3) 10.0-13.5 10.70 + 0.40(2) 10.4-11.0 ZB SAO — @) i as a za a LMAND 19.8 19.16 + 0.23 (6) 18.9-19.5 19.20 + 1.21(4) 17.7-20.5 19.08 + 1.15 (2) 18.2-19.9 LMT 6.4 6.42 +0.36(7) 60-69 6.29+0.42(4) 5.7-6.6 6.46 + 0.63(2) 6.0-6.9 HMC 7.4 8.06 +0.35(7) 7.7-8.6 7.80 + 0.62(4) 7.0-8.4 7.78 + 0.25(2) 7.6-8.0 HMR 2.8 2.65+0.11(7) 2.5-2.8 2.70+0.10(4) 2.6-2.8 2.68 +0.08(2) 2.6-2.7 ™ 18) 208 40.097), “2102233 196 EOMTG). Sea) est Ost) aie In Taxon: A. abrae (6) A. datae (2) A. datae (é) Character (Mountain Province) (Mountain Province) (Mountain Province) BB 14.95 + 0.55(7) 13.9-15.6 15.28 + 0.43 (10) 14.8-16.0 15.46 + 0.42(7) 14.7-15.9 BIF 1.98 + 0.21(7) 1.8-2.3 2.22 + 0.15(10) 2.0-2.5 2.49 + 0.15(7) 2.4-2.8 BIT 1.81 + 0.08(7) 1.7-1.9 2.02 + 0.12(10) 1.8-2.2 2.23 + 0.18 (6) 2.1-2.6 BM1 3.18 + 0.14(7) 3.0-3.4 3.17 + 0.21 (10) 2.7-3.5 3.40 + 0.10(7) 3.2-3.6 BM3 4.12 + 0.34(7) 3.44.4 4.65 + 0.27(10) 4.4-5.1 4.81 +0.19(7) 4.6-5.1 BMF 2.38 + 0.21 (7) = 2.2-2.7 2.39 + 0.28 (10) = 1.8-2.9 2.58 + 0.17 (6) 2.4-2.8 BR 5.99 + 0.15(7) 5.8-6.2 6.59 + 0.30(10) 6.3-7.3 6.92 + 0.35(7) 6.4-7.4 BZP 2298) == 0 (7) ae — SS 3.30 + 0.22(10) 2.9-3.6 3.45 + 0.18 (7) 3.2-3.6 HB 4.37 + 0.33(7) 3.9-4.7 4.73 + 0.33 (10) 4.2-5.2 4.79 + 0.24(7) 4.5-5.1 HBC 10.41 = 0.35(7) 9.7=10:7 11.15 = 0.41 (10) 10.6-12.0 11.21 = 0.52(7) 10:4=11.8 IB 5.65 + 0.14(7) 5.4-5.8 5.97 + 0.13 (10) 5.8-6.2 6.01 + 0.27(7) 5.5-6.4 IFM 1 1.68 + 0.10(7) 1.5-1.8 2.02 + 0.18 (10) 1.6-2.2 2.14 + 0.16(7) 2.0-2.4 LB 4.45 +0.19(7) 4.2-4.6 4.70 + 0.15(10) 4.4-5.0 4.76 + 0.20(7) 4.4-5.1 LD 8.46 + 0.51 (7) 7.8-9.4 9.24 + 0.39(10) 8.7-10.0 9.72 + 0.29(7) 9.4-10.1 LIF 4.68 + 0.29(7) 4.2-5.0 4.73 + 0.41 (10) 3.9-5.2 4.83 + 0.29(7) 4.4—-5.4 LM13 6.85 + 0.16(7) 6.6-7.0 6.92 + 0.20(10) 6.7-7.3 7.16 + 0.15 (7) 7.0-7.4 LN 13.49 + 0.37(7) 13.0-14.0 14.66 + 0.68 (10) 13.6-16.0 15.15 + 0.67(7) 14.3-16.3 LPB 8.22 + 0.41 (7) 7.6-8.8 9.09 + 0.28 (10) 8.6-9.5 9.38 + 0.14(7) 9.1-9.5 LR 12.54 + 0.44(7) 11.9-13.0 13.42 + 0.70(10) 12.5-15.0 14.14 + 0.48 (7) 13.3-14.8 ONL 35.62 + 0.80(7) 34.4-37.1 38.48 + 1.04(10) 36.8-40.5 39.76 + 0.60(7) 38.6-40.3 VOLUME 108, NUMBER 2 317 Appendix 1.—Continued. Taxon: A. abrae (8) A. datae (2) A. datae (6) Character (Mountain Province) (Mountain Province) (Mountain Province) PL 17.82 + 0.66(7) 17.0-19.1 19.11 + 0.60(10) 18.3-20.5 19.91 + 0.42(7) 19.4—20.8 JPARIL, 10.36 + 0.45(7) 9.8-11.1 11.61 + 0.46(10) 10.9-12.2 11.87 + 0.43 (7) 11.3-12.3 ZB 16.0 — (1) — 18.03 + 0.59 (5) 17.1-18.7 18.24 + 0.26 (4) 18.1-18.6 LMAND 19.20 + 0.46(7) 18.3-19.8 20.95 + 0.59 (10) 20.1-22.0 21.83 + 0.57(7) 21.0-22.5 LMT 6.24 +0.15(7) 6.0-6.4 6.42 + 0.23(10) 6.1-6.8 6.69 + 0.13(7) 6.5-6.9 HMC USK ae O13 (TD) Votesa3 9.13 + 0.52(10) 8.3-9.9 9.49 + 0.31(7) 9.1-10.0 HMR AAD 32 OSU) A 28) 3.06 + 0.14(10) 2.8-3.3 3.26 + 0.13 (7) 3.1-3.4 ™ 1.85 +0.10(7) 1.7-2.0 2.23 + 0.08 (10) 2.1-2.3 2.23 + 0.15(7) 2.0-2.4 Taxon: A. i. insignis (2) A. 1. insignis (6) A. insignis bardus (2) Character (Bukidnon Province) (Bukidnon Province) (Zamboanga del Norte Prov.) BB 13.32 + 0.41 (6) 12.8-13.9 13.48 + 0.30(10) 13.0-14.1 13.63 — (1) — BIF 1.98 + 0.22 (6) 1.6-2.3 2.02 +0.17(11) 1.8-2.3 2.20 — (1) — BIT GO) 101076) ale6ae8) ) 1.6m = OO), 1521-8) 1.82 = 0.20) 1.7=2.0 BM1 3.09 + 0.30 (6) 2.7-3.4 3.06 + 0.17(11) 2.8-3.3 3.29 + 0.04 (2) _ BM3 4.28 + 0.10(6) 4.2-4.5 4.19 +0.15(11) 3.9-4.4 4.23 + 0.21(2) 4.1-4.4 BMF 2.38 + 0.16(2) 2.3-2.5 2.28 + 0.15 (8) 2.1-2.5 2.46 + 0.02(2) 2.4-2.5 BR 5.43 + 0.16(6) 5.3-5.7 5.37 + 0.26(10) 5.0-5.9 Doll — (1) — BZP DEAS 101091(6) 24216 241 0.191) 2257 2.20.2 0:29.(2). 2.0-2.4 HB 3.80 + 0.24(6) 3.5-4.1 3.58 + 0.54 (9) 2.7-4.2 3.76 + 0.02(2) 3.7-3.8 HBC 9.56 + 0.13(6) 9.4-9.8 92399-09331 Q9) 9.0-9.9 9.74 + 0.07 (2) 9.7-9.8 IB 4.94 + 0.16(6) 4.7-5.2 5.07 + 0.24(11) 4.7-5.4 4.84 + 0.37(2) 4.6-5.1 IFM1 ID 2 O28 Cy 1AM SS O23 On Gh) Es LB 3.99 + 0.19(6) 3.8-4.3 3.78 + 0.30(10) 3.2-4.2 3.76 + 0.00 (2) — LD 7.28 + 0.36 (6) 6.7-7.7 7.31 +0.25(11) 6.8-7.8 7.34 + 0.16(2) 7.2-7.4 DE 3.68 + 0.49 (6) 3.0-4.3 3.71 + 0.21(11) 3.3-4.0 3.78 — (1) — LM13 6.02 + 0.10(6) 5.8-6.3 5.86 + 0.16(11) 5.6-6.2 5.74 + 0.30(2) 5.5-6.0 LN 11.49 + 0.39 (6) 11.0-12.0 11.64 + 0.70(11) 10.6-12.9 11.41 — (1) att LPB 7.41 + 0.28 (2) 7.2-7.6 7.18 + 0.02 (2) — 7.02 — (1) — LR 10.48 + 0.52 (6) 9.9-11.4 10.51 + 0.58(11) 9.3-11.5 10.72 — (1 — ONL 31.86 + 0.60 (6) 31.2-32.9 31.68 + 0.63 (10) 30.7-32.7 32.29 — (1) — PL 15.75 + 0.44 (2) 15.4-16.1 15.41 + 0.20(2) 15.3-15.6 15.49 + 0.25 (2) 15.3-15.7 PPL 9.71 = (A) uss 9.3 = (1) an 9.13 + 0.54(2) 8.8-9.5 ZB a =O) ve a (O) = ATO ss @) Be LMAND 17.26 + 0.21 (6) 17.0-17.6 17.06 + 0.44(10) 16.1-17.6 16.54 + 0.35 (2) 16.3-16.8 LMT 5.38 + 0.30(6) 5.0-5.9 5.43 + 0.17(11) 5.2-5.7 5.24 + 0.16(2) 5.1-5.4 HMC 7.25 + 0.24(6) 6.9-7.5 7.29 + 0.29(11) 6.8-7.7 6.71 + 0.16(2) 6.6-6.8 HMR 2.46 + 0.14(6) 2.2-2.6 2.42 +0.13(11) 2.1-2.7 2.45 + 0.40 (2) 2.2-2.7 ™ 1.78 + 0.10(6) 1.6-1.9 LIS ONS (1) 1.5=1-9 1.65 + 0.06 (2) 1.6-1.7 Nig ae A. gracilirostris Taxon: A. insignis bardus (6) Character (Zamboanga del Norte Prov.) 11362 (6) 1126(6) 1131) 1132(@ 1119@) Means (+SD) BB 13.29 — (1) — 15.6 16.1 S37 lid OL et 6 np G105)==3Or47A BIF 1.94 + 0.06 (2) 1.9-2.0 1.8 2.0 Dal De 1.9 1.98 + 0.145 BIT Less) as 03)(@) 1.8-1.9 1.6 LES 1.6 1.5 1.5 1.56 + 0.054 BM1 3.01 + 0.07 (2) 3.0-3.1 3.5 3.5 4.4 4.3 4.5 4.05 + 0.486 BM3 4.32 + 0.09 (2) 4.2-4.4 4.7 5.0 3) — Dell 5.24 + 0.465 BMF 2263 == 032152) 2.5-2.8 1.9 2.1 2.0 Dodd 2.3 2.09 + 0.176 BR 2 O28) (2) 5.5-5.9 6.6 6.3 6.7 6.6 6.4 6.53 + 0.145 BZP Ded O50 2612) 2.2-2.5 3.3 3.0 3.2 3.6 3.0 SZ 1F=10:223 HB 3.74 — (1) — 4.2 4.0 3.4 4.2 3.8 8) )7) ae (0)80)7/ HBC Y)oI5) — (1) — 11.1 Sse NOS Or Lt 10.88 + 0.404 IB 4.91 + 0.20 (2) 4.8-5.0 5.8 6.3 5:5 6.0 6.2 6.00 + 0.233 318 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Appendix 1.—Continued. A. gracilirostris 11362 (8) 1126 (8) 1131(4) 1132(8) 1119 @) A. insignis bardus (6) (Zamboanga del Norte Prov.) Taxon: Character Means (+SD) IFM1 esi74 se (0), (2) 1.7-1.9 2.4 Dd) 2.4 2.9 2.5 2.49 + 0.241 LB 3.99 — (1) — 5.1 5.0 4.8 Do 5.0 5408) ae O JS! LD 7.62 + 0.42 (2) 7.3-7.9 OS LON O22 hOr8 Or 1027 205450 LIF 3.80 + 0.08 (6) 3.7-3.8 4.8 4.8 4.9 5.0 4.9 4.87 + 0.102 LM13 5.54 + 0.39 (3) 5.2-6.0 6.3 6.3 6.6 6.2 6.5 6.39 + 0.246 LN WI96 = OF46;@)) > VIRG=12°3) 11458) 48 4 GO 49 S005 O66 LPB 7.44 + 0.15 (2) 7.3-7.6 OZ 8.8 9.1 9.5 9.0 9.14 + 0.259. LR 10570 7020 @)) VORG=1028 7 13:9) 139 14:0) 429 Sto Aalst Ora ONL 32.06 — (1) — 39°4 3957 39500 V4k6n 3958 139289 ee OLS PL lssisa3 O07) © Westolloss Oeil NO NOE NO NA NDS) 28 033414! PPL S741 — (1) — L9G 6 1253 el 2a ESF OeS@ ZB — — (0) _ NUS NGS USO US UoIb NYS ae O88) LMAND cys) 2s eS)" USES = AO. Mk ADA Ailes) 9 AV» AVA 26 O./50 LMT 5.42 + 0.11 (2) 5.4-5.5 6.4 Gz 6.2 6.6 6.2 6.32 + 0.183 HMC 6.70 + 0.40 (2) 6.4-7.0 8.5 8.6 8.2 9.0 8.9 8.64 + 0.298 HMR 2.40 + 0.20 (2) 2.2—2.5 De) 2.8 2.9 3.0 2.8 2.88 + 0.090 ™ 1.56 + 0.06 (2) 1.5-1.6 1.9 LE, 1.8 1.8 1.8 1.80 + 0.062 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(2):3 19-337. 1995. A review of the spiny mouse genus Scolomys (Rodentia: Muridae: Sigmodontinae) with the description of a new species from the western Amazon of Brazil James L. Patton and Maria Nazareth F. da Silva (JLP, MNFS) Museum of Vertebrate Zoology, University of California, Berkeley, California 94720 U.S.A.; (MNES) Departmento de Ecologia, Instituto Nacional de Pesquisas da Amazonia, Manaus, AM, Brazil Abstract. —Included within a collection of mammals assembled during a year- long vertebrate survey of the Rio Jurua in the western Amazon Basin of Brazil is a series of specimens of a new spiny mouse of the genus Scolomys. This genus is one of the more highly-localized and poorly-known murid rodents of the neotropical forests. Prior to the collection reported here, there were two recognized species known from a total of 15 specimens. One of these, S. me- lanops Anthony, 1924, is known only from three closely spaced localities in eastern Ecuador; the second, S. ucayalensis Pacheco, 1991, 1s known from only one locality in north-central Pert. We provide a revised diagnosis and descrip- tion of the genus while describing the third species, suggest phylogenetic affin- ities of the genus within the tribe Oryzomyini, and summarize aspects of the ecology and life history of the new species. Resumo.—Um novo catito de espinho (género Scolomys) foi coletado du- rante um levantamento da fauna de vertebrados realizado no rio Jurua, no oeste da Amazonia brasileira. Este género de roedores murideos neotropicais foi pouco estudado e possui uma distribuicao geografica muito restrita. Somente duas espécies eram até entao conhecidas: uma do leste do Equador (S. melanops Anthony, 1924; 13 espécimes provenientes de trés localidades) e outra do norte do Peru (S. ucayalensis Pacheco, 1991; dois espécimes provenientes de uma localidade). Neste estudo nos descrevemos uma terceira espécie e apresentamos uma revisao aumentada da diagnose e descricao do género. Também sugerimos afinidades filogenéticas dentro da tribo Oryzomyini, além de sumarizarmos pela primeira vez aspectos da ecologia e historia natural desses animais. The genus Scolomys contains small-bod- ied and strongly spinose mice of the tribe Oryzomyini Vorontsov, 1959 (sensu Voss & Carleton 1993) of the South American Sigmodontinae (Muridae) rodents. Each of the two known species has a very localized distribution in the forests of western Ama- zonia. The genus was described in 1924 by H. E. Anthony and the type species, S. me- lanops, is known from a total 13 specimens from three nearby localities in east-central Ecuador. A second species, S. ucayalensis, was recently described by Pacheco (1991) based on two specimens from a single lo- cality in north-central Peru. We have obtained a series of 23 speci- mens from four localities along the Rio Ju- rua in the lowland Amazonian forest of western Brazil (states of Amazonas and Acre) that represents a third species of this poorly- 320 known genus. Based on these new materials and an examination of most other speci- mens, we provide an expanded diagnosis and description of the genus and describe the new species here. We also compare Sco- lomys to the sympatric and superficially similar oryzomyine genus Neacomys as well as to other oryzomyine genera, provide re- marks on phyletic relationships based on morphological characters and comparative DNA sequences, and summarize what few facts are available on life history and ecol- ogy. Scolomys Anthony, 1924 Type species.—Scolomys melanops An- thony (1924:2). Emended diagnosis.—Members of the tribe Oryzomyini (sensu Voss & Carleton 1993) of the murid rodent subfamily Sig- modontinae (sensu Carleton & Musser 1984) with 3 pairs of mammae (1 thoracic, 1 ab- dominal, and | inguinal [following position designations given by Voss & Carleton 1993; Anthony (1924) recorded mammae as 1 pectoral, 2 inguinal]). Pelage comprised of short, stiff spines on both dorsum and ven- ter with equal-length normal hairs inter- spersed throughout, giving general spiny ap- pearance over entire body. Skull with short and blunt rostrum flanked by shallow zy- gomatic notches; supraorbital margins rounded, forming a moderately-developed shelf overhanging posterior half of orbit, and extending onto braincase as ridges; brain- case rather globular in shape; interparietal large and well-developed; palate long and wide, with well-developed and complex posterolateral pits, but with rather short and posteriorly broadened incisive foramina; al- isphenoid strut absent; carotid arterial cir- culation of Pattern 3 (of Voss 1988); sub- squamosal fenestra reduced to totally oc- cluded, tegmen tympani either not in con- tact with or only touches, but does not overlap, squamosal; incisors small, narrow, proodont to orthodont; upper and lower molars small, pentalophodont, but with low PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON cusps that wear quickly with age; procin- gulum of M1 undivided by anteromedial flexus; labial flexi deeply penetrating in all molars; lingual flexi reduced in size in M1 and M2 and obsolete in M3; upper and low- er molars with well-developed meso- loph(id)s; stomach unilocular and hemi- glandular; male phallus cylindrical with in- complete crater rim, terminally exposed urethral flaps, lateral mounds of distal bac- ulum hidden by tissue of crater rim, and an epidermis with small and widely-spaced spines. Description. — Body pelage short and close, with texture markedly spinose both above and below; color ranges from grizzled pale reddish-black to nearly totally black dor- sally and gray ventrally; dorsal hairs of two types: (1) long (averaging 12 mm), stout, flat, and broad (averaging 0.6 mm) spines with a medial trough on both surfaces, with the terminal '4 to 4 increasingly dark to the tip and proximal portion clear; and (2) long, thin hairs of length equal to spines and with tips reddish or blackish; ventral hairs of both types uniformly gray from base to tip. Mys- tacial, superciliary, genal I, submental, in- terramal, and carpal vibrissae present. Pin- nae small, appearing somewhat thickened and thus stiff, and, while appearing naked from a distance, are clothed externally and internally with short reddish-brown hairs. Manus with five large, fleshy plantar pads (two carpal and three interdigital); toes pale in color; digit I reduced but with a small nail, digits II through V long and well-de- veloped with short, stout, and curved claws. Pes rather short and broad, although meta- tarsus 1s nearly twice as long as digit III; the heel is haired and the naked sole begins at about % the length of the plantar surface (not including the digits); outer digits short- er than the middle three (with the claw of I extending to or just past the base of II and that of V to the proximal phalax of IV); conspicuous tufts of long, silvery hairs pres- ent at dorsal bases of claws extending past the tips, but the claw is visible from above; VOLUME 108, NUMBER 2 Fig. 1. Jurua, Amazonas, Brazil). claws short, stout (about twice as long as deep) and strongly curved along their dorsal surface; plantar pads five to six (thenar and four interdigital pads large, fleshy, and al- ways present; hypothenar pad either absent or only weakly developed). Tail shorter than head-and-body, appearing sparsely haired, without a terminal tuft or pencil of long hairs; 15-18 scale annuli per cm at mid- length; annular hairs broad, blackish, and 2.5—4 scale rows in length, but sparsely dis- tributed so that the tail scales are conspic- uous. The overall impression is of a small- bodied, short-tailed, and short-eared mouse with a broad head, but short and pointed rostrum and very spiny fur (Fig. 1). Skull, in dorsal view, with short and broad or tapering rostrum flanked by shallow, barely perceptible zygomatic notches; na- solacrimal capsules expanded (especially so in S. melanops, Fig. 2); interorbital region broad and hourglass-shaped, with well-de- 321 Photograph of a living Scolomys juruaense, new species (INPA 2490, Barro Vermelho, left bank Rio fined beaded ledges overhanging margins from middle of frontals, continuing along posterior margins of orbit and onto brain- case just above the squamoso-parietal su- ture as weakly to moderately-developed temporal ridges. Braincase distinctly round- ed and globular in appearance, dominating dorsal aspect of skull (length of braincase '2 length of skull). Nasals somewhat expanded and taper posteriorly to a median point that terminates well behind the premaxillary- frontal sutures. Interparietal large, 2 to 5 as deep as wide. In lateral view, nasals extend only to or just barely beyond anterior curvature of in- cisors. Zygomatic plate narrow, vertical to slightly angled posteriorly from base, and without distinct, free dorsal edge (thus, the zygomatic notch is shallow when viewed from above). Zygomatic arch thin with jugal reduced. Postglenoid foramen moderate to small; hamular process of squamosal stout; 322 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON “al Fig. 2._ Dorsal, ventral, and lateral views of cranium, lateral view of left mandible of Scolomys melanops Anthony, USNM 513581, adult male. Scale bar equals 10 mm. VOLUME 108, NUMBER 2 subsquamosal foramen reduced to totally occluded; mastoid fenestra very small to lacking. Tegmen tympani of periotic either does not contact or abuts, but does not over- lap, the squamosal. Tympanic bullae small and inflated ventrally only to level of molar series. In ventral view, incisive foramina mod- erate in size (occupying about 60% of dias- temal distance) and distinctly tear-drop in shape, pointed anteriorly with diverging sides and expanded, rounded posterior mar- gins; premaxillary-vomerine septum greatly swollen and nearly filling the entire cavity when viewed ventrally. Bony palate long and wide (sensu Hershkovitz 1962), without a medial ridge or palatal excrescences, with only weakly evident lateral folds, but with large and complex posterolateral pits. The mesopterygoid fossa wide with parallel sides and a rounded or squared anterior margin, ending well behind the third molars; bony roof of fossa complete, or perforated only by barely perceptible sphenopalatine va- cuities along the presphenoid. Paraptery- goid fossae well developed, with lateral margins straight to slightly convex and strongly divergent towards the bullae, de- void of vacuities except for a small foramen ovale, and moderately excavated, certainly not flat in appearance. Alisphenoid strut ab- sent, but only foramen ovale is present lat- erally; without anterior opening of alis- phenoid canal. A shallow trough where the masticatory-buccinator branch of the max- illary nerve courses visible; it emanates from anterior margin of the foramen ovale and obliquely crosses the alisphenoid onto the squamosal. Facial circulation apparently derived only from the internal carotid artery (Pattern 3, of Voss, 1988), as indicated by a greatly reduced to absent stapedial fora- men, no squamoso-alisphenoid groove along interno-lateral wall of braincase, and no sphenofrontal foramen (signs of supraor- bital branch of stapedial artery). Mandible short and stout; coronoid pro- cess short with a weakly to moderately- 323 curved posterior projection; capsular pro- cesses of lower incisor alveoli weakly de- veloped; lower incisors thin, elongate, with enamel essentially devoid of pigment. Up- per incisors ungrooved, with yellow to pale yellow enamel; small, deeper than wide, and proodont (S. melanops, Fig. 2) to orthodont (S. ucayalensis and the new species de- scribed below). Maxillary tooth rows slightly convergent posteriorly, and angled obliquely downward and outward at about a 40 degree angle. Teeth of nearly all known specimens mod- erately to well-worn, and, as the cusps are low, even a little wear obscures surface to- pography. Molars small, always longer (an- teriorly—posteriorly) than wide, and forming a graded series with the third molar greatly simplified. Upper teeth pentalophodont with principal cusps arranged transversely and slightly obliquely; labial and lingual reen- trant folds do not interdigitate, or contact, with major labial folds of M1 and M2 (par- aflexus and metaflexus) deep, extending at least *3 across the tooth, lingual folds re- duced, with protoflexus only evident as a shallow lateral indentation in M1 and not visible at all in M2. Procingulum of M1 and m1 well-developed but not divided into separate anterolabial and anterolingual con- ules (no anteromedial flexus [-id]); antero- flexus on M1 absent so that anteroloph not separated from labial anteroconule; anter- oloph of M2 well developed; distinct me- solophs present and extending to labial mar- gin of all three molars; posteroloph well de- veloped on M1 but barely perceptible on M2 and absent on M3. Paracone and me- tacone of M1 and M2 tall and well devel- oped with protocone and hypocone pro- portionately reduced in size and much lower in topography; only paracone and weakly- developed protocone present on M3. Content and distribution. —The genus Scolomys comprises two described species, the known ranges of which are geographi- cally restricted within the western Amazon Basin (Fig. 3). The type species, S. melan- 324 10° PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 4 eto S ee arid ated ‘ J- Fig. 3. Distributional records of spiny mice, genus Scolomys. Circles = S. melanops Anthony, triangle = S ucayalensis Pacheco, and squares = S. juruaense new species (numbered as in list of specimens; see text). ops, is known from three localities in eastern Ecuador: the type locality at Mera, 1160 m, Pastaza Province (holotype and five para- types in the American Museum of Natural History, New York [AMNH], and three topotypes in the Natural Museum of Nat- ural History, Washington D.C. [USNM)]); Huamani, Volcan Sumaco, Napo Province (above Mera, one specimen in the collection of the Escuela Politécnica Nacional, Quito [Albuja, 1991]), and Limoncocho, 250 m, Napo Province (3 specimens in USNM). Scolomys ucayalensis Pacheco is known from two specimens from its type locality, Centro de Investigaciones “‘Jenaro Herre- > 2.8 km E Jenaro Herrera, Depto. Lo- reto, right bank of Rio Ucayali, Peru, 135 m. A third species from four localities in western Amazonia of Brazil is described here; it may be known as Scolomys juruaense, new species Holotype. —MPEG 23824 (Museo Par- aense Emilio Goeldi, Belem, Para, Brazil), adult female, collected on 19 September 1991 by J. L. Patton (original number 15570); skin with skull and mandibles, in good condition, plus liver tissue preserved both deep frozen and in ethyl alcohol. Tis- sues are maintained in the collections of the Museum of Vertebrate Zoology, University of California, Berkeley. Type locality. —Seringal Condor, left bank Rio Jurua, Amazonas, Brazil 70°51'W, 6°45’S. Obtained in tree-fall slash distur- bance in otherwise primary terra firme (or upland, non-seasonally flooded) forest. Paratypes.—The total known sample of S. juruaense consists of the holotype and 22 additional specimens that are deposited in VOLUME 108, NUMBER 2 the Colecéo de Mamiferos, Instituto Na- cional de Pesquisas da Amazonia (INPA); Museu Paraense Emilio Goeldi (MPEG); and Museum of Vertebrate Zoology, Uni- versity of California at Berkeley (MVZ), list- ed here by locality (numbered as in the map, Fig. 3): Brazil—Acre: (1) Sobral, left bank Rio Jurua, 72°49’W, 8°22'S—INPA 2485, adult female, skin with skull, carcass in fluid plus karyotype; INPA 2486, adult male, in fluid plus karyotype; MPEG 24023, adult female, skin with skull, carcass in fluid plus karyotype; MPEG 24024, adult female, in fluid plus karyotype; MVZ 183172, juvenile female, in fluid plus karyotype. Amazonas: (2) type locality -MPEG 24019, adult male, skin and skull; MPEG 24020, adult male, skin and skull plus karyotype; MVZ 183167, adult female, skin and skull; MVZ 183168, adult male, skin and skull plus karyotype; (3) Penedo, right bank Rio Jurua, 70°45’W, 6°50'S—INPA 2487, adult male, skin and skull; INPA 2488, adult male, body in fluid with skull extracted; MPEG 24022, adult female, body in fluid with skull extracted; MVZ 183165, adult male, body in fluid with skull extracted; MVZ 183166, adult female, skin and skull; and (4) Barro Vermelho, left bank Rio Jurua, 68°46’W, 6°28’S—INPA 2489, subadult male, skin and skull plus karyotype; INPA 2490, adult female, skin and skull plus karyotype; INPA 2491, adult female, skin and skull plus karyotype; INPA 2492, adult male, skin and skull, karyotype; MPEG 24021, adult male, skin and skull plus karyotype; MVZ 183169, subadult male, skin and skull plus karyotype; MVZ 183170, adult male, in fluid plus karyotype; and MVZ 183171, adult male, in fluid. Liv- er tissues preserved in 95% ethyl alcohol and frozen at —76°C are available for all specimens and are deposited in the Museum of Vertebrate Zoology, as are chromosome slides for all karyotyped specimens. Distribution. —Known from three locali- ties on the left and one on the right bank in the central and upper reaches of the Rio Jurua in the Brazilian states of Acre and 325 Amazones (Fig. 3); all localities are below 400 m in elevation. Etymology. —The name refers to the known distribution along the Rio Jurua, the largest white-water tributary of the Rio Amazonas with an origin extralimital to the Andean cordillera. Diagnosis. —A small-bodied mouse (Ta- ble 1) with short, nearly naked tail (83% of body length); short, broad head with point- ed snout (Fig. 1); short and relatively broad hindfeet; hypothenar pad greatly reduced to absent (minutely present in 15 of 23 spec- imens) but thenar and interdigital pads well- developed; small and rounded ears; dorsal color varying from a grizzled pale reddish- brown (Sudan Brown to Antique Brown; capitalized color terms from Ridgway 1912) to dark reddish-black (Raw Umber) finely streaked with black; with rounded and in- flated braincase; short, basally-broad ros- trum that tapers distally; narrowed and straight zygomatic arches; narrow but long orbital openings; subsquamosal fenestra to- tally occluded by stout hamular process of squamosal; short and distally broad incisive foramina with sides distinctly ‘stepped’; wide mesopterygoid fossa with parallel sides and squared, as opposed to rounded, ante- rior margin (Fig. 4); and 2N = 50. Other characteristics are as listed above for the genus. Measurements of holotype. —Measure- ments are in millimeters and weight (mass) in grams; external measurements are those recorded on the specimen label and bilateral measurements were taken on the right side with digital calipers: Total length, 160; head and body, 86; tail, 76; hind foot (with claws), 21; ear (from notch), 16; condyloincisive length (CIL), 21.21; zygomatic breadth (ZB), 12.42: braincase breadth (BB), 11.59; least interorbital breadth (IOC), 5.84; rostral length (RL, taken from anterior orbit to tip of nasals), 7.98; nasal length (NL), 8.60; ros- tral width-1 (RW-1, across nasolacrimal capsules), 4.86; rostral width-2 (RW-2, at premaxillo-maxillary suture), 3.61; orbital 326 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Selected measurements of spiny mice of the genus Scolomys (mean + one standard error and range, with sample size). Variable Total length Tail length Hind foot length Ear height Condyloincisive length Zygomatic breadth Braincase breadth Least interorbital constriction Rostral length Nasal length Rostral width— 1 Rostral width—2 Orbital length Diastema length Maxillary tooth row length S. melanops 153.9 + 3.01 (138-167) n= 10 62.8 + 2.06 (55-77) n= 10 20.9 + 0.31 (20-23) n= 10 15.0 n= 3 19.87 + 0.30 (18.5 1-20.67) n=7 12.42 + 0.18 (11.37-12.90) n= 8 WW ae 0)(0)7/ (10.82-11.41) n= 8 4.96 + 0.07 (4.77-5.45) n=9 6.92 + 0.16 (6.34—7.78) n=9 7.64 + 0.10 (7.16-8.14) n=9 4.78 + 0.07 (4.47-5.10) n=9 3.48 + 0.08 (3.21-3.84) n=9 6.90 + 0.10 (6.45-7.21) n= 8 5.84 + 0.09 (5.50-6.34) n=9 2.70 + 0.05 (2.49-2.93) n=9 S. ucayalensis 144.0 n=1 60.0 h= 18.0 n= 13.0 n= 1 19.41 + 0.47 (18.94—-19.89) n=2 lS ae O75) (11.27-11.77) n=2 11.41 + 0.13 (11.29-11.54) n=2 5.58 + 0.09 (5.49-5.68) n=2 7.69 + 0.27 (7.42-7.96) n=2 5.59 + 0.21 (5.38-5.80) n=2 DS) se ONS (2.69-—2.94) n=2 S. juruaense 152.4 + 1.9 (142-163) n=11 69.0 + 1.50 (26-76) n= 11 20.6 + 0.20 (19-22) n= 16 15.6 + 0.18 (15-17) n= 16 20.43 + 0.22 (18.60—21.97) n= 16 130718 (11.18-13.30) n= 16 11.39 + 0.11 (10.54—12.33) n= 16 5.59 + 0.09 (4.55—6.13) n= 16 7.86 + 0.11 (6.93-8.48) n= 15 8.31 + 0.11 (7.83-9.34) n= 15 4.81 + 0.07 (4.36—5.40) n= 16 3.64 + 0.06 (2.79-4.05) n= 16 7.38 + 0.09 (6.67—7.89) n= 16 6.23 + 0.09 (5.73-6.75) n= 16 2.66 + 0.04 (2.31-2.88) n= 16 VOLUME 108, NUMBER 2 Table 1.—Continued. Variable S. melanops 3.76 + 0.09 (3.31-4.19) n=9 9.03 + 0.19 (8.42—9.69) n=6 4.41 + 0.07 (4.15-4.74) n=8 5.82 + 0.07 (5.54-6.14) n=7 LAs ae OR 1S (9.67—10.70) n=7 3.19 + 0.06 (2.98-3.38) i= 7 3.30 + 0.18 (2.98-3.68) n=4 1.79 + 0.09 (1.38—2.15) n=7 1.59 + 0.06 (1.42-1.81) n=7 Ko) ae (OIA (7.87-8.72) n=7 Incisive foramen length Palatal bridge length Alveolar width Occipital condyle width Mastoid breadth Basioccipital length Mesopterygoid fossa length Mesopterygoid fossa width Zygomatic plate width Cranial depth 77) S. ucayalensis S. juruaense 3.765 + 0.001 (3.76-3.77) =2 8.65 + 0.05 (8.60—8.70) n=2 4.475 + 0.001 (4.47-4.48) n=2 3.94 + 0.07 (3.17-4.32) n= 16 9.14+ 0.12 (8.28-9.87) n= 16 4.62 + 0.05 (4.29-4.96) n= 16 6.01 + 0.06 (5.64—6.46) n= 16 10.29 + 0.08 (9.70-10.73) n= 16 3.23 + 0.05 (2.83-3.60) n= 16 3.61 + 0.05 (3.35-3.92) n= 16 17) ae (0/08) (1.78-2.20) n= 16 Wil se O03} (1.54-1.93) n= 16 8.89 + 0.09 (8.18—9.49) n= 16 length (OL), 7.74; maxillary diastema length (D), 6.43; maxillary tooth row length (MTRL), 2.82; incisive foramen length (IFL), 4.00; palatal bridge length (PBL), 9.68; alveolar width (AW, outside of M1), 4.58; occipital condyle width (OCW), 6.46; mas- toid breadth (MB), 10.73; basioccipital length (BOL), 3.44; mesopterygoid fossa length (MPFL), 3.48; mesopterygoid fossa width (MPFW), 2.13; zygomatic plate width (ZPW), 1.66; cranial depth (CD), 8.97; mass, 26 grams. Additional measurements. —See Table 1 for additional measurements of adult spec- imens of S. juruaense and of S. melanops and S. ucayalensis. Description. —Dorsal coloration uniform from snout to rump, but individuals vary both within and among localities from griz- zled pale reddish-brown to dark reddish- black; ventral coloration uniformly clear gray. Fore and hindfeet clothed dorsally with stiff white hairs; ungual tufts of thin, silvery hairs extend to or just beyond tip of claws. Otherwise as described for the genus, above. Most features of cranial morphology (Fig. 4 and Table 1) are given above in the di- agnosis, or detailed under the extended de- 328 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 4. Dorsal, ventral, and lateral views of cranium, lateral view of left mandible of the holotype of Scolomys Juruaense, new species, MPEG 23824, adult female (original number JLP 15570). Scale bar equals 10 mm. VOLUME 108, NUMBER 2 329 Fig. 5. Occlusal views of left upper maxillary tooth row (from left to right) of S. melanops, USNM 513581 and S. juruaense, new species, MVZ 183169, INPA 2487, INPA 2485. Scale bar equals 1 mm. scription of the genus above. Features of particular note include a short rostrum dis- tally tapering from a broad base; relatively long nasals; shallow zygomatic notches; wide interorbital region with well-developed ledges extending posteriorly as ridges onto temporal region; narrow, rather straight, and nearly parallel zygomatic arches with re- duced jugals; elongated and narrow orbital openings; short (about *4 diastemal length) but posteriorly-widened incisive foramen with distinctly “‘stepped”’ lateral margins; broad mesopterygoid fossa with parallel sides and a flat, squared anterior margin; orthodont upper incisors; and short and stout hamular process of squamosal totally occluding subsquamosal fenestra. Maxillary and mandibular molar teeth are as described above for the genus (Fig. 5). The general lack of specimens with unworn molars precludes the determination of structural differences in occlusal morphol- ogy that may characterize each of the three recognized species. Phallus (Fig. 6) small, elongated, and nar- row (averaging 4.2 mm in length and 1.6 mm in width), with a distinctly cylindrical shape and straight sides. External surface of the glans rugose, sparsely covered with small spines (averaging 7 per mm) buried in ir- regular pits from the lip of the terminal cra- ter (excluding a narrow non-spinous rim) to the prepuce; without a dorsal groove or lat- eral notches, but with a thickened, spine- free midventral ridge extending from pre- puce to crater rim. A distinct, corrugated and non-spinous crater rim present, low in profile ventrally, enlarged laterally, but in- complete at dorsal midline and not circum- scribing entire crater; crater rim distinctly separated from spinous epithelium of prox- imal glans by a distinct fold or groove. Me- dial bacular mound visible distally beyond the crater rim; distinctly shorter and round- 330 -_ “ek: ee: an >* e i , A Pe 2% 3 & r ey a +. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 6. Ventral, lateral, and dorsal views of the male phallus of S. juruaense, new species (MVZ 183171). CR, crater rim; MBM, medial bacular mound; MVR, midventral ridge; P, prepuce; SE, spinous epithelium; UF, urethral flaps. Scale bar equals 2 mm. ed lateral mounds buried under laterally ex- panded crater rim. Dorsal papilla lacking spines and thickened, spatulate, and trian- gular in shape, with its tip extending just above crater rim. Urethral flaps lacking spines, but well developed, tapering distally, and varying in length but always visible above crater rim at ventral base of medial bacular mound. Distal baculum cartilagi- nous and tridigitate; proximal baculum with a stout straight shaft and laterally flared base. Stomach unilocular and hemiglandular, with a moderately deep incisura angularis (extending about one-third the depth of the corpus and antrum), with rather short but thick bordering fold, and an expanded cor- pus with a moderately small antrum (terms from Carleton 1973). Chromosome preparations are available from nine males and five females (see spec- imens examined above). The diploid num- ber is 50 and the fundamental number is 68. The karyotype (Fig. 7A) is comprised of an acrocentric X-chromosome, the largest element of the complement, a small acro- centric Y-chromosome, and 24 pairs of au- tosomes of the following size and mor- phology: 2 pairs of large subtelocentric VOLUME 108, NUMBER 2 66 a5 6@ ar 2a ©§@ @2 422 6H AM BA e & i: A Mh AO AG NA AD OA OH A HA HB HA A 331 Bs REAR MA RA An ce ce Hy >< oath, >< “ ce 2 He 2 aa RX RY KR AA 86 GR we é e@ CA 2A 40 4% @e:>ee% * B Fig. 7. melanops (USNM 513583; male, 2N = 60, FN = 78). chromosomes, 8 pairs of meta-submetacen- tric chromosomes grading in size from me- dium to small, and one large and 13 me- dium to small pairs of acrocentric chro- mosomes. Comparisons.—From S. melanops (see Table 1 and Figs. 2 and 4), S. juruaense averages larger in all cranial dimensions ex- cept maxillary tooth row length and zygo- matic breadth; and it differs by virtue of orthodont versus proodont upper incisors; longer and distally tapering rostrum; less expanded nasolacrymal capsules when viewed from above; narrow, rather straight, and nearly parallel zygomatic arches en- closing a narrow and elongated orbital opening; occluded as opposed to open subs- quamosal fenestra; more gracile mandible with longer, more curved, and narrower coronoid process; usually present but min- ute hypothenar pad on sole of hind foot; and 2N = 50 rather than 2N = 60 karyotype. Scolomys juruaense is also larger in all cra- nial dimensions (except maxillary tooth row length, least interorbital constriction, and braincase breadth) than S. ucayalensis, al- Karyotypes of (A) S. juruaense, new species (MVZ 183172; female, 2N = 50, FN = 68) and (B) S. though proportionally their skulls are more similar to one another than either is to S. melanops (ratio diagram, Fig. 8). These two species can be distinguished by the follow- ing combination of characters: pale reddish- brown to reddish-black dorsal coloration as opposed to uniformly dark gray to brown- ish-black (Pacheco 1991), longer hind foot, longer ear, greater breadth across the zy- gomatic arches, longer nasals, longer dia- stema, “‘stepped”’ lateral margins of the in- cisive foramen, and squared (rather than distinctly rounded) anterior margin of the mesopterygoid fossa. The karyotype of S. ucayalensis is not known. Little comparison can be made in relation to the occlusal surfaces of the cheek teeth. These details are not given in the original description of S. ucayalensis and the teeth are too worn in most available specimens of both S. melanops and S. juruaense (Fig. 5). Similarly, no comparison can be made in soft anatomical structures, as none have been described for species other than what we provide here for S. juruaense. The karyo- type of S. melanops (Fig. 7B) is grossly sim- 332 Relative Character Means 0.85 0.90 0.95 1.00 1.05 1.10 i111) = ele. ce. =e iy ee ". bd ba of oe? e eo? e o? o* - Pd a “ “melanops g R ‘by juruaense ucayalensis AW Fig. 8. Ratio diagram comparing averages of ex- ternal and cranial dimensions for Scolomys ucayalensis (solid circles) and S. juruaense (open squares) in re- lation to those of S. melanops (vertical line). ilar to that of S. juruaense (Fig. 7A), with a diploid number of 60 and fundamental number of 78. Its karyotype is composed of one pair of large subtelocentric, 9 pairs of medium-sized to minute meta-and sub- metacentric, and 19 pairs of acrocentric au- tosomes grading evenly from medium-sized to small. The X is a large submetacentric chromosome, the largest element of the complement, and the Y is a subtelocentric element as large as the autosomal pair. The phenotypic similarity (especially or- thodont incisors, distally tapering rostrum, less-flared and parallel zygomatic arches, long and narrow orbital openings, and more elongated braincase) of S. juruaense and S. ucayalensis suggest a closer genetic rela- tionship between them relative to S. me- lanops. This hypothesis remains to be test- ed, however. Habitat. —All specimens of SS. juruaense were obtained in primary terra firme (up- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON land, non-seasonally flooded) forest in Sher- man live traps placed on the ground and baited with a combination of ground whole oats, raisins, and peanuts. As all individuals were live-trapped, stomach contents con- sisted only of the bait used (ground whole oats, raisins, and peanuts) and a few insect parts which may have been ingested acci- dentally. An equal number of Sherman traps placed in the trees at heights of 10-15 me- ters failed to secure any specimens of this species, although species of the sigmodon- tine genera Oecomys and Rhipidomys were taken. Other sigmodontines found sympat- ric with S. juruaense include Oryzomys cap- ito, Oryzomys yunganus, Oryzomys mac- connelli, Neacomys spinosus, Neacomys sp. (a smaller-bodied form sympatric with N. spinosus, with proportionally longer tail, shorter tooth row, and 2N = 35/36 as op- posed to 2N = 64 karyotype), Nectomys Squamipes, Oecomys trinitatus, Oecomys roberti, Oecomys superans, Oecomys bicol- or, and Rhipidomys leucodactylus. Reproduction. —Scolomys juruaense ap- parently breeds in both wet and dry seasons as pregnant or perforate females were taken in March (wet season), and in August, Sep- tember, and October (dry season). Two of the three female specimens of S. melanops from the type locality were pregnant when collected in March. Litter sizes (maximum number of embryos 3 in both melanops [Anthony 1924] and juruaense [range 1-3]) are somewhat low in comparison to other sympatric small-bodied oryzomyines, no- tably Oligoryzomys microtis (modal litter size 5, range 2-8) and Neacomys spinosus (modal litter size 3, range 2-4). Phylogenetic Affinities Voss & Carleton (1993) provided a phy- logenetic diagnosis of the Oryzomyini Vo- rontsov, 1959, defining the tribe by the fol- lowing combination of characters: (1) a pec- toral pair of mammae (with mammary counts of eight or more); (2) a long palate VOLUME 108, NUMBER 2 with prominent posterolateral pits; (3) no alisphenoid strut separating the buccinator- masticatory and accessory oval foramina; (4) no posterior suspensory process of the squamosal attached to the tegmen tympani; and (5) no gall bladder. They included one extinct and 14 Recent genera within this diagnosed unit: Scolomys along with Hol- ochilus, Lundomys, +Megalomys, Melano- mys, Microryzomys, Neacomys, Nectomys, Nesoryzomys, Oecomys, Oligoryzomys, Oryzomys, Pseudoryzomys, Sigmodonto- mys, and Zygodontomys. While the place- ment of Scolomys within the Oryzomyini has never been challenged, to our knowl- edge, this genus possesses only four of these five diagnostic characters. Among oryzo- myines, Scolomys apparently is unique with a reduced mammary count, lacking both pectoral and postaxial pairs. No hypothesis of phylogenetic relation- ships among the member genera of the Ory- zomyini has as yet been proposed and a sufficient understanding of character vari- ation in the soft and hard anatomy within the tribe is, at present, too limited for such to be developed here. Scolomys and Nea- comys are the only neotropical sigmodon- tines with strongly spinose fur, and they are superficially similar in overall small size, relatively short and naked-appearing tail, and apparently strictly terrestrial habits as well. This resemblance is apparently not due to immediate common ancestry, however. For example, we could score 21 of the 25 characters listed by Voss & Carleton (1993: 23-27) in both genera. Of these, the two differ in mammary count (6 in Scolomys, the presumptive ancestral state of 8 in Nea- comys), carotid circulation (Scolomys has the derived Pattern 3, Neacomys the ances- tral Pattern 1), anteroloph on M1 (present [=ancestral state] in Neacomys, confluent with anterolabial conule in Scolomys), and protoflexus of M2 (absent in Scolomys, present [=ancestral state] in Neacomys). The two genera do share 17 characters in com- mon, but all are apparently shared-primi- 333 tive traits either for sigmodontines as a whole (11 characters; numbers 1, 2, 3, 4, 6, 9, 13, 14, 15, 16, and 24 of Voss & Carleton 1993) or for oryzomyines in particular (6 are shared broadly by other genera in the tribe; characters 7, 8, 10, 12, 21, and 25). Scolomys does share the derived carotid cir- culation pattern with Oligoryzomys (Carle- ton & Musser 1989), some but not all Ory- zomys (Gardner & Patton 1976), Nectomys, and Holochilus (Voss & Carleton 1993), which might suggest a relationship among this group of genera. As a means to assess relationships within the Oryzomyini, we examined the sequence of 801 base pairs (267 codons) of the mi- tochondrial cytochrome-b gene for 17 spe- cies of seven oryzomyine genera (Microry- zomys minutus; Neacomys spinosus and two undetermined species; Nectomys squam- ipes; Oecomys bicolor, roberti, superans, trinitatus, and an undetermined species; Oligoryzomys longicaudatus and microtis; Oryzomys capito, macconnelli, nitidus, and yunganus; and Scolomys juruaense). The thomasomyine taxa Thomasomys aureus and Rhipidomys leucodactylus were used as out-groups. Methods for DNA extraction, amplification by the polymerase chain re- action (PCR), and sequencing, as well as oligonucleotide primers used in the PCR reactions follow those given in Smith & Pat- ton (1993). All sequences are available in GenBank; those for Nectomys squamipes, Oligoryzomys longicaudatus, Oryzomys capito, and the two thomasomyine out- groups were presented in Smith & Patton (1993). Table 2 provides a matrix of average sequence divergence distances, corrected for multiple replacements (Brown et al. 1982), and of the average number of transversions at the 3rd position of codons. Distances be- tween these genera of oryzomyines are sub- stantial, averaging 33.99% in corrected se- quence divergence (range 25.6% [Oecomys versus Oryzomys] to 40.8% [Microryzomys versus Scolomys]). Moreover, Scolomys 1s consistently the most divergent, with an av- 334 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 2.— Above the diagonal: average pair-wise divergence estimates of 801 base pairs of the mitochondrial cytochrome-b gene among seven genera of oryzomyine rodents, and between them and two thomasomyine out- group taxa, corrected for multiple hits by the method of Brown et al. (1982). Below the diagonal: average number of 3rd position transversions (with estimates of times of divergence, in millions of years, in parentheses). Averages for both sequence divergence and numbers of 3rd position transversions among species within a given genus are given on the diagonal. Tho- Scolo- as- _Rhipia- Taxon® Microryzomys Neacomys Nectomys Oecomys Oligoryzomys Oryzomys mys omys omys Microryzomys — 29.6 28.5 30.3 34.0 36.4 40.8 40.5 44.4 Neacomys 46.0(7.5) 22.4/25.0 31.2 35.6 35.4 37.3 38.5 43.4 44.3 Nectomys 36.3 (5.9) 39.0(6.4) — 31.1 33.6 30.7 36.9 43.7 40.4 Oecomys 34.5 (5.6) 41.2(6.7) 38.4(6.3) 14.8/18.0 32.4 DELO) 38.5 37.7 42.3 Oligoryzomys 45.0(7.3) 44.8(7.3) 42.5 (6.9) 37.8(6.2) 17.5/20.0 30.0 Syke) SIA. 50 Oryzomys 43.3(7.1) 44.5(7.3) 39.0(6.4) 31.1(5.1) 34.9(5.7) 24.9/30.0 38.0 36.9 41.2 Scolomys 35.0 (5.7) 43.3(7.1) 45.0(7.3) 44.6(7.3) 44.5(7.3) 46.0 (7.5) — 48.6 41.7 a Microryzomys minutus (Peru: MVZ 173957), Neacomys spinosus (Brazil: MNFS 1262), Neacomys sp. 1 (Brazil: MNFS 1395), Neacomys sp. 2 (Brazil: JLP 15365), Nectomys squamipes (Peru: MVZ 166700), Oecomys bicolor (Brazil: MNFS 1499), Oecomys sp. (Brazil: J354), Oecomys roberti (Brazil: JLP 15241), Oecomys superans (Brazil: JLP 15517), Oecomys trinitatus (Brazil: MNFS 1250), Oligoryzomys longicaudatus (Argentina: MVZ 155842), Oligoryzomys microtis (Brazil: MNFS 1321), Oryzomys capito (Peru: MVZ 166676), Oryzomys mac- connelli (Brazil: MNFS 156), Oryzomys nitidus (Brazil: MNFS 1419), Oryzomys yunganus (Brazil: MNFS 1101), Scolomys juruaense (holotype, MPEG 23824), Thomasomys aureus (Peru: MVZ 170076), and Rhipidomys leucodactylus (Peru: MVZ 168938). erage of 38.67% divergence in all pair-wise comparisons. The mtDNA sequences were analyzed both by the minimum evolution tree esti- mate (using the METREE version 1.2 pro- gram; Rzhetsky & Nei 1992), based on Ki- mura 2-parameter molecular distance ma- trices (Kimura 1980), and maximum par- simony (using PAUP 3.1.1; Swofford 1993). While both approaches provide strong sup- port for the monophyly of the oryzomyine genera examined (at a confidence limit val- ue of 99% in the distance phenogram [Fig. 9A] and a bootstrap value of 95% in the parsimony cladogram [Fig. 9B]) relative to the thomasomyine out-group genera, nei- ther view provides much resolution among the oryzomyine genera or in the specific placement of Scolomys among them. All terminal branches are very long and inter- nodal distances are short. As a result, nearly all internodes linking the oryzomyine gen- era have either confidence limits (Fig. 9A) or bootstrap values (Fig. 9B) below 50%. However, with the exception of Oryzomys, both measures are above 90% in the linkage of species within polytypic genera. Scolo- mys appears as a weakly supported sister taxon to a clade composed of Nectomys, Oryzomys, and Oecomys (with a confidence of only 46%) in the distance tree and to Nectomys in the parsimony tree (but at a bootstrap value <50%). While the mtDNA sequence data do not provide strong support for relationships among this group of oryzomyine genera, these data are significant for two reasons. For one, the very short internodal distances suggest that divergence among taxa was nearly simultaneous and that, as a conse- quence, resolution of relationships by any set of characters is likely to be difficult. It may not be surpising, therefore, that Sco- lomys combines a few uniquely derived morphological characters with others that are apparently primitive for the tribe. This general lack of resolution is opposite to that observed among members of the tribe Ako- dontini of the South American sigmodon- tines, based on variation over the same se- quence of cytochrome-b (Smith & Patton 1993). Consequently, difficulties in resolv- VOLUME 108, NUMBER 2 335 84 Oecomys bicolor 52 Oecomys sp. 99 Oecomys roberti Oecomys superans 90 89 se Oecomys trinitatus Oryzomys capito Oryzomys yunganus Oryzomys macconnelli Oryzomys nitidus Nectomys squamipes Scolomys juruaense 99 | | a : Oligoryzomys longicaudatus : Oligoryzomys microtis Microryzomys minutus Neacomys spinosus Neacomys sp. 1 95° * : Neacomys sp. 2 95 Thomasomys aureus A : — Rhipidomys leucodactylus s 25 . 10 Oecomys bicolor 29 Oecomys sp. 9 Oecomys roberti Oecomys superans 14 38 us Oecomys trinitatus Oryzomys yunganus 20 =i Oryzomys macconnelli 49 7 20 24 | Oryzomys nitidus | 46 aes Oryzomys capito 15 | : a Nectomys squamipes = | 94 ‘ | 96 Scolomys juruaense 95 | = Oligoryzomys longicaudatus : Oligoryzomys microtis 3 43 92 943 Neacomys sp. 1 2g ; 17 46 | 55 ; ' 48 Microryzomys minutus \ Si Thomasomys aureus B a Rhipidomys leucodactylus Neacomys sp. 2 Neacomys spinosus Fig.9. Hypotheses of phyletic relationship between exemplar species of seven genera of the tribe Oryzomyini, with the thomasomyine genera Thomasomys and Rhipidomys as out-groups, based on 801 base pairs of the mitochondrial cytochrome-b gene. (A) Minimum evolution tree based on a Kimura 2-parameter distance matrix (METREE version 1.2; Rzhetsky & Nei 1992); branch lengths are proportional and circled numbers at each node are confidence limits. (B) Maximum parsimony cladogram, excluding 3rd position transitions (PAUP version 3.1.1; Swofford 1993); the number of character changes along each branch is indicated and circled numbers at specific nodes are bootstrap values (based on 500 iterations) above 50%. 336 ing relationships among oryzomyine genera are likely to reflect true patterns and timing of diversification rather than inadequacies of the sequence used in comparisons. Sec- ondly, the long terminal branches and over- all extensive degree of sequence divergence suggest that divergence times within the oryzomyines are deep. Based on the rate estimate of 2.3% per million years for third position transversions of the cytochrome-b gene (Table 2; rate calculation from Smith & Patton [1993] for the Akodontin1), times of divergence among the examined genera of oryzomyines average about 6.6 million years (range 5.1 to 7.5). These numbers are in general accordance with estimates of times of divergence for the Sigmodontinae as a monophyletic lineage, based on DNA-DNA hybridization analyses (reviewed in Catze- flis et al. 1993). However, they also suggest that divergences within the oryzomyines were nearly simultaneous with the diver- gence between members of that tribe and at least the Akodontini (Catzeflis et al. 1993: their fig. 12.4), if not for other sigmodon- tines as well. Acknowledgments Specimens were collected as part of the Projeto Rio Jurua, sponsored by the Con- selho Nacional de Desenvolvimento Cien- tifico e Tecnologico (CNPq) and Instituto Brasileiro do Meio Ambiente e dos Recur- sos Naturais Renovaveis (IBAMA) and fi- nanced by grants from the Wildlife Con- servation Society, the National Geographic Society, and the Museum of Vertebrate Zo- ology. We are grateful to J. M. Ayres, who served as project coordinator; A. L. Gard- ner, who shared information on S. melan- ops, including its karyotype; G. G. Musser and M. D. Carleton, who permitted access to specimens in their respective collections; S. Marques for assistance in cataloging spec- imens into the MPEG collection; V. Pache- co for examining S. juruaense and for com- menting upon the manuscript; K. Klitz for PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON preparation of some figures; our colleagues of the Jurua project for field aid; and the people of the Rio Jurua for their generous hospitality. The editorial comments of R. S. Voss, A. L. Gardner, and an anonymous reviewer greatly improved the paper, al- though none should be held responsible for its final contents. MNFS was supported by a fellowship from CNPq. Literature Cited Albuja, L. 1991. Mamiferos.—Politecnica 16:163— 203. Anthony, H. E. 1924. Preliminary report on Ecu- adorean mammals. No. 6.—American Museum Novitates 139:9 pp. Brown, W. M., E. M. Prager, A. Wang, & A. C. Wilson. 1982. Mitochondrial DNA sequences of pri- mates: tempo and mode of evolution. — Journal of Molecular Evolution 18:225-239. Catzeflis, F. M., A. W. Dickerman, J. Michaux, & J. A. W. Kirsch. 1993. DNA hybridization and rodent phylogeny. Pp. 159-172 in S. F. Szalay, M. J. Novacek, and M. C. McKenna, eds. Mam- mal phylogeny. Placentals. Springer-Verlag, New York. Carleton, M. D. 1973. A survey of gross stomach morphology in New World Cricetinae (Roden- tia, Muroidea), with comments on functional interpretations. — Miscellaneous Publications, Museum of Zoology, University of Michigan 146:43 pp. Carleton, M. D., & G. G. Musser. 1984. Muroid rodents. Pp. 289-379 in S. Anderson and J. K. Jones, Jr., eds. Orders and families of Recent mammals of the world. J. Wiley and Sons, New York. , and . 1989. Systematic studies of ory- zomyine rodents (Muridae: Sigmodontinae): a synopsis of Microryzomys.—Bulletin of the American Museum of Natural History 191:2- 83. Gardner, A. L., & J. L. Patton. 1976. Karyotypic variation in oryzomyine rodents (Cricetinae) with comments on chromosomal evolution in the Neotropical cricetine complex. — Occasional Papers, Museum of Zoology, Louisiana State University 49:1-48. Hershkovitz, P. 1962. Evolution of Neotropical cri- cetine rodents (Muridae) with special reference to the phyllotine group. — Fieldiana: Zoology 46: 1-524. Kimura, M. 1980. A simple method for estimating evolutionary rates of base substitution through VOLUME 108, NUMBER 2 comparative studies of nucleotide sequences. — Journal of Molecular Evolution 16:111-120. Pacheco, V. 1991. A new species of Scolomys (Mur- idae: Sigmodontinae) from Peru.— Publica- ciones del Museo de Historia Natural, Univ- ersidad Nacional Mayor de San Marcos, Serie A. Zoologia 37:3 pp. Rzhetsky, A., & M. Nei. 1992. A simple method for estimating and testing minimum-evolution trees.— Molecular Biology and Evolution 9:945— 967. Ridgway, R. 1912. Color standards and color no- menclature. Washington, D.C., iv + 43 pp., 53 pls. Published privately. Smith, M. F., & J. L. Patton. 1993. The diversifi- cation of South American murid rodents: evi- dence from mitochondrial DNA sequence data for the akodontine tribe. — Biological Journal of the Linnean Society 50:149-177. 337 Swofford, D. L. 1993. Phylogenetic analysis using parsimony (PAUP), version 3.1.1. University of Illinois, Champaign, Illinois. Vorontsov, N. N. 1959. The system of hamster (Cri- cetinae) in the sphere of the world fauna and their phylogenetic relations [in Russian]. — Byulleten’ Moskovskovo Obshchestva Ispyta- telei Prirody, Otdel Biologischeskii 64: 134-137. Voss, R. S. 1988. Systematics and ecology of Ichth- yomyine rodents (Muroidea): patterns of mor- phological evolution in a small adaptive radi- ation.— Bulletin of the American Museum of Natural History 188(2):259-493. Voss, R. S., & M. D. Carleton. 1993. A new genus for Hesperomys molitor Winge and Holochilus magnus Hershkovitz (Mammalia, Muridae) with an analysis of its phylogenetic relationships. — American Museum Novitates 3085:39 pp. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(2):338-369. 1995. Systematic studies of oryzomyine rodents (Muridae: Sigmodontinae): definition and distribution of Oligoryzomys vegetus (Bangs, 1902) Michael D. Carleton and Guy G. Musser (MDC) Divison of Mammals, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, U.S.A.; (GGM) Department of Mammalogy, American Museum of Natural History, New York City, New York 10024, U.S.A. Abstract. — Morphological and distributional evidence is presented to doc- ument the specific stature of Oligoryzomys vegetus (Bangs, 1902) and to clarify its identification with respect to Central American populations of O. fulvescens. The geographic range of O. vegetus covers not only the mountains of western Panama but also the cordilleras de Tilaran, Central, and Talamanca of Costa Rica, generally above 1000 meters elevation and within lower montane and montane biotic zones. Within southern Central America, populations of O. fulvescens usually occur from sea level to 1000 meters in wet tropical and subtropical associations, but numerous instances of sympatry with O. vegetus are recorded in the 1000-1500 meter zone. Two species-group taxa, both de- scribed from Costa Rica, are allocated to subjective junior synonymy under O. vegetus, namely Oryzomys fulvescens creper Goodwin, 1945 and Oryzomys fulvescens reventazoni Goodwin, 1945. With improved understanding of its © taxonomy and geographic range, Oligoryzomys vegetus emerges as another species of small terrestrial mammal restricted to the highlands of Costa Rica and western Panama, a region which has formed a modest center for mam- malian endemism in southern Central America. In 1902, Outram Bangs reported two spe- cies of Oligoryzomys occurring together at Boquete and its vicinity, on the southeast- ern slopes of Volcan de Chiriqui between 3800 and 4800 ft in western Panama. Five specimens, ranging in age from young to old adults, were identified as Allen’s (1893) Oryzomys costaricensis; 13 others, howev- er, about the same in age span, were larger- bodied with darker fur, bigger hindfeet, and longer, monocolored tails. The latter Bangs (1902) described as a new species, Oryzo- mys (Oligoryzomys) vegetus. To Bangs, the collection of Oligoryzomys from Volcan de Chiriqui was clearly separable into two spe- cies, and the comparative data he listed sup- port this view. Two years later, Allen (1904a) examined seven additional specimens from Boquete, collected by J. H. Batty in 1901, and com- pared them with Bang’s original material. Allen (1904a:69) at first did not appreciate the distinctions between the forms Bangs had identified as costaricensis and vegetus: “The type and 12 topotypes of O. vegetus, kindly sent me for examination by Mr. Bangs, do not differ appreciably from the type, 3 topotypes, and additional Costa Rica specimens of O. costaricensis. They also agree with the 7 Boquete specimens col- lected by Mr. Batty, which I unhesitatingly refer to O. costaricensis.”’ In a brief com- mentary on previously described forms of Oligoryzomys, Allen later stated (1916:526): VOLUME 108, NUMBER 2 “The Mexican and Central American forms!’ [His footnote listed fulvescens Saussure, 1860, costaricensis Allen, 1893, vegetus Bangs, 1902, and nicaraguae Allen, 1910] that have been recognized are probably only geographic races of O. fulvescens (Saussure). A reexamination of vegetus, of Chiriqui, proves it to be quite easily separable from either fulvescens or costaricensis.”’ The con- text of Allen’s remarks suggests that he had reappraised vegetus and would consider it valid at least as a subspecies, a viewpoint thereafter formalized by Goldman (1918). Goldman (1918), in his revision of North American Oryzomys, concurred with AI- len’s (1904a) original doubt about the spe- cific separation of vegetus from costaricen- sis. Nevertheless, he considered the Boquete form sufficiently distinct to retain it as a subspecies of Oryzomys fulvescens, and he did likewise for costaricensis. Goldman’s formal action considerably expanded the morphological definition of Oryzomys (Oli- goryzomys) fulvescens to embrace popula- tions distributed geographically from north- eastern Mexico, throughout Middle Amer- ica, to eastern Panama, and ranging altitu- dinally from near sea level to over 3000 m (Fig. 1). The trinomial recognition of ve- getus has been subsequently observed both in regional geographic treatments (Gold- man 1920, Goodwin 1946, Handley 1966) and in authoritative classifications of North American Mammalia (Miller 1924, Miller & Kellogg 1955, Hall & Kelson 1959, Hall 1981). The experience and observations of other museum and field workers have cast some doubt on Goldman’s (1918) interpretation of the relationship and status of Oligory- zomys populations in western Panama. In the Field Museum of Natural History, a penciled note—in the script of Wilfred H. Osgood (fide P. Hershkovitz and B. Patter- son), dated 7 November 1932, and found in a tray of Panamanian Oligoryzomys— reads: ““Goldman makes vegetus a subspe- cies of fulvescens, but its larger skull and 339 longer tail, darker coloration etc. seem to make this doubtful.”’ In the Academy of Natural Sciences, Philadelphia, Robert K. Enders deposited large series of Oligoryzo- mys that he collected from the Chiriqui re- gion of Panama in the 1930s. The extensive erasing and overwriting of identifications (fulvescens costaricensis or f. vegetus) pen- ciled on the skin tags suggest the confusion Enders encountered in trying to reconcile the variation in his samples with Goldman’s taxonomy. For example, at Rio Gariche, around 1600 m, Enders identified the two ““subspecies’” costaricensis and vegetus as occurring at the same locality (ANSP 18408- 9). In like manner, recent samples taken from western Panama have impressed field workers of the morphological distinction and sympatric overlap of vegetus and fulvescens costaricensis (USNM series from NE San Felix, collected by Ronald H. Pine in June— July, 1980). Such examples, together with our earlier restudy of the critical series in the Museum of Comparative Zoology, have led us to agree with Bangs’s (1902) original assessment of Oligoryzomys in western Panama (Carleton & Musser 1989, Musser & Carleton 1993). Bangs presented data from what are clearly samples of two species of Oligoryzomys, O. fulvescens costaricensis and O. vegetus. The purposes of this report are: 1) to consolidate the evidentiary basis for the specific rec- ognition of O. vegetus (Bangs, 1902); 2) to more fully document the geographical and altitudinal distribution of O. vegetus based on larger samples of museum specimens; and 3) to attribute Oryzomys fulvescens re- ventazoni Goodwin, 1945 and Oryzomys fulvescens creper Goodwin, 1945 as subjec- tive junior synonyms of Oligoryzomys ve- getus. Materials and Methods Specimens reported herein consist prin- cipally of skins with their associated skulls from the following institutions and muse- 340 Oryzomys fulvescens (After Hall, 1981) Fig. 1. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Central American distribution of Oryzomys (Oligoryzomys) fulvescens as set forth by the revision of Edward A. Goldman (1918). Geographic races included (adapted from Hall 1981): 1) O. f/ engraciae Osgood, 1945; 2) O. f. fulvescens (Saussure, 1860); 3) O. f. lenis Goldman, 1915; 4) O. f. pacificus Hooper, 1952; 5) O. f. mayensis Goldman, 1918; 6) O. f. nicaraguae Allen, 1910; 7) O. f. creper Goodwin, 1945; 8) O. f; reventazoni Goodwin, 1945; 9) O. f. vegetus Bangs, 1902; 10) O. f. costaricensis Allen, 1893. ums: Academy of Natural Sciences, Phila- delphia (ANSP); American Museum of Nat- ural History, New York City (AMNBH); Field Museum of Natural History, Chicago (FMNH); Louisiana State University Mu- seum of Zoology, Baton Rouge (LSUMZ), Museum of Comparative Zoology, Harvard University, Cambridge (MCZ); Museum of Natural History, University of Kansas, Lawrence (KU); University of Michigan Museum of Zoology, Ann Arbor (UMMZ); National Museum of Natural History, Smithsonian Institution, Washington, D.C. (USNM). Type specimens and original se- ries of costaricensis Allen (1893), creper Goodwin (1945), nicaraguae Allen (1910), pacificus Hooper (1952), reventazoni Good- win (1945), and vegetus Bangs (1902) were studied firsthand. Approximately 800 museum specimens of Oligoryzomys were examined (see full lo- cality information in Taxonomic Summa- VOLUME 108, NUMBER 2 ry), though only some 550 were measured, of which still smaller geographically cohe- sive subsets were identified for the various numerical analyses. Relative age was coarsely indexed by degree of molar wear to the four age-classes (juvenile, young-, full-, and old-adult) recognized by Carleton & Musser (1989). Emphasis was placed on ac- cruing samples from Costa Rica and Pan- ama to illuminate the morphological differ- entiation and distribution of O. vegetus and O. fulvescens costaricensis, but representa- tives of certain northern subspecies— name- ly, O. f. fulvescens, O. f. nicaraguae, and O. f. pacificus (according to range limits as giv- en by Hall 1981)—were included for com- parative purposes. These operational tax- onomic units (OTUs), their sample sizes, and specific localities are recognized as fol- lows. Oligoryzomys vegetus CRI: n = 13, from Costa Rica, Limon, Valle El Silencio. n = 12, from Costa Rica, Alajuela, Villa Quesada and Tapesco. n = 17, from Costa Rica, Puntare- nas, Monteverde. n= 19, from Panama, Chiriqui, Bo- quete (type locality of vegetus Bangs, 1902). n = 35, from Panama, Chiriqui, lo- calities along the upper valley of the Rio Chiriqui Viejo (Boquete Trail, Casa Tilley, Cerro Punta and en- virons, Finca Martinz). n = 68, from Panama, Chiriqui, Siola. n= 12, from Panama, Chiriqui, Rio Chebo and Cerro Pando. CR6: CR7: PA4: PAS: PA6: PA8: Oligoryzomys fulvescens costaricensis CR2: n= 24, from Costa Rica, Puntare- nas, Canas Gordas. n= 5, from Costa Rica, Puntarenas, El General (type locality of costari- censis Allen, 1893). CR3: 341 CR4: n= 25, from Costa Rica, San José, San Geronimo de Pirris. n = 26, from Costa Rica, Cartago, localities along the upper valley of the Rio Reventazon (Cartago, Cer- vantes, El Muneco, Santa Teresa Peralta, Turrialba and environs). n = 16, from Panama, Panama, Cerro Azul. n = 19, from Panama, Cocle, El Valle and vicinity. n = 20, from Panama, Chiriqui, Colorado Camp. n = 28, from Panama, Chiriqui, Finca Santa Clara and Rio Santa Clara. CRS: PAIL: PA2?: PA3: PA?7: Oligoryzomys fulvescens fulvescens FUL: n=57, from Mexico, Veracruz, var- ious localities including the restrict- ed type locality (Orizaba) of fulves- cens (Saussure, 1860). Oligoryzomys fulvescens nicaraguae NI1: n= 23, from Nicaragua, northcen- tral highland localities (following Jones & Engstrom 1986). n = 7, from Nicaragua, southwest coastal localities (following Jones & Engstrom 1986). NI2: Oligoryzomys fulvescens pacificus PAC: n= 19, from Mexico, Chiapas, Ar- iaga, Maspastepec (type locality of pacificus Hooper, 1952), and Piji- jilapan. To assist verification of place-names and geographic coordinates used in preparation of the distribution map, we consulted gaz- etteers and-or maps provided in revisionary and faunal works, principally those in Fair- child & Handley (1966), Goldman (1920), Hooper (1952), McPherson (1985), and Slud (1964). Three external and 16 skull dimensions 342 (values in millimeters, mm) were recorded and analyzed to summarize patterns of vari- ation within and between the populations sampled. Total length (TOTL), tail length (TL), and hindfoot length (HFL) are those recorded by a collector on the skin label. Crania were viewed under a dissecting mi- croscope when measuring the 14 cranial and two dental variables to 0.01 mm using hand- held digital calipers accurate to 0.03 mm. These measurements, and their abbrevia- tions as used herein, include (see Fig. 2 for landmarks): occipitonasal length (ONL); zy- gomatic breadth (ZB); breadth of braincase (BBC); breadth across the exoccipital con- dyles (BOC); least interorbital breadth (IOB); length of rostrum (LR); postpalatal length (PPL); length of bony palate (BPL); length of upper diastema (LD); length of incisive foramen (LIF); maximum breadth across incisive foramina (BIF); breadth across bony palate (BBP); breadth of zygomatic plate (BZP); coronal length of maxillary toothrow (LM 1-3); coronal width of first upper molar (WM1); and length of auditory bulla (LAB). Anatomical terminology follows Carleton & Musser (1989) for general features of the oryzomyine skull. Univariate and multivariate computa- tions were restricted to specimens assigned to the three adult cohorts (young, full, and old). Standard descriptive statistics (mean, range, standard deviation) were derived for the OTUs. One- and two-way analyses of variance, discriminant functions, and prin- cipal component analyses were computed using the three external and 16 craniodental variables, all of which were first trans- formed to natural logarithms. Principal components were extracted from the vari- ance-covariance matrix, and loadings are expressed as Pearson product-moment cor- relation coefficients of the components with the original external and cranial variables. All analytic procedures were carried out us- ing Systat (Version 5.0, 1992), a series of statistical routines programmed for micro- computers. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Comparisons of Oligoryzomys fulvescens costaricensis (Allen, 1893) and O. vegetus (Bangs, 1902) External and craniodental variation. —In diagnosing vegetus, Bangs (1902) drew at- tention to its longer hindfoot and larger, heavier skull, especially with wider zygo- matic span, as compared to examples of cos- taricensis (then recognized as a species). The better sample sizes now available bolster Bangs’s impression of their salient contrast in size, with specimens of vegetus consis- tently averaging greater in most univariate comparisons (Appendix). Lengths of tail (TL) and hindfoot (HFL) provide key char- acters for first-approximation identification in the field, with TL usually exceeding 110 mm in adult vegetus (usually <110 mm ex- cept in oldest fu/vescens) and with HFL typ- ically 22 mm or greater in adult vegetus (usually =22 mm in fulvescens). Two constellations of points, which cor- respond to our independent identification of species, are apparent within the plane of the first two principal components extracted for all intact specimens of Oligoryzomys from Costa Rica and Panama (Fig. 3). The first two components together account for 68 percent of the variation in the original variable data (Table 1); whereas, the amount explained for components three through 19 is five percent or less for each, and bivariate plots of these disclose no meaningful dis- crimination of taxa. The 95% confidence ellipse for scores of vegetus on PC I versus PC II contains not only specimens from Panama but also those from many localities in Costa Rica. Moreover, each species el- lipse circumscribes many specimens taken at the same collecting locality—such as Bo- quete, Rio Chebo, and Finca Santa Clara in Panama, and El Muneco and Santa Teresa Peralta in Costa Rica (see Specimens Ex- amined). Although individuals from sym- patric localities are not identified in Fig. 3 due to visual congestion, certain specimens that do bear on our taxonomic interpreta- VOLUME 108, NUMBER 2 343 Fig. 2. Schematic views of the cranium (dorsal, ventral, lateral), right maxillary molar row, and incisive foramina of an adult Oryzomys alfaroi illustrating the limits of the 16 craniodental measurements recorded for specimens of Oligoryzomys (see Materials and Methods for abbreviations). 344 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON o fulvescens @e vegetus 0.0 0.1 PC | Fig. 3. Projection of individual specimen scores, based on log-transformed external (3) and craniodental (16) variables, onto the first two principal components for samples of Oligoryzomys fulvescens costaricensis (n = 115) and O. vegetus (n = 134) from Costa Rica and Panama with complete measurements (see Table 1). Lower- case letters correspond to holotypes of creper (c), reventazoni (r), and vegetus (v); numbered dots depict two paratypes (1-2) of Goodwin’s (1945) reventazoni, here reassigned to O. fulvescens, and topotypes (3-4) of Allen’s (1893) costaricensis. Ellipses circumscribe 95% confidence limits for specimen scores of each species; regression lines of PC II on PC I differ significantly between species in their Y-intercepts (P < 0.001) but not their slopes 0.2 0.3 0.4 OLS OS (P = 0.556). tions are indicated. These include the ho- lotypes of Oryzomys fulvescens creper and O. f. reventazoni, which cluster among sam- ples of vegetus from Costa Rica and Pana- ma, and two topotypes each of costaricensis and reventazoni, which fall among or nearer to examples of fulvescens. Mahalanobis dis- tances derived from two-group discrimi- nant function analysis associate the two ho- lotypes as vegetus and classify all four top- Ootypes as fulvescens (even outlier number two of reventazoni) according to their pos- terior probabilities of group membership. Even though juveniles were omitted from our multivariate analyses, postweaning growth undoubtedly accounts for much of the scatter observed within each species sample and for the oblique orientation of specimen scores on the first and second principal components. A similar pattern of multivariate dispersion has been recovered for other closely related, congeneric species of Muroidea (Voss & Marcus 1992), and its ontogenetic causality explicated using lab- oratory-raised animals in which parentage, age, sex, and equality of sample sizes have been rigorously controlled (Voss et al. 1990). Such influences on variability within and between samples, particularly balanced age and sex representation, are seldom achieved with the museum series normally accessible to an investigator. Proportion of the sexes among our samples is notably biased toward males (64% of all f costaricensis; 68% of all VOLUME 108, NUMBER 2 Table 1.—Results of principal component analysis and one-way ANOVAs performed on all intact speci- mens of Oligoryzomys fulvescens (n = 115) and Oli- goryzomys vegetus (n = 134) from Panama and Costa Rica (see Fig. 3). Correlations Variable PCI PC II Ff (species) TOTL 0.94 —0.08 145.2*** TL 0.92 —0.23 223:8°"" HFL 0.81 —=(),32 247.8*** ONL 0.92 0.17 121.3" ZB 0.86 0.04 | he Seta BBC 0.76 —().33 439. S*** BOC 0.73 =), 12 bia IOB 0.09 0.29 6.4** LR 0.81 0.23 63.5" PPL 0.82 O27 i i laa BPL 0.39 0.38 0.0 LD 0.83 0.13 16.37" LIF 0.25 0.69 19.1°** BIF 0.23 0.56 11,9°** BBP 0.78 —0.06 I25A°"* BZP 0.14 0.88 60.0*** LM1-3 =(.07 0.45 57.977" WMI! 0.33 0.13 a | LAB 0.66 —0.11 23.70" Eigenvalue 0.041 0.013 % Variance a2.0 16.1 ** — P < 0.01; *** = P < 0.001. vegetus), and the number of specimens in young-, full-, or old-adult age classes does vary from OTU to OTU. In the largest, most geographically and temporally homogeneous sample available to us (91 vegetus collected by R. K. Enders in Chiriqui, Panama), consistent age-related size differences are apparent, but classifi- cation by sex and sex-age interaction reveals negligible influence on mean differences (Table 2). Total length and tail length, as expected, increase among the three adult age-classes and yield highly significant mean differences, but hindfoot length does not. Among the 16 craniodental variables, the magnitude and significance levels of f-val- ues for age-class effects are typically highest for those dimensions measured on the facial region (ZB, LR, LD, LIF), intermediate for measurements across the neurocranium 345 Table 2.— Results of two-way 4NOV4As for 3 external and 16 cranial measurements of 91 specimens (63 4, 28 °) of Oligoryzomys vegetus from the valley of the Rio Chiriqui Viejo, Chiriqui, Panama. Jflinter- Variable S(Sex) KtAge) action) TOTL 1.8 129" 0.5 TL 0.5 $.5°°* 0.8 HFL 0.6 2.6 0.6 ONL 1.4 PSS el lead 0.2 ZB 1.3 250° °° 1.1 BBC 6.9"* 4.1* 0.3 BOC 0.3 $,3°* 0.5 IOB 0.7 2.4 0.7 LR 0.5 21,9" 0.0 PPL 0.0 Z1.27"* 0.2 BPL 0.2 my ia 1.0 LD 2.1 si." 0.7 LIF 3.0 12.9°** 0.0 BIF 0.2 $.6°** 1.0 BBP 1.4 19;5*** 0.2 BZP LZ 10,17** 0.1 LM 1-3 OZ 0.7 0.2 WMI 0.8 23 0.2 LAB eA LS 1.1 (BBC, BOC, IOB), and lowest on the molars (LM1-3, WM1). These results generally parallel the patterns of covariation and growth allometry of the muroid skull doc- umented for other species of Sigmodontinae (Voss & Marcus 1992). Although relative age may sometimes confuse the identifica- tion of individual specimens, the contri- bution of age-related effects is minor com- pared to the magnitude of interspecific dif- ferentiation between O. vegetus and O. ful- vescens, for example as observed in principal component analysis (Table 3). Table 3.—One-way 4NOVAs generated for various group effects on results of principal component analysis (see Fig. 3). Variable fisex) Alage) Aispecies) PC I scores 0.6 23.2°°° 183:5"** PC II scores 0.5 26.3°°" 140,7*** **k* — P< 0.001. 346 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON a pla nes N Fig. 4. Dorsal, ventral, and lateral views of adult crania and mandibles of: left, Oligoryzomys fulvescens costaricensis (USNM 541183; Panama, Chiriqui, 24 km NNE San Felix; ONL = 21.9 mm); and right, Oligo- ryzomys vegetus (USNM 541186; Panama, Bocas del Toro, 3.5 km E Escopeta; ONL = 23.5 mm). The overall larger body size and more robust skull of vegetus relative to fulvescens costaricensis are corroborated by results of the principal component ordination. The three external and most cranial measure- ments exhibit high and positive correlations with the first principal component, which suggests a general size factor (Table 1), and higher scores were generated on average for specimens of vegetus (X = 0.12) along this axis than for those of f costaricensis (X = —0.14). Dimensions that contribute to the separation along PC II emphasize the small- er body size (TL, HFL) and narrower skull (BBC, BOC) of f. costaricensis, but also re- veal its relatively wide interorbit and zy- gomatic plate IOB, BZP), longer and wider incisive foramina (LIF, BIF), and more ro- bust molars (LM1-3) as compared to ex- amples of vegetus (Fig. 4). The compara- tively strong loadings of LIF and BIF on PC II were unexpected, but closer examination of crania confirmed that the incisive foram- ina typically span about three-quarters of the diastema in f. costaricensis in contrast to about two-thirds in vegetus. In view of the small absolute size of the skull and toothrows in Oligoryzomys, the magnitude of mean differences between the two species is not conspicuous (Appendix), even for dimensions where population var- lances are low and f-ratios for species effects are high (Table 1). Thus, simple univariate range limits, such as employed in keys, may VOLUME 108, NUMBER 2 Cap 347 cap con cor Fig.): of: top, O. vegetus (USNM 541186); and bottom, Oligoryzomys fulvescens costaricensis (USNM 541183). Ab- breviations: cap, capsular process of the lower incisor; con, condyloid process; cor, coronoid process; sn, sigmoid notch. be uninformative for practical segregation of the species. Of the variables quantified, the larger size of vegetus is best appreciated in cranial length (ONL) and especially in certain width measurements (ZB, BBC, BBP). Differences in overall form and pro- portion, such as those reflected along PC II, are easier to grasp visually. Hence, adult specimens of vegetus typically exhibit a nar- rower interorbital constriction accentuated by laterally flaring zygoma, a slimmer zy- gomatic plate and hence a shallower zygo- matic notch, a broader braincase and longer rostrum, more delicate molars for the size of skull, and shorter-narrower incisive fo- ramina as compared to those of fulvescens (Fig. 4). Development of the capsular process, among the few qualitative osteological traits which may serve to separate closely related muroid species, provides some discrimi- nation of vegetus and fulvescens costaricen- sis. This process, which represents the pos- terior alveolar terminus of the lower incisor, arises from the lateral surface of the dentary. In specimens of vegetus, the capsular pro- cess forms a distinct knob that, with the Lateral and medial views of the mandibles illustrating typical development of the capsular process ascending ramus oriented in a horizontal plane, is typically observed to extend above the ventral rim of the sigmoid notch, wheth- er viewed from a lateral or medial aspect (Fig. 5). In examples of fu/vescens, this pro- cess is weakly pronounced, forming a lateral mound or bulge that usually ends below the ventral rim of the sigmoid notch. This distinction is not absolute, for some specimens (16%) of vegetus lack the strong dorsal projection of the capsular process and a minority (7%) of fulvescens possess one. In approximately 20% of both species, the process terminates about level with the sig- moid notch (Fig. 6). Expression of the cap- sular process may correlate with age of the animal. For all individuals of vegetus scored (n = 167), there exists a strong positive as- sociation between age-class (juvenile through old-adult) and dorsal projection (below, even, above) of the capsular process (Kruskal-Wallis statistic = 38.0; P < 0.001). That is, individuals with the capsular pro- cess below or even with the sigmoid notch are predominantly restricted to the juvenile and young-adult age groups, whereas a cap- sular process extending above the notch oc- 348 100 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ae | O. fulvescens (N=139) WU, O. vegetus (N=167) 80 60 40 Percent Occurrence 20 Below Ev Above Termination of Capsular Process Relative to Sigmoid Notch Fig. 6. Distribution of character state development of the capsular process in samples of Oligoryzomys fulvescens costaricensis and O. vegetus from Costa Rica and Panama (see Fig. 5 and text). curs almost exclusively in full- and old- adults. Enigmatically, the parallel relation- ship is not supported by the same nonpara- metric ANOVA for all fulvescens costaricen- sis (n = 139; Kruskal-Wallis statistic = 4.98; P = 0.18). As with the utility of other dis- tinguishing features employed at the species level, conformation of the capsular process must be considered in concert with other traits to render confident identification. Bangs (1902) also cited the color of the incisor enamel as a diagnostic trait of O. vegetus. The difference—that is, pale yel- lowish orange in f. costaricensis and medi- um orange in vegetus—does seem valid as an average impression, but the contrast is subtle and best appreciated when compar- ing series. Pelage texture and color.—The two spe- cies differ in fur texture and coloration, es- sentially evident as a harsh and pale coat contrasted to a softer and darker one. These qualitative distinctions, although fine, are easily appreciated in freshly collected ma- terial but are less readily grasped with ex- tremely old and faded, dust-covered mu- seum skins. The dorsum of O. fulvescens is pale tawny to reddish brown, flecked with bright buffy and dark hairs that convey a more streaked or variegated appearance (somewhat like Reithrodontomys fulvescens). The streaked VOLUME 108, NUMBER 2 visual impression results from admixture of the densely black guard hairs and tricolored overhairs, which have a pronounced middle buffy band accentuated by a narrow basal plumbeous band and a very short fuscous tip. Overall, the dorsum presents a bright but pale tone and the fur is somewhat harsh- er to the touch. Examples of O. vegetus possess more somber upperparts, evenly colored dark brown and lacking conspicuous streaking. In each overhair, the basal plumbeous band is discernably long relative to the middle buffy band, which is subdued, and the ter- minal tip is dusky. The resultant tone is appreciably darker than observed in typical O. fulvescens, and the fur is soft and fine rather than harsh. The nose, forehead, and cheeks are also darker in O. vegetus, and the eye ring dusky and better defined than is common in O. fulvescens. A light versus dark venter distinguishes the two species. The underparts of O. ful- vescens are pale, a whitish gray; the over- hairs may be entirely white to their base, especially on the throat and chest, or with a light gray basal band. The ventral fur of O. vegetus, on the other hand, is dark gray in appearance, the hairs possessing a well- defined basal plumbeous band and a clear or whitish tip. A buffy overwash may occur in some specimens of either species but is more commonly encountered among sam- ples of O. fulvescens. In like manner, both species have lateral ochraceous stripes but those on the sides and flanks of O. fulvescens are generally more prominent. Coloration of the feet does not offer any distinguishing trait. The tops of the tarsus and metatarsus are covered with hairs that range from white to translucent and basi- cally present a whitish appearance in both Oligoryzomys. In summary, qualitative features of the skin and skull, augmented by the spatial structure inherent in multivariate summa- ries of the 19 mensural characters, con- vincingly sustain Bangs’s (1902) original 349 perception of the existence of two morpho- logically similar but nonetheless distinct and separable species in western Panama. Inclu- sion of samples of Oligoryzomys from Costa Rica demonstrates the occurrence of the same two morphologies and the broader distribution of the species that we recognize as O. vegetus. Relationships of costaricensis and vegetus to Other Central American Oligoryzomys At the time Bangs (1902) reported his new species Oryzomys vegetus, previously named forms of Central American pygmy rice rats were also recognized as full species, namely fulvescens Saussure (1860) and costaricensis (Allen, 1893). In his revision of North American Oryzomys, Goldman (1918) es- tablished the precedent for treating all Mid- dle American populations of Oligoryzomys, which he ranked as subgenus, as members of the single polytypic species fu/vescens, and subsequent twentieth-century descriptive activity has transpired within his taxonomic framework (see subspecies and ranges, Fig. 1). In particular, systematists have routinely classified Panamanian populations of the shorter-tailed, pale-bellied form as /fulves- cens costaricensis (Goldman 1918, 1920; Hall 1981; Handley 1966). Set against this systematic background, we naturally interpreted the morphological differentiation and sympatry of costaricen- sis and vegetus as indicative of the latter’s specific status and continued to view the former as a subspecies of fu/vescens (Carle- ton & Musser 1989, Musser & Carleton 1993). However, our casual presumption was somewhat disquieted by Joel A. Allen’s (1904a:69) provocative comment in a paper that we had initially overlooked. ‘“‘Mr. Bangs informs me [given as in litt.] that he inadvertently placed the name ve- getus on the dark form (= costaricensis vera) instead of on the light form, after 350 CV Il -5.0 -6.0 Bigs 7: PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 0.0 1.0 2.0 3.0 40 5.0 6.0 CV | Plots of first two canonical variates extracted from discrimination function analysis performed on five Central American taxa of Oligoryzomys and using all specimens with complete external and craniodental measurements (n = 332). Polygons enclose maximal dispersion of individual specimen scores around group centroids for each of five taxa (see Table 4): cos, O. fulvescens costaricensis (n = 115); ful, O. f. fulvescens (n = 47); nic, O. f- nicaraguae (n = 21); pac, O. f. pacificus (n = 15); veg, O. vegetus (n = 134). having returned the specimens of O. cos- taricensis borrowed from this Museum [that is, AMNH], so that vegetus = costaricensis and the light form was left unnamed [emphasis ours].”’ In other words, Bangs confessed to having inadvertently applied an extraneous name, vegetus, to the Boquete rodent that he con- sidered identical to Allen’s (1893) costari- censis, while the Boquete form that he had meant to describe as new has been, by de- fault, allocated to and subsequently known as fulvescens costaricensis (for example, Goldman 1920, Hall 1981, Handley 1966). Curiously, Goldman supplied no indication that he was aware of the supposed mix-up with regard to the naming of vegetus, either in his revision of Oryzomys (1918) or in his publication on the mammals of Panama (1920). Allen’s (1904a) allusion to an unfortunate lapsus by Mr. Bangs reopens the issue of which taxon, costaricensis or vegetus, if ei- ther, intergrades with northern populations of Oligoryzomys fulvescens, to which prob- lem we now turn. In assessing these rela- tionships, we included specimens from Ve- racruz, Mexico, as a comparative standard for Saussure’s (1860) fulvescens s. s. Rep- resentatives of other northern subspecies were selectively sampled: nicaraguae Allen VOLUME 108, NUMBER 2 (1910) because of its contiguous geographic position to costaricensis; and pacificus Hooper (1952) because of Jones & Engs- trom’s (1986) suggestion that it may link with more southerly-distributed, Pacific- coast populations of fu/vescens. Goldman (1918) formally recognized, aside from the nominate race, four subspe- cies of Oryzomys (Oligoryzomys) fulvescens: costaricensis Allen (1893) (with nicaraguae Allen, 1910, in full synonymy), /enis Gold- man (1915), mayensis Goldman (1918), and vegetus Bangs (1902). His remarks under the various subspecific accounts clearly affirm that he viewed costaricensis as most similar to the nominate form fulvescens. Our morphometric evaluations reinforce Goldman’s (1918) general estimation of phenetic resemblance and support a hy- pothesis of consanguinity among Central American races currently arranged under Oligoryzomys fulvescens (that is, excepting O. vegetus). Although much variation exists within and among geographic representa- tives of O. fulvescens, these subspecific taxa considerably overlap one another in mul- tivariate space, and all four are wholly set apart from specimens of O. vegetus (Fig. 7). In group assignments according to posterior probabilities of membership, specimens of the four subspecies were variously distrib- uted among one another, but none was mis- classified with vegetus, or vice versa. Clus- tering of OTU means also emphasizes the integrity of samples of O. vegetus as distinct from those representing the four geographic races of O. fulvescens (Fig. 8). The three external measurements and most cranial variables contribute more or less equally to the hiatus along the first ca- nonical variate and, by the sign and mag- nitude of their loadings (Table 4), under- score the fundamentally larger size of ve- getus aS contrasted to specimens of fu/ves- cens, including costaricensis. Certain cranial dimensions (BZP, IOB, LM1-3), as di- vulged in the PCA, depart from this pattern and appear relatively, in some OTUs ab- 351 Table 4.— Results of discriminant function analysis performed on all intact specimens of Oligoryzomys f. fulvescens (n = 47), O. f. costaricensis (n = 115), O. f. nicaraguae (n = 21), O. f. pacificus (n = 15), and Oli- goryzomys vegetus (n = 134) (see Fig. 7). Variable CV I CV Il fitaxon) TOTL =(0.63 0.41 62.0587 TL —0.69 0.44 89, 1*** HFL =0-73 0.18 Tigsee" ONL =.356 0.39 ple ik ZB —0.62 0.32 54.7" BBC =().31 0.36 139.4*** BOC =U 0.49 560.17" IOB 0.03 O75 Be A alls LR —0.48 0.33 CW ia PPL —0.37 0.28 20,5"** BPL —0.09 0.53 tis LD —0.48 0.29 i I aa LIF Oly 0.24 50°" BIF 0.16 —0.03 2 BBP —0.59 0.46 613°°" BZP 0.38 0.20 li3°°* LM1-3 0.29 0.64 34.8*** WMi1 —0.19 0.51 21 LAB —0.56 =0:07 ce ied Canonical correlations 0.93 0.73 * = P< 0.05: ** = P < 0.001. solutely, greater in certain examples of ful- vescens. For instance, Goldman (1918), in comparing costaricensis to fulvescens prop- er, cited the wider and longer molar rows of the former as its most distinctive char- acter and one that approached vegetus in size. Although the correctness of Goldman’s assessment does hold in simple univariate comparisons, a multivariate perspective demonstrates the proportionally large mo- lar rows possessed by individuals of costar- icensis, a Shape feature which clearly aligns it with other representatives of O. fulves- cens, not O. vegetus. Qualitative agreements too—like cranial shape and proportions and pelage color and texture—support the affiliation of costari- censis with fulvescens and other northern races. Termination of the incisor capsule below the sigmoid notch is the common 352 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 2 SISUEDIIL}SOO > on snjeBbea | ia ina aan (ia aaa (rman naan (Ri ain ohana (ES onan Tie aero er niin ae .08 07 06 05 04 03 02 01 0.0 Euclidean D Q BS) nn SISUBDIIE}JSOD > rs snjeBea 4 3 2 1 0.0 (SAG Fig. 8. UPGMaA clustering of log-transformed samples means for the 19 OTUs of Central American Oli- goryzomys identified in the Materials and Methods, using both a Euclidean distance coefficient (A) and the Pearson correlation coefficient (B). VOLUME 108, NUMBER 2 condition in f. fulvescens (56% of 36 spec- imens scored) and f. nicaraguae (67% of 33 specimens), but these percentages are lower than that derived for our larger sample of f. costaricensis (78% of 139 specimens). Thus, the development of the capsular pro- cess in the former populations, while in the direction of agreement with f costaricensis, is less decisive as a basis for discrimination from O. vegetus. In summary, both morphometric com- parisons and discrete character traits indi- cate the close kinship and probable conspe- cificity of costaricensis with northern O. ful- vescens. Although we advise the continued afhliation of costaricensis as a subspecies of O. fulvescens, we stress the need for stronger empirical analyses of relationship and status among populations of fulvescens (see dis- cussion below). Aside from such uncertain- ties involving O. fulvescens, and whether intended or not, Bangs (1902) did not err in selecting an individual of the “dark form” as the type of his new species Oligoryzomys vegetus. Distribution and Zoogeography In reinstating vegetus as a valid species of Oligoryzomys, Carleton & Musser (1989) at the time viewed its distribution as re- stricted to the highlands of western Panama, principally in the region of Volcan de Chi- riqui. Although zoogeographically plausi- ble, they, and later Musser & Carleton (1993), overlooked its presence in the con- tiguous mountain systems of Costa Rica. Such a distributional picture is confirmed by the renewed study of museum speci- mens, including evidence provided below for the junior synonymy of certain Oli- goryzomys —namely creper Goodwin (1945) and reventazoni Goodwin (1945)—de- scribed from the middle and upper high- lands of Costa Rica. The species Oligoryzomys vegetus, or sprightly pygmy rice rat as christened by Bangs (1902), inhabits the rugged, moun- tainous spine of central Costa Rica and 353 western Panama (Fig. 9). Specimens at hand document its geographic range from the northernmost extent of the Cordillera de Ti- laran (Monteverde and environs), through the cordilleras Central and Talamanca, to the Cordillera de Chiriqui (at least to the vicinity of Cerro Bollo). We have searched for examples of O. vegetus among other mu- seum series of O. fulvescens (in addition to those listed herein) from the highlands of Nicaragua, Guatemala, and southern Mex- ico but have found only fu/vescens proper in these regions. We therefore believe that our locality records circumscribe the prin- cipal geographic distribution of O. vegetus (Fig. 9), although it may be expected farther east in westcentral Panama and may be sought on the highest peaks of the Cordillera de Guanacaste in northern Costa Rica (but see below). The known altitudinal occurrence of Oli- goryzomys vegetus extends from 840 m (val- ley of the Rio Penas Blanca, Costa Rica) to over 3000 m (below the summits of Volcan Irazu and Cerro de la Muerte, Costa Rica). Most collecting localities lie between 1000 and 2000 m. Oligoryzomys vegetus does not penetrate the tropical lowlands of Costa Rica and Panama where O. fulvescens commonly occurs (Fig. 9). Within these countries, the latter species is found from near sea level to approximately 2000 m. Although most localities of O. f costaricensis occur from 750 m to 1500 m, as portrayed in Fig. 10, the low incidence of records below 750 m is biased by the infrequency of elevational notation by collectors (for example, the nu- merous lowland localities in eastern Pana- ma and northern Costa Rica); whereas, col- lecting stations in the middle and upper highlands more commonly bear, for what- ever reason, altitudinal information. Thus, histograms of altitudinal occurrence (Fig. 10) are dominated by the long history of mammalogical field work in the mountains of central Costa Rica and western Panama. These regions, not unexpectedly, also contain most locations of documented sym- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 354 "INOJUOD 19}9WI-QOO] 24} 0} Spuodse1109 sul] PI[Og “UloJoY pouruteXe susUIOOdS UO pase eUIeURY PUP BOTY &ISOD UI snjasaa ‘CE pur suarsaajn{ sdu0zZ41081.CQ JO UONNGLSIp o1yde1s0eH 6 “31 Asedwds ¥ SnNIeB3A @ SUsDSaAIN} O SAWOZAJOBIO oe VNOVYVOIN 3XxV1 > \ n ome BONG VOLUME 108, NUMBER 2 1000 Numbers of Collecting Localities 1000 355 O. fulvescens GN = 8S) 3000 O. vegetus (N = 66) 2000 Meters Above Sea Level Fig. 10. Frequency distributions illustrating occurrence of collecting localities with recorded elevations for samples of Oligoryzomys fulvescens and O. vegetus from Costa Rica and Panama. patry of O. vegetus and O. f. costaricensis, especially along the upper drainage basins of the Rio Reventazon, Costa Rica, and the Rio Chiriqui Viejo, Panama. In addition to Bangs’s (1902) original discernment of sym- patry at Boquete, Panama, we have en- countered examples of O. vegetus and O. f. costaricensis intermixed at several other sites, both in Costa Rica (El Muneco, Mon- teverde, Lajas) and in Panama (Rio Chebo, Rio Gariche, Finca Lerida, Finca Santa Cla- ra, Rio Santa Clara). Indeed, as argued be- low, Goodwin’s (1945) type series of Ory- zomys fulvescens reventazoni, from Santa Teresa Peralta, Costa Rica, is itself a com- posite, the holotype synonymous with O. vegetus and the four paratypes referrable to O. fulvescens. Almost all sites of sympatry or contiguous allopatry lie within the ele- vational zone from 1000 to 1500 m. In his insightful biogeographic study of Costa Rican rodents, McPherson (1985:224) characterized fulvescens as virtually ubiq- uitous within the country, occurring throughout most biotic zones, and having “geographical and elevational distributions ... greater than those of any other species in Costa Rica.’’ His geographical and eco- logical summation, like the earlier errone- ous definition of “Oryzomys fulvescens,”’ was a misleading composite based on data from two biological species. In fact, reassortment of McPherson’s, and our own, Costa Rican localities by correct specific identifications 356 reveals far narrower altitudinal and ecolog- ical occurrences of the two Oligoryzomys that correspond intelligibly with well-de- fined climatic, vegetational, and physio- graphic associations. Populations of O. ful- vescens costaricensis predominantly occur within Tropical and Premontane life zones (sensu Holdridge 1967); whereas, those of O. vegetus are principally confined to Lower Montane and Montane formations. Most instances of sympatry involve sites classi- fied as Premontane Wet Forest and Pre- montane Rain Forest, although overlap at lower (Tropical Wet Forest-Upper Transi- tion) and higher (Lower Montane Wet For- est) life zones does occur. The deeper, softer pelage of O. vegetus is consistent with its higher-elevation affinity and presumably confers greater insulative properties for a small rodent living in wet, cool environ- ments. Within these broadly characterized forest domains, pygmy rice rats prefer grassy or shrubby microhabitats placed within forest clearings or at woodland edges, often along small streams or near other aquatic settings. Specific habitat notations on skin tags com- monly reference “dense grass at edge of clearing near stream,” “‘second-growth scrub at edge of clearing,’ ““dense grass beside small stream in forest,’ and “under tall weeds and raspberry bushes.” In view of their proclivity for early successional stages, both species of Oligoryzomys seem to adapt readily to a variety of second-growth hab- itats and to prosper around active or aban- doned agricultural fields and pasturelands. We cannot glean suitably detailed infor- mation from field catalogs and skin labels that would suggest microhabitat segregation of the two Oligoryzomys at localities of sym- patry, perhaps because field workers were seldom aware that two species might be en- countered in their traplines. At one such place, El Muneco in Costa Rica (1 100-1200 m), Austin Smith trapped five specimens of Oligoryzomys, all initially identified as O. fulvescens costaricensis. Two of these prove PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON to be O. fulvescens, both caught in “‘Carib- bean rain forest in brush,” and three are O. vegetus, one each trapped in “Caribbean rain forest in sugar cane,” “pasture in cloud for- est zone,” and “Caribbean cloud forest.” Oligoryzomys, species indeterminate, has been described both as terrestrial (e.g., Han- dley 1966) and as semi-arboreal (e.g., Timm et al. 1989), and has been observed to climb nimbly upon small limbs and branches in dense undergrowth (McPherson 1985). The larger hindfeet and relatively longer tail of O. vegetus predict more highly developed scansorial abilities in this species. Beyond these natural history anecdotes and gener- alities, detailed field studies must be con- ducted to shed light on specific ecological requirements and habits of the two Oligo- ryzomys in areas of syntopy. Although the morphological similarity between Oligoryzomys vegetus and Central American O. fulvescens has engendered years of taxonomic confusion, the two may not prove to be sister species. In fact, the closest specific relative of O. vegetus may inhabit the northern Andes of South America. In his description, Bangs (1902:36) speculated that vegetus ““may prove to be only a north- ern subspecies of O. dryas humilior Thomas of Colombia, Bogota region to Santa Marta region.” The forms dryas and humilior, however, are junior synonyms of Microry- zomys minutus, a species of a very distinc- tive oryzomyine genus related to Oligory- zomys (see Carleton & Musser 1989). A more plausible candidate is suggested by Osgood (1912), who named a new pygmy Oryzomys, O. griseolus, from the Paramo de Tama in westernmost Venezuela. Os- good, who had available samples of geo- graphically contiguous Oligoryzomys like navus and tenuipes, thrice mentioned O. ve- getus of western Panama as bearing the greatest resemblance to his new species. The question of their phylogenetic stature to one another, like the need for improved delim- itation of species and phylogenetic system- atics of South American Oligoryzomys in VOLUME 108, NUMBER 2 general, still warrants much basic investi- gation (see Carleton & Musser 1989). How- ever, biogeographic consideration of other mammals, as well as other vertebrates, re- stricted to these Costa Rica-Panamanian mountains suggests that such a northern Andean connection is reasonable. The distribution of Oligoryzomys vegetus, as now understood, accords closely with a region of Middle American topography al- ready recognized by zoogeographers and variously labeled the Costa Rica-Chiriqui Highlands (Slud 1964), the Talamancan Province (Savage 1966), or the Cordilleran Faunal Area (McPherson 1985, 1986). (Middle America is here understood to in- clude Central America north of the Pana- manian isthmus and most of Mexico exclu- sive of the Mesa Central.) The middle- to high-elevation mountains in Costa Rica and western Panama, harboring cool and peren- nially moist forests, have figured promi- nently in the historical zoogeography and phyletic diversification of Central American amphibians and reptiles (Savage 1966, 1982) and birds (Slud 1964, Stiles 1983), verte- brate groups which contain a large number of species and genera endemic to the region. Small terrestrial mammals do not exhibit such high degrees of endemism as these oth- er vertebrates; nonetheless, the area has played an important role in their phyloge- netic history as well, especially as demon- strated for Rodentia by McPherson (1985, 1986). Many rodent species either reach their southern or northern limits of distribution in the Cordilleran Faunal Area or are hy- pothesized to have originated there in situ. Review and enhancement of Mc- Pherson’s (1985) taxonomic and distribu- tional data indicate that at least one genus (Syntheosciurus), 14 species, and two strongly differentiated subspecies of small mammal are limited to the mountains of Costa Rica and western Panama and their humid forest associations (Table 5). Most of these taxa occur in both the Chiriqui re- gion of western Panama and the Talaman- 357 Table 5.—Insectivora, Lagomorpha, and Rodentia endemic to highlands of Costa Rica and western Pan- ama. Cordilleras Tala- Til- Cen- man- Chiri- Taxon aran tral ca qui Soricidae Cryptotis endersi? C. gracilis” xX XxX C. nigrescens xX XX Leporidae Sylvilagus dicei4 Sciuridae Syntheosciurus brochus* x Geomyidae Orthogeomys cavator' O. heterodust x Heteromyidae Heteromys oresterus® Muridae Rheomys raptor hartmannt xX ? R. underwoodi xX Oryzomys devius 4 Oligoryzomys vegetus* xX XX Sigmodontomys aphrastus' Reithrodontomys b. brevirostris™ R. creper® xX R. rodriguezi° Scotinomys xerampelinus? ~ MK KH KH KR KY ~ AK KKK KK YK x Kh K KKK Xx Sources. —a: Choate (1970); b: Woodman & Timm (1992), N. Woodman (pers. comm.); c: Woodman & Timm (1993); d: Diersing (1981); e: Enders (1980); f: Hafner (1991); g: Rogers & Rogers (1992); h: Reid & Langtimm (1993), Voss (1988); 1: Voss (1988); j: Gard- ner (1983); k: This study; 1: Musser & Carleton (1993); m: Hooper (1952), McPherson (1985); n: Hooper (1952), Timm et al. (1989); o: McPherson (1985), Timm et al. (1989); p: Hooper (1972). cas of southcentral Costa Rica. Fewer spe- cies are known to occur throughout the four highland areas, but there is reason to doubt that they necessarily would. Geological ev- idence (summary in Castillo-Munoz 1983) supports the greater antiquity (Oligocene- Miocene) of the Talamanca-Chiriqui ranges as compared to the younger Central and T1- laran cordilleras (late Pliocene-Pleistocene). Some have attributed the high endemism to the possible isolation of the Talamanca- Chiriqui region as an island, or series of islands, within the Panamanian portal prior 358 to complete closure and late-Pliocene for- mation of the landbridge (McPherson 1985, 1986; Slud 1964; Stiles 1983). Vouchered evidence for the expected occurrence of some species within certain mountain ranges is yet lacking, an omission that prescribes the continuing need for basic faunal survey. Also lacking is recent systematic investigation of other mammalian groups with highly dis- junct Middle American distributions and distinctive geographic races limited to these southern highlands. Renewed revisionary focus on taxa such as Cryptotis parva, Pero- myscus mexicanus, Reithrodontomys mex- icanus, R. sumichrasti, and Scotinomys te- guina will likely augment the amount of en- demism thus far apparent among small mammals. Most of the 17 species-group taxa (Table 5) represent genera that are either wholly endemic to Middle America (Syntheosciu- rus, Orthogeomys, Heteromys, Rheomys, and Scotinomys), or genera whose species diversity persuasively implicates a Middle American origin and subsequent radiation into temperate North America and north- western South America (Cryptotis, Reithro- dontomys—see Choate 1970, and Hooper 1952, respectively). Except for Rheomys (see Voss 1988), these Middle American endem- ics have tribal-, subfamily-, or family-level phyletic affinity with species that are oth- erwise North American in distribution and origin. However, like Rheomys, the three species of Oryzomyini (sensu Voss & Carle- ton 1993) appear to represent Central American elements of groups that are es- sentially South American in origin and prin- cipal differentiation. Oryzomys devius is the northernmost outlier of the albigularis spe- cies-group, a complex which is predomi- nantly northern Andean in diversity and distribution (for example, see Patton et al. 1990) and whose morphology exhibits ge- neric-level differentiation from other ory- zomyines. Musser & Carleton (1993) pro- visionally associated the poorly known spe- cies aphrastus with the genus Sigmodonto- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON mys, whose type species alfari is distributed over lowland forest in southern Central America and northwestern South America, where it reaches western and northern An- dean slopes but not eastern ones. The form aphrastus actually may bear closer kinship to another enigmatic, little known species, Oryzomys hammondi of northwestern Ec- uador (for example, see Hershkovitz 1948). And the putative sister species of Oligory- zomys vegetus may be plausibly sought, as suspected by Osgood (1912), among north- ern Andean members of the genus. Biogeo- graphic evaluations of other vertebrates lend support to the possibility that the vicariant sister-groups to many taxa endemic to the Costa Rica-Chiriqui highlands occur in South America (Savage 1966, 1982; Slud 1964). The conjectural tone of the previous para- graph exposes the weaknesses of our current systematic understanding of many neotrop- ical rodents—confusion over their morpho- logical identity, meagre specimen-based knowledge of their distributions, and ab- sence of explicitly defended statements of their phylogenetic relationships. Such sys- tematic and biogeographic speculation should be tested by character data and viewed from a cladistic perspective. Im- proved species definition of forms like Oli- goryzomys vegetus is an elemental step to- ward these goals. Taxonomy Oligoryzomys vegetus (Bangs) Oryzomys (Oligoryzomys) vegetus Bangs, 1902:35 (type locality— Panama, Volcan de Chiriqui, Boquete, 4000 ft; holotype— MCZ 10298).—Miuller, 1912:177. Oryzomys costaricensis.— Allen, 1904a:69. Oryzomys fulvescens vegetus.—Goldman, 1918:93.—Goldman, 1920:102.— Miller, 1924:363.—Goodwin, 1946:396.—Mil- ler & Kellogg, 1955:441.— Hall & Kelson, 1959:568.—Handley, 1966:781.—Hall, 1981:622. VOLUME 108, NUMBER 2 Oligoryzomys vegetus.—Carleton & Mus- ser, 1989:71.—Musser & Carleton, 1993: 718. Oryzomys fulvescens creper Goodwin, 1945:2 (type locality—Costa Rica, Car- tago, Volcan Irazu, 9400 ft; holotype— AMNH 141199).—Goodwin, 1946: 396.— Miller & Kellogg, 1955:440.— Hall & Kelson, 1959:568.—Musser & Carle- ton, 1993:718. Oryzomys fulvescens reventazoni Goodwin, 1945:3 (type locality—Costa Rica, Car- tago, Santa Teresa Peralta; holotype— AMNH _ 141891).—Goodwin, 1946: 397.— Miller & Kellogg, 1955:440.— Hall and Kelson, 1959:568.— Musser & Carle- ton, 1993:718. Emended diagnosis.—A species of Oli- goryzomys characterized by relatively soft and fine pelage, dorsum a somber, dark brown, without prominent streaking or griz- zling, and underparts dark gray; tail dusky, monocolored for most of its length except slightly paler near the ventral base; size me- dium (HFL ca. 22—25 mm, ONL ca. 22—24 mm) for the genus with relatively long tail (usually >110 mm in adults), about 140% of head-and-body length; skull relatively broad across zygomatic arches (ca. 11.3- 12.5 mm) and braincase (ca. 10.5—11.0 mm), zygomatic plate and interorbital constric- tion correspondingly narrow; molar tooth- rows (2.7—3.0 mm) appear delicate for size of skull; capsular process of dentary well developed. Distribution. —Intermediate to high ele- vations (840 to <3000 m) in the mountains of Costa Rica (cordilleras Tilaran, Central, and Talamanca) and western Panama. Remarks. —Goodwin (1945) named two Costa Rican subspecies of Oryzomys ful- vescens that we herein reallocate to Oligo- ryzomys vegetus. He (1945:2) characterized creper as “A rather small, dark-colored, long- tailed Oligoryzomys, smaller than vegetus Bangs, with noticeably smaller feet and darker color.” Examples of vegetus from 359 Costa Rica do average smaller than our samples from western Panama but not in- ordinately so (Appendix), and they unam- biguously cluster with Panamanian OTUs of vegetus as opposed to geographically proximate representatives of fulvescens cos- taricensis (Fig. 8). Craniodental proportions and qualitative features, including devel- opment of the capsular process, of the type specimen of creper inarguably place it with- in the morphological range observed for Bangs’s (1902) vegetus (see Fig. 3), which condition supports their synonymy. Except for our disagreement over specific assign- ment and status, Goodwin’s description and comparisons of creper are apt, and most AMNH specimens from the central high- lands of Costa Rica that he (1945, 1946) referred to the subspecies we accept as ve- getus. Oryzomys fulvescens reventazoni, Good- win’s other Costa Rican subspecies, pre- sents a less clearcut interpretation, in part due to the uniformly young age of the five specimens composing the type series. De- fining characters, as stated by Goodwin (1945:3), recall those of creper: ““A very small, dark-colored pygmy rice rat. Similar in general appearance to O. f. creper but much smaller and tail shorter.’’? Goodwin (1945, 1946) repeatedly emphasized the striking resemblance, except for size, be- tween the type specimen of reventazoni and that of creper. In fact, many distinctions he ascribed to the two holotypes can be rea- sonably attributed to age-related size differ- ences, for the holotype of creper is a very old adult with flatly worn molars, whereas that of reventazoni is a much younger ani- mal in fresh adult pelage. Goodwin simul- taneously noted the still smaller size and paler coloration, with white underparts, of his four topotypes of reventazoni; their size and pelage impress us only as young indi- viduals of fulvescens, which is known to oc- cur with vegetus elsewhere in the upper val- ley of the Rio Reventazon (see Specimens Examined). Finally, results of principal 360 component analysis substantiate the closer morphometric agreement of the type of re- ventazoni to examples of vegetus and of the two measureable topotypes to those of ful- vescens (Fig. 3). We conclude that Good- win’s type series is a composite. The iden- tity of the type specimen (AMNH 141891), as name bearer of the taxon, dictates the relegation of reventazoni to subjective ju- nior synonymy under O. vegetus; the four topotypes (AMNH 140364-5, 141890, 141892) are reassigned to O. fulvescens cos- taricensis. Specimens examined. —352, as follows. Costa Rica: Provincia de Alajuela, Ca- taratos San Carlos (AMNH 141941-2); Monteverde Cloud Forest Reserve, valley of Rio Penas Blancas, La Esperanza, 840 m (KU 143399); Lajas Villa Quesada (AMNH 139736-43, 140363); Tapesco (AMNH 139804—6); Volcan Poas, 2000 m (UMMZ 123200); Zarcero, 6500 ft (FMNH 43990). Provincia de Cartago, Estrella de Cartago, 4500 ft (UMMZ 64131-2); Volcan Irazu, 9400 ft (AMNH 141199); Volcan Irazu, 0.25 mi N Hotel Robert, 2575 m (UMMZ 116903); Moravia, 1116 m (UMMZ 111983, 112283-4); El Muneco, 10 mi S Cartago, Rio Navarro, 3700, 3800, and 4000 ft (UMMZ 67310, 67312-3); El Muneco, 13 km S Cartago (KU 27017); Salsipuedes, Pan Am Hwy, 2730 m (UMMZ 123380); El Sauce Peralta (USNM 250394-—5); Santa Teresa Peralta (AMNH 141891). Provincia de Limon, Rio Teribé, Valle El Silencio, Rio Coton, 8000 ft (USNM 539895-907). Pro- vincia de Puntarenas, Monteverde, 1400- 1465 m (KU 142063, 142066-8, 143315, 143398; UMMZ 115425, 116904-8, 117102-4; USNM 566458); Monteverde, Cerro Amigos, 1760 m (KU 142064-—S); Monteverde, Reserva Bosque Nuboso, La Ventana (USNM 559054); Vera Cruz, 1320 m (KU 143502). Provincia de San José, El Copey de Dota, 6000 ft (UMMZ 6403940, 65071); Los Higuerones, Escazu (AMNH 137291, 138020-6, 138076-9); Cerro de Buena Vista, 10,342 ft (AMNH 9567/7909); PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Cerro de la Muerte, summit (on Pan Am Hwy), 10,200 ft (UMMZ 112280); Cerro de la Muerte, 5.5 miS (via Pan Am Hwy) sum- mit, La Georgina (UMMZ 112281-2); Cer- ro de la Muerta, Villa Mills (UMMZ 115426); 9 mi N (via Pan Am Hwy) San Isidro del General, 4800 ft (UMMZ 111970); 11 mi N (via Pan Am Hwy) San Isidro de General, 5200 ft (UMMZ 111971-S, 112279); San Joaquin de Dota, 3800 and 4000 ft (UMMZ 62882-3, 62887-8). Panama: Provincia Bocas de Toro, north- east of Boquete, near Rio Cylindro, 2380 m (USNM 516745); 3 km NE Escopeta, Que- brada de los Gatos, 1375 m (USNM 541172); 3.5 km E Escopeta, Cerro Bollo, 1800-1856 m (USNM 541186-8); 20 mi SSW Changuinola, upper Rio Changena, Rancho Mojica, 4800, 5000, and 5400-5600 ft (USNM 319316-—26); Fish Camp, 4900 ft, 08°58'N/82°40'W (USNM 520730-5); Up- per Camp 1975, 08°56’N/82°42’W (USNM 520736); 17.5 km NNW EI Hato del Vol- can, NE Cerro Pando, 2180 m (USNM 516758—-9); 25 km NNE San Félix, 1425- 1500 m (USNM 541184—5). Provincia de Chiriqui, Bambito, 5800 and 5900 ft (USNM 314355, 314777); Boquete, 4000, 4500, and 4800 ft (AMNH 18825-8, 18834— 6; FMNH 14298, 14302, 14304-6, 18518- 9; MCZ 10297-8, 10300-—4, 10308-10; USNM 242339); head of Rio Candela, 17 km NNW EI Hato del Volcan, 2000 m (USNM 516764); Casita Alta, 7000 ft (ANSP 18546-7, 18624); Rio Chebo (ANSP 18514, 18516-7, 18522); Rio Chiriqui Vie- jo, 3800 ft (ANSP 18523-5, 18945), 1600- 1850 m (UMMZ 116911-8, 117100); Co- tito Hot Springs (USNM 396546); Cylindro (ANSP 20956); Rio Gariche, 5350 ft (ANSP 18409, 18618-23); Hortigal (ANSP 20825, 20830, 20968, 21024); Finca Lerida (ANSP ZOTOl. 20768, 20803, 208715 20928 20923); Osta Clara Camp (ANSP 18334); Cerro Pando, 4100 ft (AMNH 147790; ANSP 17853-9); Pena Blanca (ANSP 20852, 21039); Cerro Punta, 1825 m (ANSP 18744, 18746-54, 18756-7; UMMZ VOLUME 108, NUMBER 2 116921); 0.5 mi W Cerro Punta, 1825 and 1880 m (UMMZ 116919-20); 0.5 mi SE Cerro Punta, Volcan de Chiriqui, 2000 m (UMMZ 116922-5); Cerro Punta, Boquete Trail, 7000, 7700, 7750, and 7800 ft (USNM 323883-97); Cerro Punta, Casa Tilley, 5300 ft (USNM 314344, 318429-30); Cerro Pun- ta, Finca Martinz, 6800 ft (USNM 314345- 54); Finca Santa Clara, 14.5 km NW El Hato de Volcan, 1200-1500 m (USNM 395547, 516742, 516757, 516761, 516763, 516767, 516774, 537618); Rio Santa Clara, 4150 ft (ANSP 18628, 18634, 18637-8); Siola, 4100-4300 ft (ANSP 17852, 18389-91, 18532, 18543, 18550-611, 18944, 18956- 8, 18960, 18964—5, 18969, 18971, 18978, 18981-3, 18989, 18991, 18994, 18998-9, 19081-2, 19088-9); locality unknown (AMNH 173903; ANSP 17825, 21046). Oligoryzomys fulvescens (Saussure) H. [esperomys] fulvescens Saussure, 1860: 102 (type locality — Mexico, Veracruz, vi- cinity of Orizaba—as subsequently re- stricted by Merriam, 1901:295). Oryzomys costaricensis Allen, 1893:239 (type locality—Costa Rica, Puntarenas, El General, 2150 ft; holotype—AMNH 9581/7922). Oryzomys delicatus Allen and Chapman, 1897:19 (type locality— Trinidad, Capa- ro; holotype—AMNH 7317/5925). Oryzomys navus Bangs, 1899:9 (type local- ity—Colombia, Magdalena, Sierra Ne- vada de Santa Marta, Pueblo Viejo, 8000 ft; holotype—MCZ 8107). Oryzomys navus messorius Thomas, 1901: 151 (type locality— British Guiana [Guy- ana], Kanuku Mountains; holotype— BMNH 1901.6.4.97). Oryzomys tenuipes Allen, 1904b:328 (type locality— Venezuela, Merida, Merida, 1630 m; holotype—AMNH 21330). Oryzomys (Oligoryzomys) nicaraguae A\- len, 1910:100 (type locality — Nicaragua, Matagalpa, Vijagua; holotype—AMNH 29543). 361 Oryzomys (Oligoryzomys) munchiquensis Allen, 1912:85 (type locality —Colombia, Cauca, La Florida, 7700 ft; holotype— AMNH 32603). Oryzomys fulvescens lenis Goldman, 1915: 130 (type locality— Mexico, Michoacan, Los Reyes; holotype—USNM 125941). Oryzomys fulvescens mayensis Goldman, 1918:92 (type locality—Mexico, Cam- peche, near Yohalttn, Apazote, 200 ft; holotype—USNM 107979). Oryzomys fulvescens engraciae Osgood, 1945:300 (type locality— Mexico, Ta- maulipas, northwest of Ciudad Victoria, Hacienda Santa Engracia, 240 m; holo- type—FMNH 54164). Oryzomys fulvescens pacificus Hooper, 1952: 23 (type locality— Mexico, Chiapas, Ma- pastepec, 150 ft; holotype—UMMZ 96764). Remarks. —While we defend our sample sizes and the coarse density of geographic sampling as adequate to clarify the status of O. vegetus, they are insufficient to evaluate geographic differentiation within O. fulves- cens and to address attendant nomencla- tural questions to their proper depth. Nev- ertheless, the several subspecies examined and the morphological divergence encoun- tered warrant some comment on intraspe- cific variation. The extensive dispersion of specimens along the second canonical variate princi- pally involves the four subspecific taxa of O. fulvescens and suggests a north-south trend of increase in average size, with in- dividuals of f. fulvescens and f. pacificus smaller, f nicaraguae intermediate, and those of f costaricensis larger (Fig. 7). Al- though a broad trend may exist (verification with more samples is desirable), individual scores within the conventional subspecies vary greatly and group polygons overlap ex- tensively. Dimensions that load heavily on CV II include IOB, BPL, LM1-3, and WM 1, all of which reach their most robust ex- pression in certain samples of costaricensis 362 (for example CR2-3, PA7—Appendix). If one ignores his inclusion of vegetus and the later additions of new geographic races, these results provide some corroboration of Goldman’s (1918) general remarks on cra- niodental variation within fulvescens. Cohesiveness of certain subspecific boundaries, however, is eroded in pheno- grams generated from clustering of sample means. Whether using a distance or corre- lation measure of phenetic similarity, OTUs of f, costaricensis fail to group exclusively of others representing fu/vescens proper, f. pacificus, and some f. nicaraguae (Fig. 8). Certain samples of f costaricensis from cen- tral Panama (PAI-3) actually link with northern subspecies instead of geographi- cally proximate OTUs from western Pan- ama (PA7) and eastern Costa Rica (CR2- 3). Pygmy rice rats from northern and cen- tral Nicaragua (NI1 = f nicaraguae) con- sistently formed a pair-group with fulves- cens from Veracruz, a resemblance in ac- cord with the opinion of Jones and Eng- strom (1986), who suggested that nicaraguae may prove inseparable from the nominate race. The most divergent OTU (NI2) among our samples of O. fulvescens is that drawn from scattered localities along the Pacific coastal region of Nicaragua. Jones & Eng- strom (1986) noted the smaller cranial size and pale buff venter of fulvescens from this region as compared to those from the cen- tral and northern highlands; they assigned, with reservation, the former to f costari- censis and the latter to f nicaraguae. We do not attach much significance to the apparent size divergence of this one OTU in view of its small sample size (n = 7), the skewed age representation (mostly young-, some full-, and no old-adults), and the coarse lumping of localities within Nicaragua necessarily employed by us and by Jones & Engstrom (1986). In particular, the reputedly smaller cranial size can be attributed to youthful age representation in this one sample; it is sig- nificant that the length of the molar rows, PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON which once erupted do not increase with age, of the two Nicaraguan OTUs are es- sentially identical (¥ LM1-3 = 2.86 and 2.87 mm in NII and NI2, respectively). Chromatic distinctions of the coastal spec- imens (paler dorsum and a white to light buff venter) do set them apart from north- central populations (darker dorsal tone and gray venter), and instead resemble examples of costaricensis from Costa Rica. As noted by Jones & Engstrom (1986), the geographic extent of this Pacific lowlands morphotype, its possible intergradation with Hooper’s (1952) pacificus to the north and with Al- len’s (1893) costaricensis to the south, de- serves amplification. Two chromosomal morphologies have been reported, both as Oryzomys fulvescens, for populations of Oligoryzomys from Cen- tral America. Gardner & Patton (1976) de- scribed a diploid complement consisting of one large and four small pairs of metacen- trics, three pairs of large subtelocentrics, and 18 pairs of small acrocentrics (2N = 54; FN = 68) for a single individual from near Santa Ana, San José, Costa Rica. In contrast, Hai- duk et al. (1979) discovered a higher diploid count (three additional acrocentric pairs; 2N = 60; FN = 74) for three individuals col- lected northeast of Catemaco, Veracruz, Mexico; they did not attach any taxonomic import to the difference. We have examined the former specimen (LSU 13169) and found it to be a typical example of O. fulvescens costaricensis. In view of their origin from Veracruz, and having encountered no other form of Oligoryzomys in the region, we pre- sume the latter to represent O. f. fulvescens. Taxonomic implication of the reported chromosomal differences, their degree of conservatism, and the existence of still oth- er karyotypic variants within populations now classified as O. fulvescens are all topics for future study. Central American populations and infra- specific taxa of O. fulvescens, now divorced of Bangs’s (1902) vegetus and Goodwin’s (1945) creper and reventazoni, appear con- VOLUME 108, NUMBER 2 specific to us. Yet questions remain—such as the divergence and relationship of pop- ulations in western Nicaragua and the sig- nificance of the disparate karyotypes re- ported —and underscore the need for further investigation. The specific limits and sys- tematic standing of forms of Oligoryzomys named from South America are far more tentative. Future research should amplify the specimen-based distribution of fulves- cens south of the Isthmus of Panama and confirm or reject the relationship and syn- onymy of South American species-group taxa that we have provisionally associated with it (Carleton & Musser 1989, Musser & Carleton 1993). Pending completion of such studies, we below list the Central American specimens examined according to recently published viewpoints on subspecies and their ranges (that is, Hall 1981, Jones & Engstrom 1986). Oligoryzomys fulvescens costaricensis (Allen) Specimens examined. —311, as follows. Costa Rica: Provincia de Alajuela, Lajas Villa Quesada (AMNH 139735); 5 mi SW San Ramon (KU 71300); Upala, Colonia Puntarenas, Route 4—KM 93, Rio Chi- murria, 80 m (KU 142726-8); Upala, San José, 45 m (KU 142729); Upala, Aguas Claras, Colonia Libertad, Finca La Anita (KU 142730). Provincia de Cartago, Agua Caliente (KU 16581-—2, 16584—6, 27015-6); Cartago, 4800 ft (KU 16535, 16583; UMMZ 66469, 66472, 66476); 3 km S Cartago, 1400 m (KU 27004); Cervantes (AMNH 123501; FMNH 35196-7; MCZ 27805; USNM 250374—-5, 256479); La Carpintera (AMNH 9565/7907); Girara (KU 16580); El Guarco (KU 165789, 16588); El Muneco, 10 mi S Cartago, Rio Navarro, 3700 and 3800 ft (UMMZ 67311, 67314); near Paraiso (UMMZ 105658); 1 mi W Paraiso, 1415 m (UMMZ 112286); Santa Teresa, Peralta (AMNH 140364—5, 141890, 141892); 3 mi SE Turrialba, Instituto Interamericano, 602 363 m (UMMZ 111976-81, 112285); 5 km SE Turrialba (KU 27005-14). Provincia de Guanacaste, 2 km SE (via rd to Cano Negro) Arenal (UMMZ 115513); Finca Jiménez, O.5 mi E headquarters, 30 m (UMMZ 115291); Cerros de San Juan, 8 mi S Santa Cruz, 1200 ft (UMMZ 65048); Hacienda Santa Maria, 3200 ft (UMMZ 65218-9); Palo Verde, 2 km S and 12 km E Bolson, 50 m (KU 143733); Rincon de La Vieja National Park, near headquarters, 780 m (KU 143748). Provincia de Puntarenas, Bo- ruca (AMNH 9572, 9573/7914, 9574/7915, 9575/7916, 9577/7918, 9580/7921; FMNH 5369); Buenos Aires (AMNH 9578/7919, 9579/7920); Canas Gordas (AMNH 142440-58, 142490-5, 142500); Osa Pen- insula, Corcovado National Park, Sirena Station (USNM 565820-1); El General, 2000 and 2150 ft(AMNH 9568/7910, 958 1- 2/7922-3; UMMZ 66470-1, 66475); Finca Helechales, 910 m (USNM 547947-9); Pal- mar Sur, 15 m (KU 88240-4); 4 km S San Vito de Java, Finca Las Cruces, 1250 m (UMMZ 116909-10); Monteverde, Pen- sion Quetzal, 1400 m (FMNH 128494). Provincia de San José, Escazu, 3000 ft (AMNH 131738); Monterrey, 22 km S San José, 1000 m (KU 60485), 1100 m (KU 39253, 60486); San Geronimo de Pirris (AMNH 1235389, 123541; FMNH 35198—- 9; MCZ 27801, 27803—4, 28864—-5; USNM 250377, 250380—-93, 256480-1); 9.3 mi W (via rd to Dominical) San Isidro del General (UMMZ 105656); 2 km NW Santa Ana (LSUMZ 13169); Sabanilla de Pirris(USNM 256449). Costa Rica, locality unknown (AMNH 10101, 10103, 19230). Nicaragua: Carazo, 3 mi NNW Diriamba (KU 71294-9); 3 km N and 4 km W Di1- riamba, 600 m (KU 110465, 115438). Chinandega, 1 km N Cosiguina, El Paraiso, 20 m (KU 115432). Granada, 10 km SE Guanacaste, Finca El Progreso, 1000 m (KU 106551-2). Panama: Canal Zone, Albrook Field (USNM 302684); Barro Colorado Island (UMMZ 59935-6; USNM 256183-5); Cu- 364 rundu (USNM 297943, 301594); Fort Clay- ton (USNM 297942); Fort Kobbe (USNM 300351); Buena Vista Peninsula, 1.75 km NNW Frijoles (USNM 503718); Rodman Naval Station (USNM 457327-9, 457917). Provincia de Chiriqui, Boquete, 3800, 4000, and 4800 ft (MCZ 10293-4, 10296, 10299, 10307); Rio Chebo (ANSP 18386-—7, 18511- 3, 18515, 18518-20, 18941-2, 19077-8); Colorado Camp, 24 km NNE San Felix, 1275-1325 m (USNM 541162-71, 541173- 83); Rio Gariche, 5350 ft (ANSP 18408); 7 km NE El Hato de Volcan, 1820 m (UMMZ 117099); Finca Lerida (ANSP 20762, 20769, 20924—5); Finca Santa Clara, 14.5 km NW El Hato del Volcan, 1200-1500 m (USNM 396541-5, 396548-5S0, 516741, 5167434, 516746-51, 516754-6, 516760, 516762, 516765, 516768-73, 537619); Rio Santa Clara, 4150 ft (ANSP 18383-4, 18627, 18629-33, 18635). Provincia de Coclé, El Valle (USNM 303417-20, 304766). Pro- vincia de Panama, Cerro Azul, 930 m (USNM 303262, 305676-94, 305697-8, 306951); Cerro Campana (USNM 303416); La Chorrera (AMNH 31431); 6 mi E El Valle (USNM 304767-92): 4 mi E and 1 mi S Pacora (USNM 305675); Panama City, Rio Chilibrillo (AMNH 36722-3). Oligoryzomys fulvescens fulvescens (Saussure) Specimens examined. — 103, as follows. Mexico: Estado de Veracruz, Achotal (FMNH 14105-8, 15882); Cerro Azul, 350 ft (KU 30559); Boca del Rio, 10 ft (KU 30570); 3 km W Boca del Rio, 10 ft (KU 24124-6); Cautlapan, 4000 ft (KU 30573- 5); Coscomatepec, 5000 ft (KU 3057 1-2); Jalapa, 4400 ft (AMNH 12536/10846-— 12541/10851, 12543/10853-12549/10859, 12583-5; FMNH 5370; USNM 93369-73, 93394); 5km N Jalapa, 4500 ft (KU 19394); Jico, 6000 ft (FMNH 13112; USNM 55032- 3); 2 km W Jico, 4200 ft (KU 19722-3); 20 km E Jesus Carranza, 300 ft (KU 24133- 4); 25 km ESE Jesus Carranza, 200 ft (KU 32156); 0.5—3 mi NE Las Minas, 1200-1400 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON m (USNM 329805-8); Mirador, 3500 ft (KU 24122; USNM 10107/12174); Monte Blan- co, 1300 m (KU 29495-6); 2 km N Mot- zorongo, 1500 ft (KU 19400); Orizaba, 4000 ft (USNM 58243-4, 58246-8, 58254-6, 58259-62, 58288), 4200 ft (USNM 63685- 8); 3 km SE Orizaba, 5500 ft (KU 19721, 19723); Ozulama, 500 ft (KU 30558); 4 km W Paso de San Juan, 250 ft (KU 24123); Planton Sanchez, 800 ft (KU 88700—2); Po- trero Viejo, 1700 ft (KU 24127-9, 24170, 30576-9, 32155); San Andreas Tuxtla (USNM 65540); 3 km E San Andreas Tux- tla, 1000 ft (KU 24130-2); 2 km SSW Te- nochtitlan, 60 m (UMMZ 116316-8); Teo- celo, 4500 ft (KU 30564—9); 12.5 mi N T1- huatlan, 300 ft (KU 88704); 5 km S Tihuat- lan, 700 ft (KU 24119); Tlacolula, 60 km WNW Tuxpan (KU 83067); 15 km ENE Tlacotepec, 1500 ft (KU 24120); 4 km W Tlacotepec, 1700 ft (KU 24121). Oligoryzomys fulvescens nicaraguae (Allen) Specimens examined. —28, as follows. Nicaragua: Boaco, 17 km N and 15 km E Santa Rosa, 300 m (KU 110461). Chon- tales, 1 km N and 2.5 km W Villa Somoza, 330 m (KU 110464). Jinotega, Yali, 860 m (KU 106546, 106549). Matagalpa, 1 km N and 5 km E Esquipulas, La Danta, 760 and 780 m (KU 115433-4); Santa Maria de Os- tuma, 1250 m (KU 106550); Vijagua (AMNH 29543). Nueva Segovia, 1.5 km N and | km E Jalapa, 660 m (KU 110455-7); 4.5 km N and 2 km E Jalapa, 680 m (KU 110453-—4); 3.5 km S and 2 km W Jalapa, 660 m (KU 110458-9). Rivas, Isla de Ome- tepe, 4 km S and 1.5 km E Alta Gracia, 40 m (KU 115439). Zelaya, Cara de Mono, 50 m (KU 115435-6); El Recreo, southern side of Rio Mico, 25 m (KU 106553-8, 110462- 3, 115437; USNM 337770). Oligoryzomys fulvescens pacificus (Hooper) Specimens examined. —22, as follows. Mexico: Estado de Chiapas, Maspaste- VOLUME 108, NUMBER 2 10 m (UMMZ 96767-9, 96798); Ariaga, 100 m (UMMZ 96770). Acknowledgments We thank the following museum staff for timely loan of specimens and-or the hos- pitality extended during the use of collec- tions under their charge: Edward B. Daesch- ler, ANSP; Bruce Patterson, FMNH; Rob- ert M. Timm and Thorvald Holmes, KU; Leanna Good and Mark Hafner, LSUMZ; Maria Rutzmoser, MCZ; Philip Myers, UMM ZZ. Patricia Wynne prepared the line drawing in Fig. 2, and Dave Schmidt, with his usual competence and concern for detail, undertook the cranial photography, distri- bution maps, and other figures. The final manuscript benefitted from the careful re- views and comments of Robert D. Fisher, Dave Schmidt, and Neal Woodman. Literature Cited Allen, J. A. 1893. Further notes on Costa Rica mam- mals, with description of a new species of Ory- zomys.— Bulletin of the American Museum of Natural History 5:237-240. . 1904a. Mammals from southern Mexico and Central and South America.—Bulletin of the American Museum of Natural History 20:29- 80. 1904b. New mammals from Venezuela and Colombia.— Bulletin of the American Museum of Natural History 20:327-335. 1910. Additional mammals from Nicara- gua.— Bulletin of the American Museum of Nat- ural History 28:87-115. 1912. Mammals from western Colombia. — Bulletin of the American Museum of Natural History 31:71-95. 1916. New mammals collected on the Roo- sevelt Brazilian Expedition.—Bulletin of the American Museum of Natural History 35:523- 530. ——.,, & F. M. Chapman. 1897. Ona second col- lection of mammals from the island of Trinidad, with descriptions of new species, and a note on some mammals from the island of Dominica, West Indies. — Bulletin of the American Muse- um of Natural History 9:13-—30. Bangs, O. 1899. A new pigmy Oryzomys from the Santa Marta region of Colombia. — Proceedings 365 of the Biological Society of Washington 13:9- 10. 1902. Chiriqui Mammalia.— Bulletin of the Museum of Comparative Zoology, Harvard Collection 39:17—51. Carleton, M. D., & G. G. Musser. 1989. Systematic studies of oryzomyine rodents (Muridae, Sig- modontinae): a synopsis of Microryzomys.— Bulletin of the American Museum of Natural History 191:1-83. Castillo-Munoz, R. 1983. Geology. Pp. 47-62 in D. H. Janzen, ed., Costa Rican natural history. The University of Chicago Press, Chicago, xi + 816 pp. Choate, J. R. 1970. Systematics and zoogeography of Middle American shrews of the Genus Cryp- totis. —University of Kansas Publications, Mu- seum of Natural History 19:195-317. Diersing, V. E. 1981. Systematic status of Sy/vilagus brasiliensis and S. insonus from North Ameri- ca.—Journal of Mammalogy 62:539-556. Enders, R. K. 1980. Observations on Syntheosciurus: taxonomy and behavior.—Journal of Mam- malogy 61:724-725. Fairchild, G. B., & C. O. Handley, Jr. 1966. Gazetteer of collecting localities in Panama. Pp. 9-22 in R. L. Wenzel & V. J. Tipton, eds., Ectoparasites of Panama. Field Museum of Natural History, Chicago, xii + 861 pp. Gardner, A. L. 1983. Oryzomys caliginosus (raton pardo, raton arrocero pardo, Costa Rican dusky rice rat). Pp. 483-485 in D. H. Janzen, ed., Costa Rican natural history. The University of Chi- cago Press, Chicago, x1 + 816 pp. —, & J. L. Patton. 1976. Karyotypic variation in oryzomyine rodents (Cricetinae) with com- ments on chromosomal evolution in the Neo- tropical cricetine complex.— Occasional Papers of the Museum of Zoology, Louisiana State Uni- versity 49:1-48. Goldman, E. A. 1915. Five new rice rats of the genus Oryzomys from Middle America. — Proceedings of the Biological Society of Washington 28:127- 130. . 1918. The rice rats of North America (Genus Oryzomys).— North American Fauna 43:1-—100. 1920. Mammals of Panama.—Smithsonian Miscellaneous Collections 69(5):309 pp. Goodwin, G. G. 1945. Four new rodents from Costa Rica.— American Museum Novitates 1293:1-3. 1946. Mammals of Costa Rica.— Bulletin of the American Museum of Natural History 87: 271-474. Hafner, M.S. 1991. Evolutionary genetics and zoo- geography of Middle American pocket gophers, genus Orthogeomys.—Journal of Mammalogy 72:1-10. Haiduk, M. M., J. W. Bickham, & D. J. Schmidly. 366 1979. Karyotypes of six species of Oryzomys from Mexico and Central America. — Journal of Mammalogy 60:610-615. Hall, E.R. 1981. The mammals of North America. — Volume 2, 2nd edition, i-vi + 601-1181. John Wiley & Sons, New York. —., & K.R. Kelson. 1959. The mammals of North America.— Volume 2:547-1083 + 79. Ronald Press, New York. Handley, C. O., Jr. 1966. Checklist of the mammals of Panama. Pp. 753-795 in R. L. Wenzel & V. J. Tipton, eds., Ectoparasites of Panama. Field Museum of Natural History, Chicago, xii + 861 pp. Hershkovitz, P. 1948. Mammals of northern Colom- bia. Preliminary report No. 3: Water rats (genus Nectomys), with supplemental notes on related forms.— Proceedings of the United States Na- tional Museum 98:49-56. Holdridge, L. R. 1967. Life zone ecology. San José, Costa Rica, Tropical Science Center, 205 pp. Hooper, E. T. 1952. A systematic review of harvest mice (Genus Reithrodontomys) of Latin Amer- ica.— Miscellaneous Publications of the Muse- um of Zoology, University of Michigan 77:255 pp. 1972. A synopsis of the rodent genus Scoti- nomys.— Occasional Papers of the Museum of Zoology, University of Michigan 665:32 pp. Jones, J. K., Jr., & M. D. Engstrom. 1986. Synopsis of the rice rats (Genus Oryzomys) of Nicara- gua.— Occasional Papers The Museum, Texas Tech University 103:1-23. McPherson, A. B. 1985. A biogeographical analysis of factors influencing the distribution of Costa Rican rodents.—Brenesia 23:97-273. 1986. The biogeography of Costa Rican ro- dents: an ecological, geological, and evolution- ary approach.—Brenesia 25—26:229-244. Merriam, C. H. 1901. Synopsis of the rice rats (genus Oryzomys) of the United States and Mexico. — Proceedings of the Washington Academy of Sci- ences 3:273-295. Miller, G. S. 1912. List of North American land mammals in the United States National Mu- seum, 1911.—Bulletin of the United States Na- tional Museum 79:xiv + 455 pp. . 1924. List of North American Recent mam- mals, 1923.—Bulletin of the United States Na- tional Museum 128:xvi + 673 pp. ——., & R. Kellogg. 1955. List of North American Recent mammals.— Bulletin of the United States National Museum 205:xiil + 954 pp. Musser, G. G., & M. D. Carleton. 1993. Family Mur- idae. Pp. 501-755 in D. E. Wilson & D. M. Reeder, eds., Mammal species of the world, a taxonomic and geographic reference, second PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON edition. Smithsonian Institution Press, Wash- ington D.C., 1206 pp. Osgood, W. H. 1912. Mammals from western Ven- ezuela and eastern Colombia.— Field Museum of Natural History, Zoological Series 10:32-67. . 1945. Two new rodents from Mexico. —Jour- nal of Mammalogy 26:299-301. Patton, J. L., P. Myers, & M. F. Smith. 1990. Vi- cariant versus gradient models of diversifica- tion: the small mammal fauna of eastern An- dean slopes of Peru. Pp. 355-371 in G. Peters & R. Hutterer, eds., Vertebrates in the tropics. Museum Alexander Koenig Research Institute and Zoological Museum, Bonn, 424 pp. Reid, F.A.,& C. A. Langtimm. 1993. Distributional and natural history notes for selected mammals from Costa Rica. — The Southwestern Naturalist 38:299-302. Rogers, D. S., & J. E. Rogers. 1992. Heteromys or- esterus. —Mammalian Species 396:1-3. Saussure, H. de. 1860. Note sur quelques mammi- feres de Mexique.— Revue et Magazine Zoolo- gique, Paris, Serie 2, 12:97-110. Savage, J. M. 1966. The origins and history of the Central American herpetofauna.—Copeia 1966: 719-766. 1982. The enigma of the Central American herpetofauna: dispersals or vicariance?.—An- nals of the Missouri Botanical Garden 69:464— 597. Slud, P. 1964. The birds of Costa Rica: distribution and ecology.— Bulletin of the American Muse- um of Natural History 128:1—430. Stiles, F.G. 1983. Birds, introduction. Pp. 502-530 in D. H. Janzen, ed., Costa Rican natural his- tory. The University of Chicago Press, Chicago, x1 + 816 pp. Thomas, O. 1901. Onacollection of mammals from the Kanuku Mountains, British Guiana.—An- nals and Magazine of Natural History, Series 7, 8:139-154. Timm, R. M., D. E. Wilson, B. L. Clauson, R. K. LaVal, & C. S. Vaughan. 1989. Mammals of the La Selva-Braulio Carillo complex, Costa Rica.— North American Fauna 75:162 pp. Voss, R.S. 1988. Systematics and ecology of ichthyo- myine rodents (Muroidea): patterns of morpho- logical evolution in a small adaptive radia- tion.—Bulletin of the American Museum of Natural History 188:259-493. —, & M. D. Carleton. 1993. A new genus for Hesperomys moliter Winge and Holochilus magnus Hershkovitz with an analysis of its phy- logenetic relationships. — American Museum Novitates 3085:39 pp. —., & L. F. Marcus. 1992. Morphological evo- lution in muroid rodents II. Craniometric factor VOLUME 108, NUMBER 2 divergence in seven Neotropical genera, with experimental results from Zygodontomys.— Evolution 46:1918-—1934. + , & P. Escalante P. 1990. Morpholog- ical evolution in muroid rodents I. Conservative patterns of craniometric covariance and their ontogenetic basis in the Neotropical genus Zy- godontomys. —Evolution 44:1568-1587. Woodman, N., & R. M. Timm. 1992. A new species of small-eared shrew, Genus Cryptotis (Insec- tivora: Soricidae), from Honduras.—Proceed- ings of the Biological Society of Washington 105: 1-12. . 1993. Intraspecific and interspecific variation in the Cryptotis nigrescens complex of small- eared shrews (Insectivora: Soricidae), with the description of a new species from Colombia. — Fieldiana: Zoology, New Series 74:30 pp. Appendix.— Descriptive statistics for selected cra- nial and external variables and OTUs of Oligoryzomys. Species and OTU n Mean Range SD Total length O. fulvescens fulvescens FUL a5 178 157-205 11.6 O. fulvescens costaricensis PA3 19 197 184-217 8.0 PA7 24 174 149-198 12.2 CR2 21 177 160-196 10.2 CR4 23 173 150-193 10.4 CRS5 24 lz 154-197 11.2 O. vegetus PA4 12 207 185-235 | bea PAS 28 192 170-213 10.7 PA6 63 211 191-238 12.6 CR1 4 196 179-216 15.3 CR7 1S 193 158-214 14.1 Tail length O. fulvescens fulvescens FUL 53 101 82-116 7.3 O. fulvescens costaricensis PA3 19 112 102-125 6.5 PAT 24 99 85-118 13 CR2 2 102 92-115 a7 CR4 23 102 88-112 6.1 CR5 22 99 90-115 5.7 O. vegetus PA4 12 119 110-130 6.6 PAS 28 113 97-126 6.3 Appendix. — Continued. Species and OTU n Mean Range SD PA6 63 126 113-142 8.3 CRI 4 113 107-125 8.3 CR7 15 114 94-129 8.5 Hindfoot length O. fulvescens fulvescens FUL 57 21.4 19-24 1.1 O. fulvescens costaricensis PA3 20 22:1 21-23 0.8 PA7 24 20.2 18-23 1.4 CR2 Z2 20.1 18-21 1.1 CR4 23 19.9 18-21 0.8 CR5 23 20.8 17-23 1.4 O. vegetus PA4 13 23.7 20-25 be PAS 30 22,9 21-26 1.0 PA6 62 24.5 23-27 1.0 CR1 11 22.4 21-24 1.1 CR7 16 23.0 20-25 1.4 Occipitonasal length O. fulvescens fulvescens FUL 50 212 19.6-23.7 0.8 O. fulvescens costaricensis PA3 20 22.0 21.5-23.0 0.4 PA7 24 21.7 20.2—23.2 0.7 CR2 14 22.0 20.8-23.9 0.9 CR4 16 21.3 19.2-23.0 1.1 CR5 15 21,5 20.1-22.5 OFF O. vegetus PA4 8 23.3 22.3-24.4 0.7 PAS 33 22.3 20.9--24.0 O.7 PA6 49 23.2 21.0-25.6 1.0 CR1 13 22.8 21.1-24.1 0.8 CR7 15 22.4 21.4-23.6 0.7 Zygomatic breadth O. fulvescens fulvescens FUL 50 11.0 10.3-12.2 0.4 O. fulvescens costaricensis PA3 20 11.3 10.8-11.9 0.4 PA7 24 11.0 10.1-11.9 0.4 CR2 14 11.3 10.6-12.4 0.5 CR4 16 11.1 10.0-11.8 0.5 CR5 16 Vee 10.4-11.8 0.4 O. vegetus PA4 8 12.3 11.8-12.9 0.4 PAS 33 11.9 10.8-12.9 0.5 368 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Appendix. — Continued. Appendix. — Continued. Species Species and OTU n Mean Range SD and OTU n Mean Range SD PA6 49 12.0 11.1-13.5 0.5 CR1 13 11.9 11.1-12.5 0.5 CR1 13 7.0 6.4-7.6 0.3 CR7 16 1 ile7/ 10.8-12.5 0.4 CR7 15 6.6 6.1-7.1 0.3 Breadth of braincase Postpalatal length O. fulvescens fulvescens O. fulvescens fulvescens FUL 50 10.0 9.4-10.5 0.3 FUL 50 7.0 6.1-7.9 0.6 — O. fulvescens costaricensis O. fulvescens costaricensis PA3 20 10.0 9.7-10.4 0.2 PA3 20 7.5 7.1-7.9 Ow PA7 24 9.9 9.4-10.4 0.2 PA7 24 val 6.5-7.7 0.3 CR2 14 10.3 9.9-10.7 0.2 CR2 14 U2 6.7-8.5 0.5 CR4 16 10.1 9.6-10.5 0.3 CR4 13 7.0 6.2-7.9 0.6 CR5 16 10.1 9.7-10.4 0.2 CR5 16 7.0 6.4-7.8 0.4 O. vegetus O. vegetus PA4 8 10.9 10.6-11.2 0.2 PA4 7 7.9 7.5-8.9 0.5 PAS 33 10.7 10.1-11.2 0.3 PAS 33 7.4 6.7-8.2 0.4 PA6 49 10.7 10.3-11.3 0.2 PA6 49 Vol 6.7-8.5 0.5 CR1 13 10.7 10.1-11.0 0.2 CR1 11 7.4 6.8-8.0 0.4 CR7 16 10.6 10.3-10.8 0.2 CR7 16 7.3 6.9-7.7 0.2 Interorbital breadth Length of diastema O. fulvescens fulvescens O. fulvescens fulvescens FUL 50 3.4 Bul=3i 0.1 FUL 50 5.1 4.2-5.8 0.3 O. fulvescens costaricensis O. fulvescens costaricensis PA3 20 a7 3.5-3.9 0.1 PA3 20 5.4 5.1-5.7 0),22 PAT 24 3.7 3.54.2 0.1 PAT 24 5.1 4.7-5.8 0.3 CR2 14 3.7 3.4-4.1 0.1 CR2 14 5.3 4.9-6.3 0.4 CR4 16 27) 2) 3) ©) 0.1 CR4 U7 5.1 4.3-5.8 0.4 CR5 16 3.6 YAY 6) (0) 0.1 CR5 16 52 4.9-5.8 0.3 O. vegetus O. vegetus PA4 8 3.7 3.5-4.0 0.1 PA4 8 5.7 5.4-6.2 0.3 PA5 33 3.6 3.1-4.0 0.1 PAS 33 5.5 4.9-6.2 0.3 PA6 49 3.6 BIB=310 Onl PA6 49 5.7 4.9-6.6 0.4 CR1 13 Qa] 3.5-4.0 0.1 CR1 13 So7/ 5.2-6.4 0.4 CR7 16 3.7 3.4-3.9 0.1 CR7 16 5.5 5.1-5.9 0.2 Length of rostrum Length of incisive foramen O. fulvescens fulvescens O. fulvescens fulvescens FUL 50 6.3 5.5-7.0 0.4 FUL 50 Si) 3.3-4.1 0.3 O. fulvescens costaricensis O. fulvescens costaricensis PA3 20 6.5 6.0-7.1 0.2 PA3 20 4.0 3.7—4.5 on PA7 24 6.4 5.8-7.1 0.3 PA7 24 3.9 3.64.3 0.2 CR2 14 Ou 6.1-7.6 0.4 CR2 14 4.0 3.7-4.3 0.2 CR4 16 6.4 5.3-7.2 0.5 CR4 17 3.7 3.1-4.0 O22 CR5 15 6.4 6.0-7.1 0.3 CRS5 16 3.8 3.5—4.2 0.2 O. vegetus O. vegetus PA4 8 6.9 6.4—7.5 0.3 PA4 8 3.8 3.7-4.1 0.1 PAS 33 6.7 5.7-7.4 0.4 PAS 33 3a// 3.1-4.0 0.2 VOLUME 108, NUMBER 2 Appendix. — Continued. Species and OTU PA6 CRI CR7 n 49 13 16 Mean Sol! 3)5// Sol Range 3.3-4.3 3.44.0 3.3-4.0 Breadth of bony palate O. fulvescens fulvescens FUL 50 3.9 O. fulvescens costaricensis PA3 20 4.0 PA7 24 3.9 CR2 14 4.0 CR4 17 4.0 CR5 16 4.1 O. vegetus PA4 8 4.3 PAS 33 4.2 PA6 49 4.3 CR1 13 4.2 CR7 16 4.0 O. fulvescens fulvescens FUL 50 1.8 O. fulvescens costaricensis PA3 20 De) PA7 24 2.0 CR2 14 2.0 CR4 17 1.8 CR5 16 1.9 3.7-4.2 3.7-4.2 3.64.3 3.8-4.4 3.8-4.2 3.8-4.4 4.2-4.5 3.9-4.4 4.0-4.8 3.9-4.5 3.8-4.2 Breadth of zygomatic plate 1.4—2.1 1.9-2.3 e222 1.8-2.2 15 =2 51 1.5-2.1 SD 0.2 0.2 0.2 Onl 0.1 0.1 0.1 0.1 0.1 O51 0.1 0.2 0.2 0.1 0.1 0.1 0.1 0.1 0.2 0.1 369 Appendix. —Continued. Species and OTU n Mean Range SD O. vegetus PA4 8 1.9 1.8-2.0 0.1 PAS 33 1.8 1.4-2.2 0.2 PA6 49 1.9 1.5-2.1 0.1 CR1 13 1.9 1.6-2.2 0.1 CR7 16 1.8 1.6-—2.0 0.1 Length of maxillary toothrow O. fulvescens fulvescens FUL 50 2.8 2.5-3.0 0.10 O. fulvescens costaricensis PA3 20 3.0 2.8-3.2 0.10 PA7 24 3.1 2.9-3.2 0.09 CR2 19 3.1 2.8-3.3 0.11 CR4 a) 2.9 2.8-3.1 0.08 CR5 24 3.0 2.8-3.1 0.08 O. vegetus PA4 12 3.0 2.8-3.1 0.08 PAS 34 2.9 2.8-3.1 0.06 PA6 58 2.9 2.7-3.1 0.08 CR1 13 2.9 2.8-3.1 0.08 CR7 16 2.8 2.6-3.0 0.10 OTU Codes.—FUL: Mexico, Veracruz, various lo- calities; CR1: Costa Rica, Limon, Valle El Silencio; CR2: Costa Rica, Puntarenas, Camas Gordas; CR4: Costa Rica, San José, San Geronimo Pirris; CR5: Costa Rica, Cartago, valley of Rio Reventazon; CR7: Costa Rica, Puntarenas, Monteverde; PA3: Panama, Chiri- qui, Colorado Camp; PA4: Panama, Chiriqui, Bo- quete; PAS: Panama, Chiriqui, Cerro Punta & vicinity; PA6: Panama, Chiriqui, Siola; PA7: Panama, Chiriqui, Finca Santa Clara. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(2):370. 1995. REVIEWERS The following people reviewed manuscripts for the Proceedings in 1994. A. Alekseev, F. Alvarez, M. Angel, A. Allison, W. Ausich, K. Baba, R. J. Baker, S. Ban, K. Banse, F. M. Bayer, I. Bayly, M. Beaton, D. Belk, J. Bierne, R. Birdsong, J. A. Blake, R. W. Bouchard, T. E. Bowman, N. L. Bruce, S. D. Cairns, E. Campos, M. R. Campos, A. Carvacho, F. A. Chace, Jr., G. Charmantier, C. A. Child, K. Coates, R. Cocroft, A. Cohen, O. Colemann, G. K. Creighton, R. I. Crombie, N. Cumberlidge, T. Daniel, M. Dardeau, P. J. F. Davie, A. Demeter, M. Dillon, S. K. Donovan, F. L. Downs, R. Drewes, L. H. Durkee, D. Duszynski, P. Dworsschak, C. Erséus, K. Fauchald, D. Fautin, A. Feduccia, D. L. Felder, W. Fender, F. Ferrara, F. D. Ferrari, C. Ferraris, R. D. Fisher, K. Fitzhugh, J. F. Fitzpatrick, O. S. Flint, Jr., A. Fosshagen, C. Fransen, J. Friend, R. J. Gagné, C. Gallegos, J. E. Garcia-Raso, A. L. Gardner, H. Griffith, M. J. Grygier, D. Guinot, N. Hairston, C. Hand, A. Harold, C. W. Hart, Jr., R. H. Heard, J. Hedgpeth, G. Hendler, M. E. Hendrickx, W. R. Heyer, J.-S. Ho, H. H. Hobbs III, R. S. Hoffmann, L. B. Holthuis, R. W. Holzenthal, P. Houde, K. Hulsemann, R. F. Inger, H. F. James, M. Jenkins, R. F. Jezerinac, B. Kensley, N. Knowlton, D. Koester, K. F. Koopman, L. S. Kornicker, R. K. Kropp, J. D. Kudenov, H. Kunz, A. Kuris, D. B. Lellinger, R. Lemaitre, D. Lindeman, P. Linke, B. Livezey, J. G. Lundberg, J. D. Lynch, S. Maas, E. Macpherson, C. Magalhaes, R. B. Manning, J. Markham, J. W. Martin, W. N. Mathis, J. Mauchline, P. A. McLaughlin, H. Michel, W. Mielke, C. Miller, R. F. Modlin, R. Mooi, J. C. Morse, T. A. Munroe, G. G. Musser, C. W. Myers, P. K. L. Ng, N. Ngoc-Ho, M. Nizinski, J. N. Norris, N. Nunomura, S. L. Olson, J. L. Patton, D. L. Pawson, T. Perkins, M. Petersen, M. Pinna, G. Poore, S. Poss, J. Pruski, J. W. Reid, J. Reynolds, S. Rice, E. A. Rickart, C. B. Robbins, S. de A. Rodrigues, G. Rodriguez, A. Rogers, G. Rouse, L. A. Ruedas, D. E. Russell, M. de Saint Laurent, S. Schaefer, H. Schmalfuss, N. J. Scott, Jr., M. Segonzac, J. Shields, P. J. Spangler, W. C. Starnes, D. W. Steadman, G. C. Steyskal, J. H. Stock, E. Suess, H. Suzuki, P. Sze, M. Telford, E. V. Thuesen, M. Tirkay, R. P. Vari, C. L. Van Dover, J. C. von Vaupel Klein, H. O. von Hagen, M. Vecchione, J. Villalobos, J.-W. Wagele, J. Wells, A. B. Williams, J. T. Williams, G. D. Wilson, L. D. Wilson, N. Woodman, K. Wouters, R. Young, G. R. Zug. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(2):371-372. 1995. INTERNATIONAL COMMISSION ON ZOOLOGICAL NOMENCLATURE Applications published in the Bulletin of Zoological Nomenclature The following Applications were published on 30 March 1995 in Vol. 52, Part 1 of the Bulletin of Zoological Nomenclature. Comment or advice on any of these applications is invited for publication in the Bulletin and should be sent to the Executive Secretary, I.C.Z.N., % The Natural History Museum, Cromwell Road, London SW7 5BD, U.K. Case No. 2901 Stictostroma Parks, 1936 (Porifera, Stromatoporoidea): proposed conser- vation, and designation of S. gorriense Stearn, 1995 as the type species. 2949 Aplysia juliana Quoy & Gaimard, 1832 (Mollusca, Gastropoda): proposed conservation of the specific name. 2922 Octopus vulgaris Cuvier, [1797] and Loligo vulgaris Lamarck, 1798 (Mol- lusca, Cephalopoda): proposed conservation of the specific names. 2899 Dodecaceria concharum Orsted, 1843 and Heterocirrus fimbriatus Verrill, 1879 (currently D. fimbriata) (Annelida, Polychaeta): proposed con- servation of the specific names by the designation of a neotype for D. concharum. Eophacops Delo, 1935 and Acernaspis Campbell, 1967 (Trilobita): proposed conservation. Diplocentrus mexicanus Peters, 1861 (Arachnida, Scorpiones): proposed con- firmation of the rediscovered holotype as the name-bearing type. Nepa rustica Fabricius, 1781 and Zaitha stollii Amyot & Serville, 1843 (currently Diplonychus rusticus and Belostoma stollii; Insecta, Het- eroptera): proposed conservation of the specific names. Aspidiphorus Ziegler in Dejean, 1821 (Insecta, Coleoptera): proposed con- servation as the correct original spelling, and ASPIDIPHORIDAE Kie- senwetter, 1877 (1859): proposed placement on the Official List. XANTHOLININI Erichson, 1839 and QUEDIINI Kraatz, 1857 (Insecta, Coleop- tera): proposed precedence over senior synonyms, and Quedius Ste- phens, 1829: proposed designation of Staphylinus levicollis Brulle, 1832 as the type species. Metablastothrix Sugonjaev, 1964 (Insecta, Hymenoptera): proposed desig- nation of Blastothrix (Metablastothrix) isomorpha Sugonjaev, 1964 as the type species. Agonus Bloch & Schneider, 1801 (Osteichthyes, Scorpaeniformes): proposed conservation; AGONIDAE Kirby, 1837 (Insecta, Coleoptera) and AGONIDAE Swainson, 1839 (Osteichthyes, Scorpaeniformes): pro- posed removal of homonymy. Proposed conservation of nine specific names of southern Afrotropical birds which are junior synonyms. 372 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Opinions published in the Bulletin of Zoological Nomenclature The following Opinions were published on 30 March 1995 in Vol. 52, Part 1 of the Bulletin of Zoological Nomenclature. Copies of these Opinions can be obtained free of charge from the Executive Secretary, I.C.Z.N. % The Natural History Mu- seum, Cromwell Road, London SW7 5BD, U.K. Opinion No. 1792. Pleurotoma meneghinii Mayer, 1868 (currently Asthenotoma meneghinii; Mollusca, Gastropoda): neotype replaced by rediscovered lectotype. 1793. Chtenopteryx Appellof, 1890 (Mollusca, Cephalopoda): confirmed as the correct original spelling. 1794. Sigara coleoptrata Fabricius, [1777] (Insecta, Heteroptera): specific name conserved, and Notonecta obliqua Thunberg, 1787: specific name placed on the Official List. Corisa sexlineata Reuter, 1882 (currently Sigara (Tropocorixa) sexlineata; Insecta, Heteroptera): specific name not conserved, and that of C. confluens Fieber, 1851 placed on Official List. Platynectes Régimbart, 1879 (Insecta, Coleoptera): conserved. Oecothea Haliday in Curtis, 1837 (Insecta, Diptera): conserved, and Helo- myza fenestralis Fallén, 1820 designated as the type species. Rivulus marmoratus Poey, 1880 (Osteichthyes, Cyprinodontiformes): given precedence over R. ocellatus Hensel, 1868, and a neotype designated for R. marmoratus. Naucrates Rafinesque, 1810 and Xyrichtys Cuvier, 1814 (Osteichthyes, Per- ciformes): conserved. Emys Dumeril, 1806 (Reptilia, Testudines): conserved. Cetiosauriscus Huene, 1927 (Reptilia, Sauropodomorpha): Cetiosauriscus stewarti Charig, 1980 designated as the type species. Dinodontosaurus Romer, 1943 (Reptilia, Synapsida): conserved. INFORMATION FOR CONTRIBUTORS Content.—The Proceedings of the Biological Society of Washington contains papers bearing on systematics in the biological sciences (botany, zoology, and paleontology), and notices of business transacted at meetings of the Society. Except at the direction of the Council, only manuscripts by Society members will be accepted. Papers are published in English (except for Latin diagnoses/descriptions of plant taxa), with an Abstract in an alternate language when appropriate. Submission of manuscripts.—Submit three copies of each manuscript in the style of the Proceedings to the Editor, complete with tables, figure captions, and figures (plus originals of the illustrations). Mail directly to: Editor, Proceedings of the Biological Society of Wash- ington, National Museum of Natural History NHB-108, Smithsonian Institution, Washington, D.C. 20560. (Do not submit manuscripts to an associate editor). Presentation.—Clarity of presentation, and requirements of taxonomic and nomenclatural procedures necessitate reasonable consistency in the organization of papers. Telegraphic style is recommended for descriptions and diagnoses. The style for the Proceedings is described in ““GUIDELINES FOR MANUSCRIPTS for Publications of the BIOLOGICAL SOCIETY OF WASHINGTON” a supplement to Volume 103, number 1, March 1990. Authors are encour- aged to consult this article before manuscript preparation. Copies of the article are available from the editor or any associate editor. The establishment of new taxa must conform with the requirements of appropriate interna- tional codes of nomenclature. Decisions of the editor about style also are guided by the General Recommendations (Appendix E) of the International Code of Zoological Nomenclature. When appropriate, accounts of new taxa must cite a type specimen deposited in an institutional col- lection. Review.—One of the Society’s aims is to give its members an opportunity for prompt pub- lication of their shorter contributions. Manuscripts are reviewed by a board of Associate Editors and appropriate referees. Proofs.—Authors will receive first proofs and original manuscript for correction and ap- proval. Both must be returned within 48 hours to the Editor. Reprint orders are taken with returned proofs. Publication charges.—Authors are required to pay full costs of figures, tables, changes in proofs ($3.00 per change or revision), and reprints. Authors are also asked to assume costs of page-charges. The Society, on request, will subsidize a limited number of contributions per volume. If subsidized manuscripts result in more than 12 printed pages, the additional pages must be financed by the author(s). Multiple authorship will not alter the 12 page limit (each author will be viewed as having used his/her 12 subsidized pages). Payment of full costs will facilitate speedy publication. Costs.—Printed pages @ $60.00, figures @ $10.00, tabular material @ $3.00 per printed inch per column. One ms. page = approximately 0.4 printed page. CONTENTS A new species of Raricirrus (Polychaeta: Ctenodrilidae) from wood collected in the Tongue of thé Ocean, Virgin Islands Harlan K. Dean Six copepodid stages of Ridgewayia klausruetzleri, a new species of copepod crustacean (Ridgewayiidae: Calanoida) from the barrier reef in Belize, with comments on appendage development Frank D. Ferrari Sinoniscus cavernicolus, a new genus and species of terrestrial isopod crustacean from a cave in China (Styloniscidae: Oniscidea) George A. Schultz Sphaerolana karenae, a new species of hypogean isopod crustacean from Nuevo Leon, Mexico Gabino A. Rodriguez-Almaraz and Thomas A. Bowman Natatolana nukumbutho, a new species (Crustacea: Isopoda: Cirolanidae) from deep water off Suwa, Fiji Niel L. Bruce and J@rgen Olesen A new species of the shrimp genus Chorocaris Martin & Hessler, 1990 (Crustacea: Decapoda: Bresiliidae) from hydrothermal vent fields along the Mid-Atlantic Ridge Joel W. Martin and Jennifer C. Christiansen A new genus and species of caridean shrimp (Crustacea: Decapoda: Bresiliidae) from hydro- thermal vents on Loihi Seamount, Hawaii Austin B. Williams and Fred C. Dobbs Remarks on the taxonomy of Sudanonautes chavanesii (A. Milne-Edwards, 1886) (Brachyura: Potamoidea: Potamonautidae) from Central Africa Neil Cumberlidge First zoea of Dissodactylus glasselli Rioja and new range and host records for species of Dissodactylus (Brachyura: Pinnotheridae), with a discussion of host-symbiont biogeography Gerhard Pohle and Fernando Marques A new crab species of the genus Pseudorhombila H. Milne-Edwards, 1837 (Crustacea: Decap- oda: Goneplacidae) Ana Rosa Vazquez-Bader and Adolfo Gracia Description of the ghost shrimp Sergio mericeae, a new species from south Florida, with reexamination of S. guassutinga (Crustacea: Decapoda: Callianassidae) Raymond B. Manning and Darryl L. Felder Contribution to the knowledge of Reynoldsia Malloch (Diptera: Muscidae) Marcia Souto Couri Rediagnosis of the brittlestar genus Ophiosyzygus and notes on its type species O. disacanthus (Echinodermata: Ophiuroidea: Ophiomyxidae) based on the type specimens from Japanese waters and new material from the Gulf of Mexico Richard L. Turner and Robyn M. Heyman Designation of a lectotype for Crocodilus siamensis Schneider, 1801 (Reptilia: Crocodylia) Charles A. Ross, Gregory C. Mayer, and Roger Bour Description of a new large-bodied species of Apomys Mearns, 1905 (Mammalia: Rodentia: Muridae) from Mindoro Island, Philippines Luis A. Ruedas A review of the spiny mouse genus Scolomys (Rodentia: Muridae: Sigmodontinae) with the description of a new species from the western Amazon of Brazil James L. Patton and Maria Nazareth F. da Silva Systematic studies of oryzomyine rodents (Muridae: Sigmodontinae): definition and distribution of Oligoryzomys vegetus (Bangs, 1902) Michael D. Carleton and Guy G. Musser Reviewers—1994 International Commission on Zoological Nomenclature 169 180 201 207 72) 220 228 238 247 254 266 281 DOP 298 302 SN) 338 370 371 Finest UTHER | - | : VOLUME 108 NUMBER 3 19 SEPTEMBER 1995 ISSN 0006-324X THE BIOLOGICAL SOCIETY OF WASHINGTON 1994-1995 i Officers President: Janet W. Reid Secretary: Carole C. Baldwin President-elect: Stephen D. Cairns Treasurer: T. Chad Walter Se Se Be ae Elected Council | Robert J. Emry Susan L. Jewett re Richard C. Froeschner Lynne R. Parenti ; Alfred L. Gardner F. Christian Thompson Custodian of Publications: Austin B. Williams PROCEEDINGS Editor: C. Brian Robbins Associate Editors Classical Languages: George C. Steyskal Invertebrates: Jon L. Norenburg Frank D. Ferrari Plants: David B. Lellinger Rafael Lemaitre Insects: Wayne N. Mathis Vertebrates: Thomas A. Munroe Membership in the Society is open to anyone who wishes to join. There are no prerequisites. Annual dues of $25.00 (for USA and non-USA addresses) include subscription to the Pro- ceedings of the Biological Society of Washington. Library subscriptions to the Proceedings are: $40.00 for USA and non-USA addresses. Non-USA members or subscribers may pay an ad- ditional $25.00 to receive the Proceedings by Air Mail. The Proceedings of the Biological Society of Washington (USPS 404-750) is issued quarterly. Back issues of the Proceedings and the Bulletin of the Biological Society of Washington (issued sporadically) are available. Correspondence dealing with membership and subscriptions should be sent to the Biological Society of Washington, P.O. Box 1897, Lawrence, Kansas 66044, U.S.A. Payment for membership is accepted in US dollars (cash or postal money order), checks on US banks, or MASTERCARD or VISA credit cards. Manuscripts, corrected proofs, editorial questions should be sent to the Editor, Biological Society of Washington, National Museum of Natural History, Smithsonian Institution, Wash- ington, D.C. 20560. Known office of publication: National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. Printed for the Society by Allen Press, Inc., Lawrence, Kansas 66044 Second class postage paid at Washington, D.C., and additional mailing office. POSTMASTER: Send address changes to PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON, P.O. Box 1897, Lawrence, Kansas 66044. This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(3):373-387. 1995. Nomenclature of the Hawaiian Akialoas and o Abstract.—The generic names Hemignathus Lichtenstein, 1839, and Het- . erorhynchus Lafresnaye, 1839, often used for the akialoas and the nukupuus, respectively, have the same type species (Hemignathus lucidus) so if the akialoas are given the status of a separate genus or subgenus, there is no generic-level name available for them. From a review of evidence concerning dates of pub- lication we conclude that for nomenclatural purposes Hemignathus and Het- erorhynchus were published simultaneously and Hemignathus has precedence according to the first reviser principle. The correct citation for the Oahu Akialoa is shown to be Drepanis ellisiana G. R. Gray (1859) rather than Hemignathus lichtensteini Wilson (1889). The name Hemignathus stejnegeri Wilson (1889) has priority over H. procerus Cabanis (1890) and is restored for the Kauai Akialoa. As no other name is available for akialoas, the new generic name Akialoa is proposed here (type species Certhia obscura Gmelin). The new names resulting from these nomenclatural changes are listed. In the spectacular Hawaiian radiation of cardueline finches of the tribe Drepanidini, there are few more distinctive birds than the akialoas and nukupuus, which consti- tute the genus Hemignathus in the sense of Amadon (1950) and other authors, before and after him. The akialoas comprise five named taxa (one fossil) of medium-sized to large drepanidines with very long, decurved bills, the upper and lower parts of which are of nearly equal length (Fig. 1). Similar prob- ing bills have evolved in various other groups of arboreal birds such as the babblers (Xiphirhynchus: Timaliidae), sunbirds (A4r- achnothera: Nectariniidae), woodcreepers (Campyloramphus: Dendrocolaptidae), and the woodhoopoes (Phoeniculidae). In the nukupuus (four named taxa, including the akiapolaau of Hawaii— Heterorhynchus wil- soni Rothschild), the upper jaw is likewise prolonged into a long, decurved probe, but the lower is much shorter (Fig. 1) and is used for pounding, prying, and pecking. The bill morphology of nukupuus is unique and has no parallel among other birds. Unfortunately, upon these birds are heaped some of the most convoluted no- menclatural problems, at both the generic and specific levels, to be found in any group of Hawaiian birds. Numerous authors, par- ticularly in the earlier literature, maintained Hemignathus and Heterorhynchus as dis- tinct genera (e.g., Rothschild 1893d, Bryan 1901). Since the revision of Amadon (1950), however, Heterorhynchus has usually been treated as a subgenus of Hemignathus (e.g., Greenway 1968, American Ornithologists’ Union 1983). As we shall see, this is no- menclaturally incorrect, as both generic names have the same type species. Pratt (1979) greatly expanded the genus Hemignathus by including in it the ama- kihis (Loxops virens, L. parva, and L. sag- ittirostris sensu Amadon 1950), which he placed in the subgenus Viridonia Rothschild 1892). At the same time he continued to 374 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON AKIALOAS Genus Akialoa, new genus a, NUKUPUUS A. obscura MOLOKAI ELS, <—s KAHOOLAWE HAWAIl Genus Hemignathus Lichtenstein H. lucidus H. wilsoni Fig. 1. Map of the Hawaiian Islands with outlines of head and bill (from Bryan 1901) of a representative akialoa (Akialoa obscura = Hemignathus obscurus auct.) and of the nukupuus Hemignathus lucidus and H. wilsoni. recognize Hemignathus and Heterorhyn- chus as separate subgenera. Although the portion of Pratt’s disserta- tion pertaining to this problem remains un- published, his nomenclature was adopted first by Berger (1981) and then by the Amer- ican Ornithologists’ Union (1983) in their Check-list, whereupon Pratt’s treatment be- came entrenched among non-taxonomists without any consideration having been giv- en to its merits. This lumping created a many new combinations that had never appeared in the entire history of Hawaiian ornithol- ogy, including two homonyms of well- known species that had to be re-named (Pratt 1979a, 1979b, 1989; Olson & James 1988). Based on myology and osteology (Olson & James 1988), we would maintain the ama- VOLUME 108, NUMBER 3 kihis, akialoas, and nukupuus as at least three genera, with the amakihis being re- ferred to the genus Loxops Cabanis. This is supported by recent genetic studies, with the proviso that the the akialoas have not yet been analyzed and that the amakihis may need to be further split (R. Fleischer, pers. comm.) It is not our intention to deal with systematic problems here, but rather to clear up some long-festering points concerning the nomenclature that should be used for these birds by those who would place them in separate generic-level taxa. The Type Species of Hemignathus The genus Hemignathus as first proposed by Lichtenstein (1839) included both an ak- ialoa and a nukupuu. The specimens avail- able to him were all from Oahu, collected by Ferdinand Deppe (Olson & James 1994a). The akialoa he considered to be the same as that of Hawaii (Certhia obscura Gmelin), whereas the nukupuu was obviously a new species and was given the name Hemigna- thus lucidus. Stejneger (1887:93, footnote) and Roths- child (1893d:87) each tried to argue that Lichtenstein’s wording made H. obscurus the type of the genus Hemignathus. Ma- thews (1930) merely followed Rothschild, as apparently did Amadon (1950:168), who stated that the type was Certhia obscura by “original designation.” The following is what Lichtenstein wrote (translation from Roths- child 1893d:87) that bears on the matter: Latham describes in the genus Certhia a species from the Sandwich Islands, which struck him particularly in so far that the lower mandible was a quarter of an inch shorter than the upper. This species was included in the system under the name of Certhia obscura; but it is so distinct from all the different subgenera into which this group has rightly been divided, that one is forced to make a new genus for it, for which I propose the name [Hemignathus]. All this says is that of the previously rec- ognized species assigned to Certhia, Lich- tenstein considered C. obscura to differ to the extent that it must be put in another 375 genus. He did not say that C. obscura was the only species to be included in this genus, as indeed he also placed his new species H. lucidus in it, and he did not specifically in- dicate either species as the type. “‘Mention of a species as an example of a genus or subgenus” does not constitute a type des- ignation (ICZN 1985:Article 67a). Shortly after Lichtenstein’s publication, G. R. Gray (1841) clearly designated H. lucidus as the type of the genus Hemignathus. Except for the authors mentioned above, all other au- thorities appear to have accepted this des- ignation (e.g., Sharpe 1885, Bryan & Green- way 1944, Greenway 1968, American Or- nithologists’ Union 1983). In the meantime, another generic name pertinent to these birds appeared nearly si- multaneously with Lichtenstein’s when Laf- resnaye (1839) described the Oahu Nuku- puu from specimens brought back by the voyage of the Venus as Mellithreptus (s.g. [subgenus] Heterorhynchus) olivaceus. A\l- though several subsequent authors were confused as to the identity of Lafresnaye’s species olivaceus, and even used this name for the Akiapolaau (Heterorhynchus wilsoni Rothschild) of Hawau, it was the same tax- on as Lichtenstein’s /ucidus (Newton 1887, Bangs 1930). The species question aside, the name Het- erorhynchus was seized upon and used by most subsequent authors, either as a genus or a subgenus, to distinguish the nukupuus from the akialoas. Steyneger (1887:93) ap- pears to be the first to have articulated this sentiment: Generally this bird [the Kauai Akialoa] is referred to the same genus as Hemignathus lucidus, but with doubtful propriety I think. The bills in this group of birds have served as the chief character for the es- tablishment of genera, and if we recognize more than one genus of Drepanine birds, the two species of Heterorhynchus with their unique bills should cer- tainly stand alone. Virtually all authors since have main- tained the akialoas and nukupuus as distinct genera or subgenera, under the names 376 Hemignathus and Heterorhynchus, respec- tively, while ignoring the fact that these ge- neric-level taxa have the same type species. A half-century ago, Bryan & Greenway (1944:128), who merged the two groups, succinctly summarized the nomenclatural consequences for those who would separate them: In our opinion, the group with long lower mandibles and those with short lower mandibles may well be considered as congeneric. If it is desired to separate them generically then the former will require a new generic name and the latter (heretofore known as Heterorhynchusj/)] will have to be called Hemigna- thus, since the two groups, as named heretofore, have the same type as designated by Gray and Lafresnaye. Rothschild’s arguments [1893d:79] have no force under the rules of zoological nomenclature (Art. 30, II, g). The message apparently never sank in. In the influential Check-list of Birds of the World, Greenway (1968), who must have forgotten what he had written previously, maintained Hemignathus and Heterorhyn- chus as separate subgenera within the genus Hemignathus, and then dutifully listed H. lucidus as the type for each! The equally influential Check-list of North American Birds (American Ornithologists Union 1983) is a later reference that acknowledges that these two generic-level taxa have the same type species. So the fact remains that if one wishes to separate the akialoas nomenclaturally, a new generic-level name will be required. After an exhaustive search of the literature of Ha- wailan birds, we found that there is no pre- viously existing supraspecific name avail- able that would be desirable to use. One very obscure name that still seems to be in limbo needs to be disposed of, however. This is the genus Falcator Temminck, 1821: 108, proposed in a footnote to an article otherwise totally unconnected to Hawaii: *“Ce nouveau genre se compose des Certhia pacifica, obscura, coccinea, et falcata de Linn. Gmel.” G. R. Gray (1869), Giebel (1875), and Dubois (1901) all listed Falcator as a syn- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON onym of Drepanis Temminck, 1820 (type Certhia pacifica Gmelin by designation of Gray, 1840), doubtless because it com- prised the very same species that were in- cluded in that genus as proposed by Tem- minck himself (1820) the preceding year. The diligent Richmond (1908) uncovered the name Falcator, but merely listed the species included by Temminck without mentioning a type designation. Dubois (1901), by assigning all the other species originally included in Falcator to other gen- era, would have made Certhia pacifica the type of the genus by elimination, but this does not constitute type fixation (ICZN 1985: Article 69b). If Falcator were restrict- ed to the Hawaii Akialoa, which is one of the included species, this could have very undesirable consequences if the akialoas were combined with practically any other taxon because the virtually unknown name Falcator would have priority over all genera of Drepanidini except Drepanis Temminck, 1820, and Psittirostra Temminck, 1820. To eliminate any potential problems that the name might thus cause, we formally des- ignate Certhia pacifica Gmelin as the type species of Falcator Temminck, 1821, so that Falcator then becomes a pure objective syn- onym of Drepanis Temminck, 1820. Dates of Publication of Hemignathus and Heterorhynchus The competing generic and specific names Hemignathus lucidus Lichtenstein versus Heterorhynchus olivaceus Lafresnaye are each now usually cited with the date of 1839 (e.g. Greenway 1968). In attempting to re- solve the nomenclature of this group it thus becomes important to try to determine why Lichtenstein’s names have traditionally been given precedence and whether they actually have priority over those of Lafresnaye. Lichtenstein’s paper is in the Abhandlun- gen of the Berlin Academy of Science for 1838, a serial that was traditionally pub- lished later than the year indicated. The an- VOLUME 108, NUMBER 3 nouncement at the beginning of Lichten- stein’s paper, which is mainly about Cali- fornia birds, states that it was read before the Academy on 27 June 1837. Hanna (1931) has pointed out that there was no meeting on that date and that the reading must have taken place on 25 May 1837. But Lichtenstein cannot have communicated anything about Hawaiian birds at that time because the collector Deppe, from whom he obtained them, did not return to his home- land until 1838, as stated elsewhere in the Lichtenstein paper (p. 448). Some early au- thors, however, have cited Lichtenstein’s names as dating from 1838 or even 1837, both of which are shown here to be erro- neous. The introductory material at the front of the Abhandlungen for 1838 has a title page that is dated 1839 and contains an annual report that carries through to the end of 1838. The volume was issued in three sep- arate sections: Physikalische (in which Lich- tenstein’s paper appeared); Mathematische, and Philologische und Historische Abhan- dlungen. Each of these parts has a separate title page, each of which is dated 1840. Thus the ostensible date of publication of Lich- tenstein’s paper is 1840, which Hanna (1931) thought to be correct. We have found no unequivocal evidence, such as library stamps, correspondence, or other archival sources that mention the receipt or appear- ance of this publication. We examined cop- ies in the Vetenskaps Akademiens Bibliotek in Stockholm, the Linnean Society of Lon- don, the Smithsonian Institution, and the British Museum (Natural History) but none is Stamped with a contemporary date of re- ceipt. There is a pencilled annotation in the British Museum copy, however, to the effect that the paper by Lichtenstein (and three others mentioned) was published in 1839 and that the proper citation should be “1838 (1840) [1839], meaning, we presume, that the volume is for 1838, with the ostensible date of 1840, but the actual publication date of 1839. 377 Enquiries made on our behalf at the Ber- lin Botanical Museum brought the response from the librarian there ‘Artikel ersch. 1840!” but with no further documentation (received in litt. from Harald Pieper, Kiel), so that this may have been based on the title page from the Physikalische Abhandlungen. What seems to be the most definitive source we have uncovered is an extensive review of the publication and contents of the Ab- handlungen der Berliner Academie in the Isis von Oken (1844, Heft XI: columns 842- 854) in which the years of appearance of the volumes is as follows: 1832=1834, 1833=1835, 1834=1836, 1835=1837, 1837=1839, 1838=1839, 1839=1841, 1840=1842, 1841=1843. Thus the volumes consistently ran two years behind except 1838, which supposedly appeared in 1839, with no volume being published in 1840. One wonders whether even this seemingly definitive indication of an 1839 publication date for the 1838 volume may be due to a misprint, as it seems unusual to have two volumes of this normally tardy serial ap- pearing in the same year. One would assume that if the 1838 volume were indeed issued in 1839, it would have been late in the year. We provisionally accept the 1839 date from the evidence of the review in Isis von Oken and the annotation in the British Mu- seum copy, each of which could be equiv- ocated. There is certainly no available ev- idence by which a more specific date can be refined, so that according to the rules of nomenclature (ICZN 1985: Article 21c) the date of Lichtenstein’s paper must be fixed as the last day of the year—31 December 1839. Otherwise we must resort to the os- tensible year of publication of 1840, like- wise with the date of 31 December. The Magasin de Zoologie, in which Laf- resnaye’s name Heterorhynchus olivaceus appeared, was issued in two series (‘‘sec- tions’’) of livraisons—the first for verte- brates and the second for invertebrates. The original wrappers for the first section that are bound with the volumes in the Smith- 378 sonian Institution Libraries have “ Livrai- son.— Année 18 ”’ set in type, with the num- ber of the livraison and the year being writ- ten in ink by hand. The numbers of the plates in each livraison were also indicated by hand beside the appropriate printed cat- egories of mammals, birds, reptiles, and fishes. According to the wrappers, the text and plate (10) for Heterorhynchus olivaceus appeared in the sixth and final livraison for 1839, along with plates 7—9 of birds (all of Lafresnaye), and 11-19 of mammals. Laf- resnaye’s plates and the recto pages of text bear the date 1839. Each of Lafresnaye’s contributions in the sixth livraison bears the date October 1839 on the verso, after his signature, but this is obviously the date the manuscripts were completed, because the seventh livraison, with the wrapper dated 1840, contains one plate (11) and text by Lafresnaye which still has October 1839 on the verso, but 1840 on the recto and the plate. The only contemporary source that we have found that might bear on the date of issue of one of the livraisons for 1839 is a notice in the Edinburgh Journal of Natural History for February 1840, which states: “The 7th Number, for the year 1839, of this elegant work has just appeared, and is oc- cupied with a description” by Geoffroy Saint-Hilaire of three new genera of birds from Madagascar. This is puzzling, because according to the wrappers in the copy we examined, that article was in the fifth liv- raison, not the seventh, the latter having the date 1840. An additional consideration is that the livraisons may not have been issued in sequence. In a card file concerning dates of publication that was assembled by C. W. Richmond in the Division of Birds, Smith- sonian Institution, there is an otherwise un- documented note saying ““Mag. de Zool. Livr. 1-15 1840—Livr. 17-18 1841.” Thus we have two extremely equivocal sources that suggest that the date of publication of Heterorhynchus may in fact be 1840. Nevertheless, all references that we have PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON seen give the date of publication of Heter- orhynchus olivaceus as 1839. The original description of Ampelis (=Xipholena) la- mellipennis Lafresnaye, which immediately preceded that of H. olivaceus and was issued in the same livraison, is also always cited with the date 1839 (e.g., Sclater 1883; Hell- mayr 1929). Therefore, in the absence of concrete evidence to the contrary, we con- sider that the year of publication of Heter- orhynchus olivaceus must be taken as 1839, but in the absence of further evidence, its date of publication must also be fixed as 31 December of that year. Thus, for purposes of nomenclature, the names Hemignathus and Heterorhynchus are here regarded as having been proposed simultaneously so that their relative precedence must be deter- mined according to the first reviser principle (ICZN 1985: Article 24). Hemignathus has traditionally been given precedence over Heterorhynchus through- out the literature of Hawaiian birds. This doubtless arose largely through various au- thors incorrectly taking the date of publi- cation of Hemignathus as being 1838 (e.g., Sharpe 1885). Gray (1841) appears to be the first author to have formally synonymized Heterorhynchus with Hemignathus, but he cannot be considered to be the first reviser because he gave the latter the publication date of 1838 and therefore did not regard the names as having been published simul- taneously. There are other historical reasons why Hemignathus was given precedence over Heterorhynchus. When Lafresnaye pub- lished his description of Heterorhynchus olivaceus, it apparently created bad blood between him and the naturalists of the voy- age of the Venus, who had collected the spe- cies and had deposited specimens at the Paris museum in August 1839 (Neboux 1840). Lafresnaye (1840) protested that he had purchased his specimen from a dealer named Dupont (who we imagine had probably re- ceived material originating in the voyage of the Venus as well) and had been ignorant of VOLUME 108, NUMBER 3 the fact that the museum had received the same species. He still maintained that the date of publication was October 1839, but went on to say that the ornithologist Nat- terer, when visiting Lafresnaye the previous month (October 1840), had informed him that the bird had already been described in Germany or Russia under a different generic name, which must be an allusion to the Lichtenstein publication. What Natterer’s reasons were for thinking that Lichten- stein’s publication had priority are no lon- ger clear, there being now no copy of that volume of the Abhandlungen in the library in Vienna where Natterer worked (Ernst Bauernfeind, Naturhistorisches Museum Wien, in litt. 28 Dec 1994). It should be recalled that Lafresnaye was used to having his names pre-empted because, as Bangs (1930:152) has remarked: Lafresnaye lived in the country, in those days a real journey away from Paris, and, therefore, was often just a little later than some one else in securing some new bird. Also, I fancy, published descriptions were slow in reaching him. Several times I have read a complaint to that effect written by him on a label. This may explain why Lafresnaye himself may have been willing to relinquish further claim to the priority of his name. Some years later, in the official report on the zoology of the voyage of the Venus, the old rivalry with Lafresnaye resurfaced when Prévost and des Murs (1849) obviously set out to administer the ‘coup de grace’ to Het- erorhynchus. They titled a separate section of their account of the birds of the voyage “Notice sur le Genre Hemignathe (Hem- ignathus, Lichtenstein), (1837). Heteror- hynchus (La Fresnaye) 1839.” Here (p. 183) they advance the 1837 date that part of Lichtenstein’s publication was read before the Academy, but then take Lafresnaye to task for having “‘forgotten”’ that Hemigna- thus had already been published in 1838 (pp. 185-186). And yet a bit further on they repeatedly give the publication date as 1839 (pp. 191-192), including in their account of the ““Hémignathe brillant,” where they list 379 Lafresnaye’s citation after Lichtenstein’s. This order of precedence has been followed ever since. Although it is not at all certain in which year Prevost and Des Murs (1849) actually believed Hemignathus to have been published, they are the first authors to as- sociate it with the year 1839 and simulta- neously to give it precedence over Heter- orhynchus, so perhaps they can be consid- ered first revisers. Should our activity in this connection make us the first revisers, we follow tradi- tion and accord precedence to Hemigna- thus. This has the unfortunate consequence of having this name, long associated with the akialoas when those birds were sepa- rated generically, attach to the nukupuus. But it has the advantage of conserving the well-known name /ucidus Lichtenstein over the long unused name olivaceus Lafresnaye. Furthermore, the scanty and inconclusive evidence as yet available suggests that if fur- ther information regarding dates of publi- cation should surface, Hemignathus is the name more likely to have actual priority. The Name for the Akialoa of Oahu G. R. Gray (1859:9), who long believed the Hawaii Akialoa to be the female of the liwi, Drepanis (Vestiaria) coccinea, named a new species of drepanidine as Drepanis (Hemignathus) ellisiana based on four ref- erences, including Lichtenstein (1839), who had included a specimen from Oahu under the name of the Hawaiian bird, H. obscurus (Gmelin). As was later established, Gray’s species was clearly a composite, as the other three references pertained to the bird from Hawaii (=H. obscurus). Nevertheless, Gray’s intention should have been clear, as he gave the “habitat”? of his new species as Oahu, and Lichtenstein’s reference was the only one unequivocally linked to that island. Wilson (1889), in his review of Hemi- gnathus, described the Oahu species as H. lichtensteini, based on the same specimen in Berlin that was figured by Lichtenstein 380 (1839), but without reference to Gray’s H. ellisiana. The following interchange be- tween the two rival camps engaged in de- scribing new Hawaiian birds at the end of the 19th century provided as much sound as light on the matter, but summarizes the opinions that have been expressed. G. R. Gray, who in more than one case erroneously considered the green birds to be the females of the red, referred H. obscurus partly to the female of Ves- tiaria coccinea, partly to his Drepanis ellisiana, which, therefore, must rank as a synonym. Wilson & Evans (1892:68). Mr. Wilson places Gray’s Drepanis ellisiana as a synonym under Hemignathus obscurus (Gmel.), and gives as his reason that Gray partly referred H. ob- scura to the female of Vestiaria coccinea and partly to his Drepanis ellisiana. ... This is entirely erro- neous, and proves that Mr. Wilson misquoted Gray, for the latter states [1859:9] that Certhia obscura, Gm., is the female of Vestiaria coccinea, while Cer- thia (Hemignathus) obscura Licht. nec Gm., is his Drepanis ellisiana; therefore I regret to have to re- duce Mr. Wilson’s name, Hemignathus lichtensteini, to synonymic rank and to reinstate Gray’s ellisiana. Rothschild 1893d:88. Mr. Rothschild [ibid.] has referred this species to the “Drepanis (Hemignathus) ellisiana”’ of Gray [1859: 9], which I have already correctly quoted as a syn- onym of H. obscurus. It is pretty clear that Mr. Gray never saw a specimen of either, and it is absolutely certain that three out of the four authorities cited by him refer to H. obscurus. Vieillot, the first of them, as I have already shown, figured [Audebert & Vieillot 1802: pl. 53] the very specimen, now at Liverpool, which was formerly in the Leverian Museum, and actually the type of Latham’s description, on which was founded the Certhia obscura of Gmelin, and hence the H. obscurus of modern ornithologists. Sim- ilarly the bird figured in Ellis’s unpublished drawings (no. 28), which from the name used by Gray is doubt- less to be regarded as the type of his supposed species, is most unquestionably H. obscurus, as anyone who examines the drawing in the British Museum may satisfy himself. The last of the authorities cited by Mr. Gray is Cassin, and he quotes Peale as saying that the species he speaks of was obtained in “‘Hawaii only,” and that according to his observations it did “not inhabit Oahu; it was accordingly also H. ob- scurus; and the mere fact of Mr. Gray’s mistakenly referring Lichtenstein’s figure, and assigning Lich- tenstein’s locality, to the so-called ’’Drepanis (Hem- ignathus) ellisiana‘* cannot remove the incontestable objection that his other references show it to be but PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON a synonym of H. obscurus. Wilson & Evans (1894: 65-66). Both ellisiana and lichtensteini have con- tinued to be used for the Oahu Akialoa, with the former generally being favored, but without any clear resolution of the matter. Because Gray’s name ellisiana is a com- posite, its disposition must be determined by designation of a lectotype, which none of the authors quoted above really do. The fact has been overlooked that Gray himself appears to have resolved the matter long before, as in his Hand-list (Gray 1869:114) under Drepanis ellisiana he gave only the Lichtenstein reference, with ‘“‘obscura, V. O. D. t. 53?” as a dubious synonym, the latter being a reference to Audebert & Vieillot (1802), which, as seen above, was shown by Wilson & Evans to refer to the Hawaii Ak- ialoa. Incidentally, this reference (Gray 1869) shows that Bryan (1901:306, foot- note) erred in considering Gray’s original query (1859) by the Vieillot work to refer to the plate number rather than the identity of the species, which was Bryan’s reason for favoring lichtensteini. Thus, we consider that Gray’s (1869:114) subsequent action re- stricted his species ellisiana to Lichten- stein’s (1839) description of a bird from Oahu, which his original designation of Oahu as the type locality also supports. In the event that others might consider that this still does not constitute proper selection of a lectotype, we specifically designate that of the four references originally given by Gray (1859), the lectotype is the bird from Oahu figured and described by Lichtenstein (1839). Therefore we use Gray’s name e/- lisiana, with H. lichtensteini Wilson as a synonym, for the Oahu Akialoa. The Name for the Akialoa of Kauai The Kauai Akialoa has been widely but unjustly known under the name Hemigna- thus procerus Cabanis. The first specimens to be studied scientifically were collected by Valdemar Knudsen and sent to the Smith- VOLUME 108, NUMBER 3 sonian Institution. They were described by Stejneger (1887), who, in the absence of comparative material, referred them to Hemignathus obscurus. On the basis of specimens he collected himself, Wilson (1889) differentiated the bird from Kauai and honored Stejneger’s contribution by naming it Hemignathus stejnegeri in a pub- lication issued 1 November 1889. In a post- script (p. 402), Wilson added: “‘Prof. Mob- ius [Director of the Royal Zoological Col- lection at Berlin] has also had the goodness to transmit two specimens of ‘Hemignathus procerus, Cab[anis]. n. spec.’ Iam not aware of any published description of this species; but the specimens sent seem to be immature males of that which I have above called H. Stejnegeri.”’ At the Berlin Museum, Jean Cabanis was not idle and proceeded to name the same taxon himself, based on the forementioned specimens from Kauai purchased from a dealer, J. Wentscher, on 2 June 1887, that were doubtless also collected by Valdemar Knudsen (Olson & James 1994a). Cabanis’s description of Hemignathus procerus was published in the October 1889 issue of Jour- nal fur Ornithologie, a periodical that Ca- banis initiated and edited for forty years. During that time the Journal was notorious for appearing considerably later than the stated date of publication, which gave rise to numerous complaints in a day when new taxa were being described at a rapid pace and the law of priority was respected. By 1876 (Anon. Nature 14:309) Cabanis was a “‘well-known sinner” at the “‘evil practice” of antedating his Journal, and he must have put this practice in effect at least as early as 1856 (Peters 1932). It continued for the rest of his tenure, despite continual protesta- tions from his colleagues. In the present in- stance, the original wrappers (copy in Smithsonian Institution Libraries) show that the issue in which the name Hemignathus procerus was proposed could not have ap- peared before January 1890. In a review in which the problem of antedating the Journal 381 fur Ornithologie is addressed at length (Anon. 1891, Ibis, ser. 6, 3:616), it was shown that the number in question was not received by the Zoological Society of Lon- don until 11 July 1890, long after Wilson’s name had been published. In this connec- tion, it is worth recording that Stejneger (1890), in a report that was issued 8 March 1890, referred two new specimens of Kauai Akialoa to Wilson’s species H. stejnegeri, with no mention of H. procerus, so that Wil- son’s paper was obviously already at hand, whereas Cabanis’s was not. Likewise, Wil- son (1890:191) in the Ibis for April 1890 (received at the Smithsonian on 21 April) again states that he had not seen a descrip- tion of H. procerus, which he regarded as “identical with H. stejnegeri.” Doubtless realizing at the time that his name stood to be forestalled, Cabanis (1890: 331) added a footnote to his description to the effect that the name H. procerus had first been published in the newspaper Vossische Zeitung for 14 September 1889. We can only speculate that he may have been comforted by the likelihood that few taxonomists would be able to check this obscure reference, for, as we shall see, the contents thereof would not have been regarded as enhancing either Cabanis’s reputation or the validity of his name. It is certain that Wilson had not seen the Vossische Zeitung when he wrote that “the species is said to have been described’”’ therein (Wilson & Evans 1892:61, foot- note—emphasis added). Furthermore, the carefully chosen wording employed in Wil- son & Evans (1892:61) in relating the events surrounding the description of this bird make it clear that Cabanis had exercised himself considerably to attach his own name to the Kauai Akialoa: As the result of inquiry regarding the various species of the genus in the Berlin Museum, information was received through Professor MO6bius that Professor Cabanis had come to a similar conclusion [that the Kauai birds were distinct from H. obscurus] on in- spection of the specimens there, and had forestalled 382 my proposed title by a few weeks; so that I must at once acknowledge his activity in securing priority for his name H. procerus. This concession of priority to H. procerus can only have been on the strength of Ca- banis’s citation of the Vossische Zeitung ar- ticle, because it must have been well known at the time that the description in Journal fur Ornithologie was published after Wil- son’s name had appeared. Yet if any sub- sequent researcher availed himself of the newspaper article, he kept his knowledge to himself. We were able to obtain a photocopy of the page containing the first use of the name Hemignathus procerus, which ap- peared in No. 429 of Vossische Zeitung, dated 14 September 1889, in a notice en- titled “‘Allgemeine deutsche ornitholo- gische Gesellschaft,” being the minutes of the meeting of that society held 9 September 1889. The pages are unnumbered, but the notice occupies about two-thirds of the middle column of the page on which it oc- curs and is initialled ““P. M.”’, for Paul Mat- schie, who was secretary of the D.O.G. from 1894 to 1907 (Prestwich 1958). The text of the portions of this notice in the original German and in English translation that per- tain to Cabanis’s discussion of Hawaiian birds follows in it’s entirety. Herr Prof. Dr. J. Cabanis beschrieb eine neue Abart eines merkwirdigen Vogels von den Sandwich in- seln. 1837 hatte der bekannte Botaniker Doppe [sic], der lange in Mexiko gesammelthatte, einige merk- wurdige kleine, sehr krummschnablige V6gel mit- gebracht, deren feuerrothe Federn den Eingeborenen zur Anfertigung von Manteln dienen. Die Thierchen haken von unten mit den fast halbkreisformigen Schnabel in den Kelch der Bliithen, um so zu den Dem Blithensaft nachgehenden Insekten zu gelan- gen. Die Sandwichinseln scheinen, ahnlich wie Neu- Zeeland, Madagaskar und Australian, einen eigenen Schopfungsheerd zu bilden und weisen ganz wud- ersame Vogelformen auf. Herr Cabanis nannte die vorliegende Art Hemignathus procerus. Herr Prof. Dr. J. Cabanis described a new variety of a remarkable bird from the Sandwich Islands. The well-known botanist Doppe [sic = Deppe], who was collecting for a long time in Mexico, brought along, in 1837, a few remarkable small birds with a strongly PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON curved beak, which provide the aboriginals with bright red feathers that are used for making cloaks. The creatures peck the calyx of blossoms from below with their almost semicircular beak to reach the nec- tar-seeking insects. The Sandwich Islands, like New Zealand, Madagascar and Australia, appear to con- stitute a separate center of creation and show won- derful forms of birds. Herr Cabanis called the species in question Hemignathus procerus. Cabanis’s claim to priority for H. procerus based on the above notice thus falls to the ground. Probably through Matschie, Ca- banis’s message had become at best garbled, with some of his introductory remarks about drepanidines being combined with the men- tion of his new name. The only bird de- scribed here possessed a “strongly curved beak”’ and bright red feathers, which can only apply to the liwi, Vestiaria coccinea (Forster). Furthermore, the mention of Deppe is irrelevant to the description of a bird from Kauai, because Deppe collected only on Oahu (Olson & James 1994a). By strict application of the rules of nomencla- ture, Hemignathus procerus Cabanis (in Matschie 1889) could be considered a junior synonym of Certhia coccinea Forster, 1781. Nevertheless, because this was clearly not Cabanis’s intent, a more charitable assess- ment would be to regard the first use of Hemignathus procerus as a nomen nudum. It is also a nomen nudum as published by Schalow on or after 7 November 1889, in another account of the meeting of the Ge- sellschaft (Schalow 1889). As we have seen, it was also published twice as a synonym by Wilson (1889, 1890), and only months later was the name H. procerus ever asso- ciated with a legitimate description. Wil- son’s name Hemignathus stejnegeri clearly has priority over H. procerus Cabanis and we have reverted to it. The Resulting Nomenclature of Akialoas and Nukupuus The nukupuus consist of two very distinct species (Olson & James, 1994b), the Nu- kupuu proper (Hemignathus lucidus), and VOLUME 108, NUMBER 3 the Akiapolaau (Hemignathus wilsoni). There are three named taxa of the former that are now considered to be subspecies of a single species, a treatment for which we can as yet offer no contrary evidence, as the only stated differences are minor plumage variations in adult males, and there is some doubt that a specimen of adult male even exists for the Oahu bird (there may be size differences between the forms, however— Thane Pratt, in litt.). Pending more detailed revisionary work we continue to rank these taxa as subspecies. Hemignathus Lichtenstein Hemignathus Lichtenstein, 1839 (31 De- cember). Type, by subsequent designa- tion (Gray 1841), Hemignathus lucidus Lichtenstein. The name has precedence over Heterorhynchus Lafresnaye accord- ing to the first reviser principle. Heterorhynchus Lafresnaye, 1839 (31 De- cember). Type, by monotypy, Heteror- hynchus olivaceus Lafresnaye 1839 = Hemignathus lucidus Lichtenstein. Hemignathus lucidus lucidus Lichtenstein Oahu Nukupuu Hemignathus lucidus Lichtenstein, 1839 (31 December) :451. The name has prece- dence over Heterorhynchus olivaceus Laf- resnaye according to the first reviser prin- ciple. Heterorhynchus olivaceus Lafresnaye, 1839 (31 December):text accompanying plate 10. Distribution and status. —Island of Oahu, extinct, evidently fairly abundant in 1837 but not collected thereafter. Hemignathus lucidus hanapepe Wilson Kauai Nukupuu Hemignathus hanapepe Wilson, 1889:401. Distribution and status. —Kauai, where 383 now presumed extinct, there having been no sightings since the 1970’s. Hemignathus lucidus affinis Rothschild Maui Nukupuu Hemignathus affinis Rothschild 1893a:112. Distribution and status. — Known histor- ically only from the upland forests of Maui, where it may still exist in extremely low numbers. Hemignathus lucidus subspp. indet. A historic specimen of this species, of in- determinate race, was collected on the is- land of Hawaii by the U.S. Exploring Ex- pedition in 1840 or 1841 (Olson & James 1994b), but the species was never again taken on that island. A fossil aimost certainly of this species was also recovered from sand dune deposits on Molokai (Olson & James 1994b). Hemignathus wilsoni (Rothschild) Akiapolaau Hemignathus olivaceus.— Wilson, 1889 (nec Heterorhynchus olivaceus Lafresnaye— Wilson was the first to recognize this spe- cies but erred in thinking that Lafres- naye’s name applied to it). Heterorhynchus wilsoni Rothschild, 1893d: 95 (key), 97. Hemignathus munroi Pratt, 1979b:1581 (new name for Heterorhynchus wilsoni Rothschild, 1893d, preoccupied by Him- atione wilsoni Rothschild, 1893c, if these taxa are regarded as congeneric). Distribution and status.—Known only from the island of Hawaii, where it is con- sidered endangered but is locally distributed in fair numbers. As we have detailed above, a new generic name is needed for the akialoas. Because there has already been so much nomencla- tural confusion generated with regard to 384 these birds, we propose to use the Hawaiian name, akialoa, which is the only name now in use for these birds that conveys an un- equivocal meaning, as a formal generic name. The four historically known taxa of akialoas were treated as full species in the early literature, but have also been listed as subspecies of a single species (e.g., Bryan & Greenway 1944). Amadon (1950) divided them into two species, keeping the Kauai birds separate because of their large size and treating the birds of Oahu and Lanai (known from a total of 5 specimens) as subspecies of obscurus of Hawaii. We find that the birds of Oahu and Lanai are closer in size to those of Kauai, whereas obscurus of Hawaii is dis- tinctly smaller than any of the others. The amount of variation between the four pop- ulations seems too great to be encompassed by a single species, but it is not yet clear how many species should be admitted, nor what the contents of each should be. Fur- thermore, we now know that the historically known akialoas were sympatric with anoth- er fossil species at least on Kauai and Oahu (James & Olson, 1991), two fossil species of akialoa appear to have been sympatric on Maui (James & Olson, unpublished data.), and yet another unnamed large spe- cies was sympatric with obscurus on Hawaii (unpublished data). Until these systematic questions can be resolved, for nomencla- tural purposes, we provisionally treat each taxon of akialoa as though specifically dis- tinct. Akialoa, new genus Type species.—Certhia obscura Gmelin, 1789. Etymology. —The Hawaiian name for these birds. Hawaiian words do not have gender but as the present name ends in “a”’ we arbitrarily treat it as feminine. Of the trivial epithets now in use, this effects the spelling only of the type species, whose orig- inal name, Certhia obscura, is feminine in any case. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Included species. —The following are the new combinations resulting from the intro- duction of the new generic name, with their principal synonyms: Akialoa obscura (Gmelin), new combination Hawai Akialoa Certhia obscura Gmelin, 1789:470. Distribution and status.— First collected on the third voyage of Captain James Cook in 1779 near Kealakekua Bay, island of Ha- wall (Medway 1981). It was collected rather frequently on that island in the 1 9th century but became extinct about 1900 (Scott et al. 1986). Akialoa lanaiensis (Rothschild), new combination Maui Nui Akialoa Hemignathus lanaiensis Rothschild, 1893b: 24. Distribution and status.—Known from three skin specimens taken in 1892 on the island of Lanai; never taken again. Extinct. Fossils of akialoas, most likely of this taxon, have been collected on Molokai and Maui (Olson & James 1982, James & Olson 1991 and unpublished data). Akialoa ellisiana (Gray), new combination Oahu Akialoa Drepanis (Hemignathus) ellisiana Gray, 1859:9. Hemignathus lichtensteini Wilson, 1889: 401. Distribution and status. —This taxon was long believed to be known only from the holotype in the Berlin Museum collected in the Nuuanu Valley, Oahu, in January 1837 by Ferdinand Deppe (Olson & James 1994a). We have identified a second specimen in the collections of the Philadelphia Academy VOLUME 108, NUMBER 3 taken at the same time and place by J. K. Townsend. The species was not encoun- tered on Oahu after 1837 and is now extinct. Akialoa stejnegeri (Wilson), new combination Kauai Akialoa Hemignathus stejnegeri Wilson, 1889:400. Hemignathus procerus Cabanis, 1890:331. Distribution and status. —Known from many specimens taken on Kauai in the 19th century. The last specimen was obtained in July 1960 (Richardson & Bowles 1964), the last observation was in 1965, and the spe- cies is now considered to be extinct (Scott et al. 1986). Akialoa upupirostris (James & Olson), new combination Hoopoe-billed Akialoa Hemignathus upupirostris James & Olson, 1991:60. Distribution and status. —Not known his- torically. The species was described from Holocene fossils from Makawehi dunes, Kauai, and from Barbers Point, Oahu. Acknowledgments For bibliographic assistance of various sorts we are most grateful to Harald Pieper (Zoologisches Museum, Kiel—especially for his providing a photocopy of a crucial page from Vossische Zeitung), Leslie Overstreet (Smithsonian Institution Libraries, Wash- ington), Ernst Bauernfeind (Naturhisto- risches Museum, Vienna), Gina Douglas (Linnean Society, London), William Lind- say (Natural History Museum, London), and Per Ericson (Swedish Museum of Natural History, Stockholm). We also thank An- drzej Elzanowski for translating the passage from Vossische Zeitung, M. Ralph Brown- ing for supplying numerous references to the problems of dating the Journal fur Orni- 385 thologie, and Richard C. Banks for com- menting on several versions of the manu- script. Literature Cited Amadon, D. 1950. The Hawaiian honeycreepers (Aves, Drepaniidae). — Bulletin of the American Museum of Natural History 95:151-262. American Ornithologists’ Union. 1983. Check-list of North American birds. Sixth ed. American Or- nithologists’ Union, [Washington, D.C.], 877 pp. Audebert, J. B., & L. P. Vieillot. 1802. Oiseaux dorés ou a reflets métalliques. Vol. 2. Histoire natu- relle et générale des grimpereaux et des oiseaux de paradis. Imprimerie de Crapelet, Paris, 164 pp. Bangs, O. 1930. Types of birds now in the Museum of Comparative ZoGlogy.— Bulletin of the Mu- seum of Comparative Zoology 70:149-426. Berger, A. J. 1981. Hawaiian birdlife. 2nd ed. Uni- versity Press of Hawaii, Honolulu, 260 pp. Bryan, E. H., Jr.,. & J. C. Greenway. 1944. Contri- bution to the ornithology of the Hawaiian Is- lands. — Bulletin of the Museum of Comparative Zoology 94:77-142. Bryan, W.A. 1901. Akey to the birds of the Hawaiian group.— Memoirs of the Bishop Museum 1:258— 332: Cabanis, J. 1890. [Description of Hemignathus pro- cerus].—Journal fiir Ornithologie 37 [for Oc- tober 1889]:331. Dubois, A. 1899-1902. Synopsis avium. Nouveau manuel d’ornithologie. Part 1. H. Lamertin, Brussels. Giebel, C. G. 1875. Thesaurus ornithologiae. Vol. 2. F. A. Brockhaus, Leipzig, 787 pp. Gmelin, J. F. 1789. Systema naturae. Vol. 1, part 1. Georg. Emmanuel Beer, Leipzig, 500 pp. Gray, G. R. 1841. A list of the genera of birds, with their synonyma, and an indication of the typical species of each genus. Second ed. Richard & John Taylor, London, 115 pp. 1859. Catalogue of the birds of the tropical islands of the Pacific Ocean in the collection of the British Museum. British Museum, London, 72 pp. . 1869. Hand-list of genera and species of birds, distinguishing those contained in the British Museum. Part 1. Accipitres, Fissirostres, Ten- uirostres, and Dentirostres. British Museum, London, 404 pp. Greenway, J.C. 1968. Family Drepanididae. Pp. 93- 103. InR. A. Paynter, Jr., ed., Check-list of birds of the world. Vol. 14. Museum of Comparative Zoology, Cambridge, Massachusetts, 433 pp. 386 Hanna, G. D. 1931. Lichtenstein on California birds.— Condor 33:21 1-213. Hellmayr, C.E. 1929. Catalogue of birds of the Amer- icas.— Field Museum of Natural History Zoo- logical Series 13, part 6. 1985. International code of zoological no- menclature. 3rd edition. International Trust for Zoological Nomenclature, London, 338 pp. James, H. F., & S. L. Olson 1991. Descriptions of thirty-two new species of birds from the Ha- waiian Islands: Part II. Passeriformes. —Orni- thological Monographs 46:1-88. Lafresnaye, F. de. 1839. [Description of Heterorhyn- chus olivaceus].—Magasin de Zoologie 1839 [Oiseaux]: plate 10, 2 unnumbered pages of text. 1840. [Corrections and additions to Neboux 1840].—Revue Zoologique 1840:321-325. Lichtenstein, M. H. K. 1839. Beitrag zur ornitholo- gischen Fauna von Californien nebst Bemer- kungen uber die Artkennzeichen der Pelicane und uber einige Vogel von den Sandwich-In- seln.—Abhandlungen der KOniglichen Akade- mie der Wissenschaften zu Berlin 1838:417—-451. Mathews, G. M. 1930. Systema avium Australasi- anarum. A systematic list of the birds of the Australian region. Part 2. British Ornithologists’ Union, London, pp. 427-1047. M[atschie], P. 1889. Allgemeine deutsche ornithol- ogische Gesellschaft [minutes of meeting of 9 September 1889].—Vossische Zeitung [Berlin] 429 [14 September]: unpaginated. Medway, D. G. 1981. The contribution of Cook’s third voyage to the ornithology of the Hawaiian Islands. — Pacific Science 35:105-175. Neboux, A. S. 1840. Descriptions d’oiseaux nou- veaux recuilles pendant Il’expedition de la Vén- us.— Revue Zoologique 1840:289-291. Newton, A. 1897. [Exhibit of holotype of Heteror- hynchus olivaceus Lafresnaye].— Proceedings of the Zoological Society of London 1896:990-99 1. Olson, S. L., & H. F. James 1982. Prodromus of the fossil avifauna of the Hawaiian Islands. — Smithsonian Contributions to Zoology 365:1- 59. 1988. Nomenclature of the Kauai Amakihi and Kauai Akialoa (Drepanidini). —Elepaio 48(2):13-14. . 1991. Descriptions of thirty-two new species of birds from the Hawaiian Islands: Part I. Non- Passeriformes. — Ornithological Monographs 45: 1-88. 1994a. A chronology of ornithological ex- ploration in the Hawaiian Islands, from Cook to Perkins. Pp. 91-102 Jn J. R. Jehl, Jr., & N. K. Johnson, eds., A century of avifaunal change in western North America.—Studies in Avian Biology 15. ICZN. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 1994b. A specimen of Nuku pu’u (Hemig- nathus lucidus) from the island of Hawaii (Aves: Drepanidini).— Pacific Science 48:33 1-338. Peters, J. L. 1932. Laterallus Gray antedates Crecis- cus Cabanis. — Proceedings of the Biological So- ciety of Washington 45:119-120. Pratt, H. D. 1979a. A systematic analysis of the en- demic avifauna of the Hawaiian Islands. Un- published Ph.D. dissertation, Louisiana State University, Baton Rouge, Louisiana, 228 pp. 1979b. [Abstract]. A systematic analysis of the endemic avifauna of the Hawaiian Is- lands. — Dissertation Abstracts 40B:1581. 1989. A new name for the Kauai Amakihi (Drepanidinae: Hemignathus).—Elepaio 49(3): 13-14. Prévost, F., and O. des Murs. 1849. Oiseaux. Pp. 177-284 in Abel du Petit-Thouars, ed., Voyage Autour du Monde sur La Frégate La Venus Commandeée par Abel du Petit-Thouars. Zool- ogie. Mammiféres, Oiseaux, Reptiles et Pois- sons. Paris: Gide et J. Baudry. [This account was first issued in 1849, and this particular vol- ume, which constitutes Volume 5, part 1, of the ““Voyage’’, was issued entire in 1855.] Prestwich, A. 1958. “I name this parrot...’ Pub- lished by the author, London, 86 pp. Richardson, F., & J. Bowles. 1964. A survey of the birds of Kauai, Hawaii.— Bernice P. Bishop Mu- seum Bulletin 227:1—-51. Richmond, C. W. 1908. Generic names applied to birds during the years 1901 to 1905, inclusive, with further additions to Waterhouse’s “Index Generum Avium’’.— Proceedings of the United States National Museum 35:583-655. Rothschild, W. 1892. Descriptions of seven new spe- cies of birds from the Sandwich Islands. —An- nals and Magazine of Natural History, series 6, 10(55):108-112. 1893a. Descriptions of three new birds from the Sandwich Islands.— Ibis, series 6, 5(17):112- 114. 1893b. [Description of Hemignathus lan- aiensis|.— Bulletin of the British Ornithologists’ Club 1:24—25. 1893c. [Descriptions of three new Hawaiian birds].— Bulletin of the British Ornithologists’ Club 1(8):41-42. 1893d. The avifauna of Laysan and the neighbouring islands: with a complete history to date of the birds of the Hawaiian possessions. Part 2. R. H. Porter, London, pp. 59-126. Schalow, H. 1889. Allgemeine deutsche-ornithol- gische Gesellschaft zu Berlin (Sitzung vom 9. September 1889).—Die Schwaibe. Mittheilun- gen des ornithologischen Vereines in Wien 13(41):533-535. VOLUME 108, NUMBER 3 387 Sclater, P. L. S. 1883. Catalogue of the birds in the Knudsen.— Proceedings of the United States British Museum. Volume 14. British Museum, National Museum 12 [for 1889]:377-386. London. Temminck, C. J. 1820. Manuel d’ornithologie. Sec- Scott, J. M., S. Mountainspring, F. L. Ramsey, & C. ond ed. Vol. 1. H. Cousin, Paris, 439 pp. B. Kepler. 1986. Forest bird communities of 1821. Account of some new species of birds the Hawaiian Islands: their dynamics, ecology, and conservation.—Studies in Avian Biology 9:1-431. Sharpe, R. B. 1885. Catalogue of the birds in the British Museum. Vol. 10. British Museum, Lon- don, 682 pp. Stejneger, L. H. 1887. Birds of Kauai Island, Ha- waiian Archipelago, collected by Mr. Valdemar Knudsen, with descriptions of new species.— Proceedings of the United States National Mu- seum 10:75-102. 1890. Notes on a third collection of birds made on Kauai, Hawaiian Islands, by Valdemar of the genera Psittacus and Columba in the mu- seum of the Linnean Society.— Transactions of the Linnean Society of London 13:107-130. Wilson, S. B. 1889. On three undescribed species of the genus Hemignathus, Lichtenstein. — Annals and Magazine of Natural History, series 6, 4:400- 402. —. 1890. On some birds of the Sandwich Is- lands. — Ibis, series 6, 2(6):170—-196 ,& A. H. Evans. 1890-1899. Aves Hawaiien- ses: the birds of the Sandwich Islands. R. H. Porter, London, 257 pp. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(3):388-391. 1995. A new lizard of the genus Sphenomorphus (Reptilia: Scincidae) from Mt. Kitanglad, Mindanao, Philippine Islands Walter C. Brown California Academy of Sciences, San Francisco, California 94118, U.S.A. Abstract.—A new lizard, Sphenomorphus kitangladensis, is described from Mindanao, Philippine Islands. It is in Group 3, the Sphenomorphus lateri- maculatus Group. It differs from other species of this group in the separation of the anterior loreal from the labials and, except for Sphenomorphus acutus, the smaller eye relative to snout-length. In color pattern and general appearance, this species is most similar to Sphenomorphus leucospilos and S. laterimaculatus from Luzon Island; but differs from the former in the greater number of scale rows between parietals and base of the tail, and from the latter in the lower number of midbody rows. Brown & Alcala (1980) recognized six species Groups (assemblages) of the scincid genus Sphenomorphus in the Philippines. Group 1 included two Philippine species characterized by very high counts for dorsal scale rows. Group 2 included six species characterized by small size and low lamellar and midbody scale-counts. Group 3 includ- ed four species characterized as relatively slender, intermediate in size (rarely exceed- ing 60 mm in snout-vent length), with 30- 36 (rarely 38—40) midbody scale rows and four supraoculars. Sphenomorphus acutus, a fifth species tentatively included in this group at the time, has six supraoculars and 26-30 midbody scale rows. Group 4 in- cluded five species characterized by a short- er, blunter snout, relatively longer limbs, usually five or six supraoculars, and usually 36-50 plus midbody scale rows. Group 5 included four species characterized by their relatively large size (usually greater than 55 mm snout—vent length) and a more slender, tapered snout than group 4. Group 6 in- cluded one species (S. fasciatus) character- ized by the very distinctive, banded color pattern, and the high number of dorsal scale rows (78-90). A recent sample from the mountain area northeast of Lake Lanao in north-central Mindanao Island differs in several charac- ters from all previously described species of Sphenomorphus. This sample represents a taxon which belongs in Group-3 as defined above. Materials and Methods Material examined included all species of Group 3: leucospilos (2), laterimaculatus (1), victoria (2), acutus (16), mindanensis (20 plus), and kitangladensis (4). Measurements on preserved specimens were determined to the nearest 0.1 mm, us- ing a Helios dial caliper. Head length (HL) was measured from posterior edge of ear opening to tip of snout; head breadth (HB) at the widest point near the angle of the jaws; snout length (SnL) from anterior edge of eye socket to tip of snout; eye diameter (ED) from anterior to posterior edge of socket; tympanum diameter (TD) from anterior to posterior edge; snout-vent length (SVL) from tip of snout to posterior edge of preanals; axilla-groin distance (Ax-Gr) from base of fore limb to base of hind limb; hind limb length from base of hind limb to tip of lon- gest toe. VOLUME 108, NUMBER 3 Scale counts were standardized as fol- lows: lamellae beneath the digits were counted down to the base or to the point where subdigital scales became less than two times the breadth of the solar scales; (2) midbody scale rows (MBSR) were counted at a point midway between fore and hind limbs, and (3) dorsal scale rows (DSR) were counted between the parietals and the row on the base of the tail at the level opposite the vent. Standard nomenclature for the headshields of the Scincidae is followed (Boulenger 1890, Smith 1935, and Brown & Alcala 1980). Sphenomorphus kitangladensis, new species Fig. 1 Holotype. —FMNH 250644, an adult male collected 9 May 1992 on Mt. Kitanglad, Bu- kidnon Province, north-central Mindanao Island by L. R. Heaney. MBSR 38; DSR 76; fourth toe lamellae 15; SVL 56.4 mm; HL 12.5 mm; HB 8.0 mm; SnL 3.7 mm; ED 2.6 mm; Ax-Gr 31.5 mm. Paratypes (3).—FMNH 250641, 250643, CAS 191084 (same locality as the holotype). Diagnosis. —This species is differentiated from other species of Sphenomorphus by the following combination of characters: (1) MBSR 34-38; (2) DSR 73-76; (3) fourth toe lamellae 15-17; (4) anterior loreal small, triangular, not in contact with upper labials; (5) prefrontals in moderate to broad con- tact; (6) frontoparietals fused; (7) interpa- rietal large; (8) SVL at maturity about 50- 60 mm. Description (based on four specimens). — SVL at maturity 55.8-56.4 mm for two males and 51.9 mm for one female; snout tapered, bluntly rounded, SnL 28-30% of HL and 43-50% of HB; HB 13-14% of SVL; ED 59-70% of SnL and 30-33% of HB; ros- tral broader than high, in broad contact with frontonasal; prefrontals in moderate to broad contact; frontal long, in contact with two or three supraoculars; four large supra- 389 Fig. 1. _Sphenomorphus kitangladensis. Dorsal head- shield pattern, CAS 191084, paratype. oculars; frontoparietals fused; interparietal large; parietals in narrow contact posteri- orly; no nuchals (Fig. 1); anterior loreal small, triangular, not in contact with upper labials; posterior loreal large, in contact with first and second or second upper labials (posterior loreal forming two superimposed scales in one specimen); seven upper labials, fifth enlarged and beneath eye; six or seven lower labials; tympanum shallow, TD 42- 57% of ED; dorsal scales smooth, 73-76 transverse rows along vertebral line be- tween parietals and base of tail; 34—38 mid- body scale rows; two enlarged preanals; limbs well developed; length of extended 390 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Standard scale counts and other pertinent characters for Philippine species referred to Group 3 of the genus Sphenomorphus. pa Species Adult SVL (mm) Midbody scale rows Dorsal scale rows Fourth toe lamellae S. mindanensis 42-56 30-32 66-78 17-20 S. laterimaculatus Syyae 40 W2 18 S. leucospilos 52-55 32 63-68 16-18 S. kitangladensis 50-57 34-38 73-76 15-17 S. victoria 45 30 64-65 18-20 Part B Fronto-parietals Anterior loreal contacts labials Prefrontals in contact Species Fused Not fused Yes No Yes No S. mindanensis xX xX > S. laterimaculatus xX xX xX S. leucospilos xX xX x S. kitangladensis XxX xX xX S. victoria xX xX » hind limb 77-82% Ax-Gr distance (proba- bly conservative since limbs were stiff); 15— 17 lamellae beneath fourth toe; 6—7 lamellae beneath first toe; tail much longer than body. Color (in preservative). — Dorsum grayish tan, variously marked with small blackish spots, especially along vertebral line; dorso- lateral area marked by a narrow, wavy, blackish band with some vague to promi- nent, pale blotches along dorsal margin; lips usually with some dark bars; lateral surfaces of body dusky; venter grayish ivory, rela- tively uniform for one specimen, mottled with black anterior to fore limbs for three specimens. Etymology. —Based on the name of the mountain which is the type locality. Range. — Known only from the type lo- cality, Mt. Kitanglad, Bukidnan Province, Mindanao Island. Comparisons Comparisons are made with the four spe- cies that might be easily confused with Sphenomorphus kitangladensis. Spheno- morphus kitangladensis is most similar to S. laterimaculatus, but differs in the some- what lower number of midbody scale rows (34-38), 40 for laterimaculatus; the small anterior loreal which is not in contact with the upper labials (Table 1); and the shorter snout relative to the head breadth. Sphen- omorphus leucospilos and_S. victoria exhibit lower counts for both midbody and dorsal scale rows and have the anterior loreal in contact with the upper labials (Table 1). Sphenomorphus mindanensis and 5S. victo- ria differ in the smaller prefrontals which are not in contact with each other and the fronto-parietals are not fused. Sphenomor- phus mindanensis further differs from this species in the lower number of midbody scale rows (Table 1). Some examples of Sphenomorphus deci- piens (a Group 4 species) resemble S. ki- tangladensis in exhibiting a small anterior loreal which is not in contact from the upper labials. However, S. decipiens differs in the blunter, shorter snout, evidenced in the SnL/ HB ratio (35-39% for decipiens and 43-50% for kitangladensis) and the HB/HL ratio (65- 81% for decipiens and 58-64% for kitang- ladensis); the smaller size (SVL 31-45 mm for 20 adult decipiens and 52—56 mm for 3 adult kitangladensis); and the lower number VOLUME 108, NUMBER 3 of dorsal scale rows (57—66 for decipiens and 73-76 for kitangladensis). Acknowledgments I wish to thank L. R. Heaney and R. F. Inger, Field Museum of Natural History, Chicago, for the opportunity to describe this species and anonymous reviewers for many helpful suggestions. The illustration was prepared by Colleen Sudekum of the Cali- fornia Academy of Sciences. 391 Literature Cited Boulenger, G. A. 1890. Fauna of British India in- cluding Ceylon and Burma. Reptilia and Batra- chia. Taylor and Francis, London, 541 pp. Brown, W. C., & A.C. Alcala. 1980. Philippine Liz- ards of the Family Scincidae (Silliman Univ. Nat. Sci. Monogr., ser. 2). Silliman University Press, Dumaguete City, Philippines, 264 pp. Smith, M. A. 1935. The fauna of British India, in- cluding Ceylon and Burma. Reptilia and Am- phibia, vol. 2. Taylor and Francis, London, 440 pp. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(3):392-394. 1995. A new species of Brachymeles (Reptilia: Scincidae) from Catanduanes Island, Philippines Walter C. Brown and Ely L. Alcala (WCB) Department of Herpetology, California Academy of Sciences, San Francisco, California 94118, U.S.A; (ELA) Center for Tropical Conservation, Silliman University, Dumaguete City, 6200 Philippines Abstract.—A new scincid lizard, Brachymeles minimus, is described from Catanduanes Island, Philippines. It differs from known species, other than Brachymeles vermis and B. apus, in the absence of external limbs. Brachymeles minimus differs from B. vermis and B. apus in its smaller size, lower dorsal scale row count, and pattern of head shields. Catanduanes, an island of low mountain terrain and about 1450 sq. km in area, lies slightly less than 10 km off the southeast coast of Luzon Island. Until recently, pub- lished information on its herpetofauna was limited to three species, one crocodile, one lizard, and one snake. Ross & Gonzales (1992), following a survey of the island in 1990, reported a total of 58 species; 13 am- phibians, 27 lizards, 17 snakes, and one crocodile. They described one amphibian, Kaloula kokacii as new, and list two lizards Brachymeles sp., Gonocephalus sp., and one snake, Oxyrhabdium sp., as being based on specimens not readily assignable to any known species. Recently, while identifying some Brachy- meles from other Philippine islands, we also examined the specimens of the unidentified Catanduanes’ Brachymeles. A sample of four specimens are now available. These, indeed, represent a previously undescribed taxon, the description of which is the subject of this paper. Materials and Methods Materials examined include examples of all 14 previously described species of the genus Brachymeles as well as the four known examples of the new species. Measurements were determined to the nearest 0.1 mm, using a Helios dial caliper. In addition to snout—vent length (SVL), oth- er measurements include head breadth (HB), made at the widest point; snout length (SnL) from anterior corner of eye to tip of snout; and eye diameter (ED) from anterior to pos- terior corner of the eye. Midbody scale counts (MBSR) were made about halfway between the tip of the snout and the vent, and the dorsal scale count (DSR) from the parietals to the transverse row opposite the vent. Standard nomenclature for head- shields in the Scincidae is followed (Bou- lenger 1890, Smith 1935, Brown & Alcala 1980). Brachymeles minimus, new species Fig. 1 Holotype. —FMNH 247990, a male (ap- parently mature based on hemipenes), col- lected in original forest at 600 m on Mt. Tungaw-tungaw (13°40’N, 124°21’E), about four km west of Gigmoto, Catanduanes Prov., Catanduanes Island, by the junior author on 18 June 1991. MBSR 22; DSR 94; ED 1.2 mm; SvL 64.0 mm; SnL 2.4 mm; HB 4.1 mm. Paratypes (3): FMNH 247991, 250817, CAS 182569 (same locality as holotype) VOLUME 108, NUMBER 3 Diagnosis.—A small Brachymeles, 56.3 and 64.0 mm for two (apparently mature) males; MBSR 22; DSR 94-102; no external vestige of limbs; no ear opening; frontopar- ietals in moderate to broad contact; eye rel- atively large, ED 43-50% of SnL. Description (based on four specimens). — A small slender Brachymeles, SVL 56.3 and 64.0 mm for two males and 38.4 and 48.5 mm for two apparent juveniles; HB 6.4— 8.2% of SVL; SnL 53.5-58.5% of HB; eye large, ED 43-50% of SnL and 23.3-29.3% of HB; rostral broader than high, having narrow contact with prefrontal; supranasals large, narrowly separated at midline; fron- tonasal broader than long; prefrontals nar- row, narrowly separated at midline; frontal about as broad as long, in contact with two anterior supraoculars; five supraoculars, separated from eye by five or six superci- liaries; frontoparietals in relatively broad contact; interparietal large; parietals in con- tact; one pair of narrow nuchals (Fig. 1); no postnasal; anterior loreal larger than pos- terior, in contact with first and second upper labials; six upper labials, anterior largest and fourth beneath center of eye; six lower la- bials; pair of contacting shields behind post- mental; no vestige of ear opening; 22 MBSR; 94-102 DSR between parietals and base of tail; no external vestiges of fore or hind limbs (Table 1). Color (in preservative).—Dorsum light brown to chocolate brown, (holotype light- est in color); ventral surface almost same 393 2.5mm Fig. 1. Brachymeles minimus. Pattern of dorsal head shields, FMNH 247991, paratype. color as dorsum for the holotype, but some- what more grayish brown in paratypes. This color pattern results from the pale margins of the ventral scales. Habitat note. —Specimens were found be- neath decaying logs and rocks lying in the duff of original forest at 600 m elevation. Range. —Known only from the type lo- cality on Catanduanes Island. Table 1.—Scale counts and other pertinent characters for Brachymeles minimus, vermis, apus, and samarensis. Fronto-parietals | super) eS ED HB External Species SVL (mm) adults DSR MBSR ciliaries Contact Separate SnL SVL limbs B. minimus 56.3-64.0 94-102 22 6 xX 43-50% 8.3% None (2) (4) (4) B. vermis 61.0-76.0 104-109 22-23 2 xX 35-44% 2.2-3.3% None (4) (6) (6) B. apus 119-131 108-113 22-24 2 xX 35% 5.7% None (2) (6) (6) B. samarensis 55.6-65.2 87-96 22-24 6 xX 25-33% 6.5-8.3% Short (14) (19) (19) Stubs 394 Etymology.—From the Latin minimus, meaning small. Discussion Skinks of the genus Brachymeles, popu- larly referred to as burrowing lizards, are rarely if ever seen except when one looks beneath the duff, rotting logs, or in top soil. In association with adaptation to these sub- terranean habitats, the species of this genus exhibit a remarkable series of limb reduc- tion stages (from relatively short, pentadac- tyl limbs to complete absence of external limb structures), and correlated body elon- gation and attenuation stages. For a discus- sion of limb loss in lizards, see Greer (1989) and Greer & Cogger (1985). Of the 15 known species of Brakhymeles, 14 are from the Philippines (Brown & AI- cala 1980) and one from Borneo (Hikada 1982). Five of the 15 species (bicolor, bou- lengeri, gracilis, schadenbergi, and talinis) are pentadactyl, but the limbs and digits are short. Seven of the species (bonitae, ce- buensis, elerae, pathfinderi, samarensis, tri- dactylus, and wrighti) exhibit further reduc- tion of the limbs and loss of digits ranging from one on either the fore or hind limbs to four or all five on both fore and hind limbs. The remaining three species (apus, vermis, and minimus) have lost all vestiges of external limbs. Brachymeles minimus differs from the other limbless species (apus and vermis) in the smaller size and lower count for DSR (Table 1). It also differs from these two spe- cies in the pattern of the head shields: (1) frontoparietals large, in contact (separated for apus and vermis) and (2) five or six su- perciliaries (two beneath anterior supraocu- lars for apus and vermis), Table 1 and Fig. 1. Brachymeles minimus is most similar to B. samarensis in the pattern of the head- shield arrangement. Both species have: (1) the fronto-parietals in contact, (2) five su- praoculars, (3) a pair of narrow nuchals, and (4) lack a postnasal. Brachymeles minimus differs from B. samarensis in the complete PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON absence of external limb structures, the number of dorsal scale rows, and the slightly larger eye relative to snout length (Table 1). The evidence from scale counts and pat- terns of head-shield development suggests that apus and vermis may be the end prod- ucts of a different evolutionary branch than that from which minimus is derived. A phy- logenetic study, using additional lines of ev- idence to those presently available, is very much needed to clarify relationships within this genus. Acknowledgments We wish to thank L. Heaney and C. Ross for suggesting that we undertake this study. We also thank R. F. Inger, A. Leviton, R. Drewes, and anonymous reviewers for help- ful suggestions during the preparation of the manuscript. Illustrations of the head shield patterns were made by Colleen Sudakum, California Academy of Sciences. Literature Cited Boulenger, G. A. 1980. Fauna of British India in- cluding Ceylon and Burma. Reptilia and Batra- chia. Taylor and Francis, London, 541 pp. Brown, W. C., & A. C. Alcala. 1980. Philippine liz- ards of the family Scincidae. (Silliman Univer- sity Natural Science Monographs, ser. 2) Silli- man University Press, Dumaguete City, Phil- ippines, 264 pp. Greer, A. E. 1989. The Biology and Evolution of Australian Lizards. Survey Beatty & Sons, Chip- ping Norton, Australia, 264 pp. —, & H. G. Cogger. 1985. Systematics of the reduced-limbed and limbless skinks currently assigned to the genus Anomalopus (Lacertilia: Scincidae).— Records of the Australian Muse- um 37:11-54. Hikada, R. 1982. A new limbless Brachymeles (Sau- ria: Scincidae) from Mt. Kinabalu, North Bor- neo.—Copeia 1982 (4):840-844. Ross, C. A., & P. C. Gonzales. 1992. Amphibians and reptiles of Catanduanes Island, Philip- pines.— National Museum Papers, Manila, Philippines 2:50-76. Smith, M. A. 1935. The fauna of British India, in- cluding Ceylon and Burma. Reptilia and am- phibia, vol. 2. Taylor and Francis, London, 440 pp. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(3):395—400. 1995. A new skink (Emoia: Lacertilia: Reptilia) from the forest of Fiji George R. Zug and Ivan Ineich (GRZ) Department of Vertebrate Zoology, National Museum of Natural History, Washington, D.C. 20560, U.S.A.; (II) Laboratoire des Reptiles et Amphibiens, Muséum National d’Histoire Naturelle, 25 rue Cuvier, 75005 Paris, France Abstract. —A moderate-sized turquoise-colored skink from the montane for- est of Vanua Levu, Fiji differs from all other central Pacific skinks in a com- bination of mensural and scalation characters. This skink, Emoia mokosari- niveikau n. sp., is a member of the Emoia samoensis species group and concolor subgroup, and its coloration and color pattern are unique among skinks of the genus Emoia. Vanua Levu is the second largest island of the Fijian group. In spite of its size and accessibility, no one has surveyed method- ically the diversity and distribution of its amphibians and reptiles. Forest lizards, such as Lepidodactylus manni and Emoia par- keri, should occur there, but no records exist of such occurrences (Zug 1991). We have begun such a survey, but as yet our efforts are small. Our first survey in 1992 provided new distributional records for a number of lizard species (e.g., Emoia parkeri) and yielded a single specimen of a new Emoia. Because this species is a forest lizard and the native forest of Vanua Levu continues to disappear, we wish to alert Fijians and biologists of this, presumably, Vanualevuan endemic by describing the lizard promptly. Forests, where they still persist, are threat- ened throughout Oceania, and the continual discovery of new species of forest plants and animals re-enforces the urgency of forest conservation and the establishment of for- est reserves. Methods Mensural and meristic characters follow the definitions and protocols of Zug (1991). Family Scincidae Genus Emoia Emoia mokosariniveikau, new species Figs. 1-2 Holotype. —USNM 322473, adult female from near Saivou, ca. 23 km (road) NW of Savusavu, Vanua Levu, Fiji, collected 23 October 1992 by I. Ineich and G. R. Zug. Diagnosis.—Emoia mokosariniveikau is a member of the samoensis group (sensu Brown 1991) and the concolor subgroup. It differs from other Fijian Emoia by: size of adult females (55 mm SVL), larger than ca- eruleocauda (42-50 mm), cyanura (39-53 mm), impar (40-47 mm) and parkeri (43- 52 mm) and smaller than campbelli (57-64 mm), concolor (59-77 mm), nigra (88-108 mm) and trossula (75-100 mm); adult col- oration (coppery brown dorsal ground color overlain by turquoise on neck and trunk and dark transverse bars on sides of trunk onto back), caeruleocauda, cyanura, impar and parkeri (light stripes on black or dark brown background), campbelli and trossula (small dark transverse bars scattered over beige to brassy background), concolor (uniform or nearly so green to olive-beige ground-color), and nigra (uniform black ground-color); number of lamellae beneath the fourth toe 396 Rigas PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON CCl > CCS DEREK ——- — > DONDNOnTANNN Hiitasenannnnnnsnyy))yo0000 Scalation of the holotype of Emoia mokosariniveikau (USNM 322473). Dorsal and lateral views of the head (left); ventral surface of the right fourth toe (right); scale bars equal 1 mm. (48-49), caeruleocauda (33-41) and parkeri (31-41), cyanura and impar (57-80), and nigra (31-39). From members of the sa- moensis group of Emoia, E. mokosariniy- eikau differs by: size of adult females (55 mm SVL), smaller than aneityumensis (=75 mm), campbelli (=68 mm), concolor (=59 mm), flavigularis (=59 mm), loyaltiensis (260 mm), nigra (=88 mm), samoensis (=79 mm), sanfordi (=68 mm) and trossula (=75 mm); scale rows around midbody (30), aneityumensis (= 39), flavigularis (= 34) and nigra (232); number of lamellae beneath fourth toe (48-49), aneityumensis (S38), campbelli (=42), flavigularis (<=48), mur- phyi (265), nigra (<39), nigromarginata (48), parkeri (=41) and sanfordi (=61); see Tables 1 & 2 for summary of measure- ments and scalation. Description of holotype.—An adult fe- male (55.1 mm SVL), slender-bodied with long tail (91 mm, regenerated, round in cross-section, tapering gradually to point) and well-developed limbs; hindlimb length 45% SVL, fourth toe tip extending nearly to axilla when adpressed to side of body; body length 48% SVL. Digits of fore- and hindfeet long and strongly clawed; hindtoes length pattern 4>5=3>2>1 (Fig. 1). Head long and slender, slightly depressed, snout pointed; head length (HL) 22% SVL; neck length (ear to axilla) 14% SVL; head width 60% HL; snout length (anterior corner of eye to tip of snout) 48% HL; and preocular length (eye to naris) 32% HL. Shiny, imbricate scales on body; dorsal scales grossly smooth but with shallow, lon- gitudinal striations visible microscopically; 61 scales from parietals to base of tail, an- teriormost pair enlarged as nuchals; mid- dorsal scales subequal in size to more lateral ones; 30 scales around midbody. Narrow lamellae on underside of all digits; 42/39 (left/right) beneath 4th finger; 48/49 be- neath 4th toe (Fig. 1). Head scales smooth (Fig. 1); rostral broad, touching frontonasal across broad suture; supranasals small, nar- rowly rectangular; prefrontals broadly in VOLUME 108, NUMBER 3 397 Table 1.—Summary of mensural characteristics of the Emoia samoensis group (sensu Brown 1991). All measurements are for sexually mature individuals; only ranges are provided and all measurements are in millimeters. Snout—vent Species of Emoia Sex length Hindlimb length Head length n aneityumensis* F 74.9-85.4 34.8-36.9 17.2-18.2 2 campbelli F 92.9 S72 18.5 1 M 68.0-89.8 31.6—42.0 15.7-20.3 4 concolor F 58.7-77.3 26.2—38.0 12.9-19.8 15 M 56.8-85.5 27.6—40.6 13.6—20.0 Di flavigularis* F&M 58.9-75.5 — — WD loyaltiensis* F&M 60.0-83.2 — _ 5 mokosariniveikau F 55.1 SA0) 12.4 1 murphyis F 59.4-65.4 25.1-30.3 13.5-14.4 3 M 65.7 33.9 15.5 1 nigra? F 87.7-107.5 42.3-53.3 18.9-23.9 8 M 88.0-112.1 44.8-53.4 20.0—25.5 10 nigromarginata‘ F&M 51.8-77.4 — — 26 parker’? F 42.9-52.0 18.9-23.7 9.9-11.7 8 M 46.9-50.0 20.4—23.0 10.6-12.4 5) samoensis* F&M 78.0-118 — — 24 sanfordi* F&M 68.3-115.0 — — 36 trossula® F 75.4-100. 1 34.7—-46.4 16.9-21.1 30 M 74.1-106.5 35.6—53.0 17.3-24.2 51 4 Emoia aneityumensis includes all concolor-like specimens from Aneityum, Futuna, and Tanna. > Data from Fijian specimens only (Zug 1991). © Data from Brown (1991). 4 Data from Samoan specimens only. contact medially; frontoparietal single; in- terparietal small with a parietal eye; parietal large and in contact behind interparietal; 4/4 supraoculars on each side bordered lat- erally by 8/8 supraciliaries; anterior loreal longer than high and equal in area to pos- terior loreal; posterior loreal longer than high; upper eyelid immoveable with 9/9 Table 2.—Summary of scale characteristics of the Emoia samoensis group. Sexual dimorphism has not been identified in scale characters; the following ranges derive from juveniles and adults. Footnote legend as in Ta- ble 1. Number of scales Subdigital lamellae Species of Emoia Dorsal Midbody 4th finger 4th toe n aneityumensis? 74-80 39-41 23-30 33-38 5 campbelli 57-64 30-34 31-38 42-51 10 concolor? 54-62 27-33 30-48 43-65 78 flavigularis< 53-64 34—40 — 36-48 2D) loyaltiensis¢ 62-71 30-34 — 51-60 5 mokosariniveikau 61 30 39-42 48-49 1 murphyis 54-58 28-31 45-51 65-74 8 nigra’ 60-69 32-40 20-24 31-39 33 nigromarginata® 56-64 28-32 — 38-48 26 parkeri’ 54-60 28-33 22-30 31-41 24 samoensis© 58-68 30-34 — 45-54 24 sanfordic 56-64 28-34 — 61-76 36 trossula® 58-77 30-40 27-42 42-65 N7/ 398 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Dorsolateral view of the holotype of Emoia mokosariniveikau (USNM 322473) in life. The dorsolateral “‘stripe”’ is a reflection from the photographic flash. scales along upper edge; lower eyelid move- able with large transparent disc surrounded by small opaque scales; 8/8 supralabial scales, enlarged 6th beneath eye; 8/8 infra- labial scales. Ear opening vertically ellipti- cal, top tilted posteriorly, with 2/2 distinctly enlarged scales on anterior edge. In life (Fig. 2), dorsal ground-color shiny coppery brown, head uniform bright cop- pery; broad turquoise bands on side of neck and trunk, bands extend upward toward midline, continuous on neck and anterior trunk, and staggered and discontinuous on posterior half of trunk; at abdominal level, small, irregular bars of dark brown or black appear on side and extend onto dorsum, bars usually edged with blue; on tail, dark bars alternate at midline and interspaced with brown then beige. Venter from mid throat onto base of tail light orangish yel- low, bordered on chin and ventrolaterally by light turquoise from lateral bands. Limbs coppery beige dorsally with dark spots, uni- form yellow ventrally; distinct dark stripe on posterior surface of thigh. In alcohol, dorsal ground-color bluish gray anteriorly and coppery beige posteriorly, lateral bars black; ventrally, chin and an- terior throat bluish ivory, throat to base of tail pale yellow, bluish tinted laterally on trunk. Etymology. — Fijians, Polynesians as well (Crombie & Steadman 1986), largely ignore their native lizards. Moko is the widely used colloquial name for lizard, whether it is a skink or gecko, big or small. A few other names do exist. Vokai refers exclusively to the Fijian iguana (Brachylophus) and is used and recognized by all Fijians. Sari and boliti refer also to lizards, although they are of uncommon usage. Sari refers to a small green lizard (Capell, 1973) when used in combination with the Fijian moko as a suf- fix, and to this Fijian word, we append the Fijian ni veikau, meaning of the forest. This combination produces mokosari ni veikau or small green lizard of the forest, and is a noun in apposition. [We have taken liberties with “green lizard” as turquoise is greenish blue rather than strictly green.] Distribution.—Emoia mokosariniveikau is known only from the type locality on the northern slope of Vanua Levu’s central mountain range. Since the type locality is not isolated ecologically or geographically, E.. mokosariniveikau likely occurs through- VOLUME 108, NUMBER 3 out the length of island where original forest remains or has regenerated sufficiently. Dick Watling (in litt., Aug 1994) described an Emoia captured (Aug 1978) at Solevu, Va- nua Levu. His description closesly matches E. mokosariniveikau but cannot be con- firmed because the specimen has not been located. One (Calif. Acad. Sci./CAS 155973; Joske’s Thumb, Viti Levu) of the paratypes of the Emoia campbelli series does not match the other members of that type series and shares some similarities with E. mokosar- iniveikau; however, its discoloration and poor preservation prevent confirmation of its specific identity. Comments on Relationships Brown first recognized the Emoia atro- costata and the E. samoensis ““complexes”’ in 1953 (Brown & Marshall 1953). He gave a set of characters defining the atrocostata complex and noted that the samoensis com- plex was closely related to the former and differed in the presence of more numerous and more highly modified digital lamellae. Later in a privately circulated key to Emoia, Brown (1970) recognized five ‘natural groups’ of Emoia: adspersa; atrocostata; baudini-cyanura; cyanogaster; physicae. Emoia concolor, murphyi, and samoensis were listed as members of the cyanogaster group. Brown later reintroduced the sa- moensis group with the recognition of E. campbelli and E. trossula (Brown & Gib- bons, 1986). The samoensis group was not characterized therein and remains defined largely by its species content. A character- ization is possible, however, by listing the identifying features from Brown’s (1991) species group key: 11 premaxillary teeth; alpha palate; unfused nasal bones; parietal eye present; anterior loreal as long or nearly as long as posterior loreal; dorsal scale rows <=87; midbody scale rows <44; rounded to moderately thin subdigital lamellae. The in- cluded species are listed in Tables 1 and 2. Assuming the samoensis group to be a 399 natural assemblage, we here recognize two species subgroups within the central Pacific (Fiji-Samoa-Tonga area) members of the samoensis group. The samoensis subgroup contains samoensis and trossula. These Emoia are moderately large (74-118 mm adult SVL), heavy-bodied skinks with 57- 77 dorsal scale rows, 30-40 midbody scale rows, and 42-65 lamellae on the fourth toe. The concolor species subgroup contains campbelli, concolor, mokosariniveikau and murphyi. These Emoia are moderate-sized (56-90 mm adult SVL), slender-bodied with 54-67 dorsal scale rows, 27-34 midbody scale rows, and 42—74 lamellae on the fourth toe. These subgroups are tentative group- ings and serve only for our initial analysis of relationships within the samoensis group. This division of species agrees with Brown’s (1991) concolor and samoensis subgroups, except for exclusion of E. nigra and E. par- keri. We question the status of these two species as members of the samoensis group. In addition to its distinctive coloration (black or dark brown ground color in con- trast to coppery beige to green), EF. nigra is behaviorally unlike the other semiarboreal to strongly arboreal samoensis group mem- bers. For example, FE. trossula forages on and above the ground and regularly escapes by climbing, in contrast FE. nigra forages on the ground and, even if resting in a tree, regularly escapes downward and away from the tree (Zug 1991). Emoia parkeri has dor- solateral stripes unlike any other juvenile or adult samoensis group member. Its snout is more attentuate and its prefrontals are not, or slightly, in contact on the midline. Our analysis of intragroup relationships is not yet sufficiently advanced to declare which character-states are synapomorphic or, for that matter, to have identified a use- ful set of characters for initiating a cladistic analysis. All scalation and mensural fea- tures of E. mokosariniveikau lie within the range of the concolor subgroup. Its dorsal and lateral coloration of neck and trunk is unique within both subgroups. Its lateral 400 dark, light-bordered bars on the trunk are similar only to the bar-pattern of some pop- ulations of Fijian E. trossula; however, its dark thigh stripe occurs only in other con- color subgroup members. No trait currently suggests a closer affinity of E. mokosariniv- eikau to any one of the other three central Pacific concolor subgroup members. Acknowledgments We appreciate and thank the following individuals and agencies for their encour- agment and assistance of our biological in- vestigations of Fijian and other Pacific liz- ards. The Ministry of Primary Industries, Forestry & Co-operatives gave us permis- sion to study lizards in Fiji and to retain vouchers in the Smithsonian (USNM). Field work was supported by the Smithsonian Scholarly Studies Program. Milika Naqa- sima and Michael Doyle of The University of the South Pacific provided the proper Fijian construction of the new specific ep- ithet. Kate Spencer skillfully illustrated the head and hindfoot of Emoia mokosariniv- eikau. Ronald Crombie, Robert Fisher, and Dick Watling reviewed and improved our early drafts of this article. Literature Cited Brown, W. C. 1970. Key to the seemingly natural groups of the genus Emoia. Privately circulated manuscript, 11 pp. . 1991. Lizards of the genus Emoia (Scincidae) with observations on their evolution and bio- geography.— Memoirs of the California Acad- emy of Sciences (15):i—vi, 1-94. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON —, & J. R. H. Gibbons. 1986. Species of the Emoia samoensis group of lizards (Scincidae) in the Fiji Islands, with description of two new species. — Proceedings of the California Acade- my of Sciences 44(4):41-53. —, & J. T. Marshall, Jr. 1953. New scincoid lizards from the Marshall Islands, with notes on their distribution. —Copeia 1953(4):201-—207. Capell, A. 1973. A new Fijian dictionary. Govern- ment Printer, Suva, 407 pp. Crombie, R.I., & D. W. Steadman. 1988. The lizards of Rarotonga and Mangaia, Cook Island Group, Oceania. — Pacific Science (1986) 40(1-4):44—57. Zug, G. R. 1991. The lizards of Fiji: natural history and systematics.— Bishop Museum Bulletin in Zoology (2):i—xil, 1-136. Appendix. 1. Specimens Examined Emoia aneityumensis. Vanuatu: BMNH 1956.1.3.64, 1973.1534-35; FMNH 69151, 69638. Emoia campbelli. Fiji: CAS 155967-70, 155972-73, 156257-58, 156710-12. Emoia concolor. Fiji: mensural and meristic data from Appendix table B in Zug, 1991; MNHN 7084, 7084A. Emoia flavigularis. Solomon Islands: mensural and meristic data from species descriptions in Brown, 1991. Emoia loyaltiensis. Loyalty Islands: mensural and me- ristic data from species descriptions in Brown, 1991. Emoia murphyi. Samoa: USNM 268363, 322743-49. Emoia nigra. Fiji: mensural and meristic data from Appendix table B in Zug, 1991. Emoia nigromarginata. Vanuatu: mensural and me- ristic data from species descriptions in Brown, 1991. Emoia parkeri. Fiji: mensural and meristic data from Appendix table B in Zug, 1991. Emoia samoensis. Samoa: mensural and meristic data from species descriptions in Brown, 1991. Emoia sanfordi. Vanuatu: mensural and meristic data from species descriptions in Brown, 1991. Emoia trossula. Fiji: mensural and meristic data from Appendix table B in Zug, 1991. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(3):401-435. 1995. Loven’s law and adult ray homologies in echinoids, ophiuroids, edrioasteroids, and an ophiocistioid (Echinodermata: Eleutherozoa) Frederick H. C. Hotchkiss 26 Sherry Road, Harvard, Massachusetts 01451, U.S.A. Abstract. —Lovén’s law, which is summarized by an AABAB rule, is found to apply to the structure of Stromatocystites and Astrocystites (edrioasteroids) and Gillocystis (an ophiocistioid) and was previously known in echinoids and in Paleozoic ophiuroids. Chance coincidence and convergent evolution are not likely explanations. Lovén’s law is interpreted from edrioasteroids as a man- ifestation of 2-1-2 symmetry, and in echinoids, ophiuroids and ophiocistioids it is interpreted as a vestige of 2-1-2 symmetry. Thus, Lovén’s law is a ple- siomorphic character that indicates ray homologies. Roman numerals as used by Lovén for echinoids are applied to edrioasteroids, ophiuroids and ophio- cistioids. The anterior ray in exocyclic echinoids is ray III, whereas it is ray II in edrioasteroids. This is reconciled by noting that ray II was anterior in the elongate tests of the Ordovician echinoids Ectinechinus and Eothuria. In echi- noids the anterior-posterior axis of the bilateral larva lines up with ray II in the imago [von Ubisch’s axis of primordial symmetry]. This relation is used to deduce that the anterior-posterior axis of edrioasteroids coincided with the axis of their larvae. Studies by Fell, Strathmann, and Smith are used to speculate that the larval type of Ordovician ophiuroids was a bilateral planktotroph but not yet an ophiopluteus. Lovén’s law is a distinctive and fundamental feature of the body plan that validates the Subphylum Eleutherozoa. There are at least two distinct constructions of 2-1-2 symmetry (Eleutherozoa and Blastozoa) and possibly a third (crinoids). This paper investigates the value of Lov- én’s law in establishing ray homologies among echinoderms. Lovén’s law of het- erotropy describes an AABAB pattern of relative sizes of echinoid basicoronal plates. In an earlier paper (Hotchkiss 1978), my a priori hypothesis was that demonstration of the AABAB pattern of Lovén’s law in the ambulacra of Paleozoic ophiuroids would establish either the madreporite or the anus as the landmark for homologous orientation of echinoderms. The unexpected finding was that possibly neither the madreporite nor the anus 1s reliable. Several a posteriori hy- potheses came out of that work, including the possibility that Lovén’s law may be fun- damental to the echinoderm ground plan. The program for this research was laid out in the earlier paper (Hotchkiss 1978). 1. Attempt to determine which, if any, of the conflicting propositions for designating ray homologies may be the correct propo- sition. 2. Investigate the location of the plane of hydrocoele closure and the orientation of the embryological or developmental axes as part of this analysis. 3. Attempt to decide whether Lovén’s law independently evolved in echinoids and ophiuroids, or whether it is indicative of a recent common ancestry between echinoids and ophiuroids [review also the embryological evidence}, or wheth- er it is fundamental to the ground plan of the phylum. 4. Investigate whether Lovén’s law is expressed in other echinoderm groups 402 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Quick reference list of the numbered hypotheses (see text). HO The null hypothesis that the arrangement of ambulacrals in the different rays is a matter of chance H1 The hypothesis of a selective advantage that led to convergent evolution of Lovén’s law in separate classes of echinoderms H2 The hypothesis that in edrioasteroids Lovén’s law is a manifestation of 2-1-2 symmetry and that in descendent groups it is a vestige of this ancient symmetry [symplesiomorphy] H3 The hypothesis that ray II was the original anterior ray H4 The hypothesis that the relation of the anteroposterior axis of the larva to the rays of the imago has remained the same in echinoids, ophiuroids, edrioasteroids, and ophiocistioids H5 The hypothesis that Lovén’s law is fundamental to the ground plan of the clade Eleuthero- zoa H6 The hypothesis that there are at least two distinct constructions of 2-1-2 symmetry, that seen in Blastozoa and that in Eleutherozoa with alternating ambulacral plates (specific example: Astrocystites). 5. Attempt to de- termine the functional significance, if any, of Lovén’s law [Jackson (1927) noted that Lovén’s law itself is without obvious mean- ing]. 6. Investigate whether the fidelity of Loveén’s law in Paleozoic ophiuroids varies between taxa (not studied). 7. Investigate whether the madreporite has a constant or statistically preferential position with re- spect to Lovén’s law in ophiuroids. The sequence and progress of the research depended on serendipity. From 1978 to 1992 I accumulated information on bila- terality in echinoderms [e.g., axes of sym- metry, situs inversus, teratology, and be- havior] but did not gain any new insight into ray homologies. In 1988 I worked out new methods for investigating the location of the madreporite in Paleozoic ophiuroids (point 7 supra). The chance discovery came in 1992 when I noticed an instance of Lovén’s law in Smith’s (1985) drawing of Stromatocys- tites walcotti. | proposed a new table of ray homologies for edrioasteroids, ophiuroids and echinoids at the second North Ameri- can Friends of Echinoderms conference and workshop in July 1992 (no published pro- ceedings). The results were confirmed and extended in 1994 when I noticed the in- stance of Lovén’s law in Jell’s (1983) draw- ing of the ophiocistioid Gillocystis. Building on previous studies (Hotchkiss & Seegers 1976, Hotchkiss 1979) the new results are used to suggest ray homologies for asteroids. This paper is the first publication of this research. To keep the presentation as direct as possible, complex supporting informa- tion is given in numbered notes. Table 1 lists the numbered hypotheses for quick ref- erence. In this paper, ray identifications that are based on Lovén’s law are labeled with Ro- man numerals [see Note 1]. The labeling that results for edrioasteroids is different from that of Bather (1914a), Regnéll (1966) and Bell (1976a): the anterior unpaired ray is ray II in my labeling, whereas it is ray III in their ray labeling. In Appendix 1 the madreporite of Paleozoic ophiuroids is shown to be located adjacent to ray IV in interradius III/IV. Lovén’s Law in Echinoderms An essential part of the program for this research was to investigate whether Loven’s law is expressed in other echinoderm groups with alternating ambulacral plates (point 4 supra). Lovén’s law is an empirical state- ment of the unequal sizes of echinoid bas- icoronal plates (Fig. 1). The five plates that begin the ambulacral columns Ia, Ila, IIIb, IVa, Vb are larger than the basicoronal plates that begin columns Ib, IIb, IIIa, IVb, Va (Loven 1874, Melville & Durham 1966:fig. VOLUME 108, NUMBER 3 403 Fig. 1. Lovén’s law in the Recent echinoid Echinocyamus bisexus (oral side). Basicoronal plates Ia, Ia, IIIb, IVaand Vb, shown vertically ruled, are larger than the unshaded partner basicoronal plates of the same ambulacral area. Interambulacral areas are stippled. Interambulacrum V/I contains the periproct (shown). The madrepore is on the apical surface (not shown); in regular echinoids the madreporic plate is formed by genital 2 in interambulacrum IJ/III. Although irregular echinoids such as Echinocyamus are elongate with ray III anterior, ray II is inferred to have been the original anterior ray in early Paleozoic echinoids and edrioasteroids. In the small insert the echinoid is rotated to have the same orientation as in Figs. 2-5. (Redrawn and annotated from Kier 1968.) 404 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Lovén’s law in the Devonian ophiuroid Eugasterella thorni (oral side). The ambulacral half-series on the anticlockwise side of each arm is labeled as either in advance (A) or behind (B) the half-series on the clockwise side of the arm. The arrangement AABAB is Lovénian. From comparisons with edrioasteroids (Figs. 3, 4), ray II is anterior and interray IV/V is posterior. Right posterior interray III/IV contains the madreporite (not shown). (Modified from Hotchkiss 1978; after Kesling 1969.) 164). This arrangement is termed Lovén’s law of heterotropy by Gordon (1929) and simply Lovén’s law by Melville & Durham (1966:U222). The unequal size of the bas- icoronal plates initiates columns of echinoid ambulacral plates that alternate according to an AABAB rule (Appendix 1). Based on this rule I showed that Lovén’s law applies to Paleozoic ophiuroids that have alternat- ing ambulacral plates (Hotchkiss 1978). Here I report Lovén’s law in two edrioas- teroids and an ophiocistioid. It could affect the inclusion of somasteroids and asteroids in the clade Eleutherozoa if there was cause to think that they obeyed a law that was different from Lovén’s law, or cause to think that they could not obey Lovén’s law. I re- port that somasteroids and asteroids had alternating ambulacral plates (primitive character state) and that it is plausible that Lovén’s law will eventually be found to ap- ply to somasteroids and to asteroids. Echinoids (Fig. 1).—Lovén’s law has been so singularly associated with echinoids that it was used by Jackson (1929:508) as part of the evidence that Bothriocidaris is prop- erly assigned to the Echinoidea. Lovén’s law was found to apply to Bothriocidaris ar- VOLUME 108, NUMBER 3 chaica by Jackson (1912:34, 241, plate 1 fig. 1; 1927:451) and to B. pahleni by Morten- sen (1930) [see also Durham 1966:figs. 214— 217]. MacBride & Spencer (1938:121, fig. 10) reported that Lovén’s law appears to apply to the Upper Ordovician Aulechinus grayae. Ophiuroids (Fig. 2).—Hotchkiss (1978) reported that Lovén’s law applies to Paleo- zoic ophiuroids and suggested that the ophiuroid madreporite is located in inter- radius III/IV. Harper (1985) described new material of Eugasterella logani (Hall), re- porting that it conformed with Lovén’s law and that “the madreporite is situated on the disc in interray III/IV, adjacent to the distal portion of the first ambulacral of IV.” The association of the madreporite with a ray (now identified as ray IV), as noted by its eccentric location in the interradius, has been documented in both the Zeugophiurina (Sollas & Sollas 1912:216) and the Lyso- phiurina (Spencer 1914:37, 1925:277, 1934: 459; Hotchkiss 1970:69, Kesling & Le- Vasseur 1971:329, Petr 1989:8, text-fig. 6). Petr (1989:8, 16) suggested that movement of the lateral plate functioned to uncover and to irrigate the madreporite. Appendix | presents additional data on Lovén’s law in Paleozoic Oegophiurida. Complete specimens that obey Lovén’s law have the madreporite in interradius III/IV. Analysis of incomplete specimens demon- strates that each arm position relative to the madreporite has a preferred character state, that the preferred character states conform with Lovén’s law, and that this pattern iden- tifies the madreporic interradius as III/IV. Somasteroids.—Lovén’s law has not yet been documented in somasteroids. As part of his research on the ancestry of sea stars, Fell (1963a,b,c) studied the morphology of the early Paleozoic somasteroids collected by Thoral (1935) and described by Spencer (1951). He found that the ambulacral os- sicles of Chinianaster and Villebrunaster form opposite pairs in the middle and base of the arm but form alternating series in 405 young specimens and at the tip of the arm in older specimens (1963c:465, fig. 5). The somasteroid Archegonaster has al- ternating ambulacrals (Spencer 1951:102, fig. 10; Smith & Jell 1990:753, fig. 40). The madreporite is typically preserved, includ- ing portions of the stone canal (Spencer 1951:105, Smith & Jell 1990:762). Thus, a careful study of the arrangement of the am- bulacral plates at the mouth frame (such as Smith & Jell 1990:fig. 47D) recorded with reference to the location of the madreporite would permit an analytical search for Lov- én’s law in the same manner as done in this paper (see Appendix). The statement of Spencer & Wright (1966: U39) that in somasteroids the ambulacrals are “generally in opposite pairs but appar- ently alternating in some forms” conveys the wrong emphasis. In addition to Arche- gonaster, Chinianaster and Villebrunaster which have been mentioned above, alter- nating ambulacrals are described for Am- pullaster (Fell 1963c:fig. 6A,D) and all of the Archophiactinidae (Spencer 1927:361). The recently described Ophioxenikos lan- genheimi has “‘proximal ambulacrals slight- ly offset, distal ambulacrals clearly offset across arm axis” (Blake & Guensburg 1993: 109). It is plausible that Archegonaster and other Paleozoic somasteroids with alternat- ing ambulacral ossicles will eventually be shown to have the ambulacrals arranged in accordance with Lovén’s law. Asteroids. —Lovén’s law has not yet been documented in asteroids. Although Spencer & Wright (1966:U13) stated “In this sub- class the ambulacrals are invariably oppo- site one another, never alternating, as in so- masteroids and early ophiuroids,” this is not truly invariable. The situation in the Paleozoic Asteroidea echoes that of the so- masteroids. The ambulacrals in some of the older species are not exactly opposite, but neither are they definitely alternating except perhaps near the tip of the arm. Spencer (1914:19) used the phrase “irregularly al- ternating”’ and expressed the view that this 406 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON I] Fig. 3. Lovén’s law in the Ordovician edrioasteroid Astrocystites ottawaensis. The cover plate half-series on the anticlockwise side of each ambulacrum is labeled as either in advance (A) or behind (B) the half-series on the clockwise side of the ambulacrum. The arrangement AABAB is Lovénian. Ray II is anterior; posterior interray IV/V contains the anus, hydropore and unpaired oro-tegminal plate. (Copied and annotated from Bather 1914b.) was the primitive stock “‘from which two sets of forms arose, namely, those with op- posite ambulacralia and those with defi- nitely alternating ambulacralia.”’ Blake & Guensburg (1989:table 1, char- acter no. 16) described the primitive char- acter state of the “‘positions of ambulacral ossicles relative to one another across the furrow” as “weakly and locally alternate’’. They listed Salteraster, Lacertasterias and Schondorfia as having this primitive char- acter state, and Hudsonaster, Devonaster and Calliasterella as having the derived char- acter state of ““opposite.”’ Other specific examples of early asteroids with “irregularly alternating’ ambulacral VOLUME 108, NUMBER 3 plates are in the genera Lanthanaster and Palasterina. The holotype of Lanthanaster cruciformis [=L. intermedius (Schuchert, 1915); see Branstrator 1979] is very well preserved but has the region of the first few ambulacrals of each arm obscured from view by adambulacral spines. According to Bran- strator (1972:68), “It is difficult to deter- mine what the natural arrangement of am- bulacral pairs across the ambulacral groove was when the animal was alive. The am- bulacral ossicles of an arm appear (arm A, E) to have occurred in an alternate biseries, but the specimen is admittedly disturbed.” Thus it is not possible to score the ambu- lacral plate arrangements of the arms as “A” or ““B’’, but this will hopefully be possible in some future material. The madreporite is in an oral interradius. According to Spen- cer (1922:228) the holotype of Palasterina bonneyi Gregory [=P. antiqua (Hisinger); see Spencer 1922:228] has ambulacrals that “are opposite or slightly alternating in the prox- imal region, distinctly alternating distally”; a larger plate in one oral interradius “‘may represent the madreporite, but one cannot be certain of the madreporiform markings.” It is plausible that Lanthanaster, Palaster- ina and other Paleozoic asteroids with “‘ir- regularly alternating’? ambulacrals will eventually be shown to have the ambula- crals arranged in accordance with Loven’s law. Edrioasteroids (Figs. 3, 4).—In Lower Cambrian Camptostroma and Stromato- cystites the ambulacra are constructed of flooring plates arranged biserially and al- ternately (Paul & Smith 1984:470). This suggests searching for Lovén’s law in ed- rioasteroids. Order Edrioblastoida (see Smith & Jell 1990): Bather’s illustration (1914b:201, fig. 6; Fay 1967:S289, fig. 172) of the exact su- tures of the Middle Ordovician edrioblas- toid Astrocystites ottawaensis Whiteaves, reproduced here as Fig. 3, shows that the arrangement of the cover plates conforms with Lovén’s law (Hotchkiss 1978:543). It 407 seems likely that the underlying ambula- crals are arranged in the same pattern as the cover plates. If the null hypothesis HO is that the arrangement of ambulacrals in the different rays is a matter of chance and that the “‘A”’ arrangement is as probable as the ““B” arrangement, then the probability of obtaining Lovén’s law by chance alone in this specimen is 5/32 = 0.165. Thus the one specimen does not allow us to reject the null hypothesis; however, the edrioasteroid that is next described turns out to provide the additional evidence that is needed to reject the null hypothesis and to draw a conclusion regarding Lovén’s law in edrioasteroids. Order Stromatocystitoida: Smith’s (1985: 724, fig. 7) camera lucida drawing of spec- imen USNM 376690 of Stromatocystites walcotti Schuchert accurately presents the precise arrangement of the proximal am- bulacral flooring plates including the oral area. [A portion of this drawing was pub- lished by Paul & Smith (1984:454, fig. 7).] Photographs of this specimen (Smith 1985: plate 88, figs. 3-5) corroborate the drawing. Essential to my interpretation is the infor- mation that Smith’s drawing shows the in- terior of the oral surface. The plate arrange- ment is thus the mirror image of the plate arrangement that would be seen looking at the exterior oral surface. Using overhead transparency sheets and a photocopy ma- chine, the image was reversed to simulate an exterior view of the oral surface (Fig. 4). Analysis of this external view shows that Lovén’s law applies to this specimen. The probability that Loveén’s law will oc- cur in this specimen by chance alone under the null hypothesis is 5/32 = 0.165. Com- parison shows that Lovén’s law in this spec- imen of Stromatocystites has the same re- lation to the posterior interradius as in the specimen of Astrocystites. The probability that the same relation of Lovén’s law to the posterior interradius will occur in this spec- imen of Stromatocystites as was observed in the specimen of Astrocystites by chance alone under the null hypothesis is (5/32) 408 vy B PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 4. Lovén’s law in the Cambrian edrioasteroid Stromatocystites walcotti. Smith’s drawing of the oral area and the proximal ambulacral flooring plates as seen from inside the test (small insert) has been reversed in the enlarged figure to simulate an external view. The flooring plate half-series on the anticlockwise side of each ambulacrum is labeled as either in advance (A) or behind (B) the half series on the clockwise side. The arrangement AABAB is Lovénian. Ray II is anterior; posterior interray IV/V contains the periproct and hy- dropore. The meaning of Lovén’s law is interpreted from Stromatocystites to be a vestige of 2-1-2 pentaradiate organization, and to indicate a triradiate ancestry. (Copied and annotated from Paul & Smith 1984.) (1/5) = 1/32 = 0.033. Thus, I reject the null hypothesis in favor of the belief that the observed arrangement of the ambulacrals is not due to chance alone. The new working hypothesis is that Lovén’s law applies to Stromatocystites walcotti and to Astrocys- tites ottawaensis and that the posterior in- terradius is interradius IV/V in both spe- cies. Ophiocistioids (Fig. 5).—Jell’s (1983:230, fig. 14) camera lucida drawing of the oral surface of the well preserved holotype of the ophiocistioid Gillocystis polypoda shows the arrangement of the podial pores in the five ambulacral areas, together with the location of the madreporite and the periproct. Ex- amination of Jell’s drawing, reproduced here as Fig. 5, shows that Lovén’s law applies to the arrangement of the podial pores, which alternate according to the AABAB pattern of Paleozoic ophiuroids. Based on Lovén’s law, the madreporite of Gil/ocystis is in in- VOLUME 108, NUMBER 3 409 i & ? (|| XN 4 4 7 pet iy ly Vy 77 477, ~~ yy 7, a LZ 42 Sita Eee {| Wy77 22 ) Bs Aes rE ~ cH vy i Sw ~ Fig. 5. Lovén’s law in the Devonian ophiocistioid Gillocystis polypoda (oral side). Podial pores with no podia preserved are cross hatched. The small insert is a tracing of the podial pores and a reconstruction of the branches from the radial water vessel to the podia. The podial pores on the anticlockwise side of each ambulacrum are labeled as either in advance (A) or behind (B) the podial pores on the clockwise side. The arrangement AABAB is Lovénian. Ray II is anterior; posterior interray IV/V contains the periproct; right posterior interray III/IV contains the madreporite (M). (Copied and annotated from Jell 1983; areas where plates are not preserved are vertically lined or blank; positions of plates on the podia are stylized.) 410 terradius III/IV (as in ophiuroids), and the periproct is in the clockwise-adjacent inter- radius IV/V (as in edrioasteroids). The madreporite and periproct of Sollasina woodwardi also occupy clockwise-adjacent interradii (Haude & Langenstrassen 1976), but there are no observations concerning Lovén’s law in this species. In summary, Lovén’s law is now known to apply to certain edrioasteroids, ophiu- roids, echinoids, Bothriocidaris (considered a stem holothurian group by Smith 1984a) and an ophiocistioid. Thus, Lovén’s law is not unique to ophiuroids and echinoids. As noted above, it is plausible that Lovén’s law may eventually be found in the Asteroidea (such as Lanthanaster) and in the Somas- teroidea (such as Archegonaster). Symplesiomorphy of Lovén’s Law Is Lovén’s law in ophiuroids, echinoids, edrioasteroids and ophiocistioids a similar- ity that is due to an ancestral character shared with a remote ancestor (symplesiom- orphy)? Because the relationships among echi- noids, ophiuroids and edrioasteroids are distant, one must consider the likelihood (the probability under the null hypothesis HO) that these three lineages developed the same pattern of ambulacral plates merely by coincidence. [Ophiocistioids are not in- cluded because they cannot be treated as distantly related to echinoids: the jaw ap- paratus is so similar as to indicate that ophiocistioids were an offshoot from the echinoid lineage (Derstler 1985). Ophiocis- tioids may be grouped with echinoids or be substituted for echinoids, in this analysis. ] Of the 32 possible arrangements of alter- nating ambulacral plates around the peri- stome of a pentamerous animal, there are just 8 distinct patterns (Hotchkiss 1978): AAAAA, AAAAB, AAABB, AABAB, BBBBB, BBBBA, BBBAA, BBABA. Only the pattern AABAB conforms with Lovén’s law. The patterns AAAAA and BBBBB can PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON occur in an animal in only one arrangement, whereas five arrangements of each of the other six patterns can occur. Assuming that each of the arrangements confers equal fit- ness, the probability that three separate lin- eages of echinoderms would coincidentally develop the same pattern (not necessarily Lovén’s law) of alternating ambulacral plates is given by [2 X<(1/32)(1/32)(1/32) + 6x (5/32)(5/32)(5/32)] = 0.023, which is an im- probable coincidence. The probability that two lineages will coincidently develop Lov- én’s law given that the third lineage already conforms to Lovén’s law is (5/32)(5/32) = 0.024, which again is an improbable coin- cidence. Therefore, I conclude that the pres- ence of Lovén’s law in echinoids, ophiu- roids and edrioasteroids is not a chance co- incidence. It is considered to be due either to acommon inheritance (indicating ray ho- mologies) or to convergent evolution (in- dicating a selective advantage to the pattern AABAB in each of its five arrangements but constituting a false guide to ray homologies because the five arrangements cannot be distinguished from one another). Lorenz (1974) has written a lucid account of how to distinguish convergent evolution (anal- ogy) from common inheritance (false anal- ogy), and here I follow the examples given in his paper very closely. In this method, if the improbable coincidental similarity has a very dissimilar function in the forms in which it occurs, then it is extremely im- probable that the resemblance is due to par- allel adaptation. The lifestyles of echinoids, ophiuroids and edrioasteroids are so obvi- ously different that the functional mor- phology of their mouthframes and their proximal ambulacral plates must likewise be different. This makes the hypothesis H1 of a selective advantage that led to the con- vergent evolution of Lovén’s law in these three classes very unlikely. On the other hand, the hypothesis H2 that in edrioaster- oids Lovén’s law is a manifestation of the 2-1-2 symmetry pattern and that in echi- noids and in ophiuroids Loveén’s law is a VOLUME 108, NUMBER 3 411 Table 2.—Proposed ray homologies for echinoids, ophiuroids, ophiocistioids, edrioasteroids and asteroids. Roman numerals I, II, II], [V and V are used for homologous ray identifications based on recognizing Lovén’s law. The numbering for edrioasteroids is different from that of previous authors (e.g., Bather 1914a, Regnéll 1966, Bell 1976a). The numbering for ophiocistioids is different from that of Jeli (1983). Carpenter letters A, B, C, D and E for edrioasteroids are shown for reference (Moore & Fell 1966, Paul & Smith 1984). The ray homologies of asteroids are based on homologizing the location of closure of the ring canal in asteroids, ophiuroids and echinoids. Letters M, H, R and P mark the location of the madreporite (M) or hydropore (H), the location of closure of the hydrocoele crescent to form the ring canal (R), and the location of the anus/periproct (P). Echinoids II M Ill Ophiuroids II Ill M Ophiocistioids II Il M Edrioasteroids II Ill Edrioasteroids A B Asteroids A B vestige of this ancient symmetry, seems highly plausible. This result applies to the ophiocistioid as well. I exclude H1 and de- duce from H2 that Lovén’s law isa common inheritance that indicates ray homologies among ophiocistioids, echinoids, ophiu- roids and edrioasteroids [see Note 2]. Ray Homologies Based on Lovén’s Law Ray homologies based on Lovén’s law are proposed in Table 2. The location of the madreporite (M) or hydropore (H), the lo- cation of closure of the hydrocoele crescent to form the ring canal (R), and the location of the anus/periproct (P) are also shown in Table 2. Ray homologies for asteroids are included in Table 2 based on the proposal of homologizing the location of hydrocoele ring closure in asteroids with the IV/V in- terradius of ophiuroids and echinoids (Hotchkiss 1978). [The proposed ray ho- mologies for asteroids are indicated by Car- penter letters in Table 2. Roman numerals are not used because asteroid ray identifi- cations are not yet based on Lovén’s law. Carpenter letters are assigned on the basis that the IV/V interradius of edrioasteroids is also the CD interradius.] Location of hydrocoele ring closure. — With the ray homologies that are proposed in Ta- ble 2, the location of closure of the ring canal coincides in echinoids, ophiuroids and ed- IV R Vv P I IV R Vv I IV R?/P Vv I IV H/R/P Vv I C H/R/P D E C R/P D M E rioasteroids and is postulated to coincide in asteroids. It is in interradius [V/V when the ray numbering is based on Lovén’s law, and in the CD interradius of Carpenter’s system. Data on the plane of hydrocoele closure for echinoids, asteroids and ophiuroids is taken from Bury (1889). Bather’s opinion on the location of the plane of hydrocoele closure in edrioasteroids 1s accepted (1915:401): “in the primitive Pelmatozoon, so far as can be inferred from the embryology of Antedon and the anatomy of early forms, the closure of the hydrocoele was in what I [Bather] have termed the M plane [the plane con- taining the hydropore].”’ I have no data on the location of ring closure in ophiocistioids (but predict that it was in interradius IV/ V). Location of the madreporite. —With the ray homologies that are proposed in Table 2, the madreporite has a different location in four out of five groups. The one exception is that the madreporite is in interradius III/ IV in both ophiuroids and ophiocistioids. Relocation of the hydropore/madreporite from interradius [V/V in edrioasteroids to interradius III/IV in ophiuroids and ophio- cistioids may mean that the hydropore was duplicated on either side of ray IV in some ancestor. I accept the assumption of Mac- Bride & Spencer (1938) that the madrepor- ite in Ectinechinus and Eothuria (Upper Or- dovician) lies in interradius II/III. Reloca- 412 tion of the madreporite in asteroids from its original position in edrioasteroids was already conceded by Bather (1915), with the explanation that the hydropore has only a secondary connection to the hydrocoele [see Note 3]. Evidence cited by Moore & Fell (1966) against the doctrine ofa “‘firmly fixed location of the madreporite in all echino- derm groups” included a documented his- tory of migrations of the madreporite to dif- ferent portions of the echinoid apical sys- tem; the fact that some asteroids carry sev- eral madreporites scattered about the aboral side; and the fact that several euryalid ophiuroids have five madreporites, or else five hydropores, disposed one in each in- terradius. Location of the periproct/anus. — With the ray homologies that are proposed in Table 2, the periproct/anus has the same location in edrioasteroids, asteroids and the ophio- cistioid. It is in interradius IV/V when the ray numbering is based on Lovén’s law, and in the CD interradius of Carpenter’s system; this is the posterior interradius of edrioas- teroids. Ophiuroids do not have an anus [see Note 4]. In echinoids the anus/peri- proct has migrated to a new location: in exocyclic echinoids it is in the posterior V/I interradius; in endocyclic Cidaroida and post-cidaroid groups the anus is incipiently displaced in the direction of the V/I inter- radius, as indicated by the tendency docu- mented by Jackson (1912, 1927) for oculars V and I to be insert (Fell in Moore & Fell 1966). In the Saleniinae and some other Echinacea there has been a tertiary postero- lateral movement of the anus within the apical system toward ocular I (Fell & Paw- son 1966:U368); in Temnopleurus reevesi the anus is markedly excentric, midway be- tween oculars I and II (Clark & Courtman- Stock 1976:fig. 251; Baranova 1982:115, fig. 1). I suspect that the position of the mad- reporite and the anus in the ophiocistioid Gillocystis represents the original condition for echinoids. Bilateral symmetries. —When oriented PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON according to Table 2, the 2-1-2 plane of bilateral symmetry of edrioasteroids and Lovén’s plane of bilateral symmetry through the bivium and trivium of echinoids do not coincide. In exocyclic echinoids ray III is anterior, and in edrioasteroids ray II is an- terior. It is clear that there have been mul- tiple changes in symmetry during the evo- lution of echinoderms, such as from bilat- eral to radiate, and from radiate to bilateral (Bell 1976a:31, Caster 1967:S583, footnote) [see Note 5]. Vestiges of such changes may be retained in the morphology and ontogeny of the organism. The 2-1-2 plan developed from an original plan of triradiate symmetry (Bather 1900a, Fell 1966a, Sprinkle 1973). The hypothesis H2, established above, is that in edrioasteroids Lovén’s law is a man- ifestation of the 2-1-2 symmetry pattern and that in ophiocistioids, echinoids and ophiu- roids Lovén’s law is a vestige of this ancient symmetry. In echinoids this result is par- ticularly significant as it shows that Lovén’s III-5 plane of bilateral symmetry, which Fell showed to be a fundamental feature of all orders of echinoids from the Cidaroida on- ward, is a new, secondary plane. The hy- pothesis H2 leads by deduction to the hy- pothesis H3 that ray II was the anterior ray in ancestral lines of the Echinoidea. The Ordovician Ectinechinus and Eothuria have elongated tests with anterior mouth and posterior anus; they are elongate not in Lov- én’s III-5 plane, but along von Ubisch’s II-4 axis of primordial symmetry, with ray II anterior (MacBride & Spencer 1938) [see Note 6]. These observations support the hy- pothesis H3 that ray II was the original an- terior ray in both echinoids and edrioaster- oids. Applied to the ophiocistioid Gillocys- tis, the hypothesis H3 that ray II was an- terior leads to the deduction that the periproct was posterior. Identification of the Axis of the Bilateral Larva in the Adult Echinoderm larvae have a definite ante- rior-posterior axis and a marked bilateral VOLUME 108, NUMBER 3 symmetry. Proposals by von Ubisch (1913, 1927), Lane & Webster (1967), Macurda (1980) and Smith & Arbizu (1987) as to which axis of an echinoderm postlarva cor- responds with the axis of the bilateral larva are summarized in Table 3 [see Note 7]. As the working hypothesis H4 for the present study I assume that the relation of the axis of anterior-posterior organization of the larva to the axis of anterior-posterior organization of the imago has remained the same in echinoids, ophiuroids, edrioaster- oids and ophiocistioids through inheritance from their common ancestor. Von Ubisch (1913, 1927) showed that the anterior-pos- terior axis of the echinopluteus larva can be traced through metamorphosis by following the fate of the pieces of the larval skeleton. In this way he showed that the II-4 axis of the adult corresponds to the anterior-pos- terior axis of the bilateral larva, with ray II ‘“‘anterior”’ in relation to the larva. His re- sults were confirmed and expanded upon by Gordon (1929), Onoda (1931) and Emlet (1988, 1989) [see Note 8]. The ray homol- ogies of Table 2 show that von Ubisch’s II-4 ‘“‘axis of primordial symmetry” in echinoids coincides with the axis of the 2-1-2 pattern of bilateral symmetry of edrioasteroids. The 2-1-2 plan of symmetry in edrioasteroids expresses a left-right and an anterior-pos- terior symmetry, with ray II anterior and the anal interradius posterior. Using hy- pothesis H4 we therefore deduce that the anterior-posterior axis of the larva of ed- rioasteroids coincided with the anterior- posterior axis of the edrioasteroid, with ray II ‘anterior’ both in relation to the larva and in the imago. [Supporting indirect evidence is found in the many invertebrate groups that carry the anterior-posterior polarity of the larva through to the adult, including hol- othurians. Exceptions, when they occur, are probably due to secondary changes of sym- metry, as shown here for postcidaroid ex- ocyclic echinoids.] Smith & Arbizu (1987) reported a situs inversus specimen of the edrioasteroid Kra- 413 Table 3.—Synopsis of some proposals concerning the identification of an axis in the adult that corre- sponds with the anterior-posterior axis of the bilateral larva [see Note 7]. Echinoids II-4 (Ubisch 1913, 1927) Crinoids E-BC (Lane & Webster 1967) Blastoids D-AB (Macurda 1980) Edrioasteroids A-CD (Smith & Arbizu 1987) ma devonica and proposed that the mirror plane for situs inversus in edrioasteroids co- incides with the anterior/posterior plane of symmetry in the adult. Although their pro- posal may have seemed intuitively correct due to the 2-1-2 pattern of bilateral sym- metry of edrioasteroids, it overlooked the fact stated by Swan (1966:414-416) that mirror images made using any axis across the animal will each produce identical situs inversus results. Thus, situs inversus by it- self does not permit identification in the adult of the axis of symmetry of the larva. Nevertheless, it turns out that Smith’s con- jecture on the relation of the larval axis to the adult edrioasteriod is supported by the present study. [Conversely, the identifica- tions of the larval axes proposed for crinoids by Lane & Webster (1967) and for blastoids by Macurda (1980) are not consistent with working hypothesis H4 and imply either an error in analysis, or a change in the axis relating the larva to the imago during the evolution of these groups. | Larval Type of Ordovician Ophiuroids It is interesting to see what can be inferred about the larvae of primitive Paleozoic ophiuroids. [Direct paleontological evi- dence on the larval forms of echinoderms is commented on in Note 9.] Smith (in Smith & Arbizu, 1987) inferred a bilateral larva for edrioasteroids based on a situs inversus specimen of Krama devonica. In the same paper Smith referred to mirror image forms of the carpoid Peltocystis cornuta. The oc- 414 currence of mirror image forms is clear ev- idence of the ‘handedness’ that comes from deriving the water vascular system from the left hydrocoele of a bilateral larva in normal larvae and from the right hydrocoele in cases of situs inversus. It seems highly likely that not only edrioasteroids but also the derived asterozoans had a bilateral larva. The extended arms of the ophiopluteus depend on the skeletal rods for their support (Strathmann 1988:60). The acquisition of skeletal rods by echinoderm larvae was ex- plained by Strathmann (1993:91) as an ex- ample of heterochrony (or adultation) as follows: ““Echinoderms have a type of cal- cite skeleton unique to the phylum. Within the echinoderms, the calcite skeleton does not develop until metamorphosis in the as- teroids, but in the echinoids and ophiuroids, deposition of calcite begins in the embryo and produces supporting skeletal rods in the larva. This distribution of traits suggests that the skeleton originated in postlarval stages and that the pluteus originated by acceler- ated skeletal development.” In comparing the pluteus skeleton of echinoplutei and ophioplutei, Strathmann (1988:62) com- mented that “the formation and branching of the skeleton is so different between classes that homology can be questioned.” He sug- gested (in litt.) that independent evolution of arm rods in ophioplutei and echinoplutei is the simplest and most plausible hypoth- esis at this time. Smith (1984a:figure 9.4) listed the devel- opment of the pluteus type of larva and also Lovén’s law of heterotropy as two of the synapomorphies [shared derived character States] between echinoids and ophiuroids [see Note 10]. This was reiterated by Paul & Smith (1984:469): ‘“‘Echinoids share a number of advanced morphological inno- vations with Ordovician ophiuroids. These include. . . oral plating in obedience to Lov- én’s law and (judging from living animals) ... the development ofa pluteus larva with a Skeletal framework.’’ Ordovician ophiu- roids belong to the prephrynophiurid orders PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Oegophiurida and Stenurida. I favor a dif- ferent hypothesis based on the proposal by Fell (1966b:131) that the ophiopluteus is a postphrynophiurid development. In a series of papers, Fell (1948, 1963b, 1963c, 1966a, 1967) expressed his view that the pluteus arose independently in ophiu- roids and echinoids and that possession of the pluteus larval form is not evidence of a close phylogenetic relation between ophiu- roids and echinoids. He supported his views with data from embryology and larval forms and from the fossil record and adult anat- omy. Differences between the ophiopluteus and the echinopluteus were summarized by Smith (1984b:452—453) as follows: ““Nei- ther the processes nor the skeleton are iden- tical in echinoids and ophiuroids. The ophiopluteus has no pre-oral processes and the main locomotory processes that develop early on are the posterolateral ones, whereas in the echinopluteus, elongate pre-oral pro- cesses are present, and the main locomotory processes are the post-oral ones. The pos- terolateral processes either appear much lat- er in development and remain small or are absent altogether. The ophiopluteus has just two centers of calcification from which cal- cite rods grow, one on either side, whereas the echinopluteus has five, two on the left, two on the right, and an anterior V-shaped rod for the pre-oral processes.” Fell (1948: 83) described the same principal differences in the ophiopluteus: “the preoral arms of the echinopluteus are not represented; the internal skeleton takes the form of a pair of calcareous rods in the body, each sending branches into the four arms on its corre- sponding side; the posterior transverse rod is not represented.’ These differences sup- port the opinion already accepted by de Beer (1940:373) that this is a case of similarity through convergence (see also Mortensen 1921:227). Fell described the development of ““Kirk’s ophiuroid” (Fell 1941a), made observations on development in Ophiomyxa (Fell 1941b), and described the development of Amphi- VOLUME 108, NUMBER 3 Dholis squamata (Fell 1946). Fell (194 1a: 416, 1948:fig. 4) observed that there was no trace of larva in the development of Kirk’s ophiuroid and initially placed this form of development as the end member of his se- ries of the regression of metamorphosis (Fell 1945:90, fig. 20, 1948:98, fig. 6). Subse- quently, he identified Kirk’s ophiuroid as Ophiomyxa sp. (Fell 1963b:fig. 15 caption, 1966b:139, 1967:fig. 25 caption), and he identified Ophiocanops fugiens as a surviv- ing oegophiurid (Fell 1962). Fell then re- examined the development of the primitive Phrynophiurida and wrote (Fell 1963c:48 1): ‘Although we still know nothing of the em- bryology of Ophiocanops, many features of its anatomy declare its affinity to the Ophio- myxidae, a group of ophiuroids in which absolutely direct development occurs, with- out any trace of a larva at all. On the other hand, those genera of Ophiuroidea which have vestigial larvae have now been shown to fall in families of relatively late deriva- tion, from groups which have pelagic larvae. They are groups with numerous secondary features in the skeleton, far removed from the archaic forms with somasteroid-like fea- tures. Thus it is now extremely probable that there are two quite distinct types of direct development in ophiuroids, one an- cient, with no vestige of larva, the other secondary and showing both by the vestigial larva and by the characters of the skeleton that it is of late origin. I now suspect that the pluteus larva will eventually be proved to be a feature evolved by ophiuroids after the separation of ophiuroids from the so- masteroid line, and that the pluteus of echi- noids is an entirely independent develop- ment of that group.” Fell (1966a:237, 1966b: 131) proposed that the ophiopluteus evolved in the order Ophiurida and was not the orig- inal larval form of ophiuroids. The Ophiur- ida are shown on paleontological and mor- phological evidence to be a late grade of evolution (Fell 1963b:410, Table 2). Fell’s work was reviewed and extended by Dorothy Patent (1970) who described the 415 early embryology of the basket star Gor- gonocephalus caryi. The embryos reached a pentagon stage, without podia, before dying. Further development may occur in nature inside polyps of the alcyonarian Gersemia. She reported (1970:262): ““At no time did cilia or other locomotory structures devel- op. The development of G. caryi does not resemble that of any other ophiuroid stud- ied. There is no trace of a pluteus, and it is postulated that the pluteus evolved after the Phrynophiurida and the Ophiurida were dif- ferentiated.”’ Strathmann (1974:334-336, 1975, 1978) has persuasively argued that non-feeding larvae appear to be derived from feeding larvae, rather than the reverse. He noted that planktotrophic development appears to have been lost entirely from all lines of de- scent in the order Phrynophiurida [see Note 11]. He reasoned that because ““members of the order Ophiurida with lecithotrophy probably do not include the ancestors of the Phrynophiurida,” lecithotrophic develop- ment must have evolved independently in the Phrynophiurida or their ancestors. In other words, the Phrynophiurida must have had a planktotrophic ancestor. Combining the proposals of Fell, Patent, Strathmann, and Smith, I speculate that the larval type of the stem phrynophiurid was a bilateral planktotroph, but not yet an ophiopluteus. This hypothetical sequence is illustrated in Fig. 6, which includes the no- tion (following Fell 1962) that Ophiocanops may be a surviving representative of the stem group of the Order Phrynophiurida. Ray Homologies and 2-1-2 Symmetry Concerning the 2-1-2 pattern of ambu- lacra in Cambrian echinoderms, Paul & Smith (1984:470) stated that “‘in all of these early pentaradiate echinoderms, the single unbranched ambulacrum lies opposite the interambulacrum that contains the peri- proct, hydropore and gonopore (when they can be recognized).”’ They make it obvious 416 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON STEM OPHIURIDA VARI OPHIURIDA PELLET regression of metamorphosis ophiopluteus direct development metamorphosis brooding/viviparous INFERRED PHRYNOPHIURIDA STEM PHRYNOPHIURID planktotrophic directlabbreviated development Kirk's ophiuroid [Ophiomyxa] n h non-pluteus brooding/viviparous metamorphosis [?Ophiocanops] STEM OEGOPHIURIDA Lovén's law EXTINCT bilateral larva OEGOPHIURIDA metamorphosis planktotrophic Fig. 6. Inferred evolution of ophiuroid larvae and ophiuroid development. The larval type of the stem phrynophiurid is inferred to have been a bilateral planktotroph, but not yet an ophiopluteus. The ophiopluteus is a postphrynophiurid development. Hence the similarity of the pluteus larvae of ophiuroids and echinoids is judged a homeomorphism. VOLUME 108, NUMBER 3 that the solution to ray homologies is to identify the 2-1-2 pattern in the descendent groups. The present study shows that echi- noids, ophiuroids, edrioasteroids and an ophiocistioid have retained the imprint of the 2-1-2 pattern of bilateral symmetry in the form of Lovén’s law. Accordingly, ray homologies based on the 2-1-2 pattern of Lovén’s law, as presented in Table 2, satisfy the requirements established by Paul & Smith. Citing Lovén (1874), Fell (in Moore & Fell 1966) interpreted Lovén’s law of het- erotropy as a correlate of Lovén’s plane of bilateral symmetry. Because the basicoronal plates of Bothriocidaris conform with Lov- én’s law of heterotropy, this (supposed) cor- relation enabled Fell to write (p. U124): “The archaic Bothriocidaris (Ord.) at least already exhibited the same anteroposterior plane of symmetry that is manifested in the Cidaroida and in post-cidaroid groups, as indicated by the potential, incipient, or con- summated migration of the anus into in- teramb 5” [see Note 12]. It now turns out from the discovery of Lovén’s law in the Cambrian edrioasteroid Stromatocystites that Lovén’s law is more ancient than Lov- én’s plane of bilateral symmetry. Lovén’s law appears to be a manifestation of the plane of symmetry of 2-1-2 pentaradiate echinoderms, and Lovén’s plane is a super- imposed secondary development in echi- noids. In Lovénian symmetry, ray III is an- terior; in 2-1-2 symmetry it is ray II that is anterior. Adult Symmetry and Phylogeny Bell (1976b:1017) studied the early growth stages of edrioasteroids and showed that “edrioasteroid ontogeny bespeaks a primi- tive triradiate symmetry that was later mod- ified to a pentaradiate plan.’’ He remarked that echinoids and Asterozoa, among oth- ers, Show no apparent signs of triradial sym- metry. He concluded (p. 1018) that “If some classes of echinoderms are primitively tri- 417 radiates ... and if, in contrast, others are indeed primitively pentaradiates, then a major phylogenetic dichotomy occurred early in the history of this phylum.” The current research helps to remove this di- chotomy and to unify the evolutionary his- tory by showing that Lovén’s law in echi- noids, ophiuroids and ophiocistioids is a manifestation of the triradiate symmetry (2- 1-2 pattern) of edrioasteroids. The current research helps to answer questions posed in my first study: whether Lovén’s law independently evolved in echi- noids and ophiuroids, or whether it is in- dicative of a recent common ancestry be- tween echinoids and ophiuroids, or whether Lovén’s law is fundamental to the ground plan of the phylum (see Hotchkiss 1978). The probabilities calculated above show that the presence of Lovén’s law in echinoids, ophiuroids, edrioasteroids and ophiocis- tioids is not a chance coincidence. Because edrioasteroids and ophiocistioids also obey Lovén’s law, it is clear that Lovén’s law is not evidence of a recent common ancestry between echinoids and ophiuroids [see Note 13]. My suggestion that Lovén’s law might be fundamental to the ground plan of the phylum was based on the ‘Treatise’ classi- fication (Fay 1967) of Astrocystites in the Class Edrioblastoidea, within the Subphy- lum Crinozoa, making it seem totally un- related to either echinoids or ophiuroids. The situation is different now that Smith & Jell (1990) have shown that Astrocystites is an edrioasteroid. The result of this change is that all of the echinoderms in which Lov- én’s law is now known belong to classes that are placed by Smith (1984b) in the Sub- phylum Eleutherozoa. Therefore, I abandon my earlier suggestion and offer the more restricted hypothesis HS that Lovén’s law is fundamental to the ground plan of the clade Eleutherozoa (sensu Smith 1984b). Mortensen’s studies led him to regard Bothriocidaris “‘as a specialized offshoot from the Diploporite Cystids” (Mortensen 1928:93, 1930), and this caused him to 418 search for Loveén’s law in diploporite cys- toids. He regretted that he did not have any definite proof of Lovén’s law in diploporite cystoids, but he argued that “‘it is not so very wild a suggestion’; he also reasoned that if Lovén’s law is due to mechanical reasons, then it might have arisen indepen- dently in Bothriocidaris (Mortensen 1928: 107, 1930:343). I have examined the illus- trations in the ‘Treatise’ and other sources, hoping to detect Lovén’s law in cystoids. I did not find Lovén’s law but found an al- together different pattern. The 2-1-2 am- bulacral areas of Blastozoa such as the cys- toids Cystoblastus, Bulbocystis, Protocrin- ites and Glyptosphaerites, and the eocri- noids Rhopalocystis, Mimocystites and Ascocystites are all of the same “handed- ness’’, so that if one gives off a brachiole on the left side before any arise on the right, then it is that way in each of the ambulacral areas. This blastozoan pattern is so regular in its own way that it leads to the hypothesis H6 that there are at least two distinct con- structions of 2-1-2 symmetry: that seen in the Blastozoa, and that seen in the Eleuth- erozoa. [Derstler (1985) maintained the cri- noids as a 2-1-2 clade distinct from blas- tozoans and edrioasteroids, and so possibly there is a third construction of 2-1-2 sym- metry (not studied here).] These observa- tions support Derstler’s working hypothesis that the transition from triradiate to 2-1-2 symmetry may have occurred indepen- dently in these groups. [Epilogue: Thanks to Dr. David L. Paw- son I received a copy of a new paper by Dr. Reimund Haude (1994) on fossil Holothu- roidea (Eleutherozoa) just before returning this manuscript to the editor (February 1995). I wish to bring this paper to the at- tention of readers and to comment on its significance to the present results. Dr. Haude has observed in the pharyngial ring of a De- vonian holothurian that the radial pieces have a left and a right anterior process, and that in each radial element one or the other of these processes is pierced by a pore. The PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON pattern of these piercings is summarized by a RRLRL rule, where R stands for the right anterior process being pierced, and L for the left anterior process. Dr. Haude noted that this pattern might very well reflect Loveén’s law in the ambulacral plates of the ancestor to the holothurians. Thus his observations support the hypothesis H5 that Lovén’s law is fundamental to the clade Eleutherozoa. The relation of the RRLRL rule to the mad- reporite or to the bivium and trivium of holothurians is not yet known, but it seems likely that this information will be deter- mined in the near future. At that time it will become possible to add holothurians to the table of ray homologies. | Notes and Acknowledgements Note 1.—When the solution to the ques- tion of ray homologies among the classes of echinoderms has been worked out, it will suffice to have a single system of ray iden- tifications. At the present time there are sev- eral conflicting suggestions that purport to answer the question of the true ray homol- ogies, including the one presented in this paper. In this circumstance it may be ad- vantageous to have both Carpenter letters and Roman numerals available for the la- beling of the rays. I recommend use of Ro- man numerals for ray identifications that are based on Lovén’s law. Roman numerals were established for echinoids by Lovén (1874). Lovén (1874: pl. 53) numbered the rays of asteroids based on identifying the location of the madreporite with the II/III interradius of echinoids; Lovén (1874:88, figure) correctly observed the location of the asteroid anus, drawing it in interradius I/II adjacent to the madreporic interradius II/ III. Gemmill (1914:276; see also Chadwick 1923:9) numbered the rays of asteroids based upon the location of the closure of the hydrocoele ring, beginning with I at the anterior (or dorsal) horn of the hydrocoele crescent, and proceeding clockwise in oral view. This placed the madreporite of aster- VOLUME 108, NUMBER 3 oids in interradius I/II and the anus in V/I. Bather (1914a; see also Regnell 1966, Bell 1976a:9) identified the ambulacra of ed- rioasteroids with Roman numerals: I is left posterior, II is left anterior, III is anterior, IV is right anterior, and V is right posterior; the anus and hydropore are in interradius V/I. Spencer used Roman numerals in his monograph of the Paleozoic Asterozoa. His text-figure 1 (Spencer 1914:5) of the “Pel- matozoan ancestor of the Asterozoa”’ shows the water pore in interradius IV/V and the numerals I through V in clockwise manner when looked at from the oral surface. His text-figure 59 (Spencer 1916:103) of Me- sopalaeaster(?) ketley has the madreporite in interradius I/II and numerals I through V applied in clockwise manner when looked at from the aboral (apical) surface. I believe that he used the latter method to number the rays of Protaster sedgwicki (Spencer 1934:459-460). Hahn & Brauckmann (1981: 9) used the madreporite of the ophiuroid Chattaster hueffneri [as C. dillensis] to iden- tify the right anterior interradius (by anal- ogy with echinoids; see Haude 1982:25). Jell (1983) numbered the rays of the ophiocis- tioid Gillocystis with the periproct in inter- radius V/I (by analogy with echinoids). Note 2.— A consequence of this finding is that Asterozoa with alternating ambulacrals must have been ancestral to those with op- posite ambulacrals. In fact, this is exactly what Fell (1963c:476) had concluded from the arrangement of the ambulacrals in Chi- nianaster and Villebrunaster: “Since we have ascertained that the ambulacral ossicles al- ternate near the tip of the arm, and in the young stages of Chinianaster, it follows from Jackson’s (1899, 1903) so-called “‘law of lo- calized stages” that the opposite condition of ambulacral ossicles must have been de- rived from an original alternating condi- tion. This alternating condition was re- tained in Archegonaster, in some Paleozoic asteroids, and in some of the Paleozoic ophiuroids.”’ Sollas & Sollas (1912:223) reported dif- 419 ficulty in deciding whether the free, partly alternating ambulacral ossicles of primitive Paleozoic ophiuroids represented alterna- tion that was natural or that was due to displacement of free opposite ambulacral ossicles either during life or after death. The current research helps to resolve this diff- culty by determining that the progression of evolution of the ambulacral ossicles in ophiuroids must have been from an alter- nating arrangement that obeyed Loveén’s law, to a partly alternating condition, ultimately leading to ossicles that are paired. As noted by Sollas & Sollas, the paired ossicles were at first free, and later they connected and fused to form vertebrae. Fell (1963b,c) removed the asterozoans from the Eleutherozoa when he proposed that somasteroids with metapinnular struc- ture evolved from crinoids. The evidence presented here places the constraint of im- probable chance coincidence or parallel evolution of Lovén’s law onto that propos- al. The cumulative evidence from the pres- ent study, from reanalysis of plate homol- ogies (Hotchkiss 1993), and from the Mid- dle Cambrian Burgess Shale crinoid Ech- matocrinus (Sprinkle 1973) is that the crinoid arm and the somasteroid ray do not appear to be comparable structures. The conclusion that Lovén’s law is a plesiom- orphic character supports the working hy- pothesis of Bather (1900b) and Smith & Jell (1990) that the ancestry of sea stars is among edrioasteroids. Note 3.—It seems that in asteroids the madreporite does not develop in the same position that it does in ophiuroids, because assigning the location of the closure of the ring canal to interradius IV/V places the asteroid madreporite in V/I. This puts an unexpected gulf between the ophiuroids and the asteroids. Bather (1915) tried to solve the problem of deriving the asteroids from edrioasteroids by invoking “‘a shifting of the whole hydrocoele [such] that each lobe of it becomes applied, not to the ray to which it would (especially on any homology with 420 Pelmatozoa) naturally belong, but to the neighboring ray.” Bather also had to pro- pose (p. 401) that “Such shifting need not involve the hydropore, with which the hy- drocoele has only a secondary connection.” Embryology is beyond the scope of this pa- per and beyond my own expertise. How- ever, it seems to me that the best explana- tions concerning the relocating of the mad- reporite during the evolution of the echi- noderm classes will most likely be associated with evolutionary changes that occurred in the larvae and in the process of metamor- phosis. Perhaps the data are already avail- able. [One of my own unsuccessful attempts at a solution may be worth noting. We know that loss or gain of rays can occur at the ends of the hydrocoele crescent during metamorphosis (Hotchkiss & Seegers 1976). It seemed to me that this offers the possi- bility of losing and gaining rays such that the ray count remains at five. This could change the relative positions of the mad- reporite and the anus in edrioasteroids, as- teroids, echinoids and ophiocistioids. How- ever, no reasonable solution came from my attempts and I am now convinced that this has not happened. Such an event is not con- sistent with retaining the imprint of 2-1-2 symmetry represented by Lovén’s law in ophiuroids, echinoids and ophiocistioids. | Note 4. —In discussing the Echinozoa and the Asterozoa, Fell wrote (1963b:426): “There 1S no anus in archaic asteroids, nor in any ophiuroid, nor is it known in any somasteroid. An aboral anus occurs as a late acquisition in post-luidiid asteroids. On the other hand, fossils show that an aboral anus is a fundamental character of echinoids, present in the earliest forms. All holothu- rians have an aboral anus.” In the asteroid lineage the anus arose “after the initial dif- ferentiation of the Astropectinidae, so it is not an original feature of the family, and is lacking in some members” (Fell 1963b: Ta- ble 1). In Smith’s (1988a:fig. 7.2, 7.3) clado- grams, the anus was present in the edrioas- teroids, became missing in the stem so- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON masteroids, reappeared later in the asteroid line, did not reappear at all in the ophiuroid line, but reappeared in the echinoids, ophio- cistioids and holothuroids as a post-stem ophiuroid development. In the experience of Smith (1988b:819): ““The presence of an anus is difficult to detect in some fossils, but does genuinely appear to be absent in prim- itive asteroids and ophiuroids’”’. Blake (1987: 499, 506-507, 522; 1988) suggested that loss of the anus occurs rather readily and that the absence of rectal caeca and of an anus in the Paxillosida is a derived character state. Kesling (1962) reported a probable anal pyr- amid in Protopalaeaster narrawayi [Middle Ordovician]; Kesling & Strimple (1966) ob- served in the central plate of Calliasterella americana [Missisippian, transferred to Ca- lyptactis by Chesnut & Ettensohn 1988] a notch that may have bordered the anus; both reports locate the anus and the madreporite consistent with Table 2. An ophiuroid from the Silurian of Gotland that is reported to have an anal cone is under study by Regnéll (1973:fig. 4, cited by Franzen 1979:220). Accepting this report means that the anus was reacquired by a Silurian ophiuroid which appears not to have left any descen- dants (at least none with an anus). From examination of Ruedemann’s specimen of Stenaster salteri (= S. obtusus) with a sup- posed anal pore (Ruedemann 1916:54, plate 11, fig. 1) (Kilfoyle 1954:199, specimen NYSM 7744), I think it probable that the pore is an artifact from weathering. A care- ful search for an anus in fossils of Paleozoic asteroids and ophiuroids should continue to be made. It should be looked for not only on the aboral surface, but also on the oral surface (as in edrioasteroids) and on the in- terradial margins of the disc, as found in the ophiocistioid Gillocystis (Jell 1983). Note 5.—Evidence of multiple changes in symmetry in echinoids includes the mark- edly oval tests of the Echinometridae (Gra- bowsky 1994). The axis of elongation plain- ly differs from that of spatangoids and other exocyclic echinoids, yet does not alter the VOLUME 108, NUMBER 3 tendency for oculars I and V to become in- sert (Fell in Moore & Fell 1966:U124). Also in echinoids, the markedly oblique peri- stome in certain cassiduloids and holecty- poids may be mentioned (Rose 1976:303). The oral side of Amblypygus merrilli figured by Clark & Twitchel (1915:plate 76) dis- plays a very evident bilateral symmetry of the peristome and the ambulacra about the II-4 plane, whereas the periproct is elongate and symmetrical in the III-5 plane. [Roman (1957) pointed out that the bilateral sym- metry of Echinolampas is just external: the internal organs are not bilaterally arranged, and upon taking measurements the test it- self is found to be not perfectly symmetri- cal.] In crinoids may be mentioned Comatula pectinata showing a relative elongation of most of the ‘anterior’ arms (i.e., those cor- responding to the side of the disc opposite the anus) (Clark 1977:Fig. 6b). Also, the recumbent bent-crown Calceocrinidae are bilateral in the E-BC homocrinid plane (Lane & Webster 1967, Brower 1985). [In Holopus rangii the position of the bivium and triv- ium of shorter and longer rays differs among specimens (Donovan 1992:668).] In holothurians may be mentioned the Antarctic psolid Ekkentropelma_ brychia which has the mouth and anus of the U-shaped body displaced at 90 degrees to the usual holothurian plane of bilateral symmetry, while the sole is morphologically ventral, as usual. ““Thus, the sole is func- tionally lateral in position, and apparently the animal is well adapted to attaching itself to vertical surfaces” (Pawson 1971:113). In asteroids may be mentioned reports of leading arms in locomotion, of arm pref- erences in righting behavior, and also bi- lateral patterns of ray addition in some mul- tiradiate starfish (see review in Hotchkiss & Seegers 1976). Also, on the aboral surface of Luidia clathrata there is a conspicuous line of dark coloration on the midline of each arm and in just one interradius (the madreporic interradius); these dark lines 421 connect not to a central point but in a bi- laterally symmetrical pattern that suggests a trivium and a bivium of rays (Gray et al. 1968: 139, figure 8A). In ophiuroids may be mentioned Aster- onyx loveni with two nonadjacent thick and long arms and three intervening thinner and shorter arms, although their orientation rel- ative to the madreporite has not been de- scribed (Fujita & Ohta 1988, Irimura 1991). [The orientation relative to the madreporite of a similar pattern found in early growth stages of Ophiophragmus filograneus and Amphiura filiformis differs among speci- mens (Turner 1974, Muus 1981).] As an- other example, the juvenile Ophiomastix flaccida illustrated by Clark (1921:138, plate 13, figure 2) shows a color pattern with un- mistakable bilateral symmetry. The color pattern of Sigsbeia lineata suggests imper- fect bilateral symmetry (Lutken & Morten- sen 1899:plate 20, figure 8). Note 6. —Concerning Paleozoic echinoids with ‘irregular’ tendencies, the orientation of the oblong test of Hyattechinus beecheri has not yet been determined (Jackson 1912). The supposedly posterior anus in the Silu- rian Palaeodiscus and Echinocystites was shown by Hawkins & Hampton (1927) to be in the usual aboral, endocylic position of the Regularia. The elongate shape of Ecti- nechinus and Eothuria, although ques- tioned as possibly due to post-mortem dis- tortion by Kier (1965:442), is provisionally accepted herein because of the exceptional interest that attaches to the observation of MacBride & Spencer (1938) that the plane of elongation follows von Ubisch’s primor- dial plane of symmetry. Also of interest is the observation by Fell (1965:6) that ““The earliest Echinoidea, such as Kothuria, poss- esed a multiplated, flexible spirally twisted body wall, similar to that of the Helicopla- coidea, and perhaps inherited from a heli- coplacoid ancestry.” Note 7.—Evidence of the axis of the bi- lateral larva in the adults of Asterias rubens has been adduced from behavioral studies 422 (Smith 1950:216): ““The tendency for arm II to dominate the locomotory pattern is not to be ascribed to any obvious organizational feature such as greater arm length or a great- er number of podia. It appears rather to reflect some intrinsic feature of nervous or- ganization such as, perhaps, the retention of traces of the bilateral symmetry of larva.” [I regret that I do not know how Smith as- signed Roman numerals to the rays.] From anatomy and embryology (Lane & Webster 1967): ““The homocrinid (E-BC) plane of bilateral symmetry in the crown of some monocyclic inadunate and flexible cri- noids is postulated to be a relict expression in adults of the dorso-ventral symmetry plane of the doliolarian, free-swimming, larval stage.” From embryology (Grave in Brooks & Grave 1899:89, pl. I, fig. 6): In Ophioderma brevispina stage ““C”’ larvae the hydrocoele is ‘a horseshoe-shaped structure astride the oesophagus,” and “half of it lies to the right of the median sagittal plane of the larva and half to the left. Radial canal 3 lies in this plane and points directly toward the ante- rior of the larva.”’ [The relation of the lobes of the hydrocoele crescent and of the plane of hydrocoele closure to the axis of the larva may provide a means of tracing the larval axis through to the adult, and should be documented wherever possible; perhaps more of such data already exist.] From teratology (Macurda 1980:1161): “Abnormalities which were present in the bilaterally symmetrical larvae would be car- ried forward during metamorphosis to the pentameral adult and should have a bilat- eral distribution.... The data from this study appear to be supportive of the AB-D plane as being the larval symmetry plane in blastoids.’’ [See also Macurda 1964, 1978] Note 8.—The left side of the larva be- comes the oral surface, and the right side becomes the aboral surface. Hence the plane of symmetry of the echinopluteus is at right angles to von Ubisch’s plane of primordial symmetry in the adult. The axes of sym- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON metry are coincident even though the planes are at right angles (Onoda 1931:126). Crys- tallographic studies of echinoid ocular and genital plates, summarized and extended by Emlet (1988, 1989), confirm von Ubisch’s plane of primordial symmetry in many gen- era for which the development has not yet been directly observed. Note 9.—Fritsch (1908) interpreted the enigmatic Ordovician fossil Furca bohem- ica as acrinoid pluteus larva, but it was later shown to be a marellamorph (Arthropoda) (Perner 1919, Mortensen 1921:233, Prokop 1989:143-144). Echinopluteus skeletal rods and baskets were described by Deflandre- Rigaud (1946) from the Upper Jurassic of France. Emlet (1985) has shown that the crystallographic orientation of ocular and genital plates in echinoids is a reliable in- dicator either of indirect development with an echinopluteus (certain of these plates growing from the larval spicules of the echinopluteus) or of direct development without a pluteus (these plates then being formed de novo); he has used this technique to determine the mode of development of fossil echinoids. Note 10.—The use by Smith (1984a) and Paul & Smith (1984) of the pluteus larva as a synapomorphy between ophiuroids and echinoids has been questioned by Smith (1984b:452-453). He concluded that “There is therefore a distinct possibility that elon- gation of the small processes common to all eleutherozoan larvae occurred indepen- dently in ophiuroids and echinoids.” Here I consider an alternative to the proposal of Paul & Smith (1984:469) that the larval type of Ordovician ophiuroids was an ophioplu- teus. Note 11.—To the best of my knowledge, a planktotrophic larval stage is still not known in the Phrynophiurida (Fell 1967: S71-S72, Patent 1970, Strathmann 1975, Hotchkiss 1978:542-543, Hendler 1991). More research on the development of the Phrynophiurida is needed. Dowidar & El- Maghraby (1970:260) listed in their plank- VOLUME 108, NUMBER 3 ton the ophiopluteus of Ophiomyxa pen- tagona, but did not provide any evidence to support this identification. It is therefore very important that the embryology and lar- val type of Ophiomyxa pentagona, and of other ophiomyxine and euryaline phryno- phiurids be worked out. Knowledge of the development of Ophiocanops could be par- ticularly rewarding because it is thought to have retained the gonadal and stomachal characters of the Oegophiurida (Fell 1963c: 481, Hotchkiss 1977, Petr 1988:38). The egg size of Ophiocanops fugiens was judged by Hendler (1975) to indicate direct or ab- breviated development. [Concerning the classification of O. fugiens: It does not have the “‘auluroid”’ vertebrae of the Oegophiur- ida. The presence of oral and adoral shields, a first ventral arm plate, and streptospon- dyline vertebrae verify that Ophiocanops is a phrynophiurid (Mortensen 1933, Hotch- kiss 1977). The extraordinary soft part char- acters support family rank and suggest (fol- lowing Fell 1962) that O. fugiens may be a living representative of the stem group for the Order Phrynophiurida.] Hendler (1975) and Strathmann (1993) caution against in- ferences based on assuming that if a species has primitive features as an adult that its larval traits are also primitive. Strathmann also cautions (in litt.) that there is a weak- ness in parts of the inferences of Fell and Patent. He mentions as counter examples: that Pteraster tesselatus has no trace of bi- lateral larval symmetry yet it is descended from an ancestor with a brachiolaria (McEdward 1992); and that those species of Ophiolepis, Ophioderma and Ophionereis that lack any vestige of a larval skeleton in their vitellaria larvae are, nevertheless, de- scended from ancestors with an ophioplu- teus (Strathmann 1988, 1993). [Hendler’s (1982) studies on Ophionereis annulata and Mladenov’s (1985) studies on Ophiocoma pumila have now demonstrated a connec- tion between the ophiopluteus and the vitel- laria.}] As another counter example, Emlet informs me (in litt.) that he has studied the 423 development of an Amphiodia with a ben- thic egg capsule and a direct development with no vestige of a larval skeleton. These comments emphasize the speculative na- ture of attempting to deduce the character- istics of ancestral ophiuroid larvae. Note 12.—Jackson (1912:33) observed that “the first oculars to become insert are the plates of the bivium, next the plates of the posterior pair of the trivium, and last, if at all, the anterior odd plate of the trivi- um.... The ocular plates therefore in many regular Echini express a bilateral symmetry in this group, and an orientation passing through ambulacrum III and interambula- crum 5, the plane of symmetry adopted by Lovén.” Jackson extended this correlation to plates of the oral surface (1927:556): ““The law of sequence of incoming oculars in Echi- ni indicates an arrangement to the right and left of the anteroposterior axis through III- 5. This is in support of Loveén’s law of the orderly arrangement of primordial ambu- lacral plates in clypeastroids, spatangoids, and young regular Echini, the Ia, Ha, IIIb, IVa, Vb are larger, while the Ib, IIb, Ula, IVb, Va are smaller. So that from them as with a key one can gather the true orien- tation of an echinoid.”’ Fell (77 Moore & Fell 1966) added the insight that the insert condition of oculars I and V is the result of rearward migration of the anus toward in- terambulacrum 5, and that this rearward migration was already evident in the Ci- daroida. [It is also noteworthy that Onoda (1933) showed a physiological anterior-pos- terior axis in Heliocidaris that coincides with the III-5 axis of Lovén’s plane.] Note 13.—A close relation between echi- noids and asterozoans, based on study of Palaeodiscus, Aulechinus and Ectinechinus, was proposed by Spencer (1904), Bather & Spencer (1934), and MacBride & Spencer (1938). Evidence for a close relation be- tween echinoids and ophiuroids was devel- oped in detail by Hyman (1955:699) and most recently by Smith (1984a). Smith’s cladogram accounts for the similarity of the 424 adult morphologies of asteroids and ophiu- roids, and the (apparent) similarity of the larval morphologies of ophiuroids and echi- noids, by showing the stem ophiuroid as most closely related to the asteroids, and the stem echinoid as most closely related to the ophiuroids. However, with the similar- ity of the larvae now reappraised as a ho- meomorphism, with Lovén’s law compre- hended as an ancestral character that was present in the edrioasteroids (symplesiom- orphy), and with a new table of ray ho- mologies showing numerous changes in the location of the madreporite, the cladogram of Smith is in need of revision. As stated by Mooi (1989) it is the “‘in- terpretations of the homologies of charac- ters deemed important” that will determine the resulting cladogram, and not the actual use of cladistic methods. I accept the con- clusions of Fell (1963b, 1963c) that ophiu- roids display vestiges of metapinnular structure; that ophiuroids therefore derive from somasteroids (which have metapin- nules); and that echinoids derive from a stock with meridional growth gradients that did not have metapinnules. Therefore, I agree with Fell that ophiuroids do not qual- ify as ancestors to the echinoids. I agree with Fell that the Asterozoa and the Echinozoa are ancient and independent lineages, and I disagree with Smith’s (1984b) combining ophiuroids with echinoids and holothurians in a new Superclass Cryptosyringida. [A re- cent study on mitochondrial gene arrange- ments supports grouping asteroids with ophiuroids into the Asterozoa, and group- ing echinoids with holothurians into the Echinozoa (Smith et al. 1993).] I interpret the fact of sharing Lovén’s law to mean that ophiuroids and echinoids belong to the same 2-1-2 clade as the edrioasteroids. Thus I agree with Smith’s (1984b) reconstitution of the Subphylum Eleutherozoa, which he expands to include stromatocystitoid and other edrioasteroids. Note 14.—The probability under the null hypothesis HO that a out of n arms will PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON have a specific character state (the A or the B that is needed to spell out Lovén’s law) is given by the binomial distribution with p=q=0.5. The probability of observing a or more of the specific character in a sample of n observations has been calculated for each of the data on arms in Table 7 [exact calculations under the null hypothesis]. Arm I: Pia 2 9, n = 9) = 1/512 Arm II: P(a = 10, n = 13) = 378/8192 Arm III: P(a = 12, n = 12) = 1/4096 Arm IV: P(a = 10, n = 11) = 12/2048 Arm V: P(a = 6, n = 8) = 37/256. Using a criterion of P<0.05, the null hy- pothesis is rejected for arms I, II, III and IV. Although the null hypothesis could not be rejected for arm V, undoubtedly that will change when more Set II observations be- come available. The working hypothesis that each arm position relative to the madre- porite has a preferred character state, that the preferred character states spell out Lov- én’s law, and that this pattern identifies the madreporic interradius as III/IV is sup- ported. The working hypothesis does not assume that any subset of rays is more (or less) faith- ful to its predominant character state than any other subset. This allows pooling the 53 observations of Set IJ and the 60 observa- tions of Set I. The number of ocurrences of the predominant character state is a = 107 in n = 113 observations. The observed fre- quency is therefore 0.947. The probability that a specimen will have the predominant character state in all five of its arms, and therefore spell out Lovén’s law, is (0.947)° = 0.762. In my previous study (Hotchkiss 1978) I reported an estimated observed fre- quency of 0.617 [with a 0.99 confidence in- terval on the estimate spanning from 0.33 to 0.85]. Conference paper.—FPortions of this pa- per were presented at the second North American Friends of Echinoderms confer- VOLUME 108, NUMBER 3 ence and workshop at the Harbor Branch Oceanographic Institution, Inc., 9-11 July 1992, with the title “Footnotes on Lovén’s law’. In the excellent atmosphere of dis- cussions among the scientists at the meet- ing, I learned from Dr. Richard Mooi that he and Dr. Bruno David had also discov- ered that Lovén’s law applies to Stroma- tocystites walcotti, and that they had used Smith’s published drawings is exactly the same way as I report here. Further, Dr. Mooi informed me that he concurs with the char- acterization of Lovén’s law in echinoids as a recapitulation of the 2-1-2 organization of Camptostroma and the edrioasteroids. Acknowledgments I thank David L. Pawson for constant help for over 25 years, without which I would not have been able to do this or any other research on echinoderms. I thank Anita P. Hotchkiss, David L. Pawson, Vaclav Petr, Richard Strathmann, Richard Emlet, and Gordon Hendler for reading and com- menting on drafts of this manuscript. I thank Andrew Smith for his stimulating papers, his extremely important and exact obser- vations on Stromatocystites, and for cor- respondence on the subject of situs inversus; Peter A. Jell for help with photographs and observations on ophiuroids used in the Ap- pendix, and for reprints and his extremely important and exact observations on Gil- locystis. 1 thank S. K. Donovan and an anonymous reviewer for their helpful re- views. I thank S. K. Donovan for informing me of the suggestion by Conway Morris (1993) that Echmatocrinus is actually a coelenterate and not a crinoid. I thank J. Sprinkle for a verbal communication con- firming that Echmatocrinus has plates and reaffirming his opinion that it is a crinoid and not a coelenterate. I thank Gordon Hendler for informing me of the bilateral symmetry in Asteronyx and of the paper by Muus (1981), and for general communica- tions and encouragement over many years; 425 Richard Emlet for correspondence on the relation of the larva to the adult in echinoids and for reprints; Richard Strathmann for his willing responses to my phone calls for help; Jon W. Branstrattor for correspondence, re- prints and advice on locating specimens for this study; David L. Meyer for being a will- ing host during my visits to the University of Cincinnati; Richard Terry of the Uni- versity of Cincinnati for showing me ma- terial that he collected; W. Bruce Gibson of Cincinnati, Ohio, Thomas Frushour of Temperance, Michigan, and Kevin Brett of West Hill, Ontario, for allowing me to ex- amine Ordovician ophiuroid materials; Ja- net Waddington, Peter von Bitter and Des- mond Collins for their kind reception and help during my visit to the Royal Ontario Museum; Greg Buckley for help during vis- its to the Chicago Field Museum; H. C. Klinger of the South African Museum and J. E. Almond of the S. A. Geological Survey for information on Bokkeveld ophiuroids under study by P. A. Jell; John Harper, J. Roman and E. P. F. Rose for correspon- dence and reprints; Vaclav Petr for his gen- erous help through correspondence, latex casts of specimens, papers (including trans- lating the paper by Perner on Furca, and for additional references on Furca); Kraig Der- stler for latex casts of material in the North Museum at Franklin and Marshall College; Niles Eldredge and Melvin T. Hinkley for assistance with the collections at the Amer- ican Museum of Natural History; Ivo Chlu- pac for information and references on Furca bohemica; David N. Lewis for latex casts of BM(NH) material; Christina Franzén- Bengtson for information on the anal cone in the Gotland ophiuroid; the Boxboro L1i- brary for introducing me to Polya’s (1954) book on plausible inference; James Doherty and Alan Doherty for their expert help in preparing rubber casts for this study; Donna Nordberg for help in final preparation of the manuscript; Rita Vine for help in obtaining the copyright permissions for the figures. I thank Dan Blake, John Dearborn, Lee Ehr- 426 man, Julian Fell, Anita P. Hotchkiss, Grace M. Hotchkiss, Peter Jell, Caleb Kaufman, John Lawrence, David Pawson, Ira B. Per- elle, Vaclav Petr and Richard Strathmann for help with literature. I thank: The Pale- ontological Society for permission to use Text. Figs. | & 2 of Hotchkiss 1978 as the basis for Figs. 1 & 2 herein; Cambridge Uni- versity Press for permission to use Fig. 6 of Bather 1914b as the basis for Fig. 3; A. B. Smith and Cambridge University Press for permission to use Fig. 7 of Paul & Smith 1984 as the basis for Fig. 4; P. A. Jell and the Association of Australasian Palaeontol- ogists for permission to use Fig. 14 of Jell 1983 as the basis for Fig. 5; D. L. Pawson for providing the final preparation of the figures. Dedication This paper is dedicated in memory of Prof. H. Barraclough Fell (1917-1994), whom I cherished as mentor and friend. Fell’s en- thusiasm and charisma, and the sincerity and collegiality with which he treated his young students (as experienced by James F. Clark and myself in the 1960s) made him one of the most important influences in my life, both personally and scientifically. I be- lieve that these remarks would be readily echoed by his other students. To prepare this manuscript I reread papers by Fell that had ignited my interest in echinoderms while an undergraduate student. Part of the pleas- ure and the challenge of writing this paper has been to try to do justice to the seminal ideas of this great scholar. Thus, in the sec- tion of this paper where I discuss the ophio- pluteus, an additional goal is to provide ac- cess to Fell’s treatment of the topic by giving detailed page references. Literature Cited Baranova, Z. I. 1982. Occurrence of the sea-urchin of the tropical family Temnopleuridae off the Paramushir Island.— Academy of Sciences of the USSR, Zoological Institute, Explorations of the PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fauna of the Seas 29(37) Marine Invertebrates of Coastal Biocenoses of the Arctic Ocean and the Pacific Ocean: 114-118. Bather, F. A. 1900a. General description of the Echi- noderma. Pp. 1-37 in E. R. Lankester, ed., A treatise on zoology, part III, the Echinoderma. Adam & Charles Black, 344 pp. . 1900b. The Edrioasteroidea. Pp. 205-216 in E. R. Lankester, ed., A treatise on zoology, part III, the Echinoderma. Adam & Charles Black, 344 pp. 1914a. Studies in Edrioasteroidea. IV. The edrioasters of the Trenton Limestone [part I].— Geological Magazine, new series, decade VI, 1:115-125, pls. 10-12. 1914b. Studies in Edrioasteroidea. V. Ste- ganoblastus. — Geological Magazine, new series, decade VI, 1:193-203, pl. 15. 1915. Studies in Edrioasteroidea. [X. The genetic relations to other echinoderms.—Geo- logical Magazine, new series, decade VI, 2:393-— 403. —., & W. K. Spencer. 1934. A new Ordovician echinoid from Girvan, Ayreshire.— Annals and Magazine of Natural History, ser. 10, 13:557- 558. Beer, G. R. de. 1940. Embryology and taxonomy. Pp. 365-393 in J. Huxley, ed., The new system- atics. Oxford University Press, London. 583 pp. Bell, B. M. 1976a. A study of North American Ed- rioasteroidea.— Memoir 21 of the New York State Museum and Science Service, Albany, NY, 447 pp. 1976b. Phylogenetic implications of onto- genetic development in the Class Edrioastero- idea (Echinodermata).— Journal of Paleontolo- gy 50:1001-1019. Blake, D. B. 1987. A classification and phylogeny of post-Palaeozoic sea stars (Asteroidea: Echinod- ermata).—Journal of Natural History 21:481- 528. 1988. Paxillosidans are not primitive aster- oids: a hypothesis based on functional consid- erations. Pp. 309-314 in R. D. Burke, P. V. Mladenov, P. Lambert, & R. Parsley, eds., Echi- noderm biology: proceedings of the Sixth Inter- national Echinoderm Conference, Victoria, 23— 28 August 1987. A. A. Balkema, Rotterdam, Netherlands, 818 pp. , & T. E. Guensburg. 1989. Two new mul- tiarmed Paleozoic (Mississippian) asteroids (Echinodermata) and some paleobiologic im- plications.—Journal of Paleontology 63:331- 340. ———,& . 1993. New Lower and Middle Or- dovician stelleroids (Echinodermata) and their bearing on the origins and early history of the VOLUME 108, NUMBER 3 stelleroid echinoderms.—Journal of Paleontol- ogy 67:103-113. Branstrator, J. W. 1972. Lanthanaster cruciformis, a new Upper Ordovician sea star from Cincinnati, Ohio.—Journal of Paleontology 46:66-69. 1979. Asteroidea (Echinodermata). Paper 1066-F [Pp. F1-F7 + pls. 1-3] in John Pojeta, ed., Contributions to the Ordovician paleon- tology of Kentucky and nearby states.—U.S. Geological Survey Professional Paper 1066-A- G. Brooks, W. K., & C. Grave. 1899. Ophiura brevis- pina.—Memoirs of the National Academy of Sciences 8:79-100 + pls. 1-3. Brower, J. C. 1985. Ontogeny and functional mor- phology of two Ordovician calceocrinids. Pp. 13-18 in B. F. Keegan & B. D. S. O’Connor, eds., Echinodermata: proceedings of the Fifth International Echinoderm Conference, Galway, 24-29 September 1984. A. A. Balkema, Rot- terdam, Netherlands, 662 pp. Bury, H. 1889. Studies in the embryology of the echi- noderms.— Quarterly Journal of Microscopical Science, new series 29:409-447 + pls. 37-39. Caster, K. E. 1967. Homoistela. Pp. S581-S627 in R. C. Moore, ed., Treatise on invertebrate pa- leontology, Part S, Echinodermata 1, volume 2. The Geological Society of America and The University of Kansas Press, S297—S650 pp. Chadwick, H.C. 1923. Asterias.—Liverpool Marine Biology Committee Memoirs on Typical British Marine Plants and Animals 25:1-—63 + pls. 1-9. Chesnut, D. R., & F. R. Ettensohn. 1988. Homber- gian (Chesterian) echinoderm paleontology and paleoecology, south-central Kentucky.— Bulle- tins of American Paleontology 95(330):1-102 + table 6. Clark, A. M. 1977. Starfishes and related echino- derms. Third edition. T.F.H. Publications, Inc., Ltd. 160 pp. [Originally published as Starfishes and their relations, first edition 1962, British Museum (Natural History).] —., & J. Courtman-Stock. 1976. The echino- derms of southern Africa. British Museum (Nat- ural History), 277 pp. [British Museum (Natural History) Publication No. 776.] Clark, H. L. 1921. The echinoderm fauna of Torres Strait: its composition and its origin. — Depart- ment of Marine Biology of The Carnegie Insti- tution of Washington 10:1—223 + 38 pls. [Car- negie Institution of Washington Publication No. 214.] Clark, W. B., & M. W. Twitchell. 1915. The Mesozoic and Cenozoic Echinodermata of the United States. — Monographs of the United States Geo- logical Survey 54:1-341. Conway Morris, S. 1993. The fossil record and the 427 early evolution of the Metazoa.— Nature 361: 219-225. Deflandre-Rigaud, M. 1946. Vestiges microsco- piques des larves d’Echinodermes de |’Oxfor- dian de Villers-sur-Mer.—Comptes Rendus hebdomadaires des Séances de |l’Académie des Sciences 222:908-9 10. Derstler, K. L. 1985. Studies on the morphological evolution of echinoderms. Unpublished Ph.D. dissertation, University of California, Davis, University Microfilms International, 438 pp.— Dissertation Abstracts International 46:2238B. Donovan, S. K. 1992. Scanning EM study of the living cyrtocrinid Holopus rangii (Echinoder- mata, Crinoidea) and implications for its func- tional morphology. — Journal of Paleontology 66: 665-675. Dowidar, N. M., & A. M. El-Maghraby. 1970. The neritic zooplankton of the south eastern Medi- terranean at Alexandria. I. Distribution and ecology of the zooplankton organisms with spe- cial reference to Copepoda.— United Arab Re- public, Ministry of Scientific Research, Bulletin of the Institute of Oceanography and Fisheries 1:225-273. Durham, J. W. 1966. Classification [of echinozoans]. Pp. U270—U297 in R. C. Moore, ed., Treatise on invertebrate paleontology, Part U, Echinod- ermata 3, volume 1. The Geological Society of America and The University of Kansas Press, 366 pp. Emlet, R. B. 1985. Crystal axes in Recent and fossil adult echinoids indicate trophic mode in larval development.—Science 230:937-940. 1988. Crystallographic axes of echinoid gen- ital plates reflect larval form: some phylogenetic implications. Pp. 299-310 (chapter 23) in C. R. C. Paul & A. B. Smith, eds., Echinoderm phy- logeny and evolutionary biology. Clarendon Press, Oxford, 373 pp. 1989. Apical skeletons of sea urchins (Echi- nodermata: Echinoidea): two methods for in- ferring mode of larval development.— Paleo- biology 15:223-254. Fay, R. O. 1967. Edrioblastoids. Pp. S289-S292 in R. C. Moore, ed., Treatise on invertebrate pa- leontology, Part S, Echinodermata 1, volume |. The Geological Society of America and The University of Kansas Press, 296 pp. Fell, H. B. 1941a. The direct development of a New Zealand ophiuroid.— Quarterly Journal of Mi- croscopical Science 82:377-441 + pls. 20-22. . 1941b. Probable direct development in some New Zealand ophiuroids.—Transactions and Proceedings of the Royal Society of New Zea- land 71:25-26. 1945. A revision of the current theory of 428 echinoderm embryology.— Transactions of the Royal Society of New Zealand 75:73-101. 1946. The embryology of the viviparous ophiuroid Amphipholis squamata Delle Chiaje.— Transactions of the Royal Society of New Zealand 75:419-464. . 1948. Echinoderm embryology and the origin of chordates. — Biological Reviews 23:81-107. 1962. Evidence for the validity of Matsu- moto’s classification of the Ophiuroidea. — Pub- lications of the Seto Marine Biological Labo- ratory 10:145-152. 1963a. A new family and genus of Somas- teroidea.— Transactions of the Royal Society of New Zealand (Zoology) 3:143-146. 1963b. The phylogeny of sea-stars.— Philo- sophical Transactions of the Royal Society of London, Series B, 246:381—435 + pls. 50-51. 1963c. The evolution of the echinoderms. — Annual Report Smithsonian Institution 1962: 457-490. 1965. The early evolution of the Echino- zoa.— Breviora 219:1-17. 1966a. Ancient echinoderms in modern seas.— Oceanography and Marine Biology An- nual Review 4:233-245. 1966b. The ecology of ophiuroids. Chapter 6, pp. 129-143 in R. A. Boolootian, ed., Phys- iology of Echinodermata. Interscience Publish- ers, New York, 822 pp. . 1967. Echinoderm ontogeny. Pp. S60-S85 in R. C. Moore, ed., Treatise on invertebrate pa- leontology, Part S, Echinodermata 1, volume 1. The Geological Society of America and The University of Kansas Press, 296 pp. —.,&D.L. Pawson. 1966. Echinacea. Pp. U367— U440 in R. C. Moore, ed., Treatise on inver- tebrate paleontology, Part U, Echinodermata 3, volume 2. The Geological Society of America and The University of Kansas Press, U367—U695 pp. Franzén, C. 1979. Echinoderms. Pp. 216-224 in Val- dar Jaanusson, Sven Laufeld and Roland Sko- glund, eds., Lower Wenlock faunal and floral dynamics— Vattenfallet Section, Gotland.— Sveriges Geologiska Undersokning, serie C nr 762, Avhandlingar och Uppsatser, Arsbok 73 nr 3. 296 pp. Fritsch, Ant. 1908. Uber eine Echinodermenlarve aus dem Untersilur BOhmens.—Zoologischer An- zeiger 33:797-798. Fujita, T., & S. Ohta. 1988. Photographic observa- tions of the life style of a deep-sea ophiuroid Asteronyx loveni (Echinodermata). — Deep-Sea Research 35:2029-2034. Gemmill, J. F. 1914. The development and certain points in the adult structure of the starfish As- terias rubens, L.— Philosophical Transactions of PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON the Royal Society of London, series B, 205:213- 294 + pls. 18-24. Gordon, I. 1929. Skeletal development in Arbacia, Echinarachnius and Leptasterias.—Philosoph- ical Transactions of the Royal Society of Lon- don, series B, 217:289-334. Grabowsky, G. L. 1994. Symmetry, locomotion and the evolution of an anterior end: A lesson from sea urchins.—Evolution 48:1130-1146. Gray, I. E., M. E. Downey & M. J. Cerame-Vivas. 1968. Sea-stars of North Carolina.—U.S. Fish and Wildlife Service Fishery Bulletin 67:127- 163. Hahn, G., & C. Brauckmann. 1981. Ein neuer Ophi- uren-Fund aus dem Kulm von Herborn (Aster- ozoa, Unter-Karbon IIIa, Hessen).—Geolo- gisches Jahrbuch Hessen 109:5-18. Hammann, W., & S. Schmincke. 1986. Depositional environment and systematics of a new ophiu- roid, Taeniaster ibericus n. sp., from the Middle Ordovician of Spain. — Neues Jahrbuch fuir Geo- logie und Paldontologie. Abhandlungen. 173:47- 74. Hansman, R. H., F. C. Shaw, & W. A. Pettyjohn. 1962. Supplement to the catalog of the type specimens of fossils in the University of Cincinnati Mu- seum. University of Cincinnati, Cincinnati, 131 pp. Harper, J.A. 1985. A new look at Eugasterella logani (Hall, 1868) (Stelleroidea: Ophiuroidea) from the Middle Devonian of New York State. — An- nals of the Carnegie Museum 54:357-373. —., & R. W. Morris. 1978. A new encrinasterid ophiuroid from the Conemaugh Group (Penn- sylvanian) of western Pennsylvania, and revi- sion of the Encrinasteridae.—Journal of Pale- ontology 52:155-163. Haude, R. 1982. Ophiuren (Echinodermata) aus dem Karbon des Rheinischen Schiefergebirges. — Geologisches Jahrbuch Hessen 110:5-25. 1994. Fossil holothurians: constructional morphology of the sea cucumber, and the origin of the calcareous ring. Pp. 517-522 in B. David, A. Guille, J. P. Feral, & M. Roux, eds., Echi- noderms through time: proceedings of the eighth international echinoderm conference, Dijon, 6- 10 Sept. 1993. A. A. Balkema, Rotterdam, Netherlands, 940 pp. —, & F. Langenstrassen. 1976. Rotasaccus den- tifer n.g.n. sp., ein devonischer Ophiocistioide (Echinodermata) mit “holothuroiden” Wand- skleriten und “‘echinoidem” Kauapparat.— Pa- laontologische Zeitschrift 50:130-150. Hawkins, H. L., & S. M. Hampton. 1927. The oc- currence, structure, and affinities of Echinocystis and Palaeodiscus.—Quarterly Journal of the Geological Society of London 83:574-603, pls. 44-46. Hendler, G. 1975. Adaptational significance of the VOLUME 108, NUMBER 3 patterns of ophiuroid development.— American Zoologist 15:691-715. . 1982. An echinoderm vitellaria with a bilat- eral larval skeleton: evidence for the evolution of ophiuroid vitellariae from ophioplutei.— Bi- ological Bulletin of the Marine Biology Labo- ratory, Woods Hole 163:43 1-437. . 1991. Echinodermata: Ophiuroidea. Chapter 6, pp. 355-511 in A. C. Giese, J. S. Pearse, & V. B. Pearse, eds., Reproduction of marine in- vertebrates, vol. VI, echinoderms and lopho- phorates, The Boxwood Press, Pacific Grove, California, 808 pp. Hotchkiss, F. H. C. 1970. North American Ordo- vician Ophiuroidea—the genus Taeniaster Bill- ings, 1858 (Protasteridae).— Proceedings of the Biological Society of Washington 83:59-76. 1977. Ophiuroid Ophiocanops (Echinoder- mata) not a living fossil.—Journal of Natural History 11:377-380. 1978. Studies on echinoderm ray homolo- gies: Lovén’s law applies to Paleozoic ophiu- roids.—Journal of Paleontology 52:537-544. 1979. Case studies in the teratology of star- fish.— Proceedings of the Academy of Natural Sciences of Philadelphia 131:139-157. 1980. The early growth stage of a Devonian ophiuroid and its bearing on echinoderm phy- logeny.—Journal of Natural History 14:91-96. . 1993. A new Devonian ophiuroid (Echinod- ermata: Oegophiurida) from New York State and its bearing on the origin of ophiuroid upper arm plates.— Proceedings of the Biological So- ciety of Washington 106:63-84. —., & P.R. Seegers. 1976. Variable symmetry in starfish. — Thalassia Jugoslavica 12(1):173-180. Hyman, L. H. 1955. The Invertebrates: Echinoder- mata. Volume IV. McGraw-Hill, New York, 763 pp. Irimura, S. 1991. Ophiuroidea. Pp. 111-152 in C. Oguro and T. Okutani, eds., Echinoderms from continental shelf and slope around Japan, Vol. II. Japan Fisheries Resource Conservation As- sociation, Tokyo, 204 pp. Jackson, R.T. 1899. Localized stages in development in plants and animals.— Memoirs of the Boston Society of Natural History 5:89-153 + pls. 16- Bye 1903. Localized stages in the development of plants and animals [paper summary].—An- nals of the New York Academy of Sciences 15: 80-81. 1912. Phylogeny of the Echini, with a revi- sion of Palaeozoic species.—Memoirs of the Boston Society of Natural History 7:1-491 + pls. 1-76. 1927. Studies of Arbacia punctulata and al- lies, and of nonpentamerous Echini.— Memoirs 429 of the Boston Society of Natural History 8:437- 565. . 1929. The status of Bothriocidaris. — Bulletin of the Museum of Comparative Zoology at Har- vard College 69:481-512. Jell, P. A. 1983. Early Devonian echinoderms from Victoria (Rhombifera, Blastoidea and Ophio- cistioidea). Pp. 209-235 in J. Roberts & P. A. Jell, eds., Memoir 1. T. Dorothy Hill jubilee memozrr: proceedings of a meeting organized by the Association of Australasian Palaeontologists at the University of Queensland, 9th and 10th September 1982. Association of Australasian Palaeontologists, Sydney, 371 pp. Kesling, R. V. 1962. Notes on Protopalaeaster nar- rawayi Hudson.—Journal of Paleontology 36: 933-942 + pls. 133-134. 1969. A new bDrittle-star from the Middle Devonian Arkona shale of Ontario.—Contri- butions from the Museum of Paleontology, The University of Michigan (Ann Arbor) 23:37-51. —., & D. Le Vasseur. 1971. Strataster ohioensis, a new Early Mississippian brittle-star, and the paleoecology of its community. — Contributions from the Museum of Paleontology, The Uni- versity of Michigan (Ann Arbor) 23:305-341. —., & H.L. Strimple. 1966. Calliasterella amer- icana, a new Starfish from the Pennsylvanian of Illinois. — Journal of Paleontology 40:1157-1166 + pls. 151-153. Kier, P. M. 1965. Evolutionary trends in Paleozoic echinoids.—Journal of Paleontology 39:436- 465. . 1968. Echinoids from the Middle Eocene Lake City Formation of Georgia.— Smithsonian Mis- cellaneous Collections 153(2):1-45. Kilfoyle, C. F. 1954. Catalog of type specimens of fossils in the New York State Museum. Supple- ment 4.— New York State Museum Bulletin 348: 1-719. Lane, N. G., & G. D. Webster. 1967. Symmetry planes of Paleozoic crinoids.— The University of Kan- sas Paleontological Contributions 25:14—-16. Lewis, D. N. 1993. Catalogue of the type and figured specimens of fossil Asteroidea and Ophiuroidea in The Natural History Museum.—Bulletin of The Natural History Museum (Geology) 49:47- 80. Lorenz, K. Z. 1974. Analogy as a source of knowl- edge.—Science 185:229-234. Lovén, S. 1874. Etudes sur les echinoidées. — Konge- lige Svenska Vetenskaps-Akademiens Handlin- gar (n. ser.) 11(7):1-91 + pls. 1-53. Litken, C. F., & Th. Mortensen. 1899. The Ophiur- idae. Reports on an exploration off the west coasts of Mexico, Central and South America, and off the Galapagos Islands, in charge of Alexander Agassiz, by the U.S. Fish Commission steamer 430 ‘‘Albatross,” during 1891, Lieut. Commander Z. L. Tanner, U.S.N., commanding. — Memoirs of the Museum of Comparative Zoology at Har- vard College 23(2):1—208 + 23 pls. MacBride, E. W., & W. K. Spencer. 1938. Two new Echinoidea, Aulechinus and Ectinechinus, and an adult plated holothurian, Eothuria, from the Upper Ordovician of Girvan, Scotland. — Philo- sophical Transactions of the Royal Society of London, series B, 229:91-136 + pls. 10-17. Macurda, D.B. 1964. The blastoid Pentremites Say — a trimerous mutant and some new occur- rences.—Journal of Paleontology 38:705-710. 1978. Abnormalities and asymmetries in blastoids—a key to larval bilateral symme- try?. Geological Society of America Abstracts with Programs 10:449. 1980. Abnormalities of the Carboniferous blastoid Pentremites. —Journal of Paleontology 54:1155-1162. McEdward, L. R. 1992. Morphology and develop- ment of a unique type of pelagic larva in the starfish Pteraster tesselatus (Echinodermata: As- teroidea). — Biological Bulletin of the Marine Bi- ology Laboratory, Woods Hole 182:177-187. Mclver, M. A., & E. P. McIver. 1955. 300-million- year-old starfishes. — Natural History (Magazine of the American Museum of Natural History) 64:159. Melville, R. V.,& J. W. Durham. 1966. Skeletal mor- phology. Pp. U220—U257 in R. C. Moore, ed., Treatise on invertebrate paleontology, Part U, Echinodermata 3, volume |. The Geological So- ciety of America and The University of Kansas Press, 366 pp. Mladenov, P. V. 1985. Development and metamor- phosis of the brittle star Ophiocomina pumila: evolutionary and ecological implications. — B1- ological Bulletin of the Marine Biology Labo- ratory, Woods Hole 168:285-295. Mooi, R. 1989. Evolutionary dissent. — Paleobiology 15:437-444. Moore, R. C., & H. B. Fell. 1966. Homology of echi- nozoan rays. Pp. U119—U131 in R. C. Moore, ed., Treatise on invertebrate paleontology, Part U, Echinodermata 3, volume |. The Geological Society of America and The University of Kan- sas Press, 366 pp. Mortensen, Th. 1921. Studies of the development and larval forms of echinoderms. G. E. C. Gad, Copenhagen. 261 pp. + 33 pls. 1928. Bothriocidaris and the origin of echi- noids.— Videnskabelige Meddelelser fra Dansk naturhistorisk forening i Kebenhavn 86:93—122. . 1930. Bothriocidaris and the ancestry of echi- noids. A reply.— Videnskabelige Meddelelser fra Dansk naturhistorisk forening 1 Kobenhavn 90: 313-352. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 1933. Papers from Dr. Th. Mortensen’s Pa- cific Expedition 1914-1916. LX. On an extraor- dinary Ophiurid, Ophiocanops fugiens Koehler. With remarks on Astrogymnotes, Ophiopteron, and on an albino Ophiocoma.— Videnskabelige Meddelelser fra Dansk naturhistorisk Forening i Kobenhavn 93:1—21 + pl. 1. Muus, K. 1981. Density and growth of juvenile Am- phiura filiformis (Ophiuroidea) in the @Ore- sund.— Ophelia 20:153-168. Onoda, K. 1931. Notes on the development of He- liocidaris crassispina with special reference to the structure of the larval body.— Memoirs of the College of Science, Kyoto Imperial Univer- sity, Series B, 7:103-—134, pls. 5-7. 1933. On the orientation of the regular sea- urchin Heliocidaris crassispina.— Japanese Journal of Zoology 5:159-164. Owen, H.G. 1965. The British Palaeozoic Asterozoa. Table of contents, supplement and index:i-vii + 541-583.—Palaeontographical Society (Lon- don) volume for 1964. Parks, W. A. 1908. Notes on the ophiuran genus Protaster, with description of a new species. — Transactions of the Canadian Institute 8:363- B22. Patent, D. 1970. The early embryology of the basket star Gorgonocephalus caryi (Echinodermata, Ophiuroidea).— Marine Biology 6:262-267. Paul, C. R. C., & A. B. Smith. 1984. The early ra- diation and phylogeny of echinoderms. — Bio- logical Reviews 59:443-481. Pawson, D. L. 1971. Ekkentropelma brychia n. g., n. sp., an Antarctic psolid holothurian with a func- tionally lateral sole.— Proceedings of the Bio- logical Society of Washington 84:113-118. Perner, J. 1919. Furca bohemica—zastupce nové ¢e- ledi korysi v ceském siluru.—Casopsis Musea Kralovstvi Ceského 93:32-33. Petr, Vaclav. 1988. A notice on the occurrence of Bohemura jahni Jaekel, 1903 (Echinodermata, Ophiuroidea) in the Bohemian Middle Ordo- vician. — Véstnik Ustiedniho ustavu geologické- ho 63:35-38 + pls. I-IV. 1989. Revision of morphology and ecology of Bohemura jahni Jaekel, 1903 (Ophiuroidea, Protasteridae) from Bohemian Middle Ordo- vician.—Sbornik Narodniho Muzea v Praze (Acta Musei Nationalis Pragae), series B—his- torie naturalis, 45:1—20 + pls. I-VIII. Polya, G. 1954. Patterns of plausible inference. [vol- ume II of Mathematics and plausible reasoning. | Princeton University Press, 190 pp. Prokop, Rudolf. 1989. Zkamenély svét. Kotva, Prague. 275 pp. Regnéll, G. 1966. Edrioasteroids. Pp. U136—U173 in R. C. Moore, ed., Treatise on invertebrate pa- leontology, Part U, Echinodermata 3, volume VOLUME 108, NUMBER 3 1. The Geological Society of America and The University of Kansas Press, 366 pp. . 1973. Tidiga tagghudingar.— Fauna och Flo- ta 4:161—165. (Stockholm). Roman, J. 1957. Les dyssymétries des Echinolampas fossiles (Echinides exocycles cassiduloides).— Bulletin de la Societé Géologique de France, 6th series 7:67-82. Rose, E. P. F. 1976. Some observations on the Recent holectypoid echinoid Echinoneus cyclostomus and their palaeoecological significance. — Thal- assia Jugoslavica 12(1):299-306. Ruedemann, R. 1916. Account of some new or little- known species of fossils (Paleontologic contri- butions from the New York State Museum). — New York State Museum Bulletin 189:7-97 + pls. 1-31. Schuchert, C. 1915. Revision of Paleozoic Stellero- idea with special reference to North American Asteroidea.— United States National Museum Bulletin 88:1—311 + pls. 1-38. Smith, A. B. 1984a. Echinoid palaeobiology. [book series “Special topics in palaeontology”: 1] George Allen & Unwin, London, 190 pp. + Fig. A. 1. 1984b. Classification of the Echinoderma- ta.— Palaeontology 27:431-459. . 1985. Cambrian eleutherozoan echinoderms and the early diversification of edrioasteroids. — Palaeontology 28:715—756 + pls. 87-89. 1988a. Fossil evidence for the relationships of extinct echinoderm classes and their times of divergence. Pp. 85-97 in C. R. C. Paul & A. B. Smith, eds., Echinoderm phylogeny and evo- lutionary biology. Published for the Liverpool Geological Society by Clarendon Press, Oxford, 373 pp. 1988b. Patterns of diversification and ex- tinction in early Palaeozoic echinoderms. — Pa- laeontology 31:799-828. — ., & M.A. Arbizu. 1987. Inverse larval devel- opment in a Devonian edrioasteroid from Spain and the phylogeny of Agelacrinitinae. — Lethaia 20:49-62. —., & P.A. Jell. 1990. Cambrian edrioasteroids from Australia and the origin of starfishes. — Memoirs of the Queensland Museum 28:715- 778. Smith, J. E. 1950. Some observations on the nervous mechanisms underlying the behaviour of star- fishes. Pp. 196—220 in Symposia of the Society for Experimental Biology, no. IV, Physiological Mechanisms in Animal Behaviour, Academic Press, New York, 482 pp. Smith, M. J., A. Arndt, S. Gorski, & E. Fajber. 1993. The phylogeny of echinoderm classes based on mitochondrial gene arrangements.—Journal of Molecular Evolution 36:545-554. 431 Sollas, I. B. J., & W. J. Sollas. 1912. Lapworthura: a typical brittlestar of Silurian Age; with sugges- tions for a new classification of the Ophiuro- idea. — Philosophical Transactions of the Royal Society of London, series B, 202:213-—232 + pls. 9-10. Spencer, W. K. 1904. On the structure and affinities of Palaeodiscus and Agelacrinus. — Proceedings of the Royal Society of London 74:31—46. . 1914. A monograph of the British Palaeozoic Asterozoa. Part 1:1-56 + pl. 1.—Palaeonto- graphical Society (London), volume for 1913. . 1916. A monograph of the British Palaeozoic Asterozoa. Part 2:57—-108 + pls. 2-5.— Palaeon- tographical Society (London), volume for 1915. . 1922. A monograph of the British Palaeozoic Asterozoa. Part 5:197-236 + pls. 14-17.—Pa- laeontographical Society (London), volume for 1920. . 1925. A monograph of the British Palaeozoic Asterozoa. Part 6:237-324. + pls. 18-22.—Pa- laeontographical Society (London), volume for 1922. . 1927. A monograph of the British Palaeozoic Asterozoa. Part 7:325-388 + pls. 23-24.—Pa- laeontographical Society (London), volume for 1925. . 1930. A monograph of the British Palaeozoic Asterozoa. Part 8:389-436 + pls. 25—28.—Pa- laeontographical Society (London), volume for 1928. . 1934. A monograph of the British Palaeozoic Asterozoa. Part 9:437-494 + pls. 29-32.—Pa- laeontographical Society (London), volume for 1933. . 1951. Early Palaeozoic starfish. — Philosoph- ical Transactions of the Royal Society of Lon- don, Series B, 235:87-129. ,& C. W. Wright. 1966. Asterozoans. Pp. U4— U107 in R. C. Moore, ed., Treatise on inver- tebrate paleontology, part U, Echinodermata 3, volume 1. The Geological Society of America and The University of Kansas Press, 366 pp. Sprinkle, J. 1973. Morphology and evolution of blas- tozoan echinoderms. — Special Publication [The Museum of Comparative Zoology, Harvard University], 284 pp. Strathmann, R. 1974. Introduction to function and adaptation in echinoderm larvae.— Thalassia Jugoslavica 10:321-339. 1975. Larval feeding in echinoderms.— American Zooologist 15:717—730. 1978. Progressive vacating of adaptive types during the Phanerozoic.—Evolution 32:907- 914. . 1988. Larvae, phylogeny, and von Baer’s law. Pp. 53-68 in C. R. C. Paul & A. B. Smith, eds., Echinoderm phylogeny and evolutionary biol- 432 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 4.—Key for processing specimens into two mutually exclusive sets for the study of Lovén’s law and the location of the madreporite in Paleozoic ophiuroids. A OralPVie were eee rere ener Mra a IN DORAL VIEW re ce ee ne ee ear ea B. Specimen has recognizable madreporite........... Madreporte not found ™, 34704 ee @ Onetommore arms scorable-) 7) ee No arms scorable DFivekanms' scorablet. = eet er ee ee ee eee One to four scorable arms...................... E. Five arms conform to Lovén’s law .............. Five arms do not conform ogy. Published for the Liverpool Geological So- ciety by Clarendon Press, Oxford 373 pp. 1993. Hypotheses on the origins of marine larvae.—Annual Review in Ecology and Sys- tematics 24:89-117. Swan, E. F. 1966. Growth, autotomy, and regener- ation. Pp. 397-434 in R. A. Boolootian, ed., Physiology of Echinodermata. Interscience Pub- lishers, New York, 822 pp. Thoral, M. 1935. Contribution a l’étude paléonto- logique de l’Ordovicien inférieur de la Mon- tagne Noire et révision sommaire de la faune Cambrienne de la Montagne Noire.—Théses présentées a la Faculté des Sciences de l’Univ- ersité de Paris, Série A, No. 1541 (No. d’ordre: 2407). Imprimerie de la Manufacture de la Charité, Montpellier. 362 pp. Turner, R. L. 1974. Post-metamorphic growth of the arms in Ophiophragmus filograneus (Echinod- ermata: Ophiuroidea) from Tampa Bay, Florida (USA).— Marine Biology 24:273-277. Ubisch, L. von. 1913. Die Anlage und Ausbildung des skeletsystems einiger Echiniden und die Symmetrieverhaltnisse von Larvae und Ima- go.— Zeitschrift fur Wissenschaftliche Zoologie (Leipzig) 104:119-156 + pls. 6—7. 1927. Uber die Symmetrieverhaltnisse von Larven und Imago bei regularen und irregularen Seeigeln.— Zeitschrift fiir Wissenschaftliche Zoologie (Leipzig) 129:541-566. Waddington, J., P. H. von Bitter, & D. Collins. 1978. Catalogue of type invertebrate, plant, and trace fossils in the Royal Ontario Museum. — Life Sci- ences Miscellaneous Publications, Royal On- tario Museum, 151 pp. Appendix I Additional Data on Lovén’s Law and the Location of the Madreporite in Ophiuroids Methods. —Evidence of Lovén’s law is obtained by recording the arrangement of the ambulacral plates (Hotchkiss 1978). Examined from the oral surface, the ambulacral series on the anticlockwise side of each arm is either in advance (A) or behind (B) the series on the clockwise side of the arm. The arrangement AABAB is Lovénian, and the labels I, II, III, IV, V apply to the arms (Fig. 2). Two methods of analysis, based on mutually exclu- sive categories of specimens, provide separate looks at the location of the madreporite in connection with Lovén’s law in Paleozoic ophiuroids. Specimens were processed into two sets following the dichotomous key given in Table 4. The SET I collection of specimens all obey Lovén’s law. Hence the arms can be labeled I, II, III, IV, V according to Lovén’s system. The location of the mad- reporite is then determined by inspection to belong to a certain interradius by name, 1.e., I/II, or I/II, etc. The working hypothesis is that the madreporite occurs almost exclusively in the III/IV interradius (Hotchkiss 1978). The null hypothesis is that there is no preferred location. The SET II collection of specimens all have a mad- reporite. We will assign interradius designation III/IV to the madreporic interradius. Hence the arms can be labeled I, I, II, IV, V, according to their placement with respect to the madreporite. The character state of each scorable arm can be determined by inspection to be either “A” or “B.”’ The working hypothesis is that each arm is associated with a particular character state and that the pattern conforms with Lovén’s law. The null hypothesis is that there is no preferred character state for any arm. Materials. — This study can use only those Paleozoic ophiuroids that have alternating ambulacral plates. The preservation must permit individual recognition of the Ist, 2nd, 3rd, etc., plates of each half series of ambu- lacral plates in order to score the character state. In some specimens the lateral plates of the two sides of the arm can be used to assist in the interpretation. To assist in proper scoring of the ambulacrals, it is nec- essary to prepare latex or silicone rubber casts of spec- imens preserved as molds. In some specimens the ab- VOLUME 108, NUMBER 3 oral view can be scored. In exceptional circumstances data can be taken from published photographs. It is fairly common for published drawings to contain re- constructed detail that would not be reliable for this study. The specimens comprising Set I and Set II all belong to the families Protasteridae and Encrinasteridae, sub- order Lysophiurina, order Oegophiurida. Set I: Twelve specimens. Details are as follows: 101. Protaster sedgwickii Forbes, lectotype, Sedg- wick Museum No. A6374, Lower Ludlow (Upper Si- lurian), Lake District, England (oral view; seen) (Hotchkiss 1978:specimen no. 1) 102. Protasteridae sp., North Museum, Franklin and Marshall College No. PE38, Devonian, from roadside near Portland, New York (oral view; seen) (Hotchkiss 1978:specimen no. 6) 103. Hamling’s ophiuroid = Protasteridae sp. juv. British Museum (Natural History) No. E13737a, Up- per Devonian, Pickard’s Down, near Barnstaple, North Devon (oral view; seen) (Hotchkiss 1978:specimen no. 11; 1980:fig. 2B; Lewis 1993:69 as Drepanaster sca- brosus) 104. Strataster ohioensis Kesling & LeVasseur, para- type, University of Michigan Museum of Paleontology No. UMMP 58332a, Meadville Shale (Early Missip- pian), Cuyahoga County, Ohio (oral view; studied from the photograph: Kesling & LeVasseur 1971:pl. 7 fig. 4) (Hotchkiss 1978:542, 1993:65) 105. Eugasterella logani (Hall), Moscow Formation (Middle Devonian), near Earlville, Madison County, New York, the AAB*AB arm and madreporite ar- rangement reported by Harper (1985:367) is visible in his photograph of Carnegie Museum of Natural His- tory No. CM 34422 (oral view; studied from the pho- tograph: Harper 1985:fig. 7A) (The lateral plates help to score the arms.) 106. Protasteridae sp., South African Museum, Cape Town, SAM K1015 (formerly SAM 69d; originally from S.A. Geological Survey collections; plaster replica only), Gydo Formation (Devonian Bokkeveld Group), near Grootrivier, Cedarberg Mountains. (oral view; studied from photography supplied by Dr. P. A. Jell; under study by Dr. Jell; locality information from Dr. J. E. Almond) [an aged rubber pull of SAM 69d is in the Geology Museum of the University of Cincinnati] 107. Protasteridae sp., South African Museum, Cape Town, SAM K1014 (formerly SAM 203/67d; origi- nally from S. A. Geological Survey collections), Gydo Formation (Devonian Bokkeveld Group), near Groo- trivier, Cedarberg Mountains. (oral view; studied from photograph supplied by Dr. P. A. Jell; under study by Dr. Jell; locality information from Dr. J. E. Almond) [an aged rubber pull of SAM 203/674 is in the Geology Museum of the University of Cincinnati] 108-112. Protasteridae sp., Devonian Bokkeveld Group, South Africa. [oral views; studied from pho- tographs loaned by Dr. P. A. Jell; material is under 433 Table 5.—Set I data. Location of the madreporite in specimens that obey Lovén’s law. VII I/II I/IV IV/V v/I 0 0 12 0 0 study by Dr. Jell; repository and registration numbers will be reported by Dr. Jell (Gnformation not available)] Set II: Nineteen specimens. Details are as follows (specimen numbers according to Table 5): 201. Encrinaster hamlingi Spencer, holotype, Insti- tute of Geological Sciences, Leeds, Nos. GSM 37360, GSM 37361, Lower Carboniferous, Croyde, North Devon (counterpart halves; seen) (Hotchkiss 1978: specimen no. 2) 202. Taeniaster spinosus (Billings), holotype of Pro- taster? granuliferus Meek, Museum of Comparative Zoology, Harvard University, No. MCZ 470, Rich- mond Group (Middle Ordovician), Moore’s Hill, In- diana (oral view; studied from the photograph: Hotch- kiss 1970:fig. 6) (Hotchkiss 1978:specimen no. 13) 203. Strataster ohioensis Kesling & LeVasseur, para- type L-25e, Meadville Shale (Early Missippian), Cuy- ahoga County, Ohio (oral view; studied from the pho- tograph: Kesling & LeVasseur 1971:pl. 4 fig. 1; pl. 10, fig. 4) (Hotchkiss 1993:65) 204. Strataster ohioensis Kesling & LeVasseur, para- type L-25i, Meadville Shale (Early Missippian), Cuy- ahoga County, Ohio (oral view; studied from the pho- tograph: Kesling & LeVasseur 1971:pl. 6 fig. 2) (Hotch- kiss 1993:65) 205. Strataster maciverorum Hotchkiss, topotype, Panther Mountain Formation (Middle Devonian), near Cooperstown, New York; New York State Museum (uncatalogued Mclver collection: rock specimen DS1 2) (oral view; seen) (Mclver & Mclver 1955, Hotchkiss 1993:73) 206. Strataster maciverorum Hotchkiss, topotype, Panther Mountain Formation (Middle Devonian), near Cooperstown, New York; New York State Museum (uncatalogued Mciver collection: rock specimen DS70) (oral view; seen) (Mclver & Mclver 1955, Hotchkiss 1993:73) 207. Taeniaster spinosus (Billings), holotype of T. schohariae Ruedemann, New York State Museum No. 7784, Schenectady beds (Middle Ordovician), near Schoharie Junction, New York. (oral view; seen) (Kil- foyle 1954:204, 639; Hotchkiss 1970:fig. 3). 208. Protasterina fimbriata Ulrich, holotype, Econ- omy Formation (Middle Ordovician), Covington, Kentucky, University of Cincinnati Geology Museum No. 25001 (oral view; seen) (Schuchert 1915:pl. 36 fig. 4) (Hansman et al. 1962) 209. Protasterina fimbriata Ulrich, Utica Slate, Cin- cinnati, Ohio, American Museum of Natural History 434 AMNH 13190 (labeled Protaster flexuosa Miller & Dyer; Faber exchange) (oral view; seen; madreporite partly covered by spines) 210. Eugasterella logani (Hall), Moscow Formation (Middle Devonian), near Earlville, Madison County, New York, photograph published by Harper (1985:fig. 2B, fig. 3), Carnegie Museum of Natural History No. CM 34421A (oral view; studied from the photograph) 211. Protaster whiteavesianus Parks, syntype, Mid- dle Ordovician Trenton Group, Kirkfield, Ontario. Royal Ontario Museum No. 23846, inked #638T spec- imen B (oral view; fragmentary; seen) (Parks 1908:368, Waddington et al. 1978:132). 212. Taeniaster ibericus Hammann & Schmincke, holotype, Museum of the Instituto Geologico y Mi- néro, Madrid, No. S 587/2, “Tristani-beds” (Middle Ordovician), near Fontanosas, Spain (oral view; stud- ied from the photograph: Hammann & Schmincke 1986:fig. 8d.) 213. Taeniaster ibericus Hammann & Schmincke, paratype, Museum of the Instituto Geologico y Minéro, Madrid, No. S 587/1, “Tristani-beds” (Middle Or- dovician), near Fontanosas, Spain (oral view; studied from the photograph: Hammann & Schmincke 1986: fig. 10.) 214. Taeniaster ibericus Hammann & Schmincke, paratype, Museum of the Instituto Geologico y Minéro, Madrid, No. S 587/5, “‘Tristani-beds” (Middle Or- dovician), near Fontanosas, Spain (oral view; studied from the photograph: Hammann & Schmincke 1986: fig. 7a, 7c.) 215-219. Protasteridae sp., Devonian Bokkeveld Group, South Africa. [oral views; studied from pho- tographs loaned by Dr. P. A. Jell; material is under study by Dr. Jell; repository and registration numbers will be reported by Dr. Jell Gnformation not available)]} Excluded: The following excluded specimens de- serve special comment. For convenience of future ref- erence they are numbered: 301. Armathyraster paradoxis Harper & Morris, Brush Creek Shale (Pennsylvanian), Punxsutawney, Jef- ferson County, Pennsylvania, ABA*AA arrangement described by Harper & Morris (1978:157) [Carnegie Museum of Natural History No. CM 33966; counter- part halves; the aboral view is easily scored.] Excluded because the location of the madreporite is doubtful: it is ““not well enough preserved for complete identifi- cation” (Harper & Morris 1978:157). The reported score ABA*AA contains three disparities with the expected AAB*AB. An alternative possibility is that the hydro- pore is not associated with a visible madreporite, and that the specimen scores as AAB(*?)AA or AAA(*?)AB with only one disparity. 302. Bohemura jahni Jaekel, figured specimen, Letna Formation, Haj near Zahorany. National Museum {Narodni Museum], Prague, No. NM L 10172. Ex- cluded because none of the arms are scorable; in the PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON published drawing (Petr 1988:fig. 1a) details of ambs around the mouth opening were filled in for the purpose of illustration. (Oral view; latex pull supplied by Dr. Petr) 303. Bohemura jahni Jaekel, lectotype, Zahorany Formation, Zahorany near Beroun. National Museum [Narodni Museum], Prague, No. NM L 10066 [for- merly registered as No. EH 146]. Excluded because none of the arms are scorable; the drawing published by Spencer (1934:473, text-fig. 305; reproduced in Spencer & Wright 1966:U86 fig. 75,2a) was found by Petr (1989:1) to be incorrect. This specimen was reil- lustrated by Petr (1989:6, text-fig. 2, pl. I) and was designated the lectotype. (Oral view; latex pull supplied by Dr. Petr) 304. Euzonosoma orbitoides Spencer, holotype, Thraive Glen Starfish Bed. British Museum (Natural History) No. BMNH E52424b [formerly No. D. 52c in Mrs. Gray’s colln.] (Owen 1965:552, Lewis 1993: 69). Excluded because the location of the madreporite is doubtful: it is not distinct enough from the other disc plates for certain identification in this specimen. The detail of amb IV in the drawing published by Spencer (1930:414, text-fig. 265, indicating the amb arrangement ???*B?; drawing reproduced in Spencer & Wright 1966:U85 fig. 74,4b) is not confirmed by the fossil. I score the fossil as AA?(*?)A?. (Oral view; latex pull supplied by D. N. Lewis) 305. Protaster piltonensis Spencer, holotype, Lower Carboniferous of Top Orchard Quarry, Pilton, Devon. British Museum (Natural History) No. BMNH E13835b [formerly No. 1292 in the Torquay Natural History Society Collection] (Owen 1965:549, Lewis 1993:75). Excluded because none of the arms are scorable. The detail of amb IV in the drawing published by Spencer (1934:470, text-fig. 304, indicating the amb arrange- ment ???*B?) is not present in the fossil. (Oral view; latex pull supplied by D. N. Lewis) Results and analysis of Set I data.—The plate ar- rangements of the twelve Lovénian specimens are used to label the arms as I, II, III, IV, V. The madreporite is found only in interradius III/IV (Table 5). Under the null hypothesis of no preferred placement of the mad- reporite, the probability of observing the madreporite in the same interradius (not specifically the III/IV interradius, but any interradius) in all twelve speci- mens 1S | (0.2)!! = 0.00000002 Therefore the null hypothesis is reyected. The madre- porite occurs in interradius III/IV more frequently than can be accounted for by chance alone. Results and analysis of Set II data.—The madre- porite is used to label interradius III/IV. The remaining arms are labeled according to their placement with respect to the madreporite. This before-the-fact label- VOLUME 108, NUMBER 3 435 Table 6.—Set II data. The madreporite is used as a landmark to label the arms I, II, III, 1V, V. The A or B score of each scorable arm is recorded in the table. Specimen I II Ill * IV V 201 Encrinaster hamlingi — B B 4 A A 202 Taeniaster spinosus _ B B > A _ 203 Strataster ohioensis A — B * A B 204 Strataster ohioensis — A — * = — 205 Strataster maciverorum A A — bd A B 206 Strataster maciverorum — — B * — = 207 Taeniaster spinosus — — B e — = 208 Protasterina fimbriata A A — A B 209 Protasterina fimbriata A B — = B A 210 Eugasterella logani A A B *, A — Pale Protaster whiteavesianus — — B - a = 212 Taeniaster ibericus — — — a A — 213 Taeniaster ibericus — — — - — B 214 Taeniaster ibericus A A B 3 — — DNS) Protasteridae sp. — A B e A = 216 Protasteridae sp. A A — A B MSY Protasteridae sp. A A B * — — 218 Protasteridae sp. A B z — B 219 Protasteridae sp. A A B = A — ing is based on having a working hypothesis and pro- vides the convenience of not having to relabel the arms after doing the analysis. The A or B score of each arm is entered in Table 6. The working hypothesis is that the arms have predominant character states, these states spell out Lovén’s law, and ray numbering based on Lovén’s law places the madreporite in interradius III/ IV. The data of Table 6 are summarized in Table 7. The predominant character states occur more frequently than can be accounted for by chance alone (one sided test; see Note 14). It is seen that the predominant char- acter states of the arms spell out Lovén’s law. Num- bering the rays based on recognizing Lovén’s law shows that the madreporite occurs in interradius III/IV. Table 7.—Analysis of Set II data. The working hypothesis is that the observed character states will spell out Lovén’s law in a way that places the madreporite in interradius III/IV. Table entries record the number of times that the working hypothesis is fulfilled and the probability (P) of observing this many or more of the stated character under the null hypothesis HO of chance alone. Arm I has arrangement A in 9 out of 9 specimens, P = 0.00195 Arm II has arrangement A in 10 out of 13 specimens, P = 0.0461 Arm III has arrangement B in 12 out of 12 specimens, P = 0.000244 Arm IV has arrangement A in 10 out of 11 specimens, P = 0.00586 Arm V has arrangement B in 6 out of 8 specimens, P = 0.145 Lovénian arrangement found in 47 out of 53 arms PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(3):436-450. 1995. Alloeocomatella, a new genus of reef-dwelling feather star from the tropical Indo-West Pacific (Echinodermata: Crinoidea: Comasteridae) Charles G. Messing Nova Southeastern University Oceanographic Center, 8000 North Ocean Drive, Dania, Florida 33004, U.S.A. Abstract. —Alloeocomatella, a new genus of comasterid feather star from tropical Indo-West Pacific reefs, contains two species: A. polycladia, a new species, and A. pectinifera (A. H. Clark), the latter reassigned from the genus Comissia. Both are cryptic during the day; at night they extend their arms for feeding or emerge completely. A/loeocomatella differs from other genera in the family Comasteridae in its pattern of development of first brachial syzygies, and in the structure of its oral pinnule combs. 4. pectinifera has rays up to 500 mm long, the longest recorded for any extant feather star. Recent collections of shallow-water cri- noids from tropical western Pacific reefs in- clude a new species of Comasteridae that cannot be assigned to any named genus. The species exhibits the same arm branching pattern as Comatella A. H. Clark, but differs in pinnule and cirrus structure, and in the placement of initial brachial syzygies. The latter changes with growth, a trait appar- ently unique among comasterids. In speci- mens with 11-20 arms, the first syzygy is as in Comanthus and several other genera. In specimens with >20 arms, placement of syzygies approaches that of Comatella. Comparison with co-occurring Comissia pectinifera A. H. Clark indicates that the two species are congeneric. In fact, specific assignment of small ten-armed specimens is problematic. The two differ substantially from the type species of Comissia (C. luet- keni A. H. Clark) and so warrant a new genus. Material used in this study is housed in the National Museum of Natural History, Smithsonian Institution (USNM), British Museum (Natural History)(BMNH), Mu- seum national d’Histoire naturelle, Paris (PM), Los Angeles County Museum of Nat- ural History (LACM), and the Institut Roy- al des Sciences naturelles de Belgique (BELG). Terms, abbreviations, measure- ments and symbols are as follows: Centro- dorsal: central aboral plate. Cirri: aboral, segmented hooks attached to centrodorsal; Roman and Arabic numerals indicate num- bers of cirri/individual and segments (cir- rals)/cirrus, respectively (a range of values is usually given); LW of cirral: length to median width ratio (:1) when viewed lat- erally. Ray: one of five branched series of ossicles radiating from center of specimen; three rays arising closest to the mouth (dis- placed to one side of visceral mass in com- asterids) are anterior. Radial: (n.) first os- sicle of a ray or (adj.) a structure associated or oriented with a ray. Axil: ossicle at which a ray branches. Brachitaxis: series of ossicles following radial or axil and including the next axil; I-IJIBr: first through third brach- itaxes; Arabic numeral immediately follow- ing indicates number of ossicles in that brachitaxis (e.g., I[Br2). Arm: unbranched series of ossicles following distalmost axil; brachial (br; plural, brr): arm ossicle; sub- script numbers indicate specific ray ossicle (brachitaxis or arm) counting from first os- VOLUME 108, NUMBER 3 sicle after preceding axil or radial (e.g., IIBr,, br7); WL of ray ossicle: median width to midaboral length ratio when viewed abor- ally. Synarthry: articulation between first two ossicles of a brachitaxis or arm consisting of two ligament bundles separated by an aboral-oral fulcral ridge, sometimes with midaboral swelling. Syzygy (+): articulation between two successive ray ossicles con- sisting of radiating ridges and grooves and appearing externally as a perforated line (e.g., br3,4); Intersyzygial interval: number of ar- ticulations between successive syzygies. Pinnules (P): unbranched segmented ap- pendages arising from alternate sides of suc- cessive brachials; subscripts count pinnules from the most proximal; numbers and let- ters refer to pinnules along exterior and in- terior side of an arm, respectively (that is, the sides away from and toward the extrap- olated axis of the ray) (e.g., P,, P.); LW of pinnule ossicles (pinnulars): length to me- dian width ratio. Comb: modification of distal pinnulars of proximal (oral) pinnules producing comblike profile. Disk: central visceral mass or, specifically, its oral sur- face; anal interambulacral area: large area on disk surrounded by food grooves and bearing anal papilla. For further discussions and examples of comatulid morphometrics, meristics, abbreviations and symbology, see A. M. Clark & Rowe (1971), Breimer (1978), Hoggett & Rowe (1986) and Messing & Dearborn (1990). In all illustrations, sparse uniform stippling indicates articulations. Genus A/loeocomatella, new genus Diagnosis.—A genus of Comasteridae with all brachitaxes of two ossicles articu- lated by synarthry; arms 10-30, always ar- ranged in single plane (rays never twisted); when present, IIIBr series developed exte- riorly; first syzygy at br,,, on arms arising from IBr series; in specimens with <20 arms, first syzygy also chiefly at br;,, on arms arising from IIBr series; in specimens with = 20 arms, syzygies at br, ,. 3,4 0r br3,4 437 on arms arising from IIBr, and chiefly br, 42344 On interior arms and br;3,, on ex- terior arms arising from IIIBr; br, occa- sionally present alone; middle brr with raised axial lines; oral pinnule combs long, of 20— 37 teeth, occurring from P, to between P, and P,; teeth confluent with lateral margin of pinnular, 22 taller than their greatest width at mid-comb, remaining tall to pin- nule tip, and arising from side of pinnular away from arm; proximal tooth not trans- verse; basal carinae absent; cirri up to 26 segments, cylindrical proximally, com- pressed and wider distally; transitional and following cirrals with transparent distal rim, and with distally-directed midaboral spine located distally or subdistally initially, grad- ually moving to middle of segment and be- coming more erect on more distal cirrals; spine usually sharp, sometimes blunt on large cirri, Sometimes broadened as a nar- row transverse ridge, occasionally forked (especially opposing spine); mouth excen- tric; anal papilla close to mouth. Type species. —Alloeocomatella polycla- dia, new species. Other included species. —Comissia pec- tinifer A. H. Clark, 1911. Distribution. —Maldive Islands, Christ- mas Island (Indian Ocean), Indonesia, New Guinea, Bismarck Archipelago, Great Bar- rier Reef (Lizard I.), Fiji, Palau, New Cal- edonia, Chuuk Atoll. From 3 to at least 25 m [one record of 100 m (A. H. Clark 1931)]. Etymology. —From the Greek alloios (adXovos) “of another kind, different” (Brown, 1978) and Comatella, a comatulid genus that it resembles. Gender is female. Remarks. —Both species of Alloeocoma- tella, when handled alive, are less “sticky” and more flexible than many other com- asterids; lack of stickiness 1s perhaps due to relatively weak development of spines and hooks on distal pinnulars. Features of Comatella distinguishing it from Alloeocomatella are as follows: arms 10 to >80; rays in single plane or twisted, with exterior branches curved to oral side 438 of radial plane; first syzygy at br,,, on arms arising from IIBr and following brachitaxes with one exception: first syzygy at br,,, on exteriormost arm of IIIBr and following brachitaxes; br3,, often absent following br, 4,2, even in specimens with 40 arms. Sin- gle confluent comb teeth arising from in- terior side of pinnule (side closest to arm); individual teeth <2 taller than greatest width at mid-comb, usually 14—18/comb, and strongly reduced on last one or two pin- nulars; sharp rounded carina usually present on basal segments of proximal pinnules. Cirri bearing aboral transverse ridges (sometimes shallow V- or Y-shaped in ab- oral view), sometimes narrowing to erect rounded or triangular prominence or spine on more distal cirrals. The type specimen of Comissia luetkeni A. H. Clark, the type of Comissia, differs from Alloeocomatella as follows: cirri stout; aboral distal margins of all but basal and distal few cirrals strongly flared and some- times dentate; [Br2 series and arm bases as far as br, closely apposed and flat-sided; no raised axial lines on brr; pinnule combs of fewer than 20 teeth, arising abruptly on all oral pinnules; most pinnulars of middle and distal pinnules with strong distal spine; dis- tal pinnulars with LW <3.0; mouth subcen- tral; anus marginal. Alloeocomatella polycladia, new species Figs. 1-2, 3a—d, h, 4 Comatella maculata. —Meyer & Macurda, 1980, pp. 63, 68, 83, 96 (part); Meyer, 1986, pp. 203, 208-209 (part?). Diagnosis. —A species of Alloeocomatella with up to 30 arms; ray length up to 205 mm; anterior:posterior ray length ratio 1.1— 1.5:1; longest segments on distal pinnules with LW usually 3.0—5.0; longest cirrals (on mature cirri) with LW 1.3—2.0; anal inter- ambulacral area usually crowded with rounded or irregular, knobbed or molari- form nodules, but not often on anal papilla PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON itself. [Small ten-armed specimens may have more elongated cirrals and distal pinnulars.] Type series. — Holotype: USNM E44632, N side of Northeast Pass, S of Quoi I., Chuuk Atoll, Federated States of Micronesia, 7°31'32’N, 151°58'11’E, =18 m, 11 Jun 93, night. Paratypes: Chuuk Atoll: USNM E44633 (1 specimen) N side of NE Pass, S of Quoi I., ~18 m, 11 Jun 1993, night. USNM E44636 (1), N side of NE Pass, S of Quoi I., 9-18 m, 11 Jun 1993, night, Patrick Colin, coll. USNM E44634 (1), N side of NE Pass, S of Quoi I., 6 m, 8 Jun 1993, Larry Sharron, coll. BMNH 1994.5850 (1), barrier reef S of Otta I., 07°08'45’N, 151°53'11’E, 11 m, 9 Jun 1993. LACM 93- 95.1 (1), barrier reef S of Otta I., 9 m, 9 Jun 93, Larry Sharron, coll. Papua New Guinea: PM ECCh-16 (1), fringing reef, N side Na- gada Harbor, Madang, 05°09'29"S, 145°49'21”E, 9 m, 10 Jul 1991, night. USNM E44642 (1), Nagada Harbor area N of Madang, no field data, Jul 1991. USNM E44635 (1), fringing reef drop off, Cape Croisilles, 4°51'30"’S, 145°48’E, 30 km N of Madang, 6 m, 13 Jun 1992, Bert Hoek- sema, coll. USNM E44637 (1), barrier reef E of Wongat I., Madang, 05°08'09’S, 145°50'51”E, 3 m, 11 Jul 1991. (All collec- tions by the author except where noted.) Other material examined. —Caroline Is- lands: USNM E44770 (4 specimens), E34983 (2), E44693 (18), E44694 (21), Pa- lau Is., D. L. Meyer, coll. Australia: USNM E44771 (3), E44768 (3), Lizard I., Great Barrier Reef, D. L. Meyer, coll. Fiji: USNM E44769 (2), D. L. Meyer, coll. Indonesia: USNM E35033 (1), Marsegoe I., N end of Ceram I., Moluccas (2°59'48"S, 128°03’E), 15m, D. L. Meyer, coll. Papua New Guinea: BELG-418 (1), Platier, Laing I., night, 23 Jul 1989, 20 m, M. C. Lahaye, coll. Description of holotype. —Centrodorsal a thick pentagonal disk with sloping sides; diam. = 9.3 mm; aboral surface rugged with deep central depression and traces of former cirrus sockets. Cirri LXV, 21-25, 16—26 mm long, of variable length and robustness, VOLUME 108, NUMBER 3 crowded two or three deep around centro- dorsal margin, with few sockets encroaching on aboral surface. First segment very short; second wider than long; third squarish; fourth or fifth to eighth, ninth or tenth cir- rals longest, LW = 1.3, slightly constricted. Following cirrals gradually decreasing in length; cirrals in distal third of cirrus shorter than wide, LW = 0.7. Twelfth to fifteenth cirral (on larger cirri) transitional, shiny dis- tally. Tip of sharp aboral spines occasionally finely divided into several tiny teeth. An- tepenultimate cirral with transverse aboral ridge, sometimes with prominent ends forming a pair of spines. Spine on preceding cirral (fourth from end including claw) may also be slightly widened transversely. Op- posing spine also a transverse ridge. Trans- verse ridges may be irregularly denticulate. Anterior rays 150—160 mm, posterior rays 140 mm long. Arm number 30. Radials hid- den by centrodorsal. Rays separated prox- imally; aboral surface of disk visible be- tween adjacent rays. Brachitaxes with low midaboral synarthrial swelling straddling articulation of first and second ossicles. IBr, short, partly hidden by centrodorsal (mostly hidden by cirri). Axil (IBr,) with very short, diverging lateral margins, WL = 2.0. IIJBr uniformly developed exteriorly. I] and IJIBr, and br, slightly longer exteriorly and united interiorly; articulations with low midaboral synarthrial swellings; each brachitaxis with gently concave lateral margin. Br, and br, (or br,,2) longer exteriorly. Br,,, oblong, WL = 1.4-1.7, diam. = 1.9-2.1 mm. Br,_, oblong, with well-developed alternating ar- ticular tubercles, WL = 1.7—2.0. Brg_, cu- neate. Following brr triangular, WL = 2.3, diam. = 2.1—2.4 mm (slightly wider than arm base). Brr become shorter by mid-arm, very strongly cuneate or triangular, WL = 2.6—3.0; distal margins thickened (raised but not everted) and spinose; thickening best developed along middle portion of arm. Distal brr almost oblong (slightly longer on one side), WL = 1.5; thickening of distal margins reduced; spines present only mid- 439 aborally. Aboral surface of arms beyond the proximal several brr with numerous fine raised axial lines projecting beyond distal margin of ossicle as distal rim of spines; these spines fewer and midaboral on distal brr. On arms arising from IIBr, syzygies at br, 42344 or br3,,4 alone. Br3,, on exterior arms and br, ,.3,,4 On interior arms arising from IIIBr. Following syzygy br,54,. to brj9429. Next intersyzygial interval chiefly 4-5 (few 3); distal intersyzygial interval chiefly 3 (some 4, 5). P, of 64 segments, 25 teeth, L = 23 mm; basal several segments shorter than wide; most segments about as long as wide, with distal rim of spines best developed on side of pinnular facing arm tip (when pinnule extends outward from arm); middle and dis- tal pinnulars also with cluster of spines on side of pinnular facing arm tip; lateral spines consolidating to form rudimentary proxi- mal comb teeth; comb teeth tall, narrow and usually triangular. P, and following oral pin- nules similar to P, but decreasing in length through last comb-bearing pinnule (P,—P,); with comb teeth developing more abruptly, few middle pinnulars longer than wide (LW to 1.3), and comb occupying relatively more segments per pinnule. P, of 54 segments, 33 teeth, L = 16 mm; P; of 45 segments, 27 teeth, L = 13 mm; P, of 30 segments, 20 teeth, L = 9 mm. P, or P, without a comb, 24 segments, L = 9 mm; basal segments short; most middle segments squarish to somewhat longer than wide, LW to 1.4; all segments except basal 3 with numerous spines on side of pinnulars facing arm tip. Middle pinnules (e.g., P59) with 29 seg- ments, L = 14 mm; similar to P, but more robust; segments with LW up to 1.4. Distal pinnules much slenderer, with up to 28 seg- ments, L = 13 mm; basal two segments short; following segments longer than broad (except near tip), becoming very slender in mid-pinnule, LW to 3.5; proximal segments with few distal spines; middle and distal segments with few weak lateral spines; last four segments with typical strong hooks. 440 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Alloeocomatella polycladia, new species. a, holotype (USNM E44632), aboral view; b, paratype (LACM 93-95.1), aboral view. Scales: 1 cm. VOLUME 108, NUMBER 3 Few large irregular nodules on large swol- len anal interambulacral area; some almost foliose. Nodules more compact on anal pa- pilla; some molariform. Anal papilla adja- cent to mouth. Other specimens. —One paratype (USNM E44633) collected with the holotype and three collected by D. L. Meyer at Lizard Island (USNM E44771) as large or larger than the holotype: centrodorsal diameter 9- 10 mm; cirri LX—LXX, the largest with 24- 25 segments, 23-27 mm long; ray length 150-205 mm; arms 28-30; P, up to 63 seg- ments. Most material smaller: 15-21 arms, ray length 105-150 mm (mostly <125 mm), and centrodorsal diameter 5.5—7.0 mm. An- terior:posterior ray length ratio 1.3-1.5:1. Central aboral depression shallower on smaller specimens. As in the holotype, cirri usually varying substantially in length, number of segments and robustness in in- dividual specimens; chiefly XXXV-LV, with largest cirri of 16—20 segments, 15-18 mm long. Longest cirrals usually fourth and fifth or sixth, with LW = 1.5—2.0 (to 2.3 on smaller cirri). P, usually 43-54 segments, 22-30 teeth, 14-20 mm long. P, or P; usually the last comb-bearing pinnule. Middle and distal pinnules correspondingly smaller than those of holotype (e.g., middle pinnules with 21 segments, 9 mm long; distal pinnules with 24—25 segments, 11-12 mm long). Middle pinnules of 20-21 segments, 9.0 mm long, not especially more robust than more prox- imal (non-combed) pinnules; LW of middle pinnulars up to 1.7. Distal pinnules differing from those of holotype in having one or few strong distal spines on both proximal and middle pinnulars, LW = 2.7-5.0, and only 0-3 weak lateral spines on distal few pin- nulars (not including distal hooks). Anal interambulacral area often crowded with rounded or irregular nodules except on anal papilla itself. Nodules usually with a knob; several sometimes coalesced into larger ridged or molariform structures, 441 sometimes forming an irregular pavement. Nodules sometimes also present on several small peripheral interambulacral areas. Oral surface completely naked in a few speci- mens. Smaller ten-armed specimens differ as follows: centrodorsal diameter 2.9—4.6 mm; central depression slight or absent; ray length up to about 110 mm; cirri usually XXITX— XXXVI (rarely less), the largest of 16-19 segments; longest cirral with LW = 1.7-2.6; only the opposing spine widened or forked; segments of distal pinnules with LW to ~6.0. Two small specimens from Palau have 6 rays and 12 arms. Description of an immature specimen. — (USNM E44693) Centrodorsal a thin disk, 3.0 mm across; aboral surface slightly con- vex, without central depression; cirri X XIII, 12-16, 11.4 mm, third to sixth segments elongated with expanded ends, fourth to fifth or sixth segments longest, with LW = 2.2- 2.4 (3.0 on a much smaller cirrus); sixth or seventh segment transitional, shorter and wider, LW = 1.6; distal several segments slightly wider than long; opposing spine slightly widened transversely. Most rays broken; anterior ray 65 mm, posterior ray ~ 45 mm; radials visible in interradial angles; IBr, completely exposed, oblong, WL = 2.2; axil pentagonal with di- verging lateral margins, WL = 2.0; br, and br, oblong, WL = 1.7; br, united interiorly only proximally; br;,, oblong, WL = 1.2, diameter = 1.0 mm; br,_, oblong, WL = 1.3-1.4; following brr becoming cuneate, none triangular; middle brr cuneate, WL = 1.0, with distal margins raised, spinose, no axial lines; distal brr almost oblong, with few coarse midaboral distal spines; WL = OFF P, of up to 35 segments, 19 teeth, 9.7 mm long; P, the last comb-bearing pinnule, smaller and slenderer than P,, up to 29 seg- ments, 19 teeth, 7.6 mm long; P;, P. and P, much smaller and slenderer than P, not present on all remaining arms; middle pin- nules of 15 segments, 6.1 mm long; distal 442 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Alloeocomatella polycladia, new species. a—b, margin of centrodorsal and base of one ray; a, large paratype with 30 arms (USNM E44633) (stippling on left side indicates reticulated tissue that obscures artic- ulations; typical of proximal portions of rays of both species); b, smaller paratype with 20 arms (USNM E44636). c, two middle brachials showing raised axial ridges and distal spinose margins, paratype (PM ECCh-16). d, centrodorsal with several cirrus bases, oblique view, paratype (USNM E44637). Scale: (left) a, b, d, 2 mm; (right) c, 1 mm. VOLUME 108, NUMBER 3 443 Fig. 3. Cirri. a—d, h. Alloeocomatella polycladia, new species. a—b, large paratype (USNM E44633); c-d, smaller paratype (USNM E44642): h, cirrus tip showing aboral spines on distal two segments (preceding terminal claw) widened as dentate transverse ridges. e-g, i, j. Alloeocomatella pectinifera (A. H. Clark). e, holotype (BMNH 87-4-26-9); fg, (USNM E44640); 1, cirrus tip showing forked aboral spine on antepenultimate cirral, and opposing spine widened as transverse dentate ridge; j, cirral following transitional cirral with aboral spine widened as sharp, narrow transverse ridge (LACM 92-160.1). Dashed lines (a—g) indicate transitional segments. Scale: a—g, 4 mm; h-i, 2 mm; j, | mm. 444 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 4. Alloeocomatella polycladia, new species. a, comb of P,, paratype (PM ECCh- 16); b-f, holotype (USNM E44632); b, P,; c, P3; d, Po; €, P») (dotted portion of distalmost enlarged pinnular obscured by tissue); f, Paistai (different arm) [first pinnular (dotted) lost]. Scale: 2 mm; enlarged pinnulars, 0.5 mm. VOLUME 108, NUMBER 3 pinnules slender, up to 20 segments, 9.9 mm long. Disk with numerous round nodules on anal interambulacral area. Color pattern. — Dark red or purplish red; pinnules sometimes beaded purplish red and pink, sometimes with yellow, orange or pink tips. Habits. —All specimens collected by SCUBA on inshore fringing reefs, lagoonal patch reefs, and barrier reefs. Cryptic during the day, often found curled under slabs of coral rubble. At night: specimens with about 20 or fewer arms concealed in crevices, with several arms extending into the water col- umn and with pinnules arranged in single plane; withdrawing rapidly and completely when illuminated; large specimens com- pletely exposed, perched on corals, forming arcuate or radial fan, also with pinnules ar- ranged in single plane. The photograph identified as Comatella maculata on p. 208 of Meyer (1986) is almost certainly this spe- cies. Distribution. —Indonesia (Ceram I.), Pa- pua New Guinea, northern Australia (Liz- ard I., Great Barrier Reef), Palau Is., Fiji, New Caledonia, Chuuk Atoll; 3-18 m. Etymology. —From the Greek polys (mod\vs) “‘many” and klados (kXdadéos) “branch,” because this species has more than the ten arms of 4. pectinifera. Remarks. —In the past, specimens of A. polycladia have been identified as Coma- tella maculata due to similarities among brachitaxes, arm number and general cirrus structure in the two species (Meyer & Ma- curda 1980, Meyer, 1986). In Hoggett & Rowe’s (1986) key to the genera of Com- asteridae, large specimens (>20 arms; first syzygy on IIBr arms at br,,,) run down to Comatella while smaller multibrachiate specimens (1 1—20 arms; first syzygy on II Br arms at br,,,) run down to either Coman- thus or Clarkcomanthus. These smaller specimens lack the diagnostic features of e1- ther of the latter two genera and cannot be assigned to a genus using this key. 445 Several small specimens of A//oeocoma- tella collected by D. L. Meyer at Palau (in- cluded in USNM E44693, E44694) are dif- ficult to assign to species. These individuals resemble A. pectinifera in having ten arms, and cirri and distal pinnules with more elon- gated segments. They are referred to A. po- lycladia because: 1) when measurable, the anterior:posterior ray length ratio is 1.1-1.5: 1, less than in A. pectinifera, and 2) most specimens have large, crowded disk nodules except on the anal papilla itself. These spec- imens were also collected with larger spec- imens clearly identifiable as A. polycladia, but both species have been collected in the same local reef habitat. It is not surprising that small specimens may be difficult to identify. It is often impossible to identify juvenile ten-armed comasterids to genus (Hoggett & Rowe 1986). Alloeocomatella pectinifera (A. H. Clark, 1911), new combination Figs. 3e-g, 1, J, 5, 6 Comissia pectinifer A. H. Clark, 1911, p. 644; 1912, p. 78; 1918, p. 19; 1931, p. 255-256, pl. 25.— Messing, 1994, p. 239. Comissia pectinifera. —A. M. Clark & Da- vies, 1965, pp. 598, 603—4.—A. M. Clark & Rowe, 1971, pp. 6, 14.—Zmarzly, 1985, pp. 348, 350—2.— Meyer, 1986, p. 203.— Bradbury et al., 1987, pp. 190-191. Comissia sp. cf. pectinifera. —Zmarzly, 1985, pp. 348, 351-2, 354-6.— Meyer, 1986, pp. 206-7. Diagnosis. —Ten arms; anterior ray length up to 500 mm; anterior:posterior ray length ratio 2.0—2.9:1; longest segments on distal pinnules (anterior rays) with LW up to 7.0; longest cirrals (on mature cirri) with LW = 1.6—2.3; anal interambulacral area usually naked, with few scattered nodules, or with numerous nodules on anal papilla. Holotype. —BMNH 87-4-26-9; Christ- mas Island, Indonesia (south of the Sunda 446 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 5. Alloeocomatella pectinifera (A. H. Clark) (USNM E44640), aboral view. Scale: 1 cm. Straits, between Sumatra and Java), H. M. S. Flying Fish (A. H. Clark, 1911, 1931). Other material examined. —Chuuk Atoll: USNM E44802 (2 specimens), N side NE Pass, S of Quoi I., Chuuk Atoll, 7°31'32’N, 151°58'11”E, 18-23 m, 11 Jun 1993, P. Col- in/C. G. Messing, coll. Papua New Guinea: USNM E44639 (1), fringing reef, N side Na- gada Harbor, Madang, E of wading beach, 05°09'29"S, 145°49'24’E, 3 m, 4 Jun 1992. USNM E44638 (1), near top of barrier reef SE of Pig I., SE of Nagada Harbor, Madang, 05°10'21"S, 145°51'47’E, 14m, 16 Jul 1991. LACM 92-160.1 (1), fringing reef, N side Nagada Harbor, Madang, 05°09'29’S, 145°49’21”E, 8 m, 12 Jun 1992. USNM E44640 (1), wall off Barracuda Rock, SE of Nagada Harbor, Madang, 05°10'20’S, 145°51'53”E, 23 m, 14 Jul 1991. USNM E44803 (1), near top of barrier reef SE of Pig I., SE of Nagada Harbor, Madang, 14 m, 16 Jul 1991. USNM E44804 (1 imma- ture). Nagada Harbor area, Madang, no field data. USNM E44641 (1), Cape Croisilles, exposed fringing reef, 4°51'30”S, 145°48’E, =30 km N of Madang, 6 m, 13 Jun 1992, B. Hoeksema, coll. Indonesia: USNM E8959 (2), Duroa Strait, Kei Is., 5°24'20’S, 132°55’E, 100 m, 15 Apr 1922, T. Morten- sen, coll. (sta. 24). (All collections by the author except where noted.) Description of holotype. — Poor condition; most rays broken near their bases. Centro- dorsal a thick pentagonal disk with sloping sides, diameter = 4.6 mm; aboral surface with shallow central depression; cirri crowded around margin, 2-3 deep, XLIV, 14-17, maximum length 14 mm; segments increasing in length from very short first to fifth; fifth and sixth segments longest, LW = 1.9; following segments decreasing in length; seventh or eighth and following seg- ments with sharp erect or distally-directed spine (blunt on some cirri); first spine-bear- ing segment with LW = 1.2, next as wide as long; following segments wider than long and becoming shorter distally; antepenul- timate with LW = 0.7; spines on few distal VOLUME 108, NUMBER 3 447 Fig. 6. Alloeocomatella pectinifera (A. H. Clark) (USNM E44803). a, P,; b, Ps; c, Py; d, P29 (dotted portions obscured by tissue); e, Puja [first pinnular (dotted) lost]. Pinnules taken from different arms. Scale: 2 mm; enlarged pinnulars, 0.5 mm. segments, including opposing spine, some- times widened as a transverse ridge which may be forked, dentate or serrate. Radials visible in interradial angles; IBr, short and separated, WL = 4.0; axils wider than IBr,, with short, diverging lateral mar- gins, WL = 1.8; br, oblong or slightly wider exteriorly, united interiorly only proximal- ly, WL = 2.8; br, wider exteriorly, WL = 2.2; br3,, diameter = 2.0-2.2 mm; br,_, oblong, WL = 2.3; proximal brr with strong- ly developed alternating articular tubercles; brr triangular or almost so by brg_,5, WL = 2.5; second syzygy at br,,, 1; brr with thick- ened, raised, spinose distal margins and raised axial lines; middle brr strongly cu- neate, WL = 1.7; distal intersyzygial inter- val 3. Most arms broken; longest intact to br53. Most pinnules broken. P, of 47 segments, with ~ 26 tall narrow teeth, 15.3 mm long; comb developing gradually (comb coiled— tip may be lost; another P, bears 32 teeth); P,, of 39 segments, about 27 teeth; comb developing more abruptly. Proximal and middle segments of oral pinnules short, with distal fringe of spines; middle segments also with few lateral spines coalescing on more distal segments to form comb rudiment. Middle pinnule of 22 segments, 9.3 mm long; segments beyond short basal few squarish or little longer than wide, with dense lateral cluster of spines. Disk absent. Description of other material. —Centro- dorsal usually a thick rounded pentagonal or circular disk with steeply sloping sides; diam. = 4.1—6.1 mm; aboral surface with 448 shallow central depression and at least a trace of radiating interradial ridges. Cirri crowd- ed, of varying length and robustness, one or two deep around centrodorsal margin; XXIX-XLVI, 16-22 (mature), maximum length on any individual 12—19 mm (chiefly 13-16 mm); only one or two segments, from the fourth to seventh, longest; LW of longest segments on largest cirri = 1.6—2.3 (chiefly 1.8—2.0); sixth to eighth segment transition- al; second segment following transitional, LW = 0.9-1.1; following segments slightly shorter; antepenultimate segment with LW = 0.6—0.8; aboral spine widened as a sharp narrow transverse ridge on some or most segments in some specimens; when present, transverse ridge may be forked (rarely tri- fid), eroded and blunt, or restricted to pen- ultimate segment. Maximum ray length chiefly 200-275 mm; anterior to posterior ray length ratio 2.0—2.5:1; one specimen with anterior rays 500 mm, posterior rays 200 mm; radials hidden or visible in interradial angles; IBr, short, oblong, separated or just touching proximally, partly hidden by centrodorsal or completely exposed, WL = 2.8—4.0; axil with short diverging lateral margins, WL = 2.0—2.2; IBr and br,_, with or without small synarthrial swelling; br, oblong or slightly longer exteriorly, united interiorly at least partly, WL = 2.0-2.8; br, longer exteriorly, usually shorter than br,, WL = 2.3-3.1; br3,,4 oblong, WL = 1.5—1.9, diameter = 1.6—-2.4 mm; br, to br, or br, oblong, WL = 1.9- 2.3; proximal brachials with moderately to strongly developed alternating articular tu- bercles; brr triangular by bry_,,, WL = 1.9- 2.6; brr with raised, spinose distal margins and raised axial lines first appearing by br7_,,; brr strongly cuneate in proximal third of arm, WL = 1.8-2.4, becoming moder- ately cuneate by mid-arm, WL = 1.5-1.9, with raised, spinose distal margins and axial lines; distal brr moderately or weakly cu- neate (almost oblong), often longer than wide, WL = 0.7-1.1, usually with distal margins weakly raised, spines strongest PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON midaborally, and axial lines weak or absent; brr near arm tip slender and elongated, WL = 0.5-0.8, with few coarse midaboral spines. Syzygies at br;,,, usually br,,,,> to bry4,,5 (rarely br,o,,, to bri¢4,7); distal intersyzy- gial interval usually 3 (sometimes 2 or 4; rarely 5 to 7). Pinnules generally similar to those of A. polycladia. P, of 49-60 segments, 23-37 teeth, L = 18-26 mm; P, of 44-57 segments, 29-39 teeth, L = 13-20 mm; P; of 43-58 segments, 29-39 teeth, L = 12-18 mm; P, to P, the last comb-bearing pinnule, up to 42 segments, 31 teeth, L = 11 mm. P, of 22 short segments (max. LW = 1.3), up to 9 mm long; proximal segments (except basal two) with distal rim of spines; most seg- ments with numerous spines on side of pin- nule facing arm tip. Middle pinnules of 19- 25 segments, L = 8-13 mm; similar to Py but with middle pinnulars more elongated; LW = 1.3-1.9. Distal pinnules up to 32 seg- ments, L = 11-17 mm; basal two segments short; remaining segments (except near tip) elongate, LW = 3.0—7.0; proximal segments with one or two strong distal spines; most segments (especially on pinnules near arm tip) smooth with expanded ends; distal seg- ments with one or two small mid-lateral spines; distal hooks weak. Disk usually lost; when present, naked, or with few scattered conical nodules on anal interambulacrum; nodules numerous on anal papilla in one specimen. Description of an immature specimen. — (USNM E44804) Centrodorsal a thin disk, 2.1 mm across; aboral surface slightly con- vex, with traces of juvenile sockets, and without central depression; cirri X XIV, 13- 15, 10 mm; third to fifth segments elongated with expanded ends; fourth segment lon- gest, with LW = 2.8-3.0; transitional sixth segment shorter and wider, LW = 1.6; distal several segments slightly wider than long; opposing spine slightly widened transverse- ly on one or two cirri. Anterior rays 130 mm, posterior rays ~ 45 mm; radials just visible beyond centro- VOLUME 108, NUMBER 3 dorsal; IBr, completely exposed, oblong, WL = 2.0; axil pentagonal with diverging lateral margins, WL = 1.5; br, and br, slightly lon- ger exteriorly, WL = 1.6; br, united inte- riorly only proximally; br;,, oblong, WL = 0.9, diameter = 0.9 mm; br,_, oblong, WL = 1.0; following brr becoming cuneate, none triangular; middle brr cuneate, longer than wide, WL = 0.7, with distal margins raised, slightly everted, spinose and with spines strongest midaborally; no axial lines; distal brr slender, constricted in middle, with few coarse midaboral distal spines; WL = 0.3. P, up to 43 segments, 29 teeth, L = 10.7 mm; P, smaller and slenderer than P,, up to 34 segments, 22 teeth, L = 7.4 mm; P; and P,. not developed; P, the last comb- bearing pinnule, much smaller and slender- er than P,, about 16-23 segments, 7-14 teeth, max. L = 4.0 mm; middle pinnules of 13-15 segments, L = 5 mm; distal pin- nules extremely slender, up to 21 segments, L = 9.2 mm. Disk naked. Color patterns.—Rays variously deep red, red-orange, reddish- or pinkish gray, ma- roon or dark red-brown, often with differ- ently pigmented articulations (red, orange or white), or a paler midaboral stripe. Some specimens with articulations and ossicle margins white. Pinnules same color as arms or darker, sometimes with paler or yellow tips, sometimes with white aboral stripe; in specimens with white-bordered ossicles, tis- sue dark brown on proximal pinnules, be- coming orange or yellow on distal pinnules; centrodorsal sometimes with white center; cirri Sometimes with white aboral stripe; disk dark red, red-brown, sometimes with nar- row white stripes. At New Caledonia (Mey- er 1986), yellowish or banded yellowish and brown, with brown pinnules. Habits. —Cryptic during the day. At night, calyx remains hidden, with four to eight arms extended more or less in parallel; pin- nules arranged in a single plane. Withdraws rapidly and completely when illuminated. Distribution. —Maldive Is., Indonesia, 449 northern Australia (Lizard I., Great Barrier Reef), Papua New Guinea, Palau Is., New Caledonia, Chuuk Atoll, Kwajalein Atoll. 3-23 m. Remarks.—As mentioned under 4. po- lycladia, small ten-armed specimens of Al- loeocomatella may be difficult to identify to species. The greater difference between an- terior and posterior ray lengths even in small specimens appears to separate 4. pectinifera from A. polycladia. The nature and distri- bution of disk nodules should be used cau- tiously as a diagnostic character because specimens of both species may lack them. Anterior:posterior ray length ratios vary within several comasterid taxa (Messing 1994), which suggests that this may not be a useful diagnostic feature. Comanthus parvicirrus (Miller), for example, tends to develop longer anterior rays and higher ra- tios in deeper and quieter habitats. Speci- mens from shallow (<8 m), higher-energy habitats tend to have rays of more nearly equal lengths. However, ray length ratio does not appear to vary with habitat in Alloeo- comatella. At Madang and Chuuk Atoll, I found both species in the same local reef habitat and observed that both extend arms or emerge completely only under relatively low energy conditions, either in deeper wa- ter (14—23 m) or on shallow (3-4 m) shel- tered inshore reefs. A single specimen from a shallow, high-energy habitat was cryptic. At New Caledonia, Meyer (1986) recorded A. polycladia (as Comatella maculata) from ‘relatively sheltered reefs” (p. 209), and A. pectinifera from “reefs subject to weak cur- rents” (p. 206). Most specimens of A. pectinifera de- scribed here are larger than previously re- corded material. The largest collected (and several measured in the field) has a maxi- mum ray length of about 500 mm, the lon- gest recorded for any comatulid. Because posterior rays are substantially shorter (about 200 mm), the total span approxi- mately equals that of the largest comatulid previously noted; the Arctic/boreal ante- 450 donid Heliometra glacialis var. maxima has rays up to 350 mm in length (Clark & Clark 1967). Acknowledgments I want to thank the following: the Chris- tensen Fund (Palo Alto, California) for awarding me the fellowship that permitted me to carry out research at the Christensen Research Institute (CRI), P.O. Box 305, Madang, Papua New Guinea (1991-1992); Matthew Jebb, former director of CRI, who smoothed the waters, parted the seas and made life and science at the antipodes a piece of cake; Patrick and Lori Colin, di- rectors of the Coral Reef Research Foun- dation (Weno, Chuuk, Federated States of Micronesia), for similar extraordinary ef- forts and accommodations at Chuuk Atoll (1993); David Pawson and Cindy Ahearn (USNM), Sheila Halsey (BMNH), Michel Jangoux (BELG) and Nadia Cominardi (PM), for making specimens available from their respective institutions; James D. Tho- mas for first recommending CRI to me, and Kim Norton and the Deep Ocean Society for providing research equipment. Many others contributed diving and boat handling support. Part of this research was carried out under the National Cancer Institute Contract for Collection and Taxonomy of Shallow-water Marine Organisms (P. Colin, principal investigator). This is Contribution No. 75 from the Christensen Research In- stitute and Contribution No. 7 from the Deep Ocean Society, Inc., Hollywood, Flor- ida. Literature Cited Bradbury, R. H., R. E. Reichelt, D. L. Meyer, & R. A. Birtles. 1987. Patterns in the distribution of the crinoid community at Davies Reef on the Central Great Barrier Reef. —Coral Reefs 5:189- 196. Breimer, A. 1978. General morphology, recent cri- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON noids. Pp. T9-T59 in R. C. Moore & C. Teich- ert, eds., Treatise on invertebrate paleontology, Part T, Echinodermata 2, vol. 1. Geological So- ciety of America, Boulder, Colorado. Clark, A. H. 1911. A new unstalked crinoid from Christmas Island.—Annals and Magazine of Natural History (8)7:644-645. . 1912. Thecrinoids of the Indian Ocean. Echi- noderma of the Indian Museum, Part VII. Cal- cutta, iv + 325 pp. 1918. The unstalked crinoids of the Siboga Expedition. —Siboga Expeditions 42b:1—300, 28 pls. 1931. A monograph on the existing crinoids 1(3).—Bulletin of the United States National Museum (82): i—vii, 1-816, 86 pls. —, & A.M. Clark. 1967. A monograph on the existing crinoids 1(5).— Bulletin of the United States National Museum (82): 1-860. Clark, A. M., & P. S. Davies. 1965. Echinoderms of the Maldive Islands.—Annals & Magazine of Natural History (13)8(94):597-612. —§, & F. W.E. Rowe. 1971. Shallow-water Indo- West Pacific Echinoderms. British Museum (Natural History), London. x + 238 pp. Hoggett, A. K. & F. W. E. Rowe. 1986. A reappraisal of the family Comasteridae A. H. Clark, 1908 (Echinodermata: Crinoidea), with the descrip- tion of a new subfamily and a new genus. — Zoological Journal of the Linnean Society 88: 103-142. Messing, C. G. 1994. Comatulid crinoids (Echinod- ermata) of Madang, Papua New Guinea, and environs: diversity and ecology. Pp. 237-243 in B. David, A. Guille, J.-P. Feral, & M. Roux, eds., Echinoderms through time. Balkema, Rot- terdam. , & J. H. Dearborn. 1990. Marine Flora and Fauna of the Northeastern United States. Echi- nodermata: Crinoidea.—NOAA Technical Re- port NMFS 91. 30 pp. Meyer, D. L. 1986. Les Crinoides. Pp. 199-225 in A. Guille, P. Laboute, & J. L. Menou, eds., Guide des etoiles de mer, oursins et autres echinod- ermes du lagon de Nouvelle-Caledonie. Editions de ’ORSTOM, Collection Faune Tropicale no. XXV, Paris. ——.,, & D. B. Macurda, Jr. 1980. Ecology and dis- tribution of the shallow-water crinoids of Palau and Guam.—Micronesica 16(1):59-99. Zmarzly, D. L. 1985. The shallow-water crinoid fau- na of Kwajalein Atoll, Marshall Islands: eco- logical observations, interatoll comparisons, and zoogeographic affinities. — Pacific Science 39(4): 340-358. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(3):451-457. 1995. Morellia dendropanacis, a new species, and other species with spotted wings: characterization and comparison (Diptera: Muscidae: Muscinae) Denise Pamplona and Marcia Souto Couri Departamento de Entomologia, Museu Nacional, Universidade Federal do Rio de Janeiro, Quinta da Boa Vista, Sao Cristovao, Rio de Janeiro, RJ. CEP 20.940-040, Brasil Abstract. —The description of a new species of Morellia Robineau-Desvoidy (1830) (Diptera, Muscidae, Muscinae) with spotted wings, morphologically similar to M. couriae Pamplona and M. maculipennis (Macquart), is given and compared with other Neotropical species with spotted wings. Among the 18 Neotropical Morellia spe- cies, 13 have spotted wings. Some species have conspicuous markings, others have faint markings. The latter includes M. affinis Malloch, with a spot on the apex of Sc and M. concacata Pamplona, M. ochricornis (Wiedemann), M. roppai Pamplona and M. sinopensis Pamplona with a spot on the hu- meral crossvein. As the spots are almost imperceptible in these species, identifica- tion using Pamplona’s key (1986) (couplet 3) requires careful comparison of the wing with the figures. The remaining eight spe- cies, VM. couriae Pamplona, M. humeralis (Stein), 17. lopesae Pamplona, M. maculi- pennis (Macquart), M. nigricosta Hough, M. semimarginata (Stein), M. xanthoptera Pamplona and M. dendropanacis, new spe- cies have conspicuous spots, facilitating their identification. The new species described here was col- lected among flies that were pollinating Dendropanax cuneatum (Araliaceae) (Pom- bal & Morellato 1995). According to these authors M. humeralis and M. dendropan- acis were the main pollinators of this plant, being specially observed from 10am until lpm, in groups of up to 24 on an inflores- cence. Pombal & Morellato considered them to be effective pollinators due to their small size, the high frequency of visitations, their visiting behavior and the large number of flies. Morellia dendropanacis, new species Figs. 1-11 Type material. —Holotype 6 [MNRJ], la- beled: Faz. Sta. Genebra [Fazenda Santa Genebra]/Campinas-SP [Sao Paulo]/03/VII/ 1992/Pombal, E.C.P. On the underface of the label: coletado em Dendropanax cunea- tum [collected on D. cuneatum]; ““HOLO- TIPO” [red label]. Left mid- and hindlegs broken; the abdomen is dissected and pre- served in glycerin in a microvial that is at- tached to the pin. Paratypes: 1 6, 42?[MNRJ] with same label data as the holotype: “PARATYPE” [green label]. All in good condition; male paratype with right hindleg glued on label, abdomen broken; one female paratype dissected and preserved in glycerin in microvial. Diagnosis. — Wings with cloudlike spots on humeral crossvein, basal radial cell, sub- costal apex, cell r, apex and crossveins r-m and dm-cu; hindtibia of 3, at anteroventral, ventral and posteroventral surfaces with many bristles at apical half. Coloration. —Metallic black-bluish with a faint whitish pruinescence on mesonotum near the head. Frons, lower half of parafa- 452 cials and lower half of face with whitish pruinescence, upper half of fronto-orbital plate and gena blackish shining. Lower half of fronto-orbital shining yellowish; pedicel and flagellum yellow-brownish; arista with basal half yellow and apical half black; pro- boscis brown and palpus yellow. Wing with spots on humeral crossvein, on the adjacent area of basal radial cell, on subcostal apex connected to a spot on r-m, on the apex of r, cell and on dm-cu (Fig. 1). Calypteres faint brownish. Halter white. Legs brown, with femur-tibial articulations yellowish. Both spiracles white. Male: Length; body, 4.5-5.0 mm; wing, 4.5-5.0 mm Head. —Eyes bare with divergent margins larger at vertex; distance between eyes at anterior ocellus level about 6—7 times less than head width; anterior-inner ommatidia larger than others. 11-13 frontal bristles. Antenna inserted below mid eye level, with flagellum about 2.2—2.4 the pedicel. Arista plumose with 11-14 hairs. Palpi a little en- larged apicad with about 10-14 setae along margin. Thorax.—Dc 0:2; Acr 0:1; 3 humeral bristles; posthumeral absent; one prs; one 1a; one pra; 3 sa; 2 psa; ipal absent. Noto- pleuron with 2 bristles. Scutellum with 1 pair of basal bristles; 2 pairs of laterals; 1 pair of apical. Prosternum wide and bare (with anterior pubescence difficult to ob- serve). Several proepisternal and proepi- meral bristles. Propleuron bare. Anterior spiracle vertical, oblong. Anepisternum covered with long setae and with 11-13 bristles on the posterior border. Ctps 1:2, supraposterior one, much larger than oth- ers. Anepimeron setulose. Lower calypter about 2.3-2.4 times the upper. Wing with M, curved to R,,; apicad; veins bare. Fore- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON leg with femur at anterior surface bare; pos- teroventral and posterodorsal surfaces with a row of bristles. Tibia at dorsal surface with a row of setulae; posterodorsal surface with 1 preapical bristle; ventral surface with sev- eral setae on apical half. Tarsomeres with short setae. Midfemur bearing some weak bristles along apical half of ventral surface; dorsal surface with 3 apical small spurs. Tibia at ventral surface with two apical bris- tles. Tarsomeres as in foreleg. Hindleg with femur at anterodorsal and anteroventral surfaces with a row of weak and short bris- tles; posterior surface bare. Tibia at antero- ventral surface with 4 bristles on apical third, ventral to anteroventral surface with 5 larg- er parallel bristles; ventral to postero-ven- tral surfaces with several setae on apical half; anterodorsal surface with 3 bristles on mid- dle third; dorsal surface with a row of very small bristles. Tarsomeres as in foreleg. Abdomen. — With few bristles on laterally areas of segments. Sternite V as in Fig. 2. Terminalia: Cercal plate small, with a ven- tral plate with 2 lateral projections with spines along inner margin (Figs. 3 and 5); surstylus with an apical projection (Fig. 4); hypandrium large (Fig. 6); aedeagal apo- deme with many spines in center, gonos- tylus membranous with small bristles, gon- ocoxite with a strong bristle; paramere with long bristles at apex (Figs. 7-8). Female: Length; body, 4.5—5.0 mm; wing, 4.5—5.0 mm. Similar to male except as fol- lows: 4 fronto-orbitals, 2 proclinate and 2 lateroclinate. All ommatidia of same size. Distance between eyes at anterior ocellus level about 2.7—2.8 times smaller than head width. Midfemur, at dorsal surface without spurs; hindtibia at anteroventral and pos- terodorsal surfaces with respectively 2 and 1 bristles on apical half. Terminalia: Ovi- = Figs. 1-7. Morellia dendropanacis. 1, wing; 2, sternite V, holotype; 3, cercal plate, holotype, dorsal view; 4, surstylus, holotype; 5, cercal plate, holotype, lateral view; 6, hypandrium, holotype; 7, aedeagus, holotype, dorsal view. VOLUME 108, NUMBER 3 453 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 454 33 Strmakae: a ¢f._—_- “ve, Morellia dendropanicus. 8, aedeagus, holotype, lateral view; 9, ovipositor, paratype °, dorsal Figs. 8-11. view; 10, ovipositor, paratype 2°, ventral view; 11, spermathecae, paratype 2°. VOLUME 108, NUMBER 3 455 6 Figs. 12-22. Wings of species of Morellia: 12, M. affinis, 13, M. concacata; 14, M. couriae; 15, M. humeralis, 16, M. lopesi; 17, M. maculipennis, 18, M. nigricosta;, 19, M. ochricornis, 20, M. roppai;, 21, M. sinopensis; po. M. xanthoptera. 456 positor long; tergite VI enlarged at base, ter- gites VII and VIII elongate; series of strong spines at sclerites above tergite VII and ter- gite VIII; cercus longer than epiproct (Fig. 8); sternite VII and VIII with 4 strong spines each (Fig. 9); spermathecae pear-shaped. Etymology. —The specific adjectival name dendropanacis refers to the plant Dendro- panax cuneatum where the specimens were collected. Comments. —The wing maculation of M. dendropanacis n. sp. 1s similar to that of M. couriae and M. maculipennis. It differs from M. couriae mainly by the presence of a spot on the humeral crossvein and another one adjacent to it on the basal radial cell. The new species differs from M. maculipennis in having the cloudlike spot on the Sc apex connected to the one of r-m crossvein. Al- though the new species can easily be distin- guished from these two species by the num- ber of frontal bristles, chaetotaxy of hind- tibia and hind spiracle shape and ciliation, the similarity of the male terminalia of these three species suggests a close relation among them. Distribution: Brazil. Sao Paulo. Key to Species of Morellia with Spotted Wings Most species of Morellia with spotted wings can be easily distinguished by the dis- tribution of the spots in association with other characters of chaetotaxy legs and tho- rax. The following key is to aid in their iden- tification: 1. Wing with no more than 2 faint small spots (Figs. 12, 13, 19-21) 2 — Wing with at least 2 conspicuous spots (Figs. 1, 14-18, 22) 2) Blagellumeyellow es ee eee Re ana M. ochricornis (Wiedemann) Flagellum brown 5) 85.0 3 WwW | Prosternum haired ............. 4 — Prosternum bare .............. 5 4. Hindtibia at anterodorsal surface with 4 bristles from the base to the 1), . De 1:2; prosternum haired PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON apex and at posterodorsal surface with 1 bristle on basal third MME ol 5 Sa M. affinis Malloch Hindtibia at anterodorsal surface with 2 bristles on apical half and at posterodorsal surface with 1 bristle on apical third a, ee ae M. concacata Pamplona oe 8 © © © © © © 8 . De 1:2; hindtibia at posteroventral with 6 bristles .. M. roppai Pamplona De 0:2; hindtibia at posteroventral without bristles are oe eee M. sinopensis Pamplona . Wing brown from base to apex along costa and connected with a dark cloud over the r-m cd le: eae M. semimarginata (Stein) Wing not as above. eee ee 7 Postpronotum yellow Ms ie eer eK Roa M. humeralis (Stein) Postpronotum background col- eo © © © © © © 8 Ored) S60 3 eee 8 . dm-cu crossvein without spot (Figs. 1:6, (1.8) nether eee eee 9 dm-cu crossvein spotted (Figs. 1, 1417, 22) 2... ere 10 . Wing with a long cloudlike spot from a little before Sc apex till wing apex (Fig. 18) .. M. nigricosta Hough Wing without the above cited cloud (Fig 16). M. lopesae Pamplona Cell r, with 1 spot (Fig. 22) Een ee M. xanthoptera Pamplona Cell r, with 2 spots (Figs. 1, 14, 17) ee ee e@ ia a M. maculipennis (Macquart) Dc 0:2; prosternum bare ....... 12 . Flagellum brownish yellow with yellow spot at the base; hindfemur at anteroventral surface with weak and short bristles only on apical third; male: anteroventral, ventral and posteroventral surfaces with few setae on apical half Se EE ieee M. couriae Pamplona Flagellum brownish yellow with- out yellow spot at the base; hind- VOLUME 108, NUMBER 3 femur at anteroventral surface with weak and short bristles from base to apex; male: anteroventral, ven- tral and posteroventral surfaces with a great number of setae on apical half .. Morellia dendropanacis Acknowledgments We are grateful to Ellen C. P. Pombal for the specimens of M. dendropanacis. Also to ‘‘Fundacao Uniprectoria José Bonifacio” (proc. 5295-7) for the financial support. 457 Literature Cited Malloch, J. R. 1923. Exotic Muscaridae (Diptera). XI.—Annals and Magazine of Natural History 12(9):505-528. Pamplona, D. 1986. Sobre Morellia R.-D., 1830 neo- tropicais II. Descrigao de cinco espécies novas (Diptera-Muscidae-Muscinae).— Revista Bras- ileira de Biologia 46(3):633-650. Pombal, E. C. P., & P. C. Morellato. 1995. Polini- zacao por moscas em Dendropanax cuneatum (Araliacea) em floresta semi-decidua no sudeste do Brasil.—Revista Brasileira de Biologia (In Press). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(3):458-464. 1995. Within-species variation in Periclimenes yucatanicus (Ives), with taxonomic remarks on P. pedersoni Chace (Crustacea: Decapoda: Caridea: Palaemonidae) Mary K. Wicksten Department of Biology, Texas A&M University, College Station, Texas 77843, U.S.A. Abstract. —The spotted cleaner shrimp Periclimenes yucatanicus is found to vary in the number of rostral teeth, length/width ratio of the carpus of the major cheliped, color pattern and cnidarian host. Similar variation has been found in other species of Periclimenes. Morphological features and type of host have been used to distinguish between P. anthophilus Holthuis & Eibl-Eibes- feldt, and P. pedersoni Chace, but these features overlap extensively; therefore, P. anthophilus is herein considered to be a junior synonym of P. pedersoni. Shrimp of the genus Periclimenes Costa, 1844 are among the most colorful carideans of coral reefs. Species inhabiting cnidarian hosts are among the best known. Four spe- cies of Periclimenes have been reported to live with cnidarians in the western Atlantic, Gulf of Mexico and Caribbean: P. yucatan- icus (Ives 1891), P. pedersoni Chace, 1958 P. anthophilus Holthuis & Eibl-Eibesfeldt, 1964 and P. rathbunae Schmitt, 1924. (See Chace 1972, for a key to the species of Per- iclimenes in the area). During studies of cleaning behavior of shrimp and fish, species of Periclimenes were observed on coral reefs near Key Largo, Florida, U.S.A., and Bonaire, Netherlands Antilles. The shrimp were identified from photographs and by sight. However, during night dives at Bonaire, shrimp with an un- usual color pattern, unlike that of known species in the area, were observed. These shrimp were found in association with the giant sea anemone, Condylactis gigantea (Weinland). To identify this unknown shrimp, speci- mens were collected and compared with previously identified material. Photographs of live animals were examined to compare color patterns of the unknown shrimp with those of species of Periclimenes known to inhabit sea anemones in the area. Records of shrimp species and their hosts were quan- tified. The results indicate that there is con- siderable within-species variation in at least two species of Periclimenes. Methods During studies of cleaning behavior, this writer and volunteer divers from the non- profit organization CEDAM International, counted numbers of shrimp per species of host in Key Largo, Florida, U.S.A., during 11-17 July 1992, and Bonaire, Netherlands Antilles, during 25 July—S August 1993 and 30 July—12 August 1994. Shrimp were ob- served during SCUBA dives at depths of 3— 22 m between 0600-2200 hours. Twenty- four photographs of all species of Pericli- menes were taken in Florida and 110 in Bonaire. Species were identified in the field according to the color patterns shown in the popular book by Humann (1992:151); un- identified ones were reported as Pericli- menes sp. Five specimens (the maximum number allowed as stated on the collecting permit) of the unidentified shrimp were collected at VOLUME 108, NUMBER 3 Fig. 1. 459 > ii e Variation in the rostrum and second pereopod of Periclimenes yucatanicus. a, left second pereopod from individual 29.8 mm in total length; b, c, left and right second pereopods from individual 16.5 mm in total length; d, rostrum from individual 29.8 mm in total length; e, rostrum from individual 13.0 mm in total length. Scales are 1 mm and refer to the drawings immediately above them. Calabas Reef, near Kralendijk, Bonaire, during 1994. An additional 13 were pho- tographed in their natural habitat, but not collected. The animals were kept alive for 5-7 days and their color patterns recorded. These specimens were compared with de- scriptions of known species and also with 13 specimens of P. yucatanicus from the collections of the National Museum of Nat- ural History Smithsonian Institution, Washington, D.C. The color patterns also were compared with those shown in pho- tographs of P. yucatanicus from Grand Cay- man, British West Indies (two slides show- ing three shrimp), Isla Cozumel, Quintana Roo, Mexico (three slides showing two shrimp), the U.S. Virgin Islands (one pho- tograph of one shrimp) and Discovery Bay, Jamaica (photograph by Colin 1978). Results The unidentified shrimp are a previously unreported color variety (hereafter called the nocturnal color phase) of P. yucatanicus. Their bodies closely match the description of P. yucatanicus as given by Holthuis (1951: 38, pl. 10, figs. a—l). In these specimens and previously collected material of the species, there is variation in the number of rostral teeth and the length-width ratio of the car- pus of the second (major) cheliped (Fig. 1). The rostrum can have 5-8 dorsal and 0-3 ventral teeth in addition to an acute apex. The length/width ratio of the carpus ranges from 2.3—7.5. A more slender carpus (4 x longer than wide or more) usually occurs in smaller shrimp (total length 16 mm or less). (The largest specimen measured 29.8 mm in total length, orbit-telson.) In three spec- imens, the second pereopods were asy- metrical. The 18 shrimp of the nocturnal color phase had translucent bodies with a few small spots of brown or white on the dorsal surface, and a translucent turquoise blotch near the gastric region. In two, there was a pale white stripe along the dorsal midline; in another two, there was a small dorsal pale brown spot on either the carapace or third abdominal somite. The pereopods were banded. Most had white antennae, but two had the outer antennal flagella banded and the inner ones white. In two, the uropods ended in faint white spots; in the others, there either were no such spots or the tail 460 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Color variation in Periclimenes yucatanicus, drawn from photographs in life. a, entire animal, showing banded appendages, from Florida, day; b, individual from Bonaire, day; c, individual from Bonaire, night; d, individual from Bonaire, day; e, same individual as seen in drawing d, night; f, individual from Bonaire, day. fan could not be observed in the photo- graph. Two of the individuals that were collected at night are shown in the nocturnal color phase in Figs. 2c and 2e. By day, both changed their color pattern to that of large white, tan, green or purple blotches, spots and saddle marks on a translucent body. The same individual is shown in Figs. 2d and 2e, illustrating the marked change in color pattern. Examination of the live spec- imens and photographs revealed consider- able color variation in P. yucatanicus. Some of the patterns are shown in Figure 2. Of the 82 photographs identifiable as P. yu- catanicus taken at Bonaire, all showed shrimp with the pereopods banded with white and a contrasting darker shade of red, purple or violet. Thirty-two shrimp (39%) had entirely white antennae; 34 shrimp (41%) had the outer flagella banded and the inner white, and 16 shrimp (20%) had both antennal flagella banded. Eleven photo- graphs showed only the front of the animal, and thus only the color of the appendages and antennae could be seen. In the other 71 photographs, at least the carapace and first three abdominal segments were visible. The VOLUME 108, NUMBER 3 461 Fig. 3. Periclimenes pedersoni at Bonaire. color pattern typically included a large dor- sal ‘“‘saddle”’ of pure white or pink, tan or green with a white border on the dorsal sur- face of the carapace, (seen in 31 shrimp, or 44% of the total), a saddle (sometimes in a figure-eight pattern) of similar color on the dorsal surface of the first and/or second ab- dominal somite (seen in 30 shrimp, or 42% of the total), and a large saddle on the third abdominal somite (seen in 46 shrimp, or 65% of the total). The tips of the uropods often were marked with white and dark spots shaped like eyespots; however, the tail fan was visible in only 21 photographs. The lower surface of the carapace or abdomen often had spots of white or white and dark pigment and the anterior parts of the ceph- alothorax could be speckled; however, it was difficult to see these marks in many pho- tographs. Many shrimp also had “‘saddle”’ marks on the fourth abdominal somite, or had marks that overlapped two abdominal somites. Two of the largest shrimp (total body length approximately 30 mm, as es- timated from photographs) had a body background color of brownish pigment; most of the shrimp were translucent except for the bands and spots. Similar color patterns were observed in photographs of shrimp from other areas except for one photograph from Grand Cayman, which showed two shrimp with the same color pattern as the nocturnal color phase from Bonaire. The photograph of P. yucatanicus by Humann (1992) shows a shrimp with a brown back- ground pigmentation. Another common shrimp, Periclimenes pedersoni, had very different markings which were not observed to change at night. This species was marked with white, blue or pur- ple-violet lines running the length of the body, similar marks on the major chelipeds, purple speckles on the maxillipeds and oth- er pereopods, and pure white antennae. The white lines and antennae are readily visible at a distance (Fig. 3). Of 302 total P. yucatanicus observed in the wild (whether photographed or not), 283 were among the tentacles of Condylactis gi- gantea, 13 were on the sea anemone Bar- tholomea annulata (Lesueur) and six were away from a cnidarian host. Usually, only 462 one shrimp was observed per host, but as many as five could be found per anemone. All of the unidentified shrimp at night were found on C. gigantea. Periclimenes peder- soni was more abundant. Of a total of 938 observed, 555 were on C. gigantea, 315 on B. annulata and 68 on other hosts or away from a cnidarian. As many as 13 P. ped- ersoni could be found living in a single sea anemone. The two species never coexisted on the same host. The hippolytid shrimp Thor amboinensis (De Man) often occupied the same individual of C. gigantea with one of the two species; snapping shrimp (Alphe- us ? armatus Rathbun) lived along the stalks of B. annulata under the tentacles where the Periclimenes lived. No individuals of P. rathbunae were ob- served at either Key Largo or Bonaire. Discussion Holthuis (1951) noted variation in the length/width ratio of the carpus of the sec- ond pereopod in specimens of P. yucatan- icus, and speculated on whether this might be due to sexual difference, a growth form or geographic variation. This asymmetry also could be due to loss and regeneration of an appendage. The results given here sug- gest that variation in the proportions of the carpus of the second pereopod exists within P. yucatanicus. Holthuis noted similar vari- ation in P. infraspinis (Rathbun). Chace & Bruce (1993) noted that several Indo-Pacific species of Periclimenes had subequal to grossly unequal second pereopods. The pattern of spots and lines in speci- mens from Bonaire and Florida was con- sistent with that shown in photographs of P. yucatanicus from other areas. Less vari- ation was seen in the color pattern of P. pedersoni (Humann 1992:151; Sefton & Webster 1986:77, figs. 117, 118; Colin 1978: 339-341; Voss 1980:87 and color plate). Periclimenes rathbunae has a pattern of white and brown dots and lines, with white dots at the tips of the uropods (Spotte et al. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 1991:fig. 1; Colin 1978:344 as “unidentified species of Periclimenes’’). It is possible that slight regional differences in color pattern may exist in these species, but such differ- ences would be difficult to see or photograph in the natural habitat. Marked changes in color patterns from day to night have been reported previously in the hippolytid shrimp Hippolyte varians Heptacarpus pictus and Heptacarpus palu- dicola (Green 1961, Bauer 1981). In these species, the shrimp showed a tranluscent blue color by night, regardless of the color pattern displayed by day. Divers at Key Lar- go noticed a similar color change in Thor amboinensis, which had a pattern of white spots against a chocolate brown background by day, and a blue background at night. There are few observations of the nocturnal coloration of shrimp, especially in their nat- ural habitat, so it is difficult to determine how common diurnal/nocturnal color changes are among species of Periclimenes. Spotte et al. (1991) reviewed associations of shrimp with cnidarians in the West Indies and Bermuda, and reported that neither P. yucatanicus nor P. pedersoni was confined to a single host. Periclimenes yucatanicus was listed to live with six species of sea anemones (Order Actiniaria), the jellyfish Cassiopeia xamachana (Bigelow) and the corallimorpharian, Rhodactis sanctaetho- masae (Duchassaing & Michelotti). Pericli- menes pedersoni was reported to live with six species of sea anemones, the same jel- lyfish and a tube anemone, Cerianthus sp.. At Key Largo, diver Rick Sammon photo- graphed Periclimenes yucatanicus in asso- ciation with another corallimorpharian, Ri- cordea florida (Duchassaing & Michelotti). I photographed the same association at Grand Cayman. Evidently, P. yucatanicus and P. peder- soni can live on multiple hosts. However, most records of either species come from the anemones C. gigantea and B. annulata. Goy (1990) found that more larval P. yu- catanicus completed metamorphosis suc- VOLUME 108, NUMBER 3 cessfully if exposed to exudates of C. gi- gantea than to those of other species or none at all; larval P. pedersoni showed better sur- vival if exposed to exudates of B. annulata. Levine & Blanchard (1980) reported that species of Periclimenes could be stung by sea anemones to which they were not ac- climated. Periclimenes anthophilus Holthuis & Eibl- Eibesfeldt (1964) was described as a distinct species from Bermuda. The original au- thors, and Chace (1972) remarked that it was very similar to P. pedersoni, but could be distinguished by the position of the he- patic spine in front of the most posterior dorsal carapace spine; the carpus of the sec- ond pereopod being less than half the length of the chela and the host being Actinia ber- mudensis or C. gigantea. Other morpholog- ical features and the color pattern were in- distinguishable from those of P. pedersoni. Chace (1972) remarked on the close mor- phological similarity between the two sup- posed species, and re-examined specimens. The only differences he found between them were the “different habits” and the “‘pro- portionately shorter carpus of the major sec- ond pereopod’’. However, proportions of the carpus of the second pereopod vary in species of Periclimenes as well as in other palaemonids. In species of freshwater shrimp of the genus Macrobrachium, for ex- ample, the entire second pereopod is mark- edly longer and more robust in adult males than in juveniles or females. As shown in Fig. 1, the length/width ratio of the carpus can vary in specimens of Periclimenes spp., as does the length of the carpus relative to the chela. Although species of Periclimenes appar- ently have preferred hosts, they are not re- stricted to a single species of cnidarian. The review by Spotte et al. (1991) shows this lack of specificity, as does a study of shrimp living with sea anemones in central Japan (Suzuki & Hayashi 1977). Both P. pedersoni and P. anthophilus have been reported to live with C. gigantea and B. annulata, as 463 well as with other hosts (Spotte et al. 1991). Lacking additional evidence of distinctive coloration or species-specific behavioral or morphological distinguishing features, P. anthophilus should be considered a junior synonym of P. pedersoni Chace, 1958. Acknowledgments I am grateful to Kalli de Meyer, Bonaire Marine Park, and the government of the Netherlands Antilles for permission to con- duct studies and collect specimens. Vol- unteer divers from CEDAM International, especially Rick Sammon, Dave Downs and Stella Covre, and Nancy Sefton of the Cay- man Islands aided immensely in photo- graphing and documenting shrimp in the natural habitat. Literature Cited Bauer, R. T. 1981. Color patterns of the shrimps Heptacarpus pictus and H. paludicola (Caridea: Hippolytidae).— Marine Biology 64:141-152. Chace, F. A., Jr. 1958. A new shrimp of the genus Periclimenes from the West Indies. — Proceed- ings of the Biological Society of Washington 71: 125-130. 1972. The shrimps of the Smithsonian-Bre- din Caribbean Expeditions with a summary of the West Indian shallow-water species (Crus- tacea: Decapoda: Natantia).—Smithsonian Contributions to Zoology 98:1-179. —., & A. J. Bruce. 1993. The caridean shrimps (Crustacea: Decapoda) of the Albatross Philip- pine expedition 1907-1910, part 6: superfamily Palaemonoidea.—Smithsonian Contributions to Zoology 543:1-152. Colin, P. L. 1978. Marine invertebrates and plants of the living reef. T.F.H. Publications, Neptune City, New Jersey, 512 pp. Costa, O.G. 1844. Su due nuovi Generi di Crostacei Decapodi Macrouri Nota.— Annali delle Acca- demia degli Aspiranti Naturalisti, Napoli 2:285. Goy, J. W. 1990. Components of reproductive effort and delay of larval metamorphosis in tropical marine shrimp (Crustacea: Decapoda: Caridea and Stenopodidea). Unpublished Ph.D. disser- tation, Texas A&M University, College Station, Texas, 177 pp. Green, J. 1961. A biology of Crustacea. H.F. & G. Witherby, Ltd., London, 180 pp. Holthuis, L. B. 1951. A general revision of the Pa- 464 laemonidae (Crustacea Decapoda Natantia) of the Americas. I. The subfamilies Euryrhynchin- inae and Pontoniinae.— Allan Hancock Foun- dation Occasional Papers | 1:1—332. ———., & I. Eibl-Eibesfeldt. 1964. A new species of the genus Periclimenes from Bermuda (Crusta- cea, Decapoda, Palaemonidae).—Senckenber- giana Biologica 45:185-192. Humann, P. 1992. Reef creature identification. Vaughan Press, Orlando, Florida, 320 pp. Ives, J. E. 1891. Crustacea from the northern coast of Yucatan, the harbor of Vera Cruz, the west coast of Florida and the Bermuda Islands. — Pro- ceedings of the Academy of Natural Sciences of Philadelphia (1891):176-207. Levine, D. M., & O. J. Blanchard, Jr. 1980. Accli- mation of two species of the genus Periclimenes to sea anemones.— Bulletin of Marine Science 30:460-466. Schmitt, W. L. 1924. The macruran, anomuran and stomatopod Crustacea. In Bijdragen Tot de PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Kennis der Fauna van Curacao. Resultaten Ee- ner Reis van Dr. J. van der Horst in 1920.— Bijdragen Tot de Dierkunde Genootschap Na- tura Artis Magistra te Amsterdam 23:61-81. Sefton, N., & S. Webster. 1986. A field guide to Ca- ribbean reef invertebrates. Sea Challengers, Monterey, California, 112 pp. Spotte, S., R. W. Heard, P. M. Bubucis, R. R. Manstan, & J. A. McLelland. 1991. Pattern and color- ation of Periclimenes rathbunae from the Turks and Caicos Islands, with comments on host as- sociations in other anemone shrimps of the West Indies and Bermuda.—Gulf Research Reports 8:301-311. Suzuki, K. & K. Hayashi. 1977. Five caridean shrimps associated with sea anemones in central Ja- pan.— Publications of the Seto Marine Biolog- ical Laboratory 24:193-208. Voss, G. L. 1980. Seashore life of Florida and the Caribbean. Banyan Books, Inc., Miami, Florida, Revised edition, 199 pp. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(3):465-476. 1995. Dittosa, a new genus of leucosiid (Crustacea: Decapoda: Brachyura) from southern Australia and New Zealand Cheryl G. S. Tan Department of Zoology, National University of Singapore, Lower Kent Ridge Road, S(0511), Republic of Singapore Abstract.—A new genus, Dittosa, is established for two southern Australian species of Philyra Leach, 1817: Philyra laevis Bell, 1855, P. murrayensis Rath- bun, 1923, and the New Zealand Ebalia cheesmani Filhol, 1886, on the basis of the male first and second gonopods which are of equal length, presence of two fissures on the roof of the orbit, structure of the male abdomen, form of the front and third maxilliped exopod, and appearance of the intestinal region of the carapace. There are 41 known species in the genus Philyra Leach, 1817 (Bell 1855, Seréne 1968, Rathbun 1924, Takeda & Nakasone 1991). These are generally characterized by their rounded carapaces, massive chelipeds, ex- panded exopod of third maxilliped and rather broad front which is not produced. For the species of Philyra in which the male first gonopod is known, this structure is stiff and stout, and about twice to three times the length of the second gonopod (Ste- phensen 1946, Dai & Yang 1991). Recently, in the course of examining leucosiids in the Smithsonian Institution collection, I found that two southern Australian species, P. /ae- vis Bell, 1855, and P. murrayensis Rathbun, 1923, and a New Zealand species, Ebalia cheesmani Filhol, 1886, have the male first and second gonopods of almost equal length, with the first gonopod long and sinuous, and the second gonopod elongate and slender. In addition, the number of fissures on the orbital roof, male abdominal segmentation, the form of the front, appearance of the in- testinal region and, to a lesser extent, form of the third maxilliped exopod in these two species differ from those in typical Philyra species. In this paper, a new genus, Dittosa, is established for P. laevis, P. murrayensis, and Ebalia cheesmani on the basis of this unique suite of characters. For purposes of comparison, eight other species of Philyra were studied and these are listed in the Appendix. The following abbreviations are used in the text: Gl, G2, the male first and second gonopods respec- tively; APE, U.S. “Albatross” Philippine Expedition and USARP, United States Ant- arctic Research Program. Measurements are given in the order carapace length times width in millimeters (mm), exclusive of the posterior intestinal spine and lateral tuber- cles. Specimens examined are deposited in the National Museum of Natural History (USNM), Smithsonian Institution, Wash- ington D.C., U.S.A., and the Zoological Reference Collection (ZRC), Department of Zoology, National University of Singapore. Family Leucosiidae Samouelle, 1819 Dittosa, new genus Type species. —Philyra laevis Bell, 1855, by present designation. Diagnosis. —Carapace rounded, with widely spaced tubercles or granules along the margins, surface smooth or granular, in- testinal region with raised region (ridge, tu- 466 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Dittosa murrayensis, male, 18.8 '/ eye- stalk width) or strongly bulbous shape seen in N. cacahuate (Fig. la). While corneal width in a small (cl 6.8 mm) topotypic fe- male of N. grandimana was about half the eyestalk width, the cornea was not as bul- bous as in comparably small specimens of N. cacahuate (Fig. 5a) and the major che- liped was clearly distinguishable as typical of N. grandimana. The major cheliped in N. cacahuate (Fig. 1b) is distinct in its more regularly arcuate, dentate inferior margin of the merus and the much longer carpus than in N. grandimana (Fig. le), while the minor chela may be distinguished by the larger rel- ative size of the carpus and presence of den- tition on opposable margins of the fingers (Fig. 3a), as compared to the minor chela of similar-sized specimens in N. grandi- mana (Fig. 3f). In addition, the mature male first and second pleopods of N. cacahuate (Figs. lc, 4a) differ from both the immature (Figs. 1f, 4e) and mature (Fig. 4d, f) forms of these appendages in N. grandimana. Neocallichirus cacahuate resembles at least two southern Caribbean congeners, JN. lemaitrei Manning, 1993 and N. nickellae Manning, 1993, in reaching mature form at a smaller size than does N. grandimana and in having a relatively long carpus on the major chela. However, the major chelae of both these species have a stronger inferior marginal serration on the carpus and prop- odus which is conspicuously visible in ex- ternal view. In N. nickellae, the broadly sul- cate gape at the base of the fixed finger, con- iform basal tooth on the opposable margin of the dactylus, and angular inferior margin on the carpus in the major chela, as well as PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON the relative lack of conspicuous teeth on the fingers of the minor chela and the relatively smaller carpus in this appendage, further distinguishes this species from WN. caca- huate. However, N. /emaitrei appears to be very close to NV. cacahuate in the shape and den- tition of the major chela, as well as in rel- ative proportions of the carpus and prop- odus of the minor chela. The strong simi- larity in these features, as well as the simi- larly bulbous corneas and nearly identical coloration of these species (based on fresh topotypic specimens of N. /emaitrei), qual- ifies NV. /emaitrei as the congener most sim- ilar to N. cacahuate. Whereas N. lemaitrei matures at smaller size than does N. gran- dimana, it reaches larger size (some with cl = 22 mm) than does N. cacahuate, and can be distinguished from the latter species in a number of features related to strength of dentition and amount of setation. In par- ticular, it can be distinguished readily from N. cacahuate by its less heavily toothed and more setose gape of the minor chela (Fig. 6a) and by the previously mentioned heavi- er setation of the inferior margins of the major chela, evident even in small, 1imma- ture specimens. On the basis of our pres- ently limited sample, the mature males of N. cacahuate and N. lemaitrei also appear to differ slightly in sculpture of the first ple- opods, the anterior lobe of which is angular in N. lemaitrei (Fig. 6b), and second pleo- pods, the appendix masculina of which is relatively broader and bears a larger vestige of the appendix interna in N. /emaitrei (Fig. 6c) than in N. cacahuate (Fig. 4a). Another recently reported southern Ca- ribbean form, listed as Neocallichirus sp. by Blanco Rambla & Linero Arana (1994) has proved to be a member of the genus Sergio rather than Neocallichirus. At the time that their record was published, the name Sergio was in press and the differences between Neocallichirus and Sergio, pointed out by Manning & Lemaitre (1994), were unknown to those authors. VOLUME 108, NUMBER 3 489 Fig. 6. Neocallichirus lemaitrei Manning, male (cl 13.8 mm) from Isla Rosario, Colombia, USLZ 3556. a, Minor chela, internal surface; b, Right first pleopod, posterolateral surface; c, Right second pleopod, posterior surface. Scale lines indicate | mm. The single existing female specimen of N. cacahuate is almost certainly immature, given the relatively sparse development of setation in the anterior pleopods, as com- pared to that in mature females of congeners (see Manning 1993:fig. 3b, d). However, gonopores are clearly evident on the coxae of the third pereopods. While, other than for setation, the second pleopod (Fig. 5d) resembles that of congeners, it is doubtful that articular proportions of this appendage or those of the first pleopod (Fig. 5c) have reached mature form. In the mature female, the terminal article will almost certainly have developed to a larger relative size than seen in this individual and will have de- veloped a more produced shoulder at mid- length and much longer and more abundant setation. The more acute frontal promi- nences on the carapace of this individual, the smallest of either sex collected, may sug- gest a character typical of immature speci- mens, perhaps for either sex. As the female specimen is missing the major cheliped, possible sexual dimorphism in this append- age cannot be evaluated. Acknowledgments We sincerely thank M. E. Rice, Director of the Smithsonian Marine Station at Link Port, who facilitated access to collecting sites and laboratory facilities at Fort Pierce, Flor- ida. Among many individuals who assisted with our field collections and laboratory ob- servations, we especially thank J. M. Felder, R. D. Felder, W. D. Lee, R. Lemaitre, E. A. Lazo-Wasem, L. K. Manning, P. M. Mikkelsen, S. F. Nates, S. Petry and P. Rothman. We also thank Lilly King Man- ning for preparation of numerous compar- ative study sketches reviewed during our preparation of the present paper. This study was supported through an ongoing program of Smithsonian Marine Station project grants to R. B. Manning and D. L. Felder. Partial support was also provided to D. L. Felder through U.S. Minerals Management 490 Service Cooperative Agreement 14-35- 0001-30470, U.S. Fish and Wildlife Service Cooperative Agreement 14-16-0009-89- 963, Task Order No. 6, and a small grant from the Coypu Foundation. This is con- tribution No. 372 for the Smithsonian Ma- rine Station and contribution No. 49 for the USL Laboratory for Crustacean Research. Literature Cited Biffar, T. A. 1971. The genus Callianassa (Crustacea, Decapoda, Thalassinidea) in south Florida, with keys to the western Atlantic species.— Bulletin of Marine Science 21(3):637-675. Blanco Rambla, J. P., & I. Linero Arana. 1994. New records and new species of ghost shrimps (Crus- tacea: Thalassinidea) from Venezuela.— Bulle- tin of Marine Science 55:16-29. Boone, L. 1927. Crustacea from tropical East Amer- ican seas. Scientific results of the first oceano- graphic expedition of the “Pawnee” 1925.— Bulletin of the Bingham Oceanographic Collec- tion 1(2):1-147. Felder, D. L., & R. B. Manning. 1994. Description of the ghost shrimp Eucalliax mcilhennyi, new species, from South Florida, with reexamination of its known congeners (Crustacea: Decapoda: Callianassidae).— Proceedings of the Biological Society of Washington 107:340-353. , & S. de A. Rodrigues. 1993. Reexamination of the ghost shrimp Lepidophthalmus louisi- anensis (Schmitt, 1935) from the northern Gulf of Mexico and comparison to L. siriboia, new species, from Brazil (Decapoda: Thalassinidea: Callianassidae).— Journal of Crustacean Biolo- gy 13:357-376. Gibbes, L. R. 1850. On the carcinological collections of the United States, and an enumeration of the species contained in them, with notes on the most remarkable, and descriptions of new spe- cies. — Proceedings of the American Association for the Advancement of Science, 3rd meeting: 167-201. Manning, R. B. 1987. Notes on western Atlantic Cal- lianassidae (Crustacea: Decapoda: Thalassinid- ea).— Proceedings of the Biological Society of Washington 100:386-401. . 1993. Two new species of Neocallichirus from the Caribbean Sea (Crustacea: Decapoda: Cal- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON lianassidae).—Proceedings of the Biological So- ciety of Washington 106:106-114. —, & D. L. Felder. 1986. The status of the cal- lianassid genus Callichirus Stimpson, 1866 (Crustacea: Decapoda: Thalassinidea).— Pro- ceedings of the Biological Society of Washington 99(3):437-443. ——., & 1991. Revision of the American Callianassidae (Crustacea: Decapoda: Thalas- sinidea). — Proceedings of the Biological Society of Washington 104:764-792. ——., & 1992. Gilvossius, a new genus of callianassid shrimp from the eastern United States (Crustacea: Decapoda: Thalassinidea). — Bulletin of Marine Science 49(1—2) [199 1]:558- 561. —, & 1995. Description of the ghost shrimp Sergio mericeae, new species, from South Florida, with reexamination of S. guassutinga (Crustacea: Decapoda: Callianassidae).—Pro- ceedings of the Biological Society of Washington 108:266—280. —., & R. W. Heard. 1986. Additional records of Callianassa rathbunae from Florida and the Ba- hamas (Crustacea: Decapoda: Callianassi- dae). — Proceedings of the Biological Society of Washington 99:347-349. —, & R. Lemaitre. 1994. Sergio, a new genus of ghost shrimp from the Americas (Crustacea: De- capoda: Callianassidae).— Nauplius (Brazil) 1:39-44. Milne Edwards, A. 1870. Révision du genre Calli- anassa (Leach).—Nouvelle Archives du Mu- seum d’Histoire Naturelle, Paris 6:75—101, pls. 1-2. Rodrigues, S. de A., & R. B. Manning. 1992. Two new callianassid shrimps from Brazil (Crusta- cea: Decapoda: Thalassinidea). — Proceedings of the Biological Society of Washington 105:324— 330. Sakai, K. 1988. A new genus and five new species of Callianassidae (Crustacea: Decapoda: Thalas- sinidea) from northern Australia.— The Beagle, Records of the Northern Territory Museum of Arts and Sciences 5(1):51-69. Say, T. 1818. An account of the Crustacea of the United States [Part 5].—Journal of the Academy of Natural Sciences of Philadelphia 1 (part 2, number 1):235-253. Schmitt, W. L. 1935. Mud shrimps of the Atlantic coast of North America.—Smithsonian Miscel- laneous Collections 93(2):1—21, pls. 1-4. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(3):491-495. 1995. Olavius nicolae, a new gutless marine tubificid species (Oligochaeta) from Belize Christer Erseus and Olav Giere (CE) Department of Invertebrate Zoology, Swedish Museum of Natural History, Box 50007, S-104 05 Stockholm, Sweden; (OG) Zoologisches Institut und Zoologisches Museum, Universitat Hamburg, Martin-Luther-King-Platz 3, D-20146 Hamburg, Germany Abstract. —Olavius nicolae, a new species from shallow water sediments at Carrie Bow Cay, on the barrier reef off Belize in Central America, is described. It belongs to an apomorphic group within the gutless genus Olavius Erséus, 1984, characterized by small, oval atria, large copulatory sacs, and dorsal sper- mathecae (the latter located in the middle of segment X). It differs from the closely related, Hawaiian species, O. parapellucidus Erséus & Davis, 1989, by its more slender body form and longer clitellum, its more numerous somatic setae, and the more dorsal position of its spermathecal pores. The barrier reef off Belize in Central America (Caribbean Sea) provides suitable habitats for several species of gutless Tu- bificidae within the genera /nanidrilus Er- seus, 1979, and Olavius Erséus, 1984 (Er- seus 1990, Diaz & Erséus 1994). The Smith- sonian Institution’s Carrie Bow Cay Field Laboratory in Belize was therefore chosen as a site for the collection of material of various gutless taxa, as part of an ongoing research project on evolutionary and bio- geographical aspects of the symbiosis of gut- less Tubificidae with sulphur oxidizing bac- teria (N. Dubilier, C. Erseus and O. Giere, principal investigators; see Giere et al. 1995). While at Carrie Bow, one previously un- known species of Olavius was found by the present authors. This species is described herein. Tubificid worms were sorted live from sieved organic material extracted from sed- iment samples collected in shallow water on the reef platform surrounding Carrie Bow Cay. Specimens to be used for light mi- croscopy examination were fixed in Bouin’s fluid overnight, preserved in 70% ethanol for some weeks, and then stained in (alco- holic) paracarmine, dehydrated in an alco- hol series and mounted whole in Canada balsam. For transmission microscopy ex- amination of body wall and symbiotic bac- teria, mid-body or posterior fragments of a few worms were fixed in Trump’s fixative (McDowell & Trump 1976), buffered with cacodylate. After embedding (Spurr’s resin) ultrathin sections were mounted on copper grids, contrasted in aqueous uranyl acetate and lead citrate and inspected in a ZEISS EM 902A. The type specimens are deposited in the United States Museum of Natural History (USNM), Smithsonian Institution, in Washington, D.C., and the Swedish Muse- um of Natural History (SMNH), in Stock- holm. Olavius nicolae, new species Figs. 1-2 Olavius n. sp.; Giere et al. 1995:table 1. Holotype. —USNM_ 171048, whole- mounted specimen. Type locality.—Immediately E of N tip of Carrie Bow Cay (16°48'N, 88°05'W), bar- 492 \ B C PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ipa another specimen; D, horizontal view of genitalia in segments X—XI. Abbreviations: a, atrium; cs, copulatory sac; mf, male funnel, prl, anterior prostate gland; pr2, posterior prostate gland; ps, penial seta; s, spermathecae; vd, vas deferens. rier reef of Belize, Central America, inner end of Thalassia bed (about 5 m from beach), barely subtidal, fine to medium, heteroge- neous sand; 6 March 1993. Paratypes. -—USNM 171049-171051, three specimens, and SMNH Type coll. 4676-4682, seven specimens; all whole- mounted, from type locality. Description. — Length of (only) three com- plete specimens, 4.6—5.0 mm, 35-38 seg- ments. Width at segment XI, 0.09—0.19 mm; but worms much narrower in other parts of body, at posterior end (of complete speci- mens) only about 0.04 mm wide. Prosto- mium generally elongate, pointed triangu- lar. Secondary annulation of body wall vari- Olavius nicolae, new species. A, Somatic seta; B, penial setae of one specimen; C, penial setae of able, not always distinct, with 5-8 annuli per segment in postclitellar part of body; in some specimens irregular, large annuli giv- ing worm a “knotty” appearance. Clitellum extending over 73X—XII plus anterior mar- gin of segment XIII. Somatic setae (Fig. 1A) bifid, with upper tooth shorter and much thinner than lower, and with subdental lig- ament. Bifid setae 21—33 um long, about |- 1.5 wm thick, three or four per bundle an- teriorly, two or three per bundle in postcli- tellar segments. Penial setae (Fig. 1 B—C, D) slender, two per bundle, ventrally in seg- ment XI, 38-56 wm long, about 2 wm thick. In front view, penial setae appearing straight, with chisel-shaped tips (Fig. 1B); in side VOLUME 108, NUMBER 3 493 Fig. 2. Olavius nicolae, new species, TEM micrograph. Cross-section through dermal-muscular layer with numerous bacteria (ba) between cuticle and epidermal tissue (ep); bc, coelomic body cavity; mu, longitudinal musculature. view, penial setae sigmoid, with tips ap- pearing sharply single-pointed and very thin (Fig. 1C). Penial setae tightly parallel within bundle, either obliquely directed towards posterior, sometimes even approaching a horizontal position in segment (a with- drawn position), or more or less erect in segment, 1.e., perpendicular to long axis of worm (a protruded position). Male pores paired, in line with ventral setae, posterior to middle of segment XI. Spermathecal pores paired, in line with dorsal setae, in middle of segment X. Alimentary canal absent. Body wall with numerous subcuticular bacteria (Fig. 2; see Remarks). Male genitalia (Fig. 1D) paired. Vas deferens non-muscular, ciliated, 6—8 wm wide, much longer than atrium, entering apical end of latter. Atrium oval, only 26- 30 um long, 16-19 wm wide, with very thin outer lining and granulated inner epitheli- um (internal ciliation probably present, but not seen). Atrium opening into inner end of complex copulatory sac. Sac with folded walls of varying thickness, generally with a distinct papilla near entrance of atrium. Sac everted in a few specimens. Anterior pros- tate gland larger than posterior one, located anterior to atrium; attachment with atrium, however, not seen. Posterior prostate gland located posterior to copulatory sac, attached to atrium by long stalk. Spermathecae (Fig. 1D) slender, generally club-shaped, 60-75 um long, consisting of very short ducts, and thin-walled ampullae; latter 15—28 um wide, with loose masses of, or somewhat bundled, sperm. Spermathecal ampullae generally di- rected towards the anterior. Etymology. — Named for Dr. Nicole Du- bilier (Biolabs, Harvard University, Cam- 494 bridge, MA), friend and collaborator in our joint symbiosis project, who participated in the collection of gutless Tubificidae at Car- rie Bow Cay. Remarks. —The subcuticular bacteria (Fig. 2) are of the same morphotype and arrangement as first described for Inanidri- lus leukodermatus (Giere, 1979) by Giere (1981), and as found also in other related species (Giere et al. 1995, for details). Distribution and habitat.—Known only from Belize. Barely subtidal heterogeneous sand associated with seagrass. At the type locality, it co-occurs with other gutless taxa: Inanidrilus reginae Erséus, 1990, Olavius (O.) tantulus Erséus, 1984, and O. imper- fectus Erséus, 1984. Discussion Olavius nicolae belongs to a group of small species within Olavius, all characterized by (1) the location of the spermathecal pores in the middle, rather than the anterior, of segment X, (2) the dorsal, rather than ven- tral or lateral, location of the same pores, and (3) the very small atria, opening at the inner end of large, convoluted copulatory sacs. Other taxa in this group are O. ten- uissimus (Erséus, 1979), O. cornuatus Da- vis, 1984, O. pellucidus Erséus, 1984, O. macer Erséus, 1984, O. bullatus Finogeno- va, 1986, O. parapellucidus Erséus & Davis, 1989, O. finitimus Erséus, 1990, O. tanner- ensis Erséus, 1991, and O. rallus Erséus, 1991, some of which are from the Atlantic, others from the Pacific Ocean. The new species appears most closely re- lated to O. parapellucidus, only known from Hawaii. This species has penial setae, two per bundle, with chisel-shaped tips (Erséus & Davis 1989), almost identical to those of O. nicolae. A number of differences are not- ed, however. Olavius nicolae is very slender, with the mid-body and posterior parts con- siderably narrower than clitellar region, whereas O. parapellucidus is more or less evenly wide throughout body (unpublished PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON observation on material in CE collection). Moreover, in O. parapelludicus the clitel- lum covers most of segment XII, but never extends into XIII as in the present species. The somatic setae (about 2 um thick in O. parapellucidus) are more numerous in O. nicolae (setae maximally three per bundle anteriorly, two per bundle posteriorly, in O. parapellucidus). Finally, the spermathecal pores are in the lines of the dorsal setae in O. nicolae, but between these lines and the lateral lines in O. parapellucidus. With the addition of O. nicolae, a total of 60 species of marine Tubificidae are known from the barrier reef off Belize (cf. Erséus 1990). Acknowledgments We are grateful to Dr. Nicole Dubilier (Harvard University) for cooperation in the gutless tubificid project; to Ms. Barbro Lof- nertz (University of Goteborg) and Ms. Christine Hammar (Swedish Museum of Natural History), for technical assistance; to the Smithsonian Institution’s Caribbean Coral Reef Ecosystems Program (CCRE), the Swedish Natural Science Research Council, and the Deutsche Forschungsge- meinschaft (Gi 100/14-1), for financial sup- port. This is CCRE Contribution No. 461 from the Carrie Bow Cay Field Laboratory. Literature Cited Davis, D. 1984. Olavius cornuatus sp. n. (Oligochae- ta, Tubificidae) from Georges Bank (NW Atlan- tic). — Zoologica Scripta 13:273-275. Diaz, R. J., & C. Erséus. 1994. Habitat preferences and species associations of shallow-water ma- rine Tubificidae (Oligochaeta) from the barrier reef ecosystems off Belize, Central America. — Hydrobiologia 278:93-105. Erséus, C. 1979. Taxonomic revision of the marine genus Phallodrilus Pierantoni (Oligochaeta, Tu- bificidae), with descriptions of thirteen new spe- cies.— Zoologica Scripta 8:187—208. . 1984. Taxonomy and phylogeny of the gut- less Phallodrilinae (Oligochaeta, Tubificidae), with descriptions of one new genus and twenty- two new species. — Zoologica Scripta 13:239-272. VOLUME 108, NUMBER 3 1990. The marine Tubificidae (Oligochaeta) of the barrier reef ecosystems at Carrie Bow Cay, Belize, and other parts of the Caribbean Sea, with descriptions of twenty-seven new species and revision of Heterodrilus, Thalassodrilides and Smithsonidrilus. —Zoologica Scripta 19:243- 303. 1991. Two new deep-water species of the gutless genus Olavius (Oligochaeta: Tubificidae) from both sides of North America.—Proceed- ings of the Biological Society of Washington 104: 627-630. . 1992. A generic revision of the Phallodrilinae (Oligochaeta, Tubificidae).— Zoologica Scripta 21:5-48. —., & D. Davis. 1989. The marine Tubificidae (Oligochaeta) of Hawaii.—Asian marine Biol- ogy 6:73-100. Finogenova, N. P. 1986. Six new species of marine 495 Tubificidae (Oligochaeta) from the continental shelf off Peru.— Zoologica Scripta 15:45—-51. Giere, O. 1979. Studies on marine Oligochaeta from Bermuda, with emphasis on new Phallodrilus species (Tubificidae). — Cahiers de Biologie Ma- rine 20:301-314. 1981. The gutless marine oligochaete Phal- lodrilus leukodermatus. Structural studies on an aberrant tubificid associated with bacteria. — Marine Ecology Progress Series 5:353-357. , C. Nieser, & C. Erséus. 1995. A comparative structural study on bacterial symbioses of Ca- ribbean gutless Tubificidae (Annelida, Oligo- chaeta).— Acta Zoologica (Stockholm) 76:000- 000 (in press). McDowell, E. M. & B. F. Trump. 1976. Histologic fixatives suitable for diagnostic light and elec- tron microscopy.—Archives of Pathology and Laboratory Medicine 100:505—514. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(3):496-501. 1995. Pionosyllis maxima Monro, 1930, P. anops Hartman, 1953, and P. epipharynx Hartman, 1953, redescribed as Eusyllis maxima (Monro, 1930), a new combination (Polychaeta: Syllidae: Eusyllinae) Maria Jiménez, G. San Martin, and E. Lopez Departamento de Biologia, Unidad de Zoologia, Facultad de Ciencias, Universidad Autonoma de Madrid, Madrid 28049, Spain Abstract.—Three Antarctic species of Pionosyllis Malmgren, 1867 (Poly- chaeta: Syllidae: Eusyllinae) including P. maxima Monro, 1930; P. anops Hart- man, 1953, and P. epipharynx Hartman, 1953, are synonymous. They are redescribed and referred to Eusyllis maxima (Monro, 1930), new combination, based on morphological characters, most notably the presence of an incomplete crown of denticles on the opening of the pharynx. We have recently begun a revision of the genus Pionosyllis Malmgren, 1867, and are currently examining all available types to redescribe poorly known species (Jiménez et al. 1994). Examination of P. maxima Monro, 1930, P. anops Hartman, 1953, and P. epipharynx Hartman, 1953, types from South Georgia reveals that they have iden- tical parapodia, aciculae, compound setae and characteristic pharynges. All species are considered to be synonymous, and are re- ferred below to Eusyllis Malmgren, 1867, because the rim of eversible pharynx is den- ticulated. Pionosyllis epipharynx and P. an- ops are considered the junior synonyms of P. maxima: the latter is redescribed as Eu- syllis maxima (Monro, 1930) new combi- nation. All the material examined is preserved specimens or fragments provided by The Natural History Museum (BMNH), Lon- don, and the Swedish Museum of Natural History (SMNH), Stockholm. The origins of these specimens are detailed under ma- terial examined. The specimens were examined using a stereomicroscope and a compound micro- scope provided with differential interfer- ence contrast optics (Nomarsky). Drawings were made by means of a camera lucida drawing tube, using the stereomicroscope for the complete specimens and the com- pound microscope for the parapodia and setae. Body width measurements were taken across the proventriculus, and exclude par- apodia or cirri; body length measurements exclude antennae, palps and anal cirri. Taxonomy Family Syllidae Grube, 1850 Subfamily Eusyllinae Rioja, 1925 Genus Eusyllis Malmgren, 1867 Eusyllis maxima (Monro, 1930), new combination Jae, J Z Pionosyllis maxima Monro, 1930: 92, fig. 29; Hartman 1964: 87, pl. 27, figs. 3-5; Hartmann-Schroder & Rosenfeldt 1988: 39. Pionosyllis anops Hartman, 1953: 24, fig. 6 a—c; 1964: 85, pl. 26, figs. 5-6. Pionosyllis epipharynx Hartman, 1953: 23, figs. 3-4; 1964: 87, pl. 27, figs. 3-5; 1967: 58. Material examined. —South Georgia: 2.7 miles S, 85°E of Jason Light, 238-270 VOLUME 108, NUMBER 3 m, grey mud, st. 45, 6 April 1926 (3 speci- mens, syntypes of P. maxima, BMNH 1930.10.8.319-—20); Swedish Antarctic Ex- pedition 1901-1903, Cumberland Bay, 252- 310 m, grey mud with few stones, st. 34 (syntypes of P. epipharynx, SMNH 611, 7 spec. and SMNH 622, 2 spec.); mouth of Westfjord, 54°15'S, 36°25’W, 250 m, st. 18 (as P. epipharynx, SMNH 3050, | spec.); May Bay, intertidal algae (syntypes of P. anops, SMNH 607, 2 spec. and as P. epi- pharynx, SMNH 3045, | ant. end); Gryt- viken, algae (as P. epipharynx, SMNH 3044, 1 spec.); Antarctic Bay, 54°12’S, 36°50’W, 250 m, stones, st. 20 (as P. epipharynx, SMNH 3032, 3 ant. ends). All examined specimens were fragmented, except SMNH 607. Description. — Body long, thick, massive, Opaque, subcylindrical anteriorly, arched dorsally, flattened ventrally, with well de- fined segments (Fig. 1A, B). Only one com- plete specimen: 9.5 mm long, 0.8 mm wide with 52 setigers (P. anops, SMNH 607); largest anterior end fragment 15 mm long with 50 setigers. Dorsum with a broad, transverse reddish brown band on anterior part of each segment surrounded by a nar- row pale area posteriorly and laterally, giv- ing an overall impression of alternating dark and light transverse stripes (Fig. 1A), some- times faded and without color in several specimens. Prostomium slightly wider than long, occipital cleft small, usually covered by dorsal flap of first setiger; four lensed eyes in almost rectangular arrangement, anterior pair slightly larger than posterior one. Me- dian antenna arising from middle of pro- stomium; lateral antennae arising anterior to anterior pair of eyes, somewhat shorter than median antenna. Antennae cylindrical, smooth generally missing or broken. Palps broad, fused basally, divergent, as long as prostomium (Fig. 1A, B). Peristomial ring strongly reduced dorsally, covered by flap from first setiger; dorsal tentacular cirri somewhat longer than body width, ventral tentacular cirri *%3 shorter. Antennae, ten- 497 tacular and dorsal cirri similar in shape, cy- lindrical, ending in conical tip (Fig. 1C), sometimes coiled over dorsum (Fig. 1A), slender, smooth, somewhat wrinkled dis- tally. Dorsal cirri about two times longer than body width, alternating with shorter cirri, about as long as body width (Fig. 1A) arising from massive cirrophores. Ventral cirri thick, rounded not extending beyond parapodial lobes. Parapodial lobes with a postsetal papilla and a smaller presetal pa- pilla (Fig. 1C, D). Compound setae heter- ogomph falcigers, numerous, 80-100 per anterior parapodia, progressively decreas- ing to about ten posteriorly; distal superior surface of shafts spinous, with three or four rows of spines (Fig. 2B, E). Compound setae of anterior parapodia (Fig. 2B) with marked dorso-ventral gradation in shape; dorsal blades bidentate, distal tooth falcate and in- creasing in size ventrally; ventral blades un- identate, distal tooth with an inconspicuous furrow on the edge; blades about 38 um dorsally, 24 um ventrally. Posterior setigers provided only with unidentate blades (Fig. 2E), distally curved, falcate; blades on pos- terior setigers 28 um dorsally, 20 um ven- trally; spines on edge of blades short, up- wardly directed. Anterior parapodia each with two or three aciculae, slightly curved distally, with blunt tips (Fig. 2C); posterior parapodia each with one or two aciculae of the same type (Fig. 2F). Solitary dorsal sim- ple seta present from setiger 39 (only found in P. anops, SMNH 607), indistinctly bi- dentate, slender, somewhat curved, finely serrated distally on inferior cutting surface (Fig. 2D). Ventral simple setae not seen. Notosetae and notoaciculae present from setiger 19. Pygidium with paired ventrolat- eral anal cirri, smooth as long as the last five segments. Pharynx extending through five to ten setigers, with a distal circlet of ten soft papillae surrounding opening and an incomplete crown of about 28 chitinous denticles on the rim; anterior pharyngeal tooth conical with enlarged base (Fig. 2A). Proventriculus barrel-shaped, extending 498 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Eusyllis maxima (Monro, 1930) n. comb.: A. anterior end, dorsal view, B. anterior end, ventral view (A, B: P. maxima, BMNH 1930.10.8.319-20), C. midbody parapodium, posterior view showing notosetae (neurosetae not drawn) (P. epipharynx, SMNH 3044), D. anterior parapodium, posterior view (P. maxima, BMNH 1930.10.8.319—20). Scale: A, B: 0.6 mm; C: 0.25 mm; D: 0.2 mm. VOLUME 108, NUMBER 3 499 F Fig. 2. Eusyllis maxima (Monro, 1930) n. comb.: A. pharynx (P. epipharynx, SMNH 611), B. dorsal, median and ventral compound setae, anterior parapodium, C. aciculae, anterior parapodium (B, C: P. maxima, BMNH 1930.10.8.319-20), D. dorsal simple seta (P. anops, SMNH 607), E. dorsal, median and ventral compound setae, posterior parapodium, F. acicula, posterior parapodium (E, F: P. maxima, BMNH 1930.10.8.319-20). Scale: A: 0.8 mm; B—F:20 um. through four to ten setigers, with about 120 nearby localities; Antarctic; Southern Ar- muscle cell rows. gentina. Type locality. —East of Jason Light, South Ecology. — Intertidal with algae to 900 m Georgia, 238-270 m. depth (Hartmann-Schroder & Rosenfeldt Distribution. —South Georgia: East of Ja- 1988, as P. maxima) in grey mud. son Light, May Bay, Cumberland Bay, and Remarks. —Pionosyllis maxima, P. epi- 500 pharynx and P. anops have identical para- podia, aciculae and compound setae, and were described from South Georgia. These three species were described as having pha- ryngeal armature composed by a solitary tooth; however, a careful examination ofthe opening of the pharynx on specimens from each the three species, shows an incomplete crown of chitinous denticles surrounded by a crown of ten soft, thick papillae that is difficult to see. For this reason, we consider these three species to be synonymous and transfer them to the genus Eusyllis. Eusyllis is characterized by having smooth to irreg- ularly wrinkled antennae and dorsal cirri, and an incomplete crown of denticles on the anterior rim of the pharynx together with a pharyngeal tooth (Malmgren 1867, Fauvel 1923). Hartman (1953) described the pharynx of P. epipharynx having a middorsal process and a crown of 23 papillae; this is a mis- interpretation of the pharynx. Her figure 4B shows the ten soft papillae and the pharyn- geal tooth, which is much larger than that of the remaining species of the genus, and her figure 4A shows the pharyngeal tooth and the crown of chitinous denticles, which she describes as papillae. In the same paper she describes a smaller specimen of the same species as P. anops; the only difference was that the last species lacks eyes which is not a good character because these sometimes disappear in alcohol. Most species of Eusyllis have been de- scribed as having compound setae with bi- dentate blades, but E. maxima has com- pound setae falcate and unidentate, except the dorsalmost setae of anterior parapodia; E.. maxima 1s close both to E. kupfferi Lan- gerhans, 1879 (Langerhans 1879, San Mar- tin 1990) from Madeira, Cuba and Canary Islands and E. transecta Hartman, 1966 (Hartman 1969) from California, in having a large pharyngeal tooth and unidentate compound setae. However, these two spe- cies are much smaller. Hartman (1953) re- ported the maximum size for P. epipharynx PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 41 mm long, 2.2 mm wide for more 75 se- tigers; E. transecta has 6.1 mm long for about 46 setigers and E. kupfferi 4.5 mm long for 35 setigers and have different colour pat- tern; moreover, they have unidentate dorsal simple seta and the shafts of the compound setae are smooth or with few, short spines; furthermore E. transecta has dorsal cirri weakly wrinkled. Acknowledgments We wish to express our gratitude to Dr. L. Sandberg, Swedish Museum of Natural History (Stockholm), Mr. A. Muir and Ms. M. Lowe, The Natural History Museum (London), who have lent us the material required in our study. Dr. M. E. Petersen, revised the manuscript and gave us very valuable advice. The comments of two anonymous referees greatly improved the quality of the paper. This paper has been supported by the project CICYT (Comision Interministerial de Ciencia y Tecnologia) ANT 93-0996: “‘Estudios de la fauna y flora bentonica de los fondos de la zona Sur de la Isla Livingston (Shetland del Sur, Antar- tida).” Literature Cited Fauvel, P. 1923. Faune de France. 5. Polychétes Er- rantes. Le Chevalier ed., Paris, 486 pp. Grube, A. E. 1850. Die Familien der Anneliden.— Archivs fiir Naturgeschichte, Berlin 16:249-364. Hartman, O. 1953. Non-pelagic Polychaeta of the Swedish Antarctic Expedition 1901-1903.— Further Zoological Results of the Swedish Ant- arctic Expedition 4(1 1):1-83. 1964. Polychaeta Errantia of Antarctica.— Antarctic Research Series 3:1-131. American Geophysical Union Publ. No. 1226. . 1967. Polychaetous annelids collected by the USNS Eltanin and Staten Island Cruises, chiefly from Antarctic Seas.—Allan Hancock. Mono- graphs in Marine Biology 2:1-387. . 1969. Atlas of the errantiate and sedontanate polychaetous annelids from California.— Allan Hancock Foundation, Los Angeles, California. Hartmann-Schréder, G., & P. Rosenfeldt. 1988. Die Polychaeten der “‘Polarstern’”-Reise ANT III/2 in die Antarktis 1984. Teil 1: Euphrosinidae bis VOLUME 108, NUMBER 3 Chaetopteridae.— Mitteilungen aus dem Ham- burgischen zoologischen Museum und Institut 85:25-72. Langerhans, P. 1879. Die Wirmfauna von Madei- ra.— Zeitschrift fur wissenchaftliche Zoologie 32: 513-592. Jiménez, M., G. San Martin, & E. Lopez. 1994. Re- descriptions of Pionosyllis neapolitana Good- rich, 1900 and Pionosyllis nutrix Monro, 1936, referred to the genus Grubeosyllis Verrill, 1900 (Polychaeta, Syllidae, Exogoninae).—Poly- chaete Research 16:52-55. Malmgren, A. J. 1867. Annulata Polychaeta Spets- 501 bergiae, Groenlandiae, Islandiae et Scandina- viae hactenus cognita.—Ofversigt af Svenska Vetenskaps Academiens F6rhandlinger 24:1- WAH. Monro, C. C. A. 1930. Polychaete worms. — Discov- ery Reports 2:1—122. Rioja, E. 1925. Anélidos poliquetos de San Vicente de la Barquera (Cantabrico). — Trabajos del Mu- seo Nacional de Ciencias Naturales. Serie Zool- Ogica 53:1-62. San Martin, G. 1990. Eusyllinae (Syllidae, Polychae- ta) from Cuba and Gulf of Mexico. — Bulletin of Marine Science 46(3):590-619. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(3):502-507. 1995. A new freshwater snail from the Coosa River, Alabama (Gastropoda: Prosobranchia: Hydrobiidae) Fred G. Thompson Florida Museum of Naturasl History, University of Florida, Gainesville, Florida 32611, U.S.A. Abstract. —Pyrgulopsis hershleri, a new species, is described from the Coosa River, Alabama. It is closely related to three species from the Tennessee River system of northern Alabama. It is distinguished by characteristics of the shell and soft anatomy. The new species was incorrectly identified in earlier literature as Amnicola olivacea Pilsbry, 1895 (=Pyrgulopsis olivacea). The hydrobiid genus Pyrgulopsis Call & Pilsbry, 1886 includes 66 described and many undescribed species of freshwater snails that are confined to temperate North America (Hershler 1994). Earlier Thomp- son (1977) reviewed eight eastern North American species under the genus name Marstonia Baker, 1926, which Hershler & Thompson (1987) later relegated to junior synonymy of Pyrgulopsis. The eastern spe- cies form a distinct clade within Pyrgulopsis distinguished by characteristics of apical whorl sculpture, operculum indentation, orientation of the terminal lobe on the pe- nis, and aspects of the oviduct coil and the bursa copulatrix duct (Hershler 1994:81). Thompson (1977) noted the existence of an undescribed species in the Coosa River, Al- abama, but the material available at the time was insufficient for taxonomic treatment. Recent field studies on the ecology of an endangered snail, Tulotoma magnifica (Conrad, 1838) provided the opportunity to collect other mollusks from a limited sec- tion of the Coosa River near Wetumpka, Alabama. The undescribed Pyrgulopsis was found to be common in this area. Its de- scription is as follows. Pyrgulopsis hershleri, new species Vernacular name: Coosa Pyrg Diagnosis. —A medium-sized species of Pyrgulopsis that is about 2.7—2.9 mm long with an elongate-conical shell consisting of about 4.5 strongly arched whorls. The shell is thin and fragile, and lacks a noticeable callus within the aperture. The verge has a square-shaped oblique apical lobe, thereby relating it to a group of species recorded from the Tennessee River system in Ala- bama (Fig 1). This group includes Pyrgu- lopsis argus (Thompson, 1977), P. ogmor- haphe (Thompson, 1977) and possibly P. olivaceus (Pilsbry, 1895). The new species is unique within this group by having a pig- mented penis filament. Shell (Figs. 2, 3A—C).—Shell thin, trans- parent when live; small, about 2.7—2.9 mm long (Table 1); light gray in color, surface glossy when fresh; elongate-conical in shape, 0.62—0.70 times as wide as long; spire slight- ly convex in outline, equal to or slightly longer than height of aperture, about 0.50- 0.58 times length of the shell. Whorls 4.5— 4.7; strongly arched but tending to be flat- tened along the periphery; suture deeply im- pressed; apical whorl 0.23-0.24 mm in di- ameter perpendicular to initial suture. Mi- crosculpture on first half of the apical whorl coarsely pitted. Surface of shell smooth; mi- crosculpture consisting of very fine irregular incremental striations and fine sparse spiral striations. Umbilicus narrowly perforate. Aperture slightly oblique, prosocline, lying at an angle of 16—20° to axis of shell in lateral profile; tending to be rhomboid in shape in VOLUME 108, NUMBER 3 503 Fig. 1. most specimens, about 0.81-0.90 times as high as wide; posterior corner bluntly an- gulate; interior with a thin internal callus along outer and basal margin. Peristome complete across parietal wall; outer lip tend- ing to be flattened compared to contour of earlier whorls; outer lip weakly arched for- ward in lateral profile. Measurements for the holotype and selected paratype are given in Table 1. Operculum (Fig. 3D).— Broadly ovate in shape; upper columellar edge slightly in- The Coosa River at the type locality of Pyrgulopsis hershleri, new species. dented; paucispiral, consisting of about 2.5 rapidly expanding whorls, nucleus subcen- tric. Attachment scar very thin and broad, about half the length of the operculum, ex- tending through nucleus; ventral callus of scar weak or absent. Male (Fig. 3E).—Base of penis com- pressed and nearly uniform in width. Penis with an enlarged oblique apical lobe that usually has a small terminal gland on its tip. Occasional specimens lack the terminal gland. Other glands are absent elsewhere on Table 1.—Pyrgulopsis hershleri, new species. Measurements in mm based on 13 adult paratypes and the holotype. SL = standard length, SW = standard width, ApH = aperture height, ApW = aperture width, Wh = whorls. SE SW ApH ApW Wh SW/SL ApH/SL ApW/ApL 6 oT 1.8 1.2 1.1 4.6 0.66 0.45 0.85 S? 0.12 0.07 0.06 0.06 0:11 0.03 0.04 0.04 min. 2.5 17 ee 0.9 45 0.61 0.42 0.79 max. 2.9 1.9 1.4 12 4.7 0.71 0.50 0.90 holotype 2.8 1.8 Le 1.1 4.5 0.64 0.44 0.85 504 mm Fig. 2. Pyrgulopsis hershleri, new species. Holotype (UF 165788). surface of penis. Penis filament moderately slender, extending beyond the apical lobe, densely pigmented with melanophores. Type locality. — Alabama, Elmore Coun- ty, Coosa River, braided island area above Moccasin Shoals, about 2.2 miles down- stream from the Jordan Hydroelectric Dam; T 18 N, R 18 E. Holotype: UF 165788; collected July 24, 1990 by Fred G. Thomp- son and Malcolm Pierson. Paratype: UF 165789 (9), UF 174347 (30), USNM 860563 (8); UF 165790 (5 in alcohol), same data as the holotype. Corn Creek Shoals, ca 7.0 mi. NW of Wetumpka UF 230595 (22), UF 230735 (10). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON The river at the type locality consists of numerous shoals and shallow intermittent pools overlying a substrata of granite-schist outcrops, boulders and gravel. The river is braided with small islands and rock reefs (Fig. 1). This section of the river fluctuates between shallow gentle flow and violent floods that occasionally rise as much as 35 feet above normal water level. Normally the river fluctuates by less than a meter on a daily cycle because of regulated discharge from the Jordan Hydroelectric Dam up- stream, but heavy rains may raise the water level to torrential levels that can scour and reorganize the river bed. Pyrgulopsis hersh- leri was found only on rootlets of the bald cypress, Taxodium disticum, growing along the edges of quiet shallow pools. Other spe- cies of aquatic angiosperms, algae mats and various types of substrata were screened for specimens, but without success. Distribution.—Specimens are available from two sections in the Coosa River. One lot (UMMZ 161894), contains two speci- mens from Etowah County, without more exact information. The second section from which specimens are known is the short seg- ment of the river between Jordan Dam and Wetumpka. This portion of river contains numerous intermittent shoals distributed Over a distance of about seven miles. This is virtually the only section of the Coosa River that remains in a nearly original state. The rest of the river above and below this section is highly modified by impound- ments. The microhabitat from which P. her- shleri was collected suggests that the species may be present in other sections of the river where there are bald cypress trees along the river edge. Available survey data is insuf- ficient to address this question. Recom- mendations concerning the conservation status of the snail are not practical until a more intensive search is made elsewhere in the Coosa River. Pyrgulopsis hershleriis the only species of the genus known from the Alabama River system. Remarks.—A popular misconception in VOLUME 108, NUMBER 3 D 505 => terminal gland x S 7—apical lobe Ie 2 tee E Fig. 3. Pyrgulopsis hershleri, new species. A, lateral view of holotype; B—C, paratypes; D, operculum; E, penis. Scale bar = I mm for Figs. A-D, 0.75 mm for Fig. E. malacology is that the freshwater molluscan faunas of the Alabama River system and its principal tributary, the Coosa River, are well known. This misconception is based pri- marily on the numerous works by Calvin Goodrich who wrote extensively on the Pleuroceridae as well as other prosobranchs (e.g., 1936, 1941, 1944a, 1944b). His stud- ies on the Pleuroceridae are marred by omission of data concerning distributions, taxonomy and the unresolved status of the numerous names that he placed in synon- ymies. His treatments of other families 1n- clude many species that are suspect of mis- identification. Indeed the systematic status of virtually every species of gastropod re- corded from the Coosa requires reexami- nation. This paper concerns one such spe- cies. Goodrich (1944a:6—10) discussed various hydrobiid gastropods from the Coosa River system, including five species of Amnicola. Specimens of the first species, Amnicola li- mosa (Say, 1817) remain as Goodrich as- signed them by virtue of the fact that the systematics of the /imosa species-group are in critical need of review. A different as- signment to species could not be made at present. Certainly they are a species of Am- nicola. The names of two other species re- 506 corded by Goodrich, A. sanctijohannis Pils- bry, 1899 and A. augustina Pilsbry, 1904, are synonyms of Cincinnatia floridana (Frauenfeld, 1863), a species endemic to the Florida peninsula. The specimens upon which Goodrich based his records as well as those for “A.” floridana from the Coosa River have been shown to be incorrectly identified (Thompson 1968:123). Speci- mens of another species were identified by Goodrich as Amnicola olivacea Pilsbry, 1895 (=Pyrgulopsis olivacea), which until then was known only from a spring in the Tennessee River system at Huntsville, Alabama. Thompson (1977:123) showed that these also were misidentified. They represent the new species that is the subject of this paper. Pyrgulopsis hershleri is a small, thin- shelled species with a simple conical shell that has 4.5 or fewer whorls. Its size, shell- shape and aperture-shape distinguish it from other eastern North American Pyrgulopsis. Anatomical features are rather monoto- nously similar among the eastern species compared to western congeners (Hershler 1994). Variation in penial morphology of- fers some useful comparisons. Penial mor- phology of Pyrgulopsis hershleri is most similar to that of two species from the Ten- nessee River system, P. argus (Thompson, 1977) and P. ogmorhaphe (Thompson, 1977). These species are alike by having a well developed oblique apical lobe on the distal left margin of the penis. The right margin of the penis ends with a slender short penis filament. The lobe has a single small terminal apocrine gland. Other glands are absent on the lobe and penis. Pyrgulopsis hershleri differs from P. argus and P. og- morhaphe by having a pigmented penis fil- ament. The filament is unmarked in the oth- er two species. It differs further from these two species by its smaller size and thinner Shell, as well as by shell shape. The adult Shell of Pyrgulopsis ogmorhaphe is about 4.0-5.1 mm long with about 5.2—5.8 whorls. It is elongate conical-terrete in shape, and it has a more broadly ovate aperture. Ju- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON venile shells are thin and transparent as in P. hershleri, but the shell becomes slightly thicker in adults. Like P. hershleri the shell is relatively fragile, it does not develop a strong callus on the inner margin of the out- er lip, and the outer lip is nearly straight when viewed in latyeral profile. The adult shell of Pyrgulopsis argus is about 3.2-3.9 mm long with about 4.6—5.4 whorls. It is ovate-conical in shape with a narrow con- ical spire. The shell is thick and opaque, and the aperture has a heavy callus ridge along the outer lip. In addition the middle of the outer lip is strongly arched forward as viewed in lateral profile (Thompson, 1977: Fig. 1 B, Fig. 6 B). The shell of P. hershleri is sim- ilar in shape to that of P. olivacea (Pilsbry, 1895). Pyrgulopsis olivacea is a much larger species, being 3.9-4.5 mm long; it has 5.0— 5.4 whorls, and the inner lip of the peri- stome is incomplete across the parietal mar- gin. It remains unknown anatomically, and it may be extinct (Thompson 1977:122- 124). Etymology.—I take pleasure in naming this species after Robert Hershler in rec- ognition of his numerous contributions to malacology. Acknowledgments This species was rediscovered during the course of an ecology study of the endangered viviparid snail, 7ulotoma magnifica, a pro- ject sponsored by the Alabama Power Com- pany (APC). Assistance in the field was pro- vided by Malcolm Pierson (APC) and Stev- en P. Christman (Florida Museum of Nat- ural History). The photograph of the Coosa River (Fig. 1) was provided by Malcolm Pierson (Alabama Power Company). The illustration comprising Fig. 2 was rendered by Barbara Harmon, Scientific Illustrator, Florida Museum of Natural History. I am grateful to Kurt Auffenberg and Elizabeth L. Raiser for assistance in the preparation of this paper. VOLUME 108, NUMBER 3 Literature Cited Goodrich, C. 1936. Goniobasis of the Coosa River, Alabama. — Miscellanious Publications Univer- sity of Michigan, Museum of Zoology 31:1-60. 1941. Distribution of the gastropods of the Cahaba River, Alabama.— Occasional Papers of the Museum of Zoology, University of Michigan 428:1-30. . 1944a. Certain operculates of the Coosa Riv- er.— Nautilus 58:1-15. . 1944b. Pleuroceridae of the Coosa River Ba- sin.— Nautilus 58:40-48. Hershler, R. 1994. A review of the North American freshwater snail genus Pyrgulopsis (Hydrobi- idae).—Smithsonian Contributions to Zoology 554:1-1v, 1-115. 507 —., & F.G. Thompson. 1987. North American Hydrobiidae (Gastropoda: Rissoacea): rede- scription and systematic relationships of 7ryon- ia Stimpson, 1865 and Pyrgulopsis Call and Pils- bry, 1886.—Nautilus 101:25-32. Pilsbry, H. A. 1895. New American freshwater mol- lusks.— Nautilus 8:114—-116. Thompson, F. G. 1968. The Aquatic Snails of the Family Hydrobiidae of Peninsular Florida. Uni- versity of Florida Press, Gainesville, Florida, 268 pp. . 1977. The hydrobiid snails of the genus Mar- stonia. — Bulletin of the Florida State Museum 21(3):113-158. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(3):508-527. 1995. The genus Laxus Cobb, 1894 (Stilbonematinae: Nematoda): Description of two new species with ectosymbiotic chemoautotrophic bacteria Jorg A. Ott, Monika Bauer-Nebelsick, and Veronica Novotny Institute of Zoology, University of Vienna, Althanstrasse 14, A-1090 Vienna, Austria Abstract.—Two new species of free-living marine nematodes belonging to the genus Laxus Cobb, 1894, L. oneistus and L. cosmopolitus, are described from shallow subtidal sands in the Belize Barrier Reef and the Adriatic Sea. Both species are covered by ectosymbiotic chemoautotrophic sulfur-oxidizing bacteria. The genus Laxus is redefined and Catanema cobbi Inglis, 1967 trans- ferred to this genus. For L. oneistus, biometrical data for juvenile stages are given. The Stilbonematinae are a subfamily within the family Desmodoridae (order Desmodorida), characterized by an obligate ectosymbiosis with sulfur-oxidizing chem- olithoautotrophic bacteria covering their cuticle in a manner that is characteristic for the genus, and even the species (Ott et al. 1991, Polz et al. 1992). The complex glan- dular sense organ described by Nebelsick et al. (1992) has so far been found only in the Stilbonematinae and is absent in species of other genera within the Desmodoridae, such as Acanthopharynx Marion, 1870, Desmo- dora DeMan, 1889 and Spirinia Gerlach, 1963 (M. Bauer-Nebelsick, pers. comm.). This is the only morphological autapo- morphic character so far known to indicate the monophyly of this taxon. Otherwise, only a few common features unite the group, most of them being negative characters, such as the lack of a buccal armature. The high variability regarding a number of taxonomically important characters, such as the amphid or the structure of cuticular reinforcement of the anterior end (“‘cephalic capsule,’ see Urbancik 1994) complicates assessment of the relationship of the Stil- bonematinae with other subfamilies of the desmodorids and between the various known genera. It is, therefore, difficult to decide whether some characters have de- veloped or have been modified (reduced) independently more than once. There is, for example, a tendency towards the develop- ment of an enlarged muscular portion at the anterior end of the pharynx, which in the genera Robbea Gerlach, 1956 and Catane- ma Cobb, 1920 is highly developed and found in all species, whereas in other genera (Leptonemella Cobb, 1920; Eubostrichus Greeff, 1869) it is only apparent in a few species. The reduction of the fovea of the amphid from the usual spiral shape to just a porus from which the corpus gelatum usu- ally protrudes seems to have occurred in- dependently at least three times (in the gen- era Leptonemella, Stilbonema Cobb, 1920 and Catanema). In addition, the type species of several genera (e.g., Laxus, Catanema, Leptone- mella, Eubostrichus) are either inadequately described or have features that are the ex- ception rather than the rule in the subse- quently described members of the genus. Liberal synonymization has contributed the rest of the problems. During studies of the ecology and eco- physiology of the symbiosis (Ott & Novak 1989, Schiemer et al. 1990, Ott et al. 1991) using material from the Caribbean and the VOLUME 108, NUMBER 3 Mediterranean Sea, a number of species new to science have been used and have been designated preliminary names. These ani- mals, designated “‘sp. 1” or “sp. 2,”’ need to be described and formally named. In ad- dition, collections have been made by the authors in various other locations. Two spe- cies from these collections belong—in our opinion—to the genus Laxus, which was de- scribed by Cobb (1894) from specimens coming from sand in the Bay of Naples (L. contortus) and Port Jackson, New South Wales (L. /ongus). Although subsequent au- thors (Gerlach 1963a, Wieser & Hopper 1967, Hopper & Cefalu 1973) have classi- fied this genus as doubtful, we have reasons to believe that we can assign our animals to Cobb’s genus and can attempt a redefinition of this taxon. Material and Methods Sediment was collected in buckets by hand and the animals extracted by shaking ali- quots of sediment in seawater and decanting the supernatant through a 63-um screen. In some cases a MgCl, solution isotonic to sea- water was used as an anaesthetic. Quanti- tative samples were taken with cores having an internal diameter of 3.5 cm down to a depth of 10 cm. For light microscopy the animals were fixed in 4% formaldehyde, transferred in glycerol:water 1:9 and slowly evaporated before mounting in pure glycerol on micro- scopic slides. Drawings and measurements were made on a Reichert Diavar, Reichert Polyvar or Leitz Diaplan, all equipped with a camera lucida. Nomarski interference contrast photos were made on the Reichert Polyvar. For scanning electron microscopy (SEM) specimens were fixed in 2.5% glutaralde- hyde in 0.1M sodium cacodylate buffer (pH 7.2) isotonic to sea water and postfixed in 2% OsO, over night. After dehydration in a graded ethanol series they were critical 509 point dried, coated with gold and examined with a JEOL JSM-35 CF. Semithin sections were cut on a Reichert Ultracut from specimens fixed for trans- mission electron microscopy (TEM) ac- cording to Eisenman & Alfert (1982) and embedded in Spurr epoxy resin, stained with toluidin blue and photographed on a Reich- ert Polyvar. Type material has been deposited in the Natural History Museum Vienna, Everte- brata Varia Collection (NHMW-EV). Laxus oneistus, new species Figs. 1-4, Table 1 Catanema sp. in Ott & Novak 1989, Schie- mer et al. 1990, Polz et al. 1992, Ott et al. 1991, Nebelsick et al. 1992. NHMW-EV Nr.3406 IL, S 8.22 nim W198 b= 77 C= 120 Allotype: female, NHMW-EV Nr.3407 L=8.40mm a=168 b=75 c= 130 NHMW-EV Nr.3408 Holotype: male, Paratypes: male, L=689 mm a= 156 b=58 c=99 male, NHMW-EV Nr.3409 L=9.94mm a=202 b=76 c=99 male, NHMW-EV Nr.3410 L=10.25mm a=195 b=78 c= 144 male, NHMW-EV Nr.3411 L=9.50mm a=181 b=81 c=140 female, NHMW-EV Nr.3412 L979 mmys a — 222 (b= 1/3 6¢— 122 female, NHMW-EV Nr.3413 L=8.9l1mm a=175 b=79 c=117 female, NHMW-EV Nr.3414 L=9.83 mm a= 200 b=77 c=145 Etymology. —from the greek oneistos, meaning “most useful,’”’ because of its value as an experimental animal during ecological and ecophysiological studies. Type locality.—Carrie Bow Cay, Belize, Caribbean Sea; coarse, poorly-sorted, cor- alline sand at the northern tip of the island; 0.2 to 0.5 m depth. Extremely slender worms, with a cylin- 510 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON eee ea 23 cs AO Ee HOO LY O ee ST ON, 4OF Ze ==> NSQES ZA9 wR RBSS s a Ne ie Se A \ AG Dy oY? IN ie Ke \ 5508 OP 88x rn a 50 100 150 200um | O 50 B-D /--——_+#@ -——__+4_____+——__+—_ igi Xe) 2 O sOum i fey, | . fb y/ bba Sy Y O65) 6) ae 3) Ae IST S157 91cronIO 6 een RON IT STOIOIS a Fig. 1. Laxus oneistus, male, holotype. A. Anterior body region, lateral view. B. Anterior end and pharyngeal region. C. Posterior end with spicular apparatus and tail containing caudal glands. D. Beginning of bacterial coat showing reduction of body diameter. E. Midbody region showing longitudinal muscles and two sublateral rows of glandular sensory organs. A-E, lateral views. Abbreviations:am, amphid; bac, ectosymbiotic bacteria; VOLUME 108, NUMBER 3 511 Fig. 2. Laxus oneistus, female, allotype. A. Total view. B. Anterior end and pharyngeal region. C. Tail. A-— C, lateral views. (For abbrevations see Fig. 1.) —_— bbac, begin of bacterial coat; bl, block layer; ca, cardia; cg, corpus gelatum; cgso, concentration of gso; cs, cephalic sensillae; cut, cuticula; de, ductus ejaculatorius: fa, fovea amphidialis; fp, fingerprint region; gr, granules of gso; gso, glandular sense organ; gu, gubernaculum; 1, intestine; m, longitudinal muscles; nr, nerve ring; ov, egg; ovr, Ovary; re, rectum; rs, receptaculum seminis; scs, subcephalic setae; sp, spiculum; ss, somatic setae; te, testis; ut, uterus; vu, vulva. lA PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Hhae yok pert “ bass _ a PAT Fig. 3. Laxus oneistus. A. Anterior end and pharyngeal region showing the block layer of the cephalic capsule, the tripartite pharynx and several glandular sensory organs. Light microphotograph of a semithin section stained with toluidin blue. B-E. Nomarski interference contrast microphotographs of whole-mounts in glycerol. B. Anterior end showing cephalic capsule with block layers and granules in glandular sense organs. C. Beginning of bacterial coat. D. Spicular apparatus. E. Surface view of bacterial coat. (For abbrevations see Fig. 1.) drical body. Body covered with radially ar- ranged, rod-shaped bacteria (2.1 x 0.6 wm) except for the anterior region (Figs. 1A, 2A, 3C, E, 4F, G) which appears clear. Bacteria- covered remainder of the body white in in- cident and dark in transmitted light. Bac- teria-free portion is 1010-1090 um (8-9 times the pharynx length) in males, but only 340-500 wm (3-3.6 times the pharynx length) in females (Fig. 2A). Anterior region containing pharynx slightly swollen (Figs. 1A, B, 2A, B), maximum body diameter at =~ Fig. 4. Laxus oneistus. SEM photographs. A. Anterior body region with bacterial coat removed showing rows of somatic setae. B. En-face view of anterior end showing the three circles of cephalic sensillae (6 + 6 + 4), 8 subcephalic setae, the first circle of 8 somatic setae and the amphids. The mouth opening protrudes slightly VOLUME 108, NUMBER 3 513 exposing the first circle of cephalic sensillae. C. En-face view of amphid with corpus gelatum surrounded by cuticular “fingerprint region.” D. Close-up of first circle of cephalic sensillae showing terminal pores. E. Tip of tail with openings of caudal glands. F. Transition between bacteria-free and bacteria-coated region. G. Rod- shaped bacteria on cuticle. (For abbrevations see Fig. 1.) 514 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Biometrical data for selected characters of the type specimens in Laxus oneistus (n = 10) and L. cosmopolitus (n = 8). All measurements are in um. Character L. oneistus L. cosmopolitus Diameter at end of cc. 52.5-66. 1 42.4-49.2 Mid-body diameter 44.1-52.5 49.2-55.9 Diameter at anus 50.1-61.0 42.4-56.1 Amphid width 13.6—-20.3 11.9-15.3 Amphid (cbd) 27.1-42.4 35.6—37.3 Nerve ring (dfae) 66.1-83.1 84.8-110.2 Nerve ring (cbd) 52.5-69.5 47.5-59.3 Pharynx length 112.0-134.0 141.0-168.0 Bulbus width 28.8—40.7 28.8-32.2 Bulbus (cbd) 52.5-67.8 47.5-54.2 Vulva (dfae) 4290.0-4450.0 3644—-5847.0 Vulva (% body length) 51.5-56.0 60.3-72.3 Spiculum length 67.8-89.8 67.8-84.8 Gubernaculum length 32.2-35.6 35.6-50.9 Tail length 64.4-79.7 84.8—113.0 dfae = distance from anterior end. cbd = corresponding body diameter. level of nerve ring. Body diameter is prac- tically constant from a short distance pos- terior to end of pharynx all the way to the anus. This midbody diameter has been used to calculate a. Tail is conical, 1.4—1.5 anal diameters long in both sexes, curved ven- trally in males (Fig. 1C), nearly straight in females (Fig. 2C). Cuticle faintly annulated (Figs. 1C, 4E, F, G); annuli 0.25 wm wide. Annulation in the head region irregular forming a char- acteristic “fingerprint” pattern (Fig. 4C). Cephalic capsule consisting of a special lay- er of blocks of a homogeneous material, giv- ing the anterior end a reticulated appearance if one focusses slightly below the surface (Figs. 1B, 2B, 3A, B). Tip of tail free of annulations and caudal glands open ter- minally through a complicated pore com- plex (Fig. 4E, see also Nebelsick et al. 1992: fig. 2). Cuticle 3 wm thick in bacteria-free body region. Cuticle thins at the beginning of bacterial coat and body diameter be- comes smaller to an extent that the bacterial coat does not increase the overall thickness of the animal (Figs. 1D, 4F). Eight rows (four sublateral, four submedian) of somatic setae (5 um long) in anterior body region (Figs. 1B, 2B) are outlets of glandular sense organs (gso, Nebelsick et al. 1992) which lie in double rows in lateral (Fig. 1E), dorsal, and ventral positions. Posterior to end of pharynx ventral and dorsal double rows of gso merge gradually into a single row each. Consequently only six rows of setae (four sublateral, one mediodorsal and one me- dioventral) and gso are present over most of the body. Near posterior end the sub- median rows become distinct from each other again (Fig. 1C). Cephalic sensillae consist of a circle of six blunt papillae with apical openings surrounding the mouth opening (Fig. 4B, C), often appearing re- tracted and not visible from the outside ex- cept in en-face view (Fig. 4D). Second circle formed by six short setae (2.2 wm) with swollen tips, the third by four long, conical setae (32-35 um). A circle of eight subce- phalic setae, 10-13 um long, apparently not connected to glandular sensory organs is flanked by first circle of regular somatic se- tae. Two or three more circles of eight setae VOLUME 108, NUMBER 3 Table 2.— Laxus oneistus. Distribution of males, fe- males and juveniles in 10 cores. Core # Males Females Juveniles 1 10 17 6 2 19 24 3 3 1 2 1 4 5 4 4 5 21) 26 30 6 6 i 5 7 38 28 6 8 7 6 y) 9 1 7 19 10 — 1 4 Total 114 122 86 situated on the reinforced cuticle of the head. Amphids close to the anterior end consist- ing of a simple ventrally wound spiral. Buccal cavity small and tubular, tri-par- tite pharynx consisting of a distinctly swol- len corpus occupying almost the anterior half, twice as wide as the isthmus, and a round terminal bulb leading into intestine without cardia (Figs. 1B, 2B, 3A). Isthmus and bulb occupy approximately 30 and 24% of pharynx length, respectively. Nerve ring encircling isthmus approximately at two- thirds of the length of the pharynx. No ven- tral gland or excretory pore seen. Single testis in males at 39-46% of body length. Spicula cephalate proximally, arcu- ate, with a velum (Figs. 1C, 3D), about one anal diameter long (chord) or up to 1.3 anal diameters (arch). Gubernaculum simple, with a dorsally directed apophysis. Ovaries paired, symmetrical, antidromous reflexed, uteri containing extremely long (up to 620 um) ova. Vulva at 51-56% of body length. Diagnosis. —Characters of the genus; cor- pus occupying anterior 46% of the pharynx, twice as wide as the isthmus, bulbus 24% of pharynx length; subcephalic setae ca. 40% of the length of the cephalic setae; annula- tion very faint; coat of bacterial symbionts begins at a distinct level several hundred microns behind the anterior end. Biometry of developmental stages. —Ten Table 3.—Laxus oneistus. Biometrical data for juvenile stages (j) 1 + 2, 3 and 4, males, females and pooled adults from the cores (only 61 adults included). Length SD Range SD Mean SD Range Mean SD Range Mean Range Mean Stage 13.6-44.3 37.2-71.0 6.9 29.8 13.9-35.6 26.6-49.3 44.1-64.4 42.4-75.9 a7 6.6 2.1 29.8-88.0 2 13.1 SEZ 1.08-2.92 3.05-4.91 5.09-6.86 5.36-8.71 0.55 0.55 0.50 0.90 1.06 1.02 1.93 3715 5.81 39 Za1] fil te oF 16.8 53.6 37.1 68.0-135.0 15.8 86.8 124.0 134.3 130.9 132.0 46.8-109.4 75.8-125.3 56.0-132.8 8.6 7 93.0 Doi 9.3 54.4 93.7-154.1 101.6-161.1 16.2 18.1 20 j4 12.8 61.7 6.86 TEAS: 7.08 21 Males 19.2 94.0 36.4-99.0 4.2 Deed] 1 1 91.9-176.4 66.6 91.9-176.4 5.18-9.36 222 5.18-9.36 Females Adults 56.0-132.8 17.2 9357 36.4-99.0 64.8 20.9 61 Ss 516 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON O oe ee ee es 1000 2000 3000 400 Fig. 5. juvenile stage 5000 6000 7000 8000 9000 length (uum) Laxus oneistus. Size-frequency plot of juveniles (shaded) and adults (white). Separation of juvenile stages 1/2, 3 and 4 by fitting normal distributions using Bhattacharaya’s method (Elefan). quantitative cores taken in 1991 contained a total of 322 specimens, 236 of which were adults (114 males, 122 females) (Table 2). The 86 juveniles ranged in size from 1.1 mm to 6.9 mm (Table 3). A size frequency plot (Fig. 5) revealed four modes among the juveniles. Bhattacharaya’s method for sep- arating frequency distributions, however, gave only three normal distributions. We interpret these as the pooled stages | and 2, stage 3 and stage 4 juveniles. The stage 4 juveniles length distribution strongly over- laps with that of the adults. The statistical method used was unable to separate stage 4 juveniles by length from the adults, which were identified by the presence of fully de- veloped gonads and treated separately. Plotting a (body length divided by max- imum body width) against length (Fig. 6) shows that juveniles become progressively more slender during growth. With the adult molt the worms become stouter, especially the females due to the development of the gonads. The midbody region elongates more strongly during growth than either the phar- ynx and the tail as is evident from plots of b (body length divided by pharynx length) or c (body length divided by tail length) against body length (Figs. 7, 8). The index c shows little correlation to body size in adult males probably due to different de- grees of curvature of the tail. Remarks. —Laxus oneistus is extremely common in subtidal coarse sands in the Be- lize Barrier Reef system, from just below the waterline to at least 7 m depth. It in- habits the deeper, microxic to anoxic sedi- ment layers. By migration through the chemocline it exposes its symbiontic bac- teria alternately to sulfide and oxygen. The VOLUME 108, NUMBER 3 517 200 150 100 + juveniles 50 : females males 0 Lani 1. ia pie ee Ic Ce 1 2 3 4 5 6 i 8 9 length (mm) Fig.6. Laxus oneistus. Biometry of development: Relationship between body length and ratio a (body length/ maximum body width). Linear regression. chemoautotrophic bacteria constitute prob- ably the main bulk of the food for these worms (for details of the biology see Ott and Novak, 1989 and Ott et al., 1991). When extracted from the sediment it tends to curl up its body except for the anteriormost part (Fig. 2A). It is very thigmotactic and in dish- es forms tightly knotted aggregations with conspecifics or other long nematodes from which only the anterior parts protrude ex- ecuting elegant undulating movements. Laxus cosmopolitus, new species Synonym. — “undescribed genus from the Adriatic Sea’”’ in Ott et al. 1991, “‘unde- scribed genus from the Mediterranean” in Polz et al. 1992, Eubostrichus exilis (Cobb, 1920) in Gerlach 1963a, probably E. exilis (Cobb, 1920) in Gerlach 1964 Holotype: male, NHMW-EV Nr.3415 L=6.53 mm a= 133 b= 46 c=59 Allotype: female, NHMW-EV Nr.3416 L=7.71 mm a= 138 b=45 c=76 Paratypes: male, NHMW-EV Nr.3417 L=6.01 mm a=115 b=42 c=57 male, NHMW-EV Nr.3418 L= 6.99 mm a= 128 5=45 c= 83 female, NHMW-EV Nr.3419 L=7.29 mm a= 148 b= 48 c = 66 female, NHMW-EV Nr.3420 L=6.22mm a= 122 b=42 c=73 female, NHMW-EV Nr.3421 518 100 80 60 40 20 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON juveniles v females QO males 0 ij oa: fe ee es Tee Oe rs 1 Z 3 4 5 6 i 8 9 length (mm) Fig. 7. Laxus oneistus. Biometry of development: Relationship between body length and ratio b (body length/ pharynx length). Linear regression. L=7.97 mm a female, L = 9.07 mm a= G45) — 40 5G — 3 8S NHMW-EV Nr.3422 L781 — Ole —-80 Etymology. —from the greek kosmopol- ites, meaning “‘citizen of the world,” be- cause of its wide distribution. Type locality.—Bay of Vestar south of Rovinj, Croatia, Adriatic Sea, Mediterra- nean; moderately well sorted coarse calcar- eous sand in 3—4 m depth. Cylindrical, long, robust worms (Figs. 9A, 10A), body completely covered by radially arranged rod shaped bacteria (1.8 CTE AS > Fig. 9. A-—C. Laxus cosmopolitus, male, holotype. A. Total view showing concentrations of gso ventrally in postpharyngeal and preanal region. B. Anterior end and pharyngeal region, showing cephalic capsule and pharynx. C. Posterior end with spicular apparatus and tail with caudal glands. D. Male, paratype. Spicular apparatus. A— D, lateral views. (For abbrevations see Fig. 1.) VOLUME 108, NUMBER 3 521 iO Gl A 3 oe | 6) Fig. 10. Laxus cosmopolitus, femaie, allotype. A. Total view. B. Anterior end and pharyngeal region, optical section. C. Head, surface view showing amphid and honeycomb pattern of block layer. D. Vulvar region and part of anterior ovary. E. Tail with caudal glands. A—E, lateral views. (For abbrevations see Fig. 1.) 522 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 11. Laxus cosmopolitus, Nomarski interference contrast microphotographs. A. Anterior body region showing concentration of gso. B. Anterior end and pharyngeal region. C. Head showing amphid, block layer of cephalic capsule and beginning of bacterial coat. D. Posterior end of cephalic capsule, annulated cuticle and microbial coat. E. Optical section through midgut region showing gut filled with bacteria similar to the ecto- symbiotic bacteria on the cuticle. F. Posterior end of female showing rectum, anus and caudal glands. (For abbrevations see Fig. 1.) VOLUME 108, NUMBER 3 the Maldives (3.98-6.6 mm, ¥ = 4.95 + 0.85 mm, 7 = 8) (Kruskal-Wallis one-way analysis by ranks, P = 0.005, pairwise test of significance of differences after Conover 1980). The tail is strongly curved ventrally in males from Rovinj and regularely coni- cal, in the Red Sea specimens it is almost straight, conical with a distinctly set-off cy- lindrical end part, whereas in the Maldives the tail is almost cylindrical over most of its length and narrows abruptly only near the tip. The two rows of subventral setae on the male tail are slender in the Rovinj spec- imens, stouter in those from the Red Sea and triangular thorns in males from the Maldives. The spicula in the Maldive males are more distinctly cephalate, less curved and have only a small velum (Fig. 13). Ad- ditional specimens provided by G. Boucher (Paris) from New Caledonia resemble the Red Sea animals. Diagnosis. —Characters of the genus; cor- pus occupying 50% of the length of the phar- ynx, being 30-50% wider than the isthmus (34% of pharynx length), bulbus small (16%); subcephalic setae ca. 30% of length of ce- phalic setae; annulation relatively coarse (1.3 um); concentration of glandular sense or- gans ventrally, in males in the postphar- yngeal and praecloacal region, in females in the pre- and post-vulvar region; bacterial coat extends into region of cephalic capsule. Remarks. — Laxus cosmopolitus is regu- larely found in coarse subtidal sand at the type locality and in several locations in the vicinity of Rovinj (Bay of Cisterna, Punta Croce). It is, however, never abundant. Both the Red Sea and the Maldive Island mate- rial comes from shallow subtidal coralline sand. The association of the worms with sulfidic sediments and the similarity be- tween the bacteria found in the gut and on the cuticle together with their ultrastructure suggest a similar biology as in L. oneistus. Discussion The genus Laxus was proposed by Cobb in 1894 for two nematodes collected from 523 such different regions of the world as the Gulf of Naples and the Australian coast. Although the description of the type species, L. contortus, 1s on the basis of a female only and is not accompanied by a figure, it is sufficiently precise to warrant the assump- tion, that it is congeneric with the second species, L. longus. In fact, Cobb himself des- ignated the latter as the genotype in Stiles & Hassal (1905) (see also Baylis & Daubney 1926). This allows an identification on the genus level with our species described above. In particular, the genus Laxus is recognized by the special structure of the tripartite pharynx with its swollen corpus, which is not as clearly set off as in the genera Ca- tanema Cobb, 1920 or Robbea Gerlach, 1956; the lack of cervical papillae (which distinguishes it from Robbea); and the spic- ular apparatus with a large cephalate spic- ulum having a velum in its distal half and a simple, slightly curved gubernaculum with a straight, dorsally directed apophysis (which distinguishes it from Catanema). Addition- al characters are the four long setae of the third circle of cephalic sensillae, which are flanked by shorter setae (“two setae of un- equal size placed on each submedian line,” Cobb 1894:415) and the presence of addi- tional setae adding up to the eight elongated setae that constitute the first circle of so- matic setae; the apparent lack of an excre- tory system; the shape and position of the amphid; the shape of the tail and the prom- inent rows of subventral setae pre- and post- anally. Although none of the forgoing char- acters is unique for the genus, the combi- nation is. Cobb’s fig. 11/II also indicates the reinforcement of the cephalic cuticle by the block layer. Laxus septentrionalis, described by Cobb in 1914 from Antarctica, however, does cer- tainly not belong to this genus. Neither does the conical head nor the amphid depicted in the description resemble the foregoing species. The pharynx (“oesophagus” in Cobb’s description) is explicitely described as ‘“‘cylindroid,”’ there is a distinct excretory 524 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 12. Laxus cosmopolitus, SEM microphotographs. A. En-face view of anterior end showing fingerprint region, amphids, mouth opening surrounded by second circle of six cephalic sensillae (double arrow), third circle of four cs (triple arrow) and subcephalic setae. Note that bacterial coat reaches the fingerprint region. B. Close- up of amphid and mouth opening with first (single arrow), second (double arrow) and third circle (triple arrow) of cs. C. Posterior end of female showing rows of somatic setae. The disruption of the bacterial coat is an artefact VOLUME 108, NUMBER 3 525 Fig. 13. Laxus cosmopolitus, Nomarski interference contrast microphotographs. Variations in the shape of the tail, postanal setae and spicular apparatus between male specimens from the Mediterranean Sea (A, D), the Red Sea (B, E) and the Maldive Islands (C, F). (For abbrevations see Fig. 1.) system, and the gubernaculum is described to have an apophysis perpendicular to “‘the part which is applied to the spicula.”’ In contrast, Catanema cobbi Inglis, 1967 certainly has to be transferred to the genus Laxus on the basis of the special structure of the pharynx and the distinct block layer reinforcing the head cuticle. Inglis also men- tiones that the annulation extends over the region of the block layer. It is very similar _— to L. longus because of the length of the first circle of eight setae and the relative pro- portions of the three regions of the pharynx, but differs in the number of postcloacal sub- ventral setae (or “‘tube-like organs’’). In the present state of knowledge and in the ab- sence of type specimens it is better to refrain from synonymizing these two species. The genus Laxus may be redefined as fol- lows: Desmodoridae, Stilbonematinae, cu- of preparation. D. Close-up of bacterial coat in the midbody region showing closely packed rods standing perpendicular to the annulated surface of the cuticle. (For abbrevations see Fig. 1.) 526 ticle finely annulated, with “fingerprint” pattern on the anterior end. In this region a cephalic capsule is developed by a special ‘‘block-layer’’ inserted between the median and basal zone of the cuticle; pharynx tri- partite, with a swollen corpus that is not distinctly set off from the isthmus; spicula curved, cephalate; gubernaculum without apophysis, straight to slightly S-shaped. Symbionts are rod-shaped bacteria ar- ranged in a single layer in which the lon- gitudinal axis of the microorganisms is per- pendicular to the cuticle surface. The following species belong to the genus: Laxus longus Cobb, 1894 (type species), Laxus contortus Cobb, 1894, Laxus cobbi (Inglis, 1967) (syn. Catanema cobbi Inglis, 1967), Laxus oneistus new species., Laxus cosmopolitus new species. It is peculiar, that neither Cobb nor In- glis—both being keen observers— mention the bacterial ectosymbiosis. Most of the stil- bonematids that have been described pre- viously and practically all specimens in our collection (comprising all genera so far de- scribed) show the conspicuous and species- specific coat of microorganisms. Also, in Cobb’s descriptions of the type species of the genera Sti/lbonema, Leptonemella, Ca- tanema and Laxonema (Cobb 1920), no mention 1s made of the ectosymbionts. Es- pecially in Sti/bonema the multilayered bac- terial coat cannot be overlooked. Since no type material exists, we cannot be sure whether the bacterial ectosymbiosis is ab- sent in these species (which seems improb- able), whether the bacteria had been lost during preparation, or whether they were deliberately ignored because the authors be- lieved them to be contaminants. The small proportion of juveniles in the collection is consistent with other indica- tions of a slow development, low reproduc- tion rate and longevity of adults in L. oneis- tus. Both juveniles and adults could be kept alive for up to three weeks in dishes during which time neither molting or egg-laying was observed. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON There is a notable difference between the mean body length of the type specimens and the adults from the quantitative cores, the latter being significantly smaller. This could be an indication of seasonal or year to year variation in size (the type specimens were collected in May 1989, the cores were taken in February 1991). Another explanation could be an unconscious bias when choosing type specimens from qualitative collections where larger “‘representative” specimens may be preferentially selected. Therefore, size differences between haphazardly se- lected material and type specimens should not be overrated. Acknowledgments We thank Werner Urbancik, Rudolf No- vak and Martin Polz for providing several SEM pictures and Hubert Keckeis for help with the biometrical analysis. This study was supported by grants #7814 BIO and #9189 BIO of the FFWF (Austria) and grants from the CCRE project of the National Mu- seum of Natural History, Washington, D.C. The generous help of Klaus Ruetzler and the staff of the Carrie Bow Cay Laboratory is gratefully acknowledged. Collections in the Adriatic were made possible through the help of the Center for Marine Research, Rovinj, Croatia. This is contribution #438 from the Carrie Bow Cay Laboratory (CCRE). Literature Cited Baylis, H. A., & R. Daubney. 1926. A synopsis of the families and genera of Nematoda.— London (British Museum) 1-277. . Cobb, N. A. 1894. Tricoma and other new nematode genera. — Proceedings of the Linnean Society of New South Wales 8(2):389-421. . 1914. The North American free-living fresh- water nematodes. — Transactions of the Amer- ican microscopical Society. 33:69-119 (reprint- ed in Contributions to a Science of Nematology (Baltimore) 2:35-99). . 1920. One hundred new nemas (type species of 100 new genera). — Contributions to a Science of Nematology (Baltimore) 9:217-343. VOLUME 108, NUMBER 3 Conover, W. J. 1980. Practical nonparametric sta- tistics. — 2nd edition John Wiley, New York, 494 pp. Eisenman, E. A., & M. Alfert. 1982. A new fixation procedure for preserving the ultrastructure of marine invertebrate tissues.—Journal of Mi- croscopy 125:117—120. Gerlach, S. A. 1956. Die Nematodenbesiedlung des tropischen Brandungsstrandes von Pernambu- co. Brasilianische Meeres-Nematoden II.— Kie- ler Meeresforschung 12:202-218. . 1963. Freilebende Meeresnematoden von den Malediven II.— Kieler Meeresforschung 18:8 1— 108. 1964. Freilebende Nematoden aus dem Ro- ten Meer.—Kieler Meeresforschung 20 (Son- derheft): 1 8-34. Hopper, B. E., & R. C. Cefalu. 1973. Free-living ma- rine nematodes from Biscayne Bay, Florida v. Stilbonematinae; Contribution to the taxonomy and morphology of the genus Eubostrichus Greef and related genera. — Transactions of the Amer- ican Microscopical Society 92(4):578-591. Inglis, W. G. 1967. Interstitial Nematodes from St. Vincent’s Bay, New Caledonia.—Expéditions francaise sur les recifs coralliens de la Nouvelle Calédonie, Paris, (Editions de la Foundation Signer Polignac.) 2:29-74. De Man, J. G. 1889. Espéces et genres nouveaux ou peu connus.— Mémoires de la Societe Zoolo- gique de France 2:1-10. Marion, A. F. 1870. Recherches zoologiques et an- atomiques sur des Nématodes non parasites, marins.—Annales des Sciences Natureiles 13(14):1-100. Nebelsick, M., M. Blumer, R. Novak, & J. Ott. 1992. A new glandular sensory organ in Catanema sp. SI (Nematoda, Stilbonematinae).— Zoomorphol- ogy 112:17-26. Ott, J., & R. Novak. 1989. Living at an Interface: Meiofauna at the oxygen/sulfide boundary of marine sediments. pp. 415—422 in J. S. Ryland & P. A. Tyler, eds., Reproduction, genetics and distribution of marine organisms. Olsen & Ol- sen, Fredensborg, Danmark. , , F. Schiemer, U. Hentschel, M. Ne- belsick, & M. Polz. 1991. Tackling the sulfide gradient: A novel strategy involving marine nematodes and chemoautotrophic ectosym- bionts.—P.S.Z.N.I: Marine Ecology 12(3):261- 279. Polz, M., H. Felbeck, R. Novak, M. Nebelsick, & J. Ott. 1992. Chemoautotrophic, sulfur-oxidiz- ing symbiotic bacteria on marine nematodes: morphological and biochemical characteriza- tion.— Microbial Ecology 24:313-329. Schiemer, F., R. Novak, & J. Ott. 1990. Metabolic studies on thiobiotic free-living nematodes and their symbiotic microorganisms.— Marine Bi- ology 106:129-137. Stiles, C. W., & A. Hassal. 1905. The determination of generic types and a list of roundworm genera, with their original and type species.— Bulletin of the Bureau of Animal Industries of the United States Department for Agriculture 79:1-150. Urbancik, W. 1994. The ultrastructure of the body- and head cuticle of Stilbonematinae (Nematoda, Desmodoridae). Unpublished M.Sc. Thesis, University of Vienna, 78 pp. Wieser, W., & B. Hopper. 1967. Marine Nematodes of the east coast of North America. I. Florida. — Bulletin of the Museum of Comparative Zool- ogy, Harvard 135:239-344. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(3):528-532. 1995. Neotropical Monogenoidea. 26. Annulotrematoides amazonicus, a new genus and species (Dactylogyridae: Ancyrocephalinae), from the gills of Psectrogaster rutiloides (Kner) (Teleostei: Characiformes: Curimatidae) from the Brazilian Amazon Delane C. Kritsky and Walter A. Boeger (DCK) College of Health Professions, Idaho State University, Pocatello, Idaho 83209, U.S.A.; (WAB) Departamento de Zoologia, Universidade Federal do Parana, Caixa Postal 19020, Curitiba, Parana 81531-970, Brazil, and Conselho Nacional de Desenvolvimento Cientifico e Tecnologico (CNPq) Abstract. —Annulotrematoides, a new genus, is proposed to include species with the following characteristics: presence of a longitudinal ventral groove on the posterior trunk; incomplete tegumental annulations; a vaginal aperture on the left margin of the trunk; overlapping gonads; dorsal and ventral anchor- bar complexes; 14 hooks with ancyrocephaline distribution; ventral bar with anteromedial process; and articulated accessory piece and male copulatory organ. The first two characters are unique to the new genus; Annulotrematoides is monotypic. Annulotrematoides amazonicus, new species, is described from the gills of Psectrogaster rutiloides (Kner) from Furo do Catalao near Manaus, Amazonas, Brazil. Although phylogenetic analyses are want- ing, Current evidence suggests that some Monogenoidea from Neotropical fishes have ancient evolutionary links to those of Af- rica. Gusev (1976a) considered Jainus Miz- elle, Kritsky & Crane, 1968, from Neotrop- ical Characiformes to be a senior synonym of the Ethiopian Characidotrema Paperna & Thurston, 1968, and used this synonymy In part to support a postulated relationship of the two biogeographical faunas. While Kritsky et al. (1987) did not recognize the synonymy, they indicated that the two gen- era may be sister taxa based on common morphological features of the haptoral ar- mament. Kritsky & Kulo (1992) suggested that a Neotropical-Ethiopian link might also be supported by relationships of species of the Neotropical Trinigyrus Hanek, Molnar & Fernando, 1974, the marine Hamatope- duncularia Yamaguti, 1953 and Chauha- nellus Bychowsky & Nagibina, 1969, and the Ethiopian Schilbetrema Paperna & Thurston, 1968, all parasites of siluriform fishes. In the present paper a new species of Annulotrematoides, new genus, is figured and described from Neotropical Characiformes. This species appears to have an evolution- ary relationship with species currently in- cluded in Annulotrema Paperna & Thurs- ton, 1968, from Ethiopian characiforms. Materials and Methods Hosts, Psectrogaster rutiloides (Kner), were collected by seine from the Furo do Catalao (Amazon River System) near Ma- naus, Amazonas, Brazil (5 January 1989). Methods of parasite collection from the hosts’ gills, preparation of the helminths for study, measurement, and numbering of hook pairs are those of Kritsky et al. (1986). Ter- minology is that of Mizelle & Kritsky (1967) and Kritsky & Mizelle (1968). Measure- VOLUME 108, NUMBER 3 ments, in wm, include the average followed by the range and number (7) of structures measured in parentheses. Type specimens are deposited in the helminthological col- lections of the Instituto Oswaldo Cruz, Rio de Janeiro, Brazil (IOC), the United States National Museum, Beltsville, Maryland (USNPO), the University of Nebraska State Museum, Lincoln, Nebraska (HWML). Class Monogenoidea Bychowsky, 1937 Order Dactylogyridea Bychowsky, 1937 Dactylogyridae Bychowsky, 1933 Ancyrocephalinae Bychowsky, 1937 Annulotrematoides, new genus Diagnosis. —Body comprising cephalic region, trunk, peduncle, haptor; longitudi- nal ventral groove between ends of incom- plete tegumental annulations on posterior trunk. Tegument thin. Cephalic lobes, head organs, cephalic glands present. Eyes pres- ent. Mouth ventral; pharynx muscular, glandular; esophagus short to nonexistent; intestinal ceca 2, confluent posterior to go- nads, lacking diverticula. Genital pore mid- ventral. Gonads intercecal, overlapping; testis dorsal to germarium. Seminal vesicle a dilation of vas deferens; 2 prostatic res- ervoirs; copulatory complex comprising tu- bular copulatory organ, accessory piece; ac- cessory piece comprising proximal articu- lation process, distal rod. Oviduct short; uterus delicate; vaginal aperture on left mar- gin of trunk; seminal receptacle a dilation of vaginal duct. Vitellaria present in trunk, absent in regions of reproductive organs. Haptor armed with ventral, dorsal anchor/ bar complexes; 14 hooks with ancyroce- phaline distribution (Mizelle 1936, see Miz- elle & Price 1963). Anchors simple; ventral bar with anteromedial projection; hooks similar, each with erect thumb, shank com- prising 2 subunits. Parasites of gills of Neo- tropical curimatid fishes. Type species, host, and locality. —Annu- lotrematoides amazonicus, new species, 529 from gills of Psectrogaster rutiloides, Furo do Catalao, Manaus, Amazonas, Brazil. Etymology.—The generic name reflects similarity of the genus to Annulotrema. Annulotrematoides amazonicus, new species Figs. 1—7 Description (based on 46 specimens).— Body 354 (274-438; n = 32) long, fusiform; greatest width 86 (73-117; n = 33) usually in posterior trunk or near midlength. Teg- umental annulations unscaled, each termi- nating at margin of longitudinal ventral groove. Cephalic margin broad; 2 indistinct terminal, 2 bilateral cephalic lobes. Eyes 4, equidistant; eye granules subspherical to su- bovate, variable in size; accessory granules absent. Pharynx spherical, 19 (16-22; n = 33) in diameter. Peduncle narrow; haptor subhexagonal, 54 (42-64; n = 32) long, 71 (62-85; n = 31) wide. Ventral anchor 42 (38-45; n = 10) long, with broad well-dif- ferentiated roots, evenly curved shaft, point; superficial root longer than deep root; point tip recurved; base width 17 (16-19; n = 10). Dorsal anchor 27 (21-30; n = 10) long, with elongate superficial root, short broad deep root, curved shaft, recurved point; base width 17 (15-18; n = 10). Ventral bar 28 (25-30; n = 27) long, platelike, with slightly enlarged ends, elongate anteromedial pro- jection; dorsal bar 31 (28-35; n = 25) long, slender, with medial bend. Each hook with erect thumb, delicate point; proximal sub- unit of shank expanded; hook pr. 1—18 (15- 21; n= 4); pr. 2—20 (17-23; n = 4); pr. 3— 21 (19-23; n = 7); pr. 4—22 (20-25; n = 5); pr. S—16-17 (n = 2); pr. 6—24 (23-26; n = 4); pr. 7—23 (22-26; n = 6) long; fila- mentous hooklet (FH) loop extending to just short of union of shank subunits. Male cop- ulatory organ 41 (34—49; n = 10) long, an arcuate tube with slightly flared termina- tion; base with elongate variable proximal flange. Rod of accessory piece 30 (26-32; n = 10) long, acute terminally, with variable 530 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 6 i Figs. 1-7. Annulotrematoides amazonicus, new species: 1, Composite drawing of whole mount (ventral view; hook pairs numbered respectively); 2, Hook (of pair 3); 3, Copulatory complex (dorsal view); 4, Ventral bar; 5, Dorsal bar; 6, Ventral anchor; 7, Dorsal anchor. Abbreviations: mco, male copulatory organ; a, distal bar of accessory piece; b, proximal articulation process; c, spatulate arm of accessory piece. All figures are drawn to the 25 um scale except Fig. 1 (100 wm). VOLUME 108, NUMBER 3 flattened projection near midlength; spat- ulate arm arising from distal end of short articulation process located between base of copulatory organ and rod of accessory piece. Margins of testis undefined; seminal vesicle small; prostatic reservoirs saccate. Germar- ium elongate, 90 (68-113; m = 25) long, 27 (20-35; n = 23) wide; ootype not observed; vagina pyriform, with thick proximal wall, opening into large central seminal recepta- ele: Specimens studied. —Holotype, IOC 33664; 45 paratypes, USNPC 84822, HWML 38340. Etymology. —The specific name reflects the Brazilian state from which the species was collected. Discussion Gusev’s (1976a, 1976b, 1978) thesis that the monogenoidean faunas of the Neotrop- ical and Ethiopian biogeographical regions have ancient evolutionary relationships may be supported by the discovery of Annulo- trematoides amazonicus. This species pos- sesses characters and host preferences sug- gesting that it may share acommon ancestor with species of the African Annulotrema. Both genera are characterized by species with tegumental annulations, and while restrict- ed to their respective biogeographical regions, species of Annulotrema and An- nulotrematoides are exclusively parasites of characiform fishes. With the exception of Paperna’s (1979) drawings, description of the internal anat- omy has been lacking in studies dealing with Annulotrema (see Guégan et al. 1988, Er- gens 1988, Paperna 1973). However, Pap- erna’s (1979) depictions of the internal anat- omy are not clear regarding the dorsoventral orientation of the internal organs. Exami- nation of four specimens of two unidentified Annulotrema species in the collection of the senior author and collected from the gills of Alestes cf. nurse (Rtippell) in Togo indicated anatomical features that serve to separate 53.1 Annulotrematoides from Annulotrema. A\l- though Paperna (1979; 110) states that the vagina opens on the left in all African An- nulotrema species, three of the above spec- imens clearly show the vaginal aperture on the right margin (vagina opens on left mar- gin in Annulotrematoides amazonicus). The tegumental annulations in Annulotrema are complete ventrally, and the gonads are slightly overlapping (testis dorsoposterior to germarium) while in Annulotrematoides amazonicus the testis is dorsal to the ger- marium. Within the Neotropics, Annulotrema- toides amazonicus is probably most closely related to the complex of Ancyrocephalinae reported by Boeger & Kritsky (1988) from the gills of the red-breasted piranha, Pygo- centrus nattereri (Characiformes). Species in the complex comprise Amphithecium Boeger & Kritsky, 1988, Notothecium Boe- ger & Kritsky, 1988, and Notozothecium Boeger & Kritsky, 1988, and possess com- parable internal anatomy, haptoral organi- zation, and structure of the copulatory com- plex to those of A. amazonicus. In addition, some species in all of these genera have an annulated tegument, which 1s usually scaled in those from piranha. The copulatory com- plexes in Annulotrematoides amazonicus and the piranha’s species include a distal rod in the accessory piece, a possible syn- apomorphy for the four genera. Acknowledgment The authors are grateful to Michel Jegu, Convention ORSTOM/CNPg, Instituto Nacional de Pesquisas da Amazonia, Ma- naus, Amazonas, Brazil, for identifying the hosts. Literature Cited Boeger, W. A., & D.C. Kritsky. 1988. Neotropical Monogenea. 12. Dactylogyridae from Serras- almus nattereri (Cypriniformes, Serrasalmidae) and aspects of their morphologic variation and distribution in the Brazilian Amazon.—Pro- 532 ceedings of the Helminthological Society of Washington 55:188-213. Ergens, R. 1988. Four species of the genus Annulo- trema Paperna et Thurston, 1969 (Monogenea: Ancyrocephalinae) from Egyptian freshwater fish.— Folia Parasitologica 35:209-215. Guégan, J.-F., A. Lambert, & E. Birgi. 1988. Obser- vations sur le parasitisme branchial des Char- acidae du genre Hydrocynus en Afrique de Ouest. Description d’Annulotrema pikoides n. sp. (Monogenea, Ancyrocephalidae) chez Hy- drocynus vittatus (Castelnau, 1861).—Annales de Parasitologie Humaine et Comparée 63:9 1- 98. Gusev, A. V. 1976a. Systematics, composition of the Indian fauna, zoogeography and evolution of Monogenoidea from freshwater fishes. — Trudy Biologo-Pochvennogo Instituta, Novaya Seriya 35:5-32. 1976b. Freshwater Indian Monogenoidea, principles of systematics, analysis of the world faunas and their evolution.—Indian Journal of Helminthology 25 & 26:1-241. 1978. Monogenoidea of freshwater fishes. Principles of systematics, analysis of the world faunas and their evolution. — Parazitologiches- kii Sbornik 28:96-198. Kritsky, D. C., & S.-D. Kulo. 1992. A revision of Schilbetrema (Monogenoidea: Dactylogyridae), with descriptions of four new species from Af- rican Schilbeidae (Siluriformes).— Transactions of the American Microscopical Society 111:278— 301. , & W.A. Boeger. 1987. Resurrection PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON of Characidotrema Paperna and Thurston, 1968 (Monogenea: Dactylogyridae) with description of two new species from Togo, Africa.—Pro- ceedings of the Helminthological Society of Washington 54:175-184. —, & J. D. Mizelle. 1968. Studies on monoge- netic trematodes. XXXV. Some new and pre- viously described North American species of Gyrodactylus.—American Midland Naturalist 79:205-215. , V. E. Thatcher, & W. A. Boeger. 1986. Neo- tropical Monogenea. 8. Revision of Uroclei- doides (Dactylogyridae, Ancyrocephalinae).— Proceedings of the Helminthological Society of Washington 53:1-37. Mizelle, J.D. 1936. New species of trematodes from the gills of Illinois fishes.—American Midland Naturalist 17:785-806. —, & D.C. Kritsky. 1967. Studies on monoge- netic trematodes. X XX. Five new species of Gy- rodactylus from the Pacific tomcod, Microgadus proximus (Girard). — Journal of Parasitology 53: 263-269. —.,&C.E. Price. 1963. Additional haptoral hooks in the genus Dactylogyrus.—Journal of Parasi- tology 49:1028-1029. Paperna,I. 1973. New species of Monogenea (Vermes) from African freshwater fish. A preliminary re- port. — Revue de Zoologique et de Botanique Af- ricaines 87:505-518. 1979. Monogenea of inland water fish in Af- rica.—Annales-Serie IN-8°-Sciences Zoolo- giques, Musee Royal de |’Afrique Centrale 226: 1-131, 48 plates. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(3):533-550. 1995. New records of azooxanthellate stony corals (Cnidaria: Scleractinia and Stylasteridae) from the Neogene of Panama and Costa Rica Stephen D. Cairns Department of Invertebrate Zoology, NHB-163, W-329, Smithsonian Institution, Washington, D. C. 20560, U.S.A Abstract. — Five new species of azooxanthellate Scleractinia are described from the Panamanian-Costa Rican Neogene: Septastraea altispina, Antillocy- athus gracilis, Paracyathus adetos, Oxysmilia pliocenica, and Asterosmilia ir- regularis. Three additional species of azooxanthellate stony corals are also reported for the first time in the fossil record: Gardineria minor, Schizocyathus fissilis, and Stylaster roseus. The records of S. roseus are the first fossil occur- rence of a stylasterid in the western Atlantic. The specimens that form the basis for this paper were collected in the first seven years (1986-1992) of the Panama Paleontology Project, a multiphyletic survey of the ma- rine invertebrate Neogene fauna of the Neo- tropics, especially southern Central Amer- ica. The purpose of the first phase of the project was to determine the biotic response to the Neogene closure of the Caribbean- eastern Pacific seaway through a detailed analysis of the stratigraphic and geographic distributions of various invertebrate groups. An early review of this project was pub- lished (Coates et al. 1992, Collins 1993), which included a general description of the Panama and Costa Rica localities, their lith- ostratigraphy, and their biostratigraphic correlations. A much more detailed analysis of the stratigraphy of various groups, in- cluding Scleractinia, is now in preparation (Collins & Coates in lett) as an edited vol- ume. In that volume I (Cairns 1996) list and document the 18 species of azooxanthellate stony corals that are known to occur in the Panamanian and Costa Rican Neogene, eight of which are discussed herein, i.e., the five new species and the three previously described species that are new to the fossil record. Species synonymies are considered to be complete or give a reference to a complete synonymy. In the Material Examined sec- tions, each record begins with a PPP locality number, followed by the number of speci- mens in that lot, followed by its catalog number. The PPP collection sites, with their original “‘CJ’’ field numbers, are listed in the Appendix. Absolute ages were derived from biostratigraphic dating using planktic foraminifera, calcareous nannoplankton, and the geological time scale of Berggren, et al (1985). The following abbreviations are used in the text: GCD, Greater Calicular Diameter; GCD:LCD, Ratio of greater calicular di- ameter to lesser calicular diameter; H:W, Ratio of height to width of a corallum; PD: GCD, Ratio of pedicel diameter to greater calicular diameter; PPP, Panama Paleon- tology Project; Sx, Cx, Px, Cycle of septa, costae, or pali (respectively) designated by numerical subscript; USGS, United States Geological Survey; USNM, United States National Museum (part of the National Mu- 534 seum of Natural History, Smithsonian In- stitution, Washington, D.C.). Systematic Account Class Anthozoa Order Scleractinia Suborder Faviina Family Rhizangiidae d’Orbigny, 1849 Septastraea d’Orbigny, 1849 Diagnosis. —Encrusting or ramose colo- nies having closely spaced cerioid corallites. Intercorallite coenosteum sparse. Corallites with two, rarely three, cycles of septa. Inner edges of septa finely dentate. Pali absent; fossa shallow; columella a solid fusion of inner septal edges. Miocene to Pleistocene: eastern and southern U.S., Central Ameri- ca, Colombia, Europe. Type species. —Hinde (1888) argued that d’Orbigny’s (1849) originally designated type species (and only species listed by him in 1849) of Septastraea subramosa could not be considered the type species of the genus because it was not described by d’Orbigny in 1849. Because Milne Edwards & Haime (1849) described four species of Septastraea later in the same year, including S. forbesi, and because d’Orbigny (1852) later synon- ymized S. forbesi with his S. subramosa, Hinde logically reasoned that S. forbesi must be the type species, being the only apparent valid species of the pair. However, the ICZN (1985), written long after Hinde’s intellec- tual exercise, clearly states in Article 12b6 that a description of a new genus before 1931 that includes reference to a new spe- cies without further description of that spe- cies does constitute availability of the spe- cies name through indication, 1.e., the ge- neric description also serves as the species description. The type species of Septastraea must therefore be S. subramosa d@’Orbigny, 1849 (=S.. forbesi Milne Edwards & Haime, 1849; =S. marylandica (Conrad, 1841)). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Septastraea altispina, new species Figs. 1-4 Diagnosis. —Corallum small and cerioid, encrusting gastropod shells. Intercalicular coenosteum well developed. Septa hexam- erally arranged in two cycles (12 septa). CS1 highly exsert, each rising to a massive, red- dish-brown coensteal spine that is triangu- lar in cross section. Columella massive and flat. | Description. —Corallum cerioid, known only to encrust gastropod shells, the largest colony examined (the holotype) a spherical mass about 11 mm in diameter consisting of approximately 25 corallites. Calices cir- cular to elliptical, larger calices up to 3.3 mm in GCD. Coenosteum well developed: corallites not directly adjacent but separated by 0.50-0.75 mm of coenosteum. In one specimen (USNM 95738), a shallow sulcus surrounds several corallites. Otherwise, coenosteum finely granular (nonspinose) and white, the exsert portions of the 6 CS1 being reddish-brown. Septa hexamerally arranged in two cycles (12 septa), with no indication of additional S3. Sl highly exsert (up to 1.1 mm), the distal portion of each S| forming a massive spine that is triangular in basal cross-section (up to 0.3 mm thick), resulting in a crown of 6 CS1 guarding each calice. Inner edges of S1 finely dentate and inclined obliquely toward columella. S2 nonexsert and some- times rudimentary, their inner edges also finely dentate and oriented obliquely to hor- izontally, fusing with the columella lower in fossa. Septa relatively thin (0.05—0.11 mm) and well separated from one another by a distance of about four times a septal width (0.3-—0.4 mm). Fossa shallow. Columella a solid, massive central structure up to 1.2 mm in diameter, which is flat to slightly concave and granular like the coenosteum. Discussion. —It may seem inadvisable to describe a new species of Septastraea when most, if not all, of the five or six nominal species in this genus have been previously VOLUME 108, NUMBER 3 535 Figs. 1-7. (1-4, Septastraea altispina: 1, 3-4, paratype from PPP326, USNM 95738; 2, holotype, USNM 95735): 1, Stereo view of corallum showing highly exsert costoseptal spines, S3>S4). One crown on 12 P3. Description. —Corallum ceratoid (edge angle 14°-19°), compressed, and usually slightly curved about 45° in plane of GCD. Largest specimen examined (USNM 95539) 13.5 x 10.3 mm in calicular diameter and 27.0 mm in height; holotype slightly smaller and somewhat damaged, but otherwise in excellent preservation, measuring 10.3 x 7.7 mm in calicular diameter and 18.6 mm in height. Calice elliptical, GCD:LCD = 1.12—1.33. Pedicel quite small, only 0.4-0.9 mm in diameter, and invariably detached from substrate. Costae equal, convex, and highly granular. Costae 0.40-0.45 mm wide, separated by shallow intercostal furrows about 0.15 mm wide. Thecal edge crests present on lower 2—3 mm of corallum, best preserved in juvenile coralla before they be- come broken or worn. Edge crests quite thin, semi-circular in shape, and up to 2.6 mm in height, occurring on both concave and convex thecal edges, but usually more prominent on convex edge. On upper con- vex thecal edge of many coralla a second, lower crest may be present or the principal costa on that edge may be slightly more pro- duced. Corallum white to reddish-brown. Septa hexamerally arranged in four cycles (some large coralla have some pairs of S4, up to 60 septa) according to the formula S1l— 2>S3>S4. S1-2 slightly exsert (about 1 mm) and have straight, vertical inner edges that reach about three-quarters distance to col- umella. S3 less exsert and about half width of S1—2. Each S3 bordered by a wide (up to 1.6 mm), lamellar palus, the 12 P3 forming a distinct elliptical crown encircling the col- umella. S4 about half width of an S3. Fossa moderate in depth, containing an elongate, VOLUME 108, NUMBER 3 Table 1.—Diagnostic characters of the four species of Antillocyathus. Character Typical size (GCD) Shape; lateral edge angle GCD:LCD Edge crests Septal arrangement Pali Columella Age and Distribu- tion A. maoensis 13-18 mm Straight, com- pressed; 45°-80° 1.75-2.50 Basally only 19-24:19-24:38—48: 0-8 (78-80-104) 19-24 secondary pali (P2) Solid, lamellar Late Miocene to early Pliocene of Dominican Re- A. alatus 9-11 mm Straight, full; paral- lel in upper coral- lum 1.3-1.75 Basally only 14—16:14-16:24-32: 0-8 (48-64-72) 14-16 P2 Solid, lamellar Late Miocene to early Pliocene of Dominican Re- A. cristatus 22-26 mm Curved, full; 36°—44° 1.6-1.7 Basally and convex edge >S5 (106 septa) 24 P4 Lamellar to spongy Late Miocene to early Pliocene of Dominican Re- A. gracilis 10-13 mm Curved, com- pressed; 14°-19° (ceratoid) 1.1-1.3 Basally and convex edge S1-2 > S3 > S4 (48-60) 12 P3 Fused papillose to sublamellar Early to middle Pliocene of Panama public public papillose to lamellar columella consisting of three or four fused papillae or a thin, dis- sected lamella. Discussion. —Antillocyathus gracilis is compared to the three other species in the genus in Table 1. To summarize, it is dis- tinguished by having a curved, slender cor- allum; a low GCD:LCD; hexamerally ar- ranged septa in four cycles; and 12 P3. Etymology.—The species name gracilis (from the Latin gracilis, slender) refers to the slender ceratoid growth form of the spe- cies. Material examined. — Holotype: PPP63, 1 corallum, USNM 95469. Paratypes: REE So, ls, USNM 95524; PPP57, 3, USNM 95522; PPP63, 18, USNM 95470; BRE I96, 27, USNM 95523; PPP345, 17, USNM 95521. Nontypes: PPP55, 2, USNM 95525; PPP193, 27, USNM 95527; PPP194, 2, USNM 95526; PPP197, 8, USNM 95528; PebIos 25, USNM 95529; PPP208, 1, USNM 95530; PPP294, 7, USNM 95532; REEDS 2, USNM 95533; PPP298, 1, USNM 95534; PPP306, 1, USNM 95535; PeroOs, 8 USNM 95536; PPP308, 2, USNM 95537; PPP311, 3, USNM 95538; public; Pliocene of Panama and Costa Rica PPP335, 1, USNM 95539; PPP346, 6, USNM 95540; PPP348, 1, USNM 95541; PPP350, 3, USNM 95542; PPP355, 3, USNM 95543; PPP357, 1, USNM 95544; PPP423, 1, USNM 95545. Type locality.—PPP63: Caribbean Pan- ama, Cayo Agua, small island offshore (9°10'44.0"N, 82°03'11.0”W); Cayo Agua Formation, 3.5—3.6 Ma (early Late Plio- cene). Distribution and Age. — Most records from Cayo Agua, Bocas del Toro, Caribbean Pan- ama; also known from Isla Colon and Isla Popa, both also Caribbean Panama. Cayo Agua Formation (early Late Pliocene). Paracyathus Milne Edwards & Haime, 1848a Diagnosis. —Corallum solitary; trochoid to turbinate; fixed or free. Base often poly- cyclic. Paliform lobes usually multilobate, occurring before all but last cycle of septa. Columella papillose, elements often indis- tinguishable from lower paliform lobes. Eo- cene to Recent: cosmopolitan. 538 Type species. —Paracyathus procumbens Milne Edwards & Haime, 1848a, by sub- sequent designation (Milne Edwards & Haime 1850: xv). Paracyathus adetos, new species Figs. 8-14 Diagnosis. —Corallum tympanoid, cylin- drical, or trochoid, often overgrowing its substrate of attachment to become free. Base polycyclic. Septa hexamerallay arranged in four cycles: S1>S2>S4=2S3. Each S1-3 bears several narrow paliform lobes. Colu- mella papillose or solid. Description. —Corallum solitary, at- tached when young but often incorporating the substrate into its base (Fig. 13) and thus becoming free in adult stage. Young coralla cylindrical and squat (tympanoid); older coralla persist as short cylinders if the sub- Strate 1s too large to overgrow, but become bowl-shaped to trochoid or turbinate if the substrate is overgrown. Largest specimen examined (holotype) 6.7 mm in circular cal- icular diameter and 3.9 mm in height, the height usually being about half the calicular diameter. Base polycyclic (Figs. 10, 12, 14), larger coralla having up to five concentric thecal rings occurring at the following ap- proximate diameters and septal comple- ments: first ring, 0.5—O.7 mm (6 septa); sec- ond, 1.05—1.20 mm (12 septa); third, 1.4— 1.9 mm (about 18 septa); fourth, 2.2—-3.6 mm (24 septa); and fifth, 2.6-6.7 mm (48 septa). Coralla usually attach to a small bi- valve shell or piece of a bivalve shell or to a small gastropod shell, and less frequently to another coral, a brachiopod shell, or a bryozoan colony. Costae highly ridged and dentate, the Cl—2 up to 0.4 mm in height, the C3—4 less prominent. Septa hexamerally arranged in 4 complete cycles above a GCD of about 4.3 mm ac- cording to the formula: S1 >S2>S4=2S3. Sl highly exsert (up to 0.7 mm) and have Straight inner edges that extend about 0.8 distance to columella. Inner edges of each PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON S1 bears one small, blunt paliform lobe ad- jacent to the columella. S2 less exsert and extend only 0.6 distance to columella, each S2 internally bordered by two or three small, blunt paliform lobes. S3-4 equally exsert, the S3 extending about half the distance to columella, the S4 equal to or slightly wider than the S3. Each S3 internally bordered by three or four tall, blunt paliform lobes; inner edges of S4 finely serrate. Inner edges of each pair of S4 within a half-system fuse to their common S83, each pair of S3 within a system, in turn, fuses to its common S2 through its innermost teeth, directly adja- cent to the columella. All septa covered with tall granules. Fossa quite shallow. Colu- mella papillose in small coralla, the papillae fusing in larger coralla to form a massive, slightly concave solid structure. Discussion. — Paracyathus adetos is sim- ilar to P. vaughani Gane, 1895, a species known from the Late Miocene of the south- eastern United States from Virginia to Flor- ida (see Vaughan 1904, Weisbord 1971). Both species have low, tympanoid coralla, polycyclic bases that are often attached to bivalve shells, and similarly shaped pali- form lobes. Paracyathus vaughani, howev- er, differs in attaining a larger size (e.g., up to 11 mm GCD with 88 septa and eight thecal rings), always remaining attached, having a papillose (never solid) columella, and having a deeper fossa (Figs. 15, 18). Paracyathus adetos differs from P. henekeni Duncan, 1863 in having a larger corallum and more septa; a lower H:W; and a more solid columella. Etymology.—The species name adetos (from the Greek adetos, loose, free, un- bound) refers to the tendency of mature cor- alla of this species to become free attach- ment by incorporating the substrate into its base. Material examined. — Holotype: PPP196, 1, USNM 95575. Paratypes: PPP56, 6, USNM 95576; PPP57, 3, USNM 95577; PPP63, 13, USNM 95578; PPP65, 14, USNM 95579; PPP66, 1, USNM 95580; VOLUME 108, NUMBER 3 PPP194, 13, USNM 95581; PPP195, 16, USNM 95582; PPP196, 42, USNM 95583; PPP197, 7, USNM 95584; PPP198, > 100, USNM 95585; PPP205, 2, USNM 95586; PPP306, 32, USNM 95588; PPP307, 53, USNM 95589; PPP311, 11, USNM 95590; PPP352, 22, USNM 95591; PPP475, 28, USNM 95587. Type locality. —PPP196: Caribbean Pan- ama, Cayo Agua, SW Pt. Norte (9°10'42.5”"N, 82°03'9.0”W); Cayo Agua Formation, 3.5-3.6 Ma (early Late Plio- cene). Distribution and Age. —All records from Cayo Agua, Bocas del Toro, Caribbean Pan- ama (early Late Pliocene). Oxysmilia Duchassaing, 1870 Diagnosis (emended).—Corallum soli- tary; ceratoid to trochoid; attached or free. Base composed of concentric, partitioned thecal rings achieved by formation of ex- othecal dissepiments over ridged costa, or the base may be narrow and unattached. Septotheca costate. Rudimentary paliform lobes occasionally present before S3. Fossa deep. Columella variable, ranging from massive granular papillae to lamellar or aligned fused papillae. Dissepiments ab- sent. Late Pliocene to Recent: Central America, western Atlantic (46-640 m). Type species.—Lophosmilia rotundifolia Milne Edwards & Haime, 1848c, by mono- typy. Oxysmilia pliocenica, new species Figs. 16-17 Diagnosis. —Corallum ceratoid and free, with a narrow pedicel. Septa hexamerally arranged in five cycles, the last cycle incom- plete: S1-2>S3>S4>S5. Rudimentary pal- iform lobes (P3) sometimes present. Colu- mella papillose or lamellar. Description. —Corallum ceratoid and free, the largest specimen examined (holotype) 14.2 x 11.8 mm in calicular diameter and 18.4 mm in height. Pedicel very narrow, 539 only 1.0—1.2 mm in diameter, in most cases partially incorporating a small gastropod shell into its base. Calice elliptical: GCD: LCD = 1.20-1.25. Costae poorly defined on lower third of corallum; broad, slightly con- vex, and granular on middle third; and ridged and rather thin in upper third adja- cent to calice, where intercostal furrows are correspondingly broader. Upper half to third of corallum reddish-brown. Theca quite thin, only about 0.05 mm in thickness. Septa hexamerally arranged in five cycles, the last cycle incomplete. The holotype has 92 septa and the largest paratype of GCD 13.9 mm has 82 septa, consisting of seven half-systems with both pairs of S5, three with one pair of S5, and two half-systems with no S5. Septal formula: Sl- 2>S3>S4>S5. Relative exsertness of sep- tal cycles unknown due to poor preservation of calicular edges. Sl1-2 broad, extending 0.8-—0.9 distance to columella, and have straight, vertical, slightly thickened inner edges. S3 about three-quarters width of an Sl-—2 and have finely sinuous inner edges, some of which extend to the columella low in fossa as irregular trabecular ribbons. S4 half width of S3 and have finely dentate inner edges. S5 rudimentary, about one-third width of an S4. Faces of SI-4 homoge- neously covered with low, pointed granules. Fossa moderate in depth. Columella vari- able in shape: in the holotype and three paratypes it consists of an alignment of slen- der, fused papillae, but in one paratype (USNM 95594) the columella is a thin, dis- sected lamella. Discussion. —Oxysmilia pliocenica 1s quite similar to the only other species in the ge- nus, O. rotundifolia (Milne Edwards & Haime, 1848c), a species known only from the Recent western Atlantic from North Carolina to Surinam and the western Gulf of Mexico at 46-640 m (Cairns 1979). Points of similarity include septal number, ar- rangement, and size; costal shape; and vari- ation in columella shape, the columella of O. rotundifolia also ranging from elongate PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Figs. 8-18. (8-14, Paracyathus adetos: 8, holotype, USNM 95575; 9, paratype, PPP 196, USNM 95583; 10- 12, paratype, PPP198, USNM 95585; 13, paratype, PPP63, USNM 95578; 14, paratype, PPP475, USNM 95587): 8, Calice of holotype, x 6.4; 9, Stereo view of a paratype, x 7.8; 10-12, Young stages attached to bivale shells (calices of fig. 10 illustrating 2 thecal rings, calices of fig. 12, 3 rings), x28, x6.3, x26, respectively; 13, VOLUME 108, NUMBER 3 papillose to solid lamellar. Also, in both species some specimens bear trabecular processes on the inner edges of their S3 that resemble small paliform lobes. Oxysmilia pliocenica differs in having a simple base (not partitioned concentric rings similar to polycyclic development), being smaller in size, and having a reddish-brown corallum. Lophosmilia Milne Edwards & Haime, 1848b is a closely related genus of three or four species known from the Cretaceous to Eocene of Europe, Texas, and Peru, differ- ing from Oxysmilia in having a deep-seated lamellar columella (Wells 1956). But, given the range of columellar variation found in the two known species of Oxysmilia, this difference hardly seems significant. In fact, O. rotundifolia was originally described in the genus Lophosmilia. Etymology.—Named for the geological epoch in which it occurs. Material examined. — Holotype: PPP311, USNM 95593. Paratypes: PPP196, 1, USNM 95595; PPP311, 3, USNM 95594. Type locality. —PPP311: Caribbean Pan- ama, Cayo Agua, S of Pt. de Nispero (9°09'57.4”"N, 82°01'48.6”W); Cayo Agua Formation, 2.9-3.6 Ma (early Late Plio- cene). Distribution and age. — Known only from Cayo Agua, Bocas del Toro, Caribbean Pan- ama (early Late Pliocene). Asterosmilia Duncan, 1867 Diagnosis. —Corallum solitary; ceratoid, trochoid, or flabellate; free or attached. Sep- totheca costate. Paliform lobes present be- fore penultimate septal cycle. Columella la- mellar or fascicular. Vesicular endotheca abundant. Oligocene to Recent: West In- dies, Central America. — 541 Type species. — Trochocyathus abnormal- 1s Duncan, 1864, by subsequent designation (Vaughan 1919:354). Asterosmilia irregularis, new species Figs. 19-20 Asterosmilia abnormalis.—Cairns & Wells, 1987: 37 (in part: USGS 8321). Diagnosis. —Corallum ceratoid, elongate, and gently curved. Septa arranged in 15-18 sectors, with the following septal comple- ments: 15—18:15-—18:30—-36:0—-10 (up to 76 septa). Paliform lobes present on secondary septa and tertiary septa that are flanked by quaternary septa. Columella lamellar. Description.—Corallum ceratoid, elon- gate, and usually gently curved about 90° in basal region. Largest specimen examined (USGS 20468) 19.4 x 16.8 mm in calicular diameter and 72.9 mm in length; holotype 16.3 $3 > S4>S5 (96-108 septa), S1-2 highly exsert 24 P4 Corallum size; shape PD:GCD Costae Septal arrangement Paliform lobes Vesicular dissepiments Abundant, inclined, 1—5 “rings” Columella Coarse papillose or cris- pate GCD max. = 19.4 mm; ceratoid, free GCD max. = 10.4 mm; ceratoid, free 0.06-0.12 0.06-0.11 All costae equal Cl-2 usually broader than others 15-18:15—18:30-36:0-10 S1-—2 > S3 > S4 (48 sep- (68-76 septa), not exsert _ta), not very exsert 16-18 lobes before sec- 12 P3 ondaries and some ter- tiaries Less common, inclined, Sparse, horizontal, one 1-2 “rings” ring Plate-like (lamellar or Lamellar to crispate labyrinthiform) septa and containing a medial secondary septum, two tertiary septa and 0-2 quater- nary septa, resulting in a calicular total of 68—76 septa. Septal formula of holotype: 15: 15:30:10 (70 septa, 17 well-developed pali and three rudimentary ones); other speci- mens having: 17:17:34:4 (72 septa, 17 well- developed pali), USNM 95611; 17:17:34:6 (74 septa, 17 well-developed pali and three rudimentary), USNM 95615; and 18:18: 36:4 (76 septa, 18 well-developed pali and two rudimentary), USNM 95615. Preser- vation of most coralla does not allow anal- ysis of septal or palar number. Primary sep- ta only slightly exsert (1.0—1.2 mm), of mod- erate thickness (about 0.4 mm), and have Straight, vertical inner edges that extend about three-quarters distance to columella. Secondary septa less exsert and less thick, and only about half width ofa primary. Each secondary septum bordered internally by a broad, lamellar paliform lobe of equal width to the secondary septum. Inner edges of pal- iform lobes fuse to columella. In sectors lacking quaternaries, tertiary septa are about one-third width of a secondary, but if a pair of S4 flank a tertiary, the tertiary is doubled in width and sometimes bears a broad pal- iform lobe (P3) of variable size, sometimes as large as a P2 but occasionally rudimen- tary. S4 rudimentary. Thin vesicular dis- sepiments occur in upper corallum, inclined and structured as in A. profunda, but oc- curring less abundantly, 1.e., only one or two dissepiments present at a level in any in- terseptal space. Fossa of moderate depth, containing a plate-like columella of variable construction. Often the plate is a single, thin, medial lamella; occasionally it is expressed as several parallel, slightly overlapping plates; and in one case (USNM 95615) the plates were labyrinthiform in arrangement. Discussion. —Asterosmilia irregularis dif- fers from its congeners in having nonhex- ameral septal symmetry. It is intermediate in size between the two other Neogene spe- cies known from Panama and differs from them in other characters as well (Table 2). Etymology.—The species name irregu- laris (from the Latin in regularis, not ac- cording to rule) refers to the variable num- ber of septal sectors contained by specimens of this species. Material examined. — Holotype: PPP1119, USNM 95609. Paratypes: PPP627, 38, USNM 95618; PPP757, 4, USNM 95617; PPP1101, 12, USNM 95610; PPP1102, 2, USNM 95611; PPP1103, 1, USNM 95612; VOLUME 108, NUMBER 3 PPP1104, 1, USNM 95613; PPP1105, 10, USNM 95614; PPP1107, 13, USNM 95615; PPP1118, 1, USNM 95616; USGS 8321, 1, USNM 65323; USGS 20468, 69, USNM 64024; Limon Centro, Costa Rica, 3, USNM 81310; Moin Formation, Costa Rica, 4, USNM 72351. Type locality. —PPP1119: Lomas del Mar, Limon, Costa Rica, (construction site): 9°59'31.0”N, 83°02'12.2'W; Moin Forma- tion, 1.7—1.9 Ma (Late Pliocene). Distribution and Age. —All specimens from Lomas del Mar, Limon, Caribbean Costa Rica (Late Pliocene). Family Flabellidae Bourne, 1905 Gardineria Vaughan, 1907 Diagnosis. —Corallum solitary, ceratoid to turbinate; transverse division lacking. Corallum attached through a pedicel as well as having a massive lateral secondary root- let. Epitheca transversely wrinkled. Upper, outer septal edges separated from smooth calicular edge by a deep notch. P2 usually present; columella papillose. Late Pliocene (Costa Rica) to Recent (tropical western At- lantic, Hawaiian Islands, Philippines, Ant- arctica, South Africa). Type species. —Gardineria hawaiiensis Vaughan, 1907, by original designation. Gardineria minor Wells, 1973 Figs. 21-26 Gardineria minor Wells, 1973: 49-53, figs. 36a-g; Cairns, 1979: 162-163, pl. 31, figs. 7-9 (complete synonymy). Description. —Corallum small, subcy- lindrical to ceratoid, and firmly attached to substrate by a small pedicel 0.9-1.3 mm in diameter and a larger, irregularly-shaped lateral rootlet, which issues from the lower side of the theca and appears as a broad thecal adhesion to the substrate. Base poly- cyclic, usually with only two rings—the pro- tothecal and the outer epithecal—the latter being smooth but circumferentially finely 543 wrinkled. The seven fossil specimens ex- amined range in calicular diameter from 2.6— 4.9 mm. Corallum white. Septa hexamerally arranged in two or three cycles, depending on calicular diameter. Coralla 2.6-3.0 mm in diameter have only 12 septa (Figs. 23, 26), whereas those over 3.3 mm usually have 24 septa arranged: S1>S2>S3 (Figs. 21-22, 24-25). S1 exsert, thick (about 0.25 mm), and have entire in- ner edges that attain the columella. S2 no- nexsert, much thinner than SI, and have sloping, dentate to laciniate inner edges that also merge with the columella. S3 rudimen- tary, each represented only as a series of spines. Tall, slender P2 sometimes present. Fossa shallow. Columella small and papil- lose. Discussion. —The fossil specimens re- ported herein are indistinguishable from Recent G. minor (Figs. 24-25), a species that is common throughout the Caribbean. Large living specimens (e.g., over 5 mm GCD) are known to have a fourth cycle of septa. Material examined. —PPP466, 1, USNM 95689; PPP710, 6, USNM 95690; type se- ries, USNM; specimens reported by Cairns (1979). Types.—The holotype and 11 paratypes of G. minor are deposited at the USNM (53503-53506) (Cairns 1991). Type locality. —off Yallahs, Jamaica, 15 m. Distribution and age. —Santa Rita and Li- mon, Caribbean Costa Rica; Moin For- mation (Late Pliocene). Previously known only from the Recent throughout the Ca- ribbean and Bahamas, including off Pana- ma, at 2—241 m (Cairns 1979). Family Guyniidae Hickson, 1910 Schizocyathus Pourtalés, 1874 Diagnosis. —Corallum solitary, ceratoid, and invariably attached to a fragment of its parent corallum through longitudinal par- ricidal budding. Smooth epitheca bears 12 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Figs. 19-29. (19-20, holotype of Asterosmilia irregularis, USNM 95609): 19-20, Stereo calicular and lateral views of holotype, x 2.5, x0.9, respectively. (21-26, Gardineria minor: 21-23, 26, PPP710, USNM 95690; 24— 25, Recent, Jamaica, 31 m, USNM 80892): 21-22, Side and calicular views of a specimen with 3 cycles of septa, VOLUME 108, NUMBER 3 rows of mural pores (spots), a pair of rows flanking each S2. P1 often present; P3 al- ways present, each pair within a system uniting in a V-shaped structure. Columella absent. ?Paleocene, ?Eocene, Late Pliocene to Recent: ?Russia, ? South Australia, Costa Rica, Atlantic (88-1300 m). Type species. —Schizocyathus fissilis Pourtalés, 1874, by monotypy. Discussion. — The illustrations of the spe- cies described from the Danian of Russia, S. daschsalachlyensis Kuzmicheva, 1987, are not typical for the genus. Likewise, the description and figures of Cyathosmilia ve- lata Dennant, 1902 (Eocene, South Austra- lia), doubtfully placed in Schizocyathus by Vaughan & Wells (1943), are also inconsis- tent with the type species of the genus. Therefore the early Tertiary records of this genus are strongly doubted. Schizocyathus fissilis Pourtalés, 1874 Figs. 27-29 Schizocyathus fissilis Pourtalés, 1874: 36—- 37, pl. 6, figs. 12—13; Cairns, 1979: 166— 167, pl. 32, figs. 4-7 (complete synony- my); Zibrowius, 1980: 166, pl. 85, figs. A-O. Discussion. —The single specimen re- ported herein measures only 1.9 mm in cal- icular diameter and 1.5 mm in height, but is attached to a one-sixth sector of its parent corallum, which is 1.6 mm long. Both par- ent fragment and budded corallum display external mural pores about 0.11 mm in di- ameter. The septal pattern is obscured by sediment within the calice, but the specimen appears to have three cycles of septa, the S1 being quite exsert. —_ 545 Even though the Panamanian specimen is small and poorly preserved, there is little doubt it is S. fissilis. Asexual fragmentation from a wedge of a parent corallum, usually a longitudinal one-sixth of the corallum, is the most common mode of reproduction in this species (Figs. 27-29). The Panamanian specimen is the first fossil occurrence of this species. Material examined. —PPP362, 1, USNM 95675; syntype series. Types. — Forty-one syntypes are deposit- ed at the MCZ (5470 and 2791). Type locality. —off Barbados, 183 m. Distribution and age.—Escudo de Vera- guas, Bocas del Toro, Caribbean Panama; Escudo de Veraguas Formation, 1.8—1.9 Ma (Late Pliocene). Previously known only from the Recent of Caribbean and Gulf of Mex- ico, including off Honduras, and northeast- ern Atlantic, at 88-1300 m (Cairns 1979, Zibrowius 1980). Class Hydrozoa Family Stylasteridae Gray, 1847 Stylaster Gray, 1831 Diagnosis. —Gastro- and dactylopores ar- ranged in cyclosystems. Cyclosystems vari- able in location, ranging from a uniform coverage of all branch surfaces to a strictly sympodial arrangement. Coenosteal texture also variable, but usually linear-imbricate or reticulate-granular. Gastro- and dacty- lostyles present; gastrostyles usually ridged; ring palisade often present. Ampullae su- perficial. Oligocene to Recent: cosmopoli- tan. Type species.—Madrepora rosea Pallas, 1766, by subsequent designation (Milne Ed- wards & Haime 1850: xxi1). x 13.8, 12.2, respectively; 23, 26, A smaller specimen with only 2 cycles of septa, both x 16.2; 24-25, Oblique calicular and calicular views of a Recent specimen included for comparison to fossil specimens, x14, x15, respectively. (27-29, Schizocyathus fissilis: 27, Recent, off Barbados, 200 m, USNM 61747; 28, PPP362; 29, Recent, Gulf of Mexico off Florida, 221 m, USNM 61744): 27-28, Small coralla budded asexually from a parent fragment, x11, x 18, respectively; 29, A three-generation budded corallum, x 10.4. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Figs. 30-36. (Stylaster roseus: 30-34, PPP710, USNM 95724; 35-36, Recent, Carrie Bow Cay, Belize, 3 m, USNM 47807): 30, Stereo view of a branch illustrating a cyclosystem and a female ampulla, x32; 31, 33, Longitudinal fracture through cyclosystems revealing gastrostyles and elements of ring palisade, x92, x 190, respectively; 32, Stereo view of cross section of a cyclosystem just above gastrostyle tip, illustrating ring palisade and circle of dactylopores, x80; 34, A cyclosystem, x50; 35-36, Cyclosystem and gastrostyle of a Recent specimen, x50, x 155, respectively. VOLUME 108, NUMBER 3 Stylaster roseus (Pallas, 1766) Figs. 30-36 Madrepora rosea Pallas, 1766: 312-313. Stylaster roseus. —Cairns, 1986: 61-65, pl. 27, figs. A-H, pl. 28, figs. A-C, pl. 53, fig. D (complete synonymy and distribution map). Description. —Only relatively small branch fragments are reported herein, the longest (USNM 95724) 9 mm long consist- ing of nine cyclosystems. Coenosteum com- posed of slightly convex, worn strips 80-90 um wide arranged in a reticulate fashion. Cyclosystems circular to slightly ellipti- cal, 0.8-1.0 mm in greater diameter, and arranged in a strictly sympodial manner. Based on 14 cyclosystems, the range of dac- tylopores per cyclosystem is 10-13, average = 11.1 (o = 0.86), and mode = 11. Gastro- pores circular (about 0.25 mm in diameter) and contain a well-developed, diffuse ring palisade consisting of robust, cylindrical el- ements up to 35 um in height and 14-21 um in diameter. Gastrostyles lanceolate, highly ridged, and very spinose. Illustrated gastrostyle (Figs. 31, 33) 0.24 mm in height and 0.11 mm in diameter, bearing slender spines up to 35 wm long and only 3-4 wm in basal diameter. Dactylotomes 60-65 wm wide; pseudosepta wedge-shaped and about two times width of a dactylotome. An ad- cauline diastema three times dactylotome width is often present. Dactylostyles rudi- mentary, composed of widely-spaced, lin- early arranged cylindrical elements up to 25 um in height and only about 7 um in di- ameter. Female ampullae 0.55-0.65 mm in di- ameter, sometimes clustered. Lateral effer- ent pore 0.15—0.18 mm in diameter. Male ampullae not observed. Discussion. —Direct comparison of the Neogene stylasterids to Recent S. roseus shows no significant differences, S. roseus being the only shallow-water stylasterid known from the western Atlantic (Cairns 1986). Although the genus Sty/asteris known 547 from the fossil record, these records are be- lieved to be the first fossil stylasterids of any kind reported from the western Atlantic. Material examined.—PPP55, 1, USNM 95720; PPP634, 3, USNM 95721; PPP639, 2, USNM 95722; PPP708, 3, USNM 95723; PPP710, 12, USNM 95724; PPP720, 30, USNM 95725; PPP738, 10, USNM 95726. Types. — Not traced. Type locality. —off Santo Domingo, depth unknown. Distribution and age. —Isla Colon, Bocas del Toro, Caribbean Panama; Pueblo Nue- vo, Limon, and Santa Rita, Caribbean Cos- ta Rica; Moin Formation (early Late to Late Pliocene). Previously known from Recent throughout Caribbean and Bahamas, in- cluding off Caribbean Panama, Costa Rica, Honduras, and Belize, at 0.5—73 m (Cairns 1986). Acknowledgments The specimens were collected by the Pan- ama Paleontology Project, and were pre- pared by Yira Ventocilla (PPP, Smithsonian Tropical Research Institute (STRI), Pana- ma), and Rene Panchaund (Naturhisto- risches Museum, Basal, Switzerland). Lau- rel Bybell and Harry Dowsett (USGS) are acknowledged for their biostratigraphic dat- ing of the PPP samples. The PPP has been funded by the National Science Foundation, STRI, the National Geographic Society, and the Kugler Fund, Basel. Literature Cited Berggren, W. A., D. V. Kent, J. J. Flynn, & J. A. Van Couvering. 1985. Coenozoic geochronolo- gy.— Bulletin of the Geological Society of Amer- ica 96:1407-1418. Bourne, G. C. 1905. Report on the solitary corals collected by Professor Herdman, at Ceylon, in 1902.—Ceylon Pearl Oyster Fisheries, Supple- mentary Reports 29:187—242, pls. 1-4. Cairns, S. D. 1979. The deep-water Scleractinia of the Caribbean Sea and adjacent waters. —Stud- ies on the Fauna of Curacao 57(180):341 pp. . 1986. A revision of the Northwest Atlantic Stylasteridae (Coelenterata: Hydrozoa).— 548 Smithsonian Contributions to Zoology 418:131 pp. 1991. Catalog of the type specimens of stony corals (Milleporidae, Stylasteridae, Scleractinia) in the National Museum of Natural History, Smithsonian Institution. — Smithsonian Contri- butions to Zoology 514:59 pp. 1996. Stratigraphic distributions of Neogene Caribbean azooxanthellate corals (Scleractinia and Stylasteridae). Pp. xxx in L. S. Collins & A. G. Coates, eds. Marine Biodiversity of Neogene southern Central America, Part 1: Caribbean. — Bulletins of American Paleontology (in press). ——., & J. W. Wells. 1987. Neogene Paleontology in the northern Dominican Republic. Part 5. The suborders Caryophylliina and Dendro- phylliina (Anthozoa: Scleractinia). — Bulletins of American Paleontology 93(328):23-43, 52-55, 68-74. Coates, A. G., et al. 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. Collins, L. S. 1993. Neogene paleoenvironments of the Bocas del Toro Basin, Panama. —Journal of Paleontology 67(5):699-710. Conrad, T. A. 1841. [Twenty-six new species of fossil shells, discovered in the Medial Tertiary de- posits of Calvert Cliffs, Maryland].—Proceed- ings of the Academy of Natural Sciences of Phil- adelphia 1:28-33. Dana, J. D. 1846. Zoophytes. United States Explor- ing Expedition during the years 1838-1842 un- der the command of Charles Wilkes, Philadel- phia, Lea & Blanchard 7:vi + 740 pp. Dennant, F.G.S. 1902. Descriptions of new species of corals from the Australian Tertiaries, Part 5.— Transactions of the Royal Society of South Australia 26:255-264, pls. 5-6. Duchassaing, P. 1870. Revue des Zoophytes et des Spongiaires des Antilles, Paris, 52 pp., 2 pls. Duncan, P. M. 1863. On the fossil corals of the West Indies. Part 1.—Quarterly Journal of the Geo- logical Society of London 19:406-458, pls. 13- 16. . 1864. On the fossil corals of the West Indies. Part 2.— Quarterly Journal of the Geological So- ciety of London 20:20-45, pls. 2-5. 1867. On the genera Heterophyllia, Palaeo- cyclus, and Asterosmilia. — Philosophical Trans- actions of the Royal Society of London 157: 643-656, pls. 31-32. Gane, H. S. 1895. A contribution to the Neocene corals of the United States.—Johns Hopkins University Circulars, 15(121):8-10. Gray, J. E. 1831. Description of a new genus (Sty- laster) of star-bearing corals.— Zoological Mis- cellany: 36-37, London. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 1847. An outline of an arrangement of stony corals.—Annals and Magazine of Natural His- tory (1)19:120-128. Hickson, S. J. 1910. Ona new octoradiate coral, Pyr- ophyllia inflata. — Memoirs and Proceedings of the Manchester Literary and Philosopical So- ciety 54(3):1-7, 4 figs. Hinde, G. J. 1888. On the history and characters of the genus Septastraea, d’Orbigny (1849), and the identity of its type species with that of Gly- phastraea, Duncan (1887).— Quarterly Journal of the Geological Society of London 44:200- MSI. fol, International Code of Zoological Nomenclature (ICZN). 1985. Ride, W. D. [Chairman of editorial com- mittee]. International Code of Zoological No- menclature, Third edition, adopted by the XX General Assembly of the International Union of Biological Sciences. International Trust for Zoological Nomenclature, London, xx + 338 pp. Kuzmicheva, E. U. 1987. Upper Cretaceous and Pa- leogene corals from the USSR. Academy of Sci- ences, Moscow, 187 pp. [in Russian]. Meeder, J. F. 1987. The paleoecology, petrology and depositional model of the Pliocene Tamiami Formation, southwest Florida, Parts 1-2. Un- published Ph.D. dissertation, University of Mi- ami, Miami, Florida, 748 pp. Milne Edwards, H., & J. Haime. 1848a. Recherches sur les Polypiers. Deuxiéme mémoire. Mono- graphie des Turbinolides. — Annals des Sciences Naturelles, Zoologie, Paris, (3)9:211-344, pls. 7-10. OL 1848b. Observations sur les Po- lypiers de la famille des Astéides.—Comptes Rendu, Académie des Sciences, Paris, 27(19): 465-470. , & 1848c. Recherches sur les Poly- piers. Quatrieme Mémoire. Monographie des Astréides (1). — Annales des Sciences Naturelles, Zoologie, Paris (3)10:209-—320, pls. 5-9. , & . 1849. Recherches sur les Polypiers. Quatriéme Mémoire. Monographie des As- tréides (quatriéme section).—Annales des Sci- ences Naturelles, Zoologie, Paris (3)12:95-197. OC 1850. A monograph of the British fossil corals. Part 1. Introduction. —Paleonto- graphical Society of London, Ixxxv + 71 p. Orbigny, A. C. V. D.d’ 1849. Note sur des re fossiles. Paris, 12 pp. 1852. Prodrome de paléontologie stratigrap- hique universalle des animaux Mollusques et Rayonneés, faisant suite au cours élementaire de paléontologie. vol. 1, Paris. Pallas, P. S. 1766. Elenchus Zoophytorum. Hague- Comitum. 451 pp. Pourtalés, L. F. 1874. Zoological results of the Hass- VOLUME 108, NUMBER 3 ler expedition. Crinoids and corals.—Memoirs of the Museum of Comparative Zoology 4:33- 50, pls. 6-9. Vaughan, T. W. 1904. Anthozoa. Maryland Geolog- ical Survey, Systematic Paleontology, Miocene, pp. 438-447, pls. 122-129. . 1907. Recent Madreporaria of the Hawaiian Islands and Laysan.—Bulletin of the United States National Museum 59:427 pp., 96 pls. 1919. Fossil corals from Central America, Cuba, and Porto Rico, with an account of the American Tertiary, Pleistocene, and Recent cor- al reefs.— Bulletin of the United States National Museum 103:189-524, pls. 68-152. ——., & J. E. Hoffmeister. 1925. New species of fossil corals from the Dominican Republic. — Bulletin of the Museum of Comparative Zool- ogy 47:315-326, pls. 1-4. , & J. W. Wells. 1943. Revision of the sub- orders families, and genera of the Scleractinia. — Geological Society of America Special Papers 44:363 pp. Weisbord, N. E. 1971. Corals from the Chipola and Jackson Bluff formations of Florida.—Geolog- ical Bulletin of the Florida Bureau of Geology 53:100 pp. Wells, J. W. 1937. Coral studies. Part 2. Five new genera of the Madreporaria.— Bulletins of American Paleontology 23(79):242-249. 1956. Scleractinia, Pp. F328—-F444 in R. C. Moore, ed., Treatise on Invertebrate Paleontol- ogy, Pt. F, Coelenterata. Geological Society of America and University of Kansas Press, Law- rence. . 1973. New and old scleractinian corals from Jamaica.— Bulletin of Marine Science 23:16-55. Zibrowius, H. 1980. Les Scléractiniaires de la Méd- iterranee et de |’Atlantique nord-oriental. — Meé- moires de I’Institut Océanographique 11:283 pp. Appendix: PPP Collection Sites, Including Original CJ Field Numbers PPP55 (CJ86-31-1) Isla Colon, Bocas del Toro: 9°25'23.2”N, 82°15'36.8”W; age and formation unknown. PPP56 (CJ86-32-1) Cayo Agua, Bocas del Toro, SW of Pt. Norte: 9°10'39.1”N, 82°03’8.8”W; Cayo Agua Formation, 3.5—3.6 Ma (early Late Pliocene). PPP57 (CJ86-33-1) Ibid.: 9°10'48”’N, 82°03'6.7’W; Cayo Agua Formation, 3.5—3.6 Ma (early Late Pliocene). PPP63 (CJ8639-1) Cayo Agua, Bocas del Toro, small island offshore: 9°10'44.0’N, 82°03'11.0”W; Cayo Agua Formation, 3.5—3.6 Ma (early Late Plio- cene). PPP65 (CJ86-40-2) Cayo Agua, Bocas del Toro, S of 549 Pt. de Nispero: 9°09'57.4”N, 82°01'48.6”W; Cayo Agua Formation, 3.4—10.8 Ma (age uncertain). PPP66 (CJ86-40-3) Ibid. PPP193 (CJ87-27-1) Cayo Agua, Bocas del Toro, SW of Pt. Norte: 9°10'35.1”N, 82°03'8.7”W; Cayo Agua Formation, 3.5-3.6 Ma (early Late Pliocene). PPP194 (CJ87-28-1) Ibid. PPP195 (CJ87-29-1) Ibid.: 9°10’37.0’N, 82°03'9.0’W. PPP 196 (CJ87-29-2) Ibid.: 9°10'42.5”N, 82°03'9.0”W. PPP197 CJ87-29-3) Ibid.: 9°10'45.8”N, 82°03'8.3”W. PPP198 (CJ87-29-4) Ibid.: 9°10’48.0’N, 82°03'6.7’W. PPP205 (CJ87-33-4) Cayo Agua, Bocas del Toro, SW of Pt. Piedra Roja: 9°08’29.3”N, 82°00'43.8’W; age and formation unknown. PPP208 (CJ87-33-7) Ibid.: 9°08'33.3”N, 82°00'38.4’”W; ?Cayo Agua Formation (age uncertain). PPP294 (CJ88-18-2) Cayo Agua, Bocas del Toro, W side of Pt. de Tiburon: 9°09'18.0”N, 82°01'34.1”W; Cayo Agua Formation, 2.9-3.6 Ma (early Late Pliocene). PPP295 (CJ88-18-3) Ibid. PPP298 (CJ88-18-6) Ibid. PPP306 (CJ88-20-4) Cayo Agua, Bocas del Toro, E side Pt. de Nispero: 9°10'3.1”N, 82°01'48.2’W; Cayo Agua Formation, 2.9-3.6 Ma (early Late Pliocene). PPP307 (CJ88-20-5) Ibid. PPP308 (CJ88-20-6) Ibid. PPP311 (CJ88-21-1) Cayo Agua, Bocas del Toro, S side Pt. de Nispero: 9°09'57.4’”N, 82°01'48.6’W; Cayo Agua Formation, 2.9-3.6 Ma (early Late Pliocene). PPP326 (CJ88-25-12) Cayo Agua, Bocas del Toro, N side Pt. de Nispero: 9°10'4.8’N, 82°02'0.6”W; Cayo Agua Formation, 2.9-3.6 Ma (early Late Plio- cene). PPP335 (CJ88-26-2) Cayo Agua, Bocas del Toro, E tip Pt. de Tiburon: 9°09'11.4”N, 82°01'21.6”W; Cayo Agua Formation, 2.9-3.6 Ma (early Late Plio- cene). PPP345 (CJ88-27-1) Cayo Agua, Bocas del Toro, be- tween Tiburon and Piedra Roja: 9°08'50.2’N, 82°00'54.4”W; ?Cayo Agua Formation (age un- certain). PPP346 (CJ88-27-2) Ibid.: 9°08'48.5”N, 82°00'56.6"W. PPP348 (CJ88-27-4) Ibid.: 9°08’45.3”N, 82°01'2.0"W. PPP350 (CJ88-28-1) Ibid.: 9°08'48.6’N, 82°00'53.1”W. PPP352 (CJ88-28-3) Ibid.: 9°08'48.4’”N, 82°00'50.3’W. PPP355 (CJ88-28-6) Ibid.: 9°08'47.1”N, 82°00'46.9’W. PPP357 (CJ88-29-2) Cayo Agua, Bocas del Toro, E end Pt. Piedra Roja: 9°08'35.3”N, 82°00'30.1"W; ?Cayo Agua Formation (age uncertain). PPP362 (CJ88-30-5) Escudo de Veraguas, Bocas del Toro, NW coast: 9°06'4.5”N, 81°34'18.3”W; Es- cudo de Veraguas Formation, |.8-1.9 Ma (Late Pliocene). PPP379 (CJ88-32-4) Valiente Peninsula, Bocas del Toro, NW side Bruno Bluff: 9°02'32.2’N, 550 8 1°44'42.0”W; Shark Hole Point Formation, 3.5- 3.6 Ma (early Late Pliocene). PPP442 (CJ88-59-1) Isla Popa, Bocas del Toro, NE coast: 9°12'57.7"”N, 82°06'24.7”"W; ?Cayo Agua Formation (age uncertain). PPP423 (CJ888-59-2) Ibid. PPP466 (CJ88-00-35) Limon, Costa Rica, near Pro- gressive Baptist Church: 9°59'28.2’"N, 83°02'29.9”W; age and formation unknown. PPP475 (CJ88-00-44) Cayo Agua, Bocas del Toro, SE of Pt. Norte: 9°10'26.3”N, 82°02'26.9”W; Cayo Agua Formation, 4.6—5.0 Ma (Early Pliocene). PPP627 (CJ89-14-1) Pueblo Nuevo, Limon, Costa Rica (Cerro Mocho subdivison): 9°59'51.0’N, 83°02'36.0”W; Moin Formation, 1.5—3.5 Ma (ear- ly Late to Late Pliocene). PPP634 (CJ89-16-1) Ibid. (cemetery): 9°59'28.2’N, 83°02'29.2”W; Moin Formation, |.9—2.4 Ma (Late Pliocene). PPP639 (CJ89-17-2) Lomas del Mar, Limon, Costa Rica (construction site): 9°59'46.3’N, 83°02'26.2”W; Moin Formation, |.7—1.9 Ma (Late Pliocene). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON PPP708 (CJ89-33-1) Santa Rita, Limon, Costa Rica (near stream): 9°58'24.5”N, 83°07'22.4’W; for- mation unknown, 2.4—3.4 Ma (early Late Plio- cene). PPP710 (CJ89-35-1) Limon, Costa Rica, Santa Edu- viges subdivision: 10°00'14.9”N, 83°02'30.9"W; Moin Formation, less than 1.8 Ma (Late Pliocene). PPP720 (CJ89-39-1) Santa Rita, Limon, Costa Rica (near stream): 9°58'10.3”N, 83°07'48.9"W; for- mation unknown, 2.2—-3.4 Ma (Late Pliocene). PPP738 (CJ89-47-1) Limon, Costa Rica, Hotel Olas: 10°00'41.1”N, 83°02'49.3”W; formation un- known, 0.01—1.7 Ma (Pleistocene). PPP757 (CJ89-00-19) Limon, Costa Rica aia del Mar construction site): 9°59’46.3”N, 83°02'26.2”W; Moin Formation, 1.7—1.9 Ma (Late Pliocene). PPP1101-1119 (CJ92-00-21 to 40) Ibid.: 9°59'31.0’N, 83°02'12.2”W; Moin Formation, |.7—1.9 Ma (Late Pliocene). USGS 8321 Costa Rica, north shore of Provision Is- land, Gatun Formation; Late Miocene. USGS 20468 Puerto Limon, Costa Rica; ?Late Plio- cene. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(3):551. 1995. BIOLOGICAL SOCIETY OF WASHINGTON 122nd Annual Meeting, 12 May 1995 The meeting was called to order at 12:10 p.m. in the Waldo Schmitt Room, National Museum of Natural History, by President Janet Reid. Janet asked Treasurer, Chad Walter, and Editor, Brian Robbins, for a summary of the year’s activities. Income for the period of 1 January 1994 to 31 December 1994 was $67,877.14, of which $37,190.00 was received for publication charges, $23,107.00 from dues and sub- scriptions, $4,331.00 from the sale of back issues and bulletins, and $3,249.14 from interest on the Society’s accounts. Expenditures were $66,153.57, of which $56,818.22 was as- sociated with publishing the Proceedings, the remainder with management costs and bank charges. The net income for the period was $1,723.57, but Chad noted that this figure does not reflect all income and expenditures associated with the publication of issue 107- 4. Financial loss per issue of Volume 107 resulting from unpaid page charges ranged from $2,626.76 for 107-3 to $8,003.54 for 107-4. Income from dues, subscriptions, and sales of past volumes of the Proceedings and Bulletin compensated for these losses, and there was no significant increase or decrease in the net worth of the Society. Four issues of Volume 107 of the Proceedings were published, comprising 78 papers and 779 pages. There were 87 submissions in 1994, the same as in 1993. As of 1 May 1995, there were 36 submissions, up slightly from 34 in 1994. Brian announced that effective 1 October 1995, Dr. Thomas A. Munroe, our Vertebrate Zoology editor will resign. He will be replaced by Dr. Gary R. Graves, Curator of Birds, NMNH. Dr. David L. Pawson has been named Chairman of the Nominating Committee for the officers and council members for the 1996 election. The Society notes with sorrow the death on May 3, 1995, of past president John W. Aldrich, who served from 1947-1949. Among his many services to the Society, Dr. Aldrich was the author of ““The Biological Society of Washington: a Centennial History 1880- 1980,” Bulletin of the Biological Society of Washington, No. 4:1—40, published 1980. The meeting was adjourned at 12:25 p.m. Respectfully submitted, Carole C. Baidwin, Secretary nD co rs 7 > le i a aay she aha * j - + > ated ou om - } = | TOT Sa 16 VTACO SA, ity ae eo, ° *, ) | ' : ; 7 a ete a : * - 7 : ice i 5 the ; 5 ) ax =e INFORMATION FOR CONTRIBUTORS Content.—The Proceedings of the Biological Society of Washington contains papers bearing on systematics in the biological sciences (botany, zoology, and paleontology), and notices of business transacted at meetings of the Society. Except at the direction of the Council, only manuscripts by Society members will be accepted. Papers are published in English (except for Latin diagnoses/descriptions of plant taxa), with an Abstract in an alternate language when appropriate. Submission of manuscripts.—Submit three copies of each manuscript in the style of the Proceedings to the Editor, complete with tables, figure captions, and figures (plus originals of the illustrations). Mail directly to: Editor, Proceedings of the Biological Society of Wash- ington, National Museum of Natural History NHB-108, Smithsonian Institution, Washington, D.C. 20560. (Do not submit manuscripts to an associate editor). Presentation.—Clarity of presentation, and requirements of taxonomic and nomenclatural procedures necessitate reasonable consistency in the organization of papers. Telegraphic style is recommended for descriptions and diagnoses. The style for the Proceedings is described in ““GUIDELINES FOR MANUSCRIPTS for Publications of the BIOLOGICAL SOCIETY OF WASHINGTON” a supplement to Volume 103, number 1, March 1990. Authors are encour- aged to consult this article before manuscript preparation. Copies of the article are available from the editor or any associate editor. The establishment of new taxa must conform with the requirements of appropriate interna- tional codes of nomenclature. Decisions of the editor about style also are guided by the General Recommendations (Appendix E) of the International Code of Zoological Nomenclature. When appropriate, accounts of new taxa must cite a type specimen deposited in an institutional col- lection. __ Review.—One of the Society’s aims is to give its members an opportunity for prompt pub- lication of their shorter contributions. Manuscripts are reviewed by a board of Associate Editors and appropriate referees. Proofs.—Authors will receive first proofs and original manuscript for correction and ap- proval. Both must be returned within 48 hours to the Editor. Reprint orders are taken with returned proofs. Publication charges.—Authors are required to pay full costs of figures, tables, changes in proofs ($3.00 per change or revision), and reprints. Authors are also asked to assume costs of page-charges. The Society, on request, will subsidize a limited number of contributions per volume. If subsidized manuscripts result in more than 12 printed pages, the additional pages must be financed by the author(s). Multiple authorship will not alter the 12 page limit (each author will be viewed as having used his/her 12 subsidized pages). Payment of full costs will facilitate speedy publication. Costs.—Printed pages @ $60.00, figures @ $10.00, tabular material @ $3.00 per printed inch per column. One ms. page = approximately 0.4 printed page. CONTENTS Nomenclature of the Hawaiian Akialoas and Nukupuus (Aves: Drepanidini) Storrs L. Olson and Helen F. James A new lizard of the genus Sphenomorphus (Reptilia: Scincidae) from Mt. Kitanglad, Mindanao, Philippine Islands Walter C. Brown A new species of Brachymeles (Reptilia: Scincidae) from Catanduanes Island, Philippines Walter C. Brown and Ely L. Alcala A new skink (Emoia: Lacertilia: Reptilia) from the forest of Fiji George R. Zug and Ivan Ineich Lovén’s law and adult ray homologies in echinoids, ophiuroids, edrioasteroids, and an ophio- cistioid (Echinodermata: Eleutherozoa) Frederick H. C. Hotchkiss Alloeocomatella, a new genus of reef-dwelling feather star from the tropical Indo-West Pacific (Echinodermata: Crinoidea: Comasteridae) Charles G. Messing Morellia dendropanacis, a new species, and other species with spotted wings: characterization and comparison (Diptera: Muscidae: Muscinae) Denise Pamplona and Marcia Souto Couri Within-species variation in Periclimenes yucatanicus (Ives), with taxonomic remarks on P. pedersoni Chace (Crustacea: Decapoda: Caridea: Palaemonidae) Mary K. Wicksten Dittosa, a new genus of leucosiid (Crustacea: Decapoda: Brachyura) from southern Australia and New Zealand Cheryl G. S. Tan Neocallichirus cacahuate, a new species of ghost shrimp from the Atlantic coast of Florida, with reexamination of N. grandimana and N. lemaitrei (Crustacea: Decapoda: Callianassidae) Darryl L. Felder and Raymond B. Manning Olavius nicolae, a new gutless marine tubificid species (Oligochaeta) from Belize Christer Erséus and Olav Giere Pionosyllis maxima Monro, 1930, P. anops Hartman, 1953, and P. epipharynx Hartman, 1953, redescribed as Eusyllis maxima (Monro, 1930), a new combination (Polychaeta: Syllidae: Eusyllinae) Maria Jiménez, G. San Martin, and E. Lopez A new freshwater snail from the Coosa River, Alabama (Gastropoda: Prosobranchia: Hydro- biidae) Fred G. Thompson The genus Laxus Cobb, 1894 (Stilbonematinae: Nematoda): Description of two new species with ectosymbiotic chemoautotrophic bacteria Jorg A. Ott, Monika Bauer-Nebelsick, and Veronica Novotny Neotropical Monogenoidea. 26. Annulotrematoides amazonicus, a new genus and species (Dac- tylogyridae: Ancyrocephalinae), from the gills of Psectrogaster rutiloides (Kner) (Teleostei: Characiformes: Curimatidae) from the Brazilian Amazon Delane C. Kritsky and Walter A. Boeger New records of azooxanthellate stony corals (Cnidaria: Scleractinia and Stylasteridae) from the Neogene of Panama and Costa Rica Stephen D. Cairns Biological Society of Washington: 122nd Annual Meeting STS 388 892 395 401 436 451 458 465 477 49] 496 502 508 528 IS 55 Il ISSN 0006-324Xx THE BIOLOGICAL SOCIETY OF WASHINGTON 1994-1995 Officers President: Janet W. Reid Secretary: Carole C. Baldwin President-elect: Stephen D. Cairns Treasurer: T. Chad Walter Elected Council ~ Robert J. Emry Susan L. Jewett Richard C. Froeschner Lynne R. Parenti Alfred L. Gardner F. Christian Thompson Custodian of Publications: Austin B. Williams PROCEEDINGS Editor: C. Brian Robbins Associate Editors Classical Languages: George C. Steyskal Invertebrates: Jon L. Norenburg Frank D. Ferrari Plants: David B. Lellinger Rafael Lemaitre Insects: Wayne N. Mathis Vertebrates: Thomas A. Munroe Membership in the Society is open to anyone who wishes to join. There are no prerequisites. Annual dues of $25.00 (for USA and non-USA addresses) include subscription to the Pro- ceedings of the Biological Society of Washington. Library subscriptions to the Proceedings are: $40.00 for USA and non-USA addresses. Non-USA members or subscribers may pay an ad- ditional $25.00 to receive the Proceedings by Air Mail. The Proceedings of the Biological Society of Washington (USPS 404-750) is issued quarterly. Back issues of the Proceedings and the Bulletin of the Biological Society of Washington (issued sporadically) are available. Correspondence dealing with membership and subscriptions should be sent to the Biological Society of Washington, P.O. Box 1897, Lawrence, Kansas 66044, U.S.A. Payment for membership is accepted in US dollars (cash or postal money order), checks on US banks, or MASTERCARD or VISA credit cards. Manuscripts, corrected proofs, editorial questions should be sent to the Editor, Biological Society of Washington, National Museum of Natural History, Smithsonian Institution, Wash- ington, D.C. 20560. Known office of publication: National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. ; Printed for the Society by Allen Press, Inc., Lawrence, Kansas 66044 Second class postage paid at Washington, D.C., and additional mailing office. POSTMASTER: Send address changes to PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON, P.O. Box 1897, Lawrence, Kansas 66044. © This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(4):553-559. 1995 Paraturbanella solitaria, a new psammic species (Gastrotricha: Macrodasyida: Turbanellidaeg), from the coast of California : M. Antonio Todaro Department of Zoology and Physiology, Louisiana State Uni crgity, | es Baton Rouge, Louisiana 70803, U.S.A. ‘ = NARIES X, Abstract.—A new species of marine gastrotrich is described. Adults of Par- aturbanella solitaria new species are simultaneous hermaphroditic, ribbon shaped, up to 530 wm long and up to 64 wm wide. The head bears a pair of sensorial piston pits and its anterior-most portion is hexagonally shaped. The adhesive apparatus consists of anterior and posterior adhesive tubules in ad- dition to bilateral ‘“‘SeitenfiiRchen”’ organs that take origin in the middle of the pharyngeal region. The new species is placed and discussed within the “‘teis- sieri’’ species assemblage. This is the first gastrotrich reported from the coast of California. Marine gastrotrichs in recent years have been the object of an increasing number of taxonomical and biogeographical studies; extensive faunistic surveys have been car- ried out in northern Europe (Hummon & Warwick 1992, Jouk et al. 1992), the Med- iterranean Sea (Hummon et al. 1990, 1992, 1993, 1994; Balsamo et al. 1992; Todaro 1992; Todaro et al. 1992), the Atlantic coast of the United States (Todaro et al. 1991; Evans 1992, 1994), and the northern Gulf of Mexico (Todaro 1994, Todaro et al. 1995). Unlike fauna of these regions, gas- trotrichs of the Pacific coast of the United States are poorly known, the only contri- butions being those of Wieser (1957) and Hummon (1966, 1969, 1972). Accounts of these authors were limited respectively to gastrotrichs from Puget Sound and the San Juan Archipelago, both within the state of Washington. In an attempt to expand the knowledge of the gastrotrich-fauna of the west coast of the United States, I arranged to obtain some sediment from California beaches. This paper deals with the descrip- tion of a new Paraturbanella, the only gas- trotrich species found in a sample of marine sand collected near Los Angeles. Materials and Methods Sand was collected on 26 November 1994, from Huntington Beach, California (Fig. 1). After digging a 30 cm deep hole at MLLW, 200 cm?’ of sediment was re- moved from the wall and bottom of the hole, placed in a plastic bag and shipped to the laboratory within 72 h. In the laboratory the sediment was kept in a cold-room at 14°C and processed within 3 days. Speci- mens were extracted by the narcotization- decantation technique using an isosmotic magnesium chloride solution (Pfannkuche & Thiel 1988). Supernatant was poured in 5-cm plastic dishes and gastrotrichs were located under a M 5 Wild dissecting micro- scope. Twenty-five sexually mature, living, relaxed individuals were transferred by a glass micropipette to slides and observed using either differential interference con- trast optics with a Microphot-FXA Nikon microscope or phase contrast with a Wild M 20 microscope. At that time gastrotrichs were photographed and/or recorded on S- VHS video tape. Measurements of all spec- imens observed were obtained form the mi- croscope using an ocular micrometer, or 554 e Los Angeles Santa Ana Fig. 1. Location of study site, Huntington Beach (arrow), in southern California. from photographs or video images. A type specimen has been deposited at the Nation- al Museum of Natural History, Washington, D.C., U.S.A. (USNM). Photographs and video images of several individuals are kept in the author’s collection (ref. no. PS23-28/ 94). Granulometric analysis of the sediment was carried out according to Giere et al. (1988). Mean grain size, sorting coefficient, kurtosis, and skewness were calculated by a computerized program based on the equa- tions of Seward-Thompson & Hail (1973). Order Macrodasyida Rao & Clausen, 1970 Family Turbanellidae Remane, 1925 Genus Paraturbanella Remane, 1927 Paraturbanella solitaria new species jee, Z Holotype.—An adult 460 wm long, for- malin-glycerin wholemount (USNM ref. no. 169956). Type locality.—Huntington Beach, Cali- fornia, U.S.A. (33°41'N; 118°24’W). Etymology.—The specific name solitaria Latin meaning lonely, alludes to the fact that this was the only gastrotrich species found in the sample. Diagnosis.—Paraturbanella with adults up to 531 wm long and 64 wm wide. Head bearing lateral sensorial piston pits and lacking ventral sensorial papillae. Anterior- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON most portion of the head hexagonally shaped. Eight to ten anterior and ten pos- terior adhesive tubules; lateral adhesive tu- bules absent. Caudal cone 7—9 pum long. Caudal lobe 40 wm long. “‘SeitenfiiBchen”’ organs at the middle of the pharyngeal re- gion, shorter tubules 13-17 pm, longer 25— 28 wm. Large, heavily cuticularized buccal cavity; pharynx up to 150 pm long with pharyngeal pores 14—16 wm from the pha- ryngeal-intestinal junction. 7 Description.—Miature specimens attain a total body length of 460-531 pm and a width of 47-64 wm (Fig. 2; Table 1). The body is transparent, dorsoventrally flattened and slightly tapering towards the posterior end where it is 23—33 wm wide. The head is 29-30 wm wide and somewhat hexagonal in shape (Fig. 2A). It bears marginal cilia as well as several sensory bristles 10-14 wm long. The posterior margin of the head is demarcated from the rest of the body by a slight constriction, 25—26 wm wide. Lo- cated just posterior to the constriction, on each side, is a sensorial piston pit 3.8 wm in diameter (Fig. 2A, C). No other sensorial organ (i.e., ventral sensorial papillae) is present. The adhesive apparatus consists of anterior and posterior adhesive tubules in addition to bilateral “‘SeitenfiiBchen”’ ad- hesive organs. Eight to ten anterior tubules (2-5 pm long) are arranged in two ventral symmetrical, hand-shaped, groups (Fig. 2C). Ten posterior tubules (6—21 jm long) are part of the 39-41 wm long caudal lobes. A caudal cone, 7-9 wm long, occurs be- tween the symmetrical caudal lobes (Fig. 2A, B). The two “‘SeitenfiiRchen’’ adhesive organs, literally “‘lateral foot,’ referred also as ‘“‘dorni” tube groups by Evans & Hum- mon (1991), consist each of two tubules of unequal length originating ventrolaterally, about in the middle of the pharyngeal re- gion, and directed backwards; the longer tube is 25—28 ym in length while the short- er one is 13-17 pm (Fig. 2C). The body, dorsally and laterally, bears 18—20 pairs of sensory bristles (9-12 wm long); ventrally it bears the locomotory cil- VOLUME 108, NUMBER 4 555 iii a ELLE EU {| An Cc || \\ Fig. 2. Paraturbanella solitaria new species. A. Habitus; B. Caudal lobes, ventral; C. Anterior end, ventral. D. Mid intestine, ventral. An, anus; At, anterior adhesive tubules; Bc, buccal cavity; Cc, caudal cone; Et, SeitenfiiBchen organs; Pp, pharyngeal pores; Sp, Sensorial piston pits. Scale bars represent 50 wm. 556 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Morphometrical parameters of Paraturbanella solitaria new species (measurements are in wm). Range x SD n Total body length 477.0—531.0 505.9 21.6 25 Caudal lobe length 35.2-41.1 BHD 2.4 24 Head width 29.4—30.4 29.8 0.5 23 Head constriction width 25.0—27.0 DSS) 0.8 23 Maximum trunk width 47-64 SIL) 6.6 19 Posterior trunk width 23.5—33.4 28.8 3.9 DS Pharynx length 125.0-144.0 132.0 DH 20 Distance of the pharyngeal pores from the pharyngo-intestinal junction 14.6—-17.6 15.6 1.1 21 Mouth length 19.4-21.6 19.6 123 20 Mouth width 10.5-11.7 10.9 0.4 19 SeitenfiiBchen longer tubule length 25-28.4 ZA 2 od) 7 SeitenfiiRchen shorter tubule length 11.7-17.6 15.6 D2) Wy Piston pits diameter 3.0-3.9 3.6 0.3 20 Caudal cone length 7.0-9.8 8.9 oll 19 Number of anterior tubules 8-10 9.6 0.8 25) Number of posterior tubules 10.0 10.0 0.0 25 X, mean value. SD, standard deviation. n, number of specimens studied. ia. The latter are arranged in two bands that run parallel to each other from the head to the base of the caudal lobes. The buccal cavity is spacious with a thick cuticular wall and measures 19-21 pm in length by 10.7—12 pm in width. The phar- ynx 1s 125-150 pm long and 15-16 pm wide; the pharyngeal pores open about 14— 16 wm from the pharyngeal-intestinal junc- tion (Fig. 2A). The intestine is straight and morphologically recognizable in two parts. The anterior part is wider and has a thinner wall compared to the posterior part. The anus opens ventrally at 24—30 pm from in- dentation between the caudal lobes. The re- productive system is made up of paired bi- lateral testes and probably bilateral ovaries. Testes originate 32—38 wm posterior to the pharyngeal-intestinal junction, thence ta- pering in sperm ducts (vasa deferentia) that run backward laterally to the intestine until the mid-trunk, where apparently they turn anteriorly to join at the midline, ventrally to the intestine, 65—70 wm posterior to the pharyngeal-intestinal junction (Fig. 2D). A single mature ovocyte, 55—70 wm in length, was visible dorsal to the mid-intestine in most of the specimens studied. Neither ad- ditional accessory reproductive organs or genital orifices have been observed. Habitat.—A large population of individ- uals representative of all age classes inhab- its the fine, silicious, clean sand of the in- tertidal zone of Huntington Beach shore. There, sand granules are of low sphericity, sub-angular and moderately well sorted (Fig. 3). At the time of the collection, water temperature and salinity were 19°C and 35 ppt respectively. Major associated taxa were nematodes and turbellarians. Remarks.—In these specimens the head lacks laterally projecting tentacles, the buc- cal capsule does not extend beyond the mouth, the anterior tubules are borne on fleshy projections (hands), and the ‘’Seiten- fiiBchen”’ organs are located in the mid- pharyngeal region. These characteristics, according to Evans & Hummon (1991), af- filiate them with the genus Paraturbanella. Among the thirteen species so far ascribed to this genus, because of the morphology of the cephalic region, the absence of lateral adhesive tubes and the morphology of the intestine, the present specimens resemble more closely to species that fall within the “teissier’’ group, namely: P. mesoptera VOLUME 108, NUMBER 4 Weight (%) 100 M = 2.11 SD = 0.73 891 Sk = -0.04 Ku = 3.5 seo, Vr = 0.31 40 20 igeS: 3)// Grain size (Phi) Granulometric analysis: Histogram by weight and cumulative curve. M, mean grain size; SD, sorting coefficient; Sk, skewness; Ku, kurtosis; Vr, roundness value. Rao, 1970, P. microptera Wilke, 1954, and P. teissieri Swedmark, 1954. The speci- mens from California differ from the ones described from India as P. mesoptera, in that they are of a larger size, 460-531 wm vs. 360-380 p.m, and in that they bear few- er adhesive tubules in both the anterior groups, 4—5 vs. 9, and in the caudal lobes, 5 vs. 9. The present specimens differ from those affiliated with the European P. teis- sieri in that they lack the ventral sensorial papillae and also because they bear a small- er number of anterior as well as posterior adhesive tubules, 8—10 vs. 12 and 10 vs. 12—20 respectively. Recently Kisielewski (1987), discussed synonymizing P. microp- tera (originally found near Naples, Italy) with P. teissieri noticing that the two taxa were described almost simultaneously and that both species were considered by their authors as the second within the genus, and consequently only differences from P. dor- ni Remane, 1927, were included in their discussions. This clearly indicates that Swedmark and Wilke were unaware of each other’s findings. The possible synonymy of P. teissieri and P. microptera was men- tioned previously by Schmidt & Teuchert (1969), and an oversight of piston pits (one of the possible differences between the two species) by Wilke was suggested by these authors. On this subject Kisielewski (1987), pointed out that the piston pits were not mentioned either in the original description of P. teissieri, and thus, their later findings could not be used as an argument for sep- arating both forms. Kisielewski’s conclu- sions were agreed with by Todaro et al. (1993) on the basis of an extensive faunistic survey around the Italian peninsula that yielded only specimens whose morpholog- ical characteristics were in full accordance with the description of the north European P. teissieri (M. Balsamo, W. Evans, W. D. 558 Hummon, M. A. Todaro & P. Tongiorgi, un- publ. data). Among the morphological traits shared between the Italian and the north Eu- ropean specimens, were also sensorial pis- ton pits and sensorial papillae. Since neither one of these organs were reported in the original description of P. microptera it was assumed by Todaro et al. (1993), that they were both overlooked by Wilke. On this ground, therefore, I think it is appropriate to affiliate the Californian specimens with a new taxon, proposing the name Paratur- banella solitaria new species for it. Acknowledgments I am indebted to Neil Scott, surfer and friend, for providing me with the sand from Huntington Beach. I am grateful to Dr. John W. Fleeger for his encouragement and sup- port. The paper benefitted from the com- ments of Maria Balsamo, Wayne Evans, William Hummon and Paolo Tongiorgi. Literature Cited Balsamo, M., M. A. Todaro, & P. Tongiorgi. 1992. Marine gastrotrichs from the Tuscan Archipel- ago (Tyrrherian Sea). II. Chaetonotida, with de- scription of three new species.—Bollettino di Zoologia 59:487-498. Evans, W. A. 1992. Five new species of marine Gas- trotricha from the Atlantic coast of Florida.— Bulletin of Marine Science. 51:315—328. 1994. Morphological variability in warm- temperate and subtropical population of Macro- dasys (Gastrotricha: Macrodasyida: Macroda- syidae) with description of seven new spe- cies.—Proceedings of the Biological Society of Washington 107:239—255. , & W. D. Hummon. 1991. A new genus and species of Gastrotricha from the Atlantic coast of Florida, U.S.A.—Transactions of the Amer- ican Microscopical Society 110:321—327. Giere, O., A. Eleftheriou, & D. J. Murison. 1988. Abi- otic factors. Pp. 134-145 in R. P. Higgins & H. Thiel, eds., Introduction to the study of meio- fauna. Smithsonian Institution Press, Washing- ton, D.C. Hummon, W. D. 1966. Morphology, life history, and significance of the marine gastrotrich Chaeton- otus testiculophorus n. sp.—Transactions of the American Microscopical Society 85:450-457. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 1969. Musellifer sublitoralis, a new genus and species of Gastrotricha from the San Juan Archipelago, Washington.—Transactions of the American Microscopical Society 88:282—286. . 1972. Dispersion of Gastrotricha in a marine beach of the San Juan Archipelago, Washing- ton.—Marine Biology 16:349-355. , & R. M. Warwick. 1990. The marine flora and fauna of the Isles of Scilly—Gastrotricha.— Journal of Natural History 24:519-525. , M. Balsamo, & M. A. Todaro. 1992. Italian marine Gastrotricha: I. Six new and one rede- scribed species of Chaetonotida.—Bollettino di Zoologia 59:499_-5 16. , M. R. Hummon, & M. H. Mostafa. 1994. Marine Gastrotricha of Mediterranean Egypt.— American Zoologist 34:10A (Abstract). , M. A. Todaro, & M. R. Hummon. 1990. Ma- rine Gastrotricha of the central Mediterranean Sea.—American Zoologist 31:20A (Abstract). , & P. Tongiorgi. 1993. Italian marine Gastrotricha: II. One new genus and ten new species of Mascrodasyida.—Bollettino di Zool- ogia 60:109-127. Jouk, P. E. H., W. D. Hummon, M. R. Hummon, & E. Roidou. 1992. Marine Gastrotricha from the Belgian coast: species list and distribution.— Bulletin de I’Istitut Royal des Sciences Natu- relles de Belgique 62:87—90. Kisielewski, J. 1987. New records of marine Gastro- tricha from the French coasts of Manche and Atlantic. I. Macrodasyida with description of seven new species.—Bulletin Museum National d’ Histoire Naturelle, Paris 9:837—877. Pfannkuche, O., & H. Thiel. 1988. Sample process- ing. Pp. 134-145 in R. P. Higgins & H. Thiel, eds., Introduction to the study of meiofauna. Smithsonian Institution Press, Washington, D.C. Schmidt, P., & G. Teuchert. 1969. Quantitative unter- suchungen zur okologie der gastrotrichen im gezeit-sandstrand der insel Sylt-—Marine Biol- ogy 4:4—23. Seward-Thompson, B. L., & J. R. Hails. 1973. An appraisal of the computation of statistical pa- rameters in grain size analysis.—Sedimentology 20:161—-169. Todaro, M. A. 1992. Contribution to the study of the Mediterranean meiofauna: Gastrotricha from the Island of Ponza, Italy—Bollettino di Zool- ogia 59:321—333. 1994. Chaetonotus triacanthus and Heter- oxenotrichula texana, two new chaetonotid gas- trotrichs from the Gulf of Mexico.—Transac- tions of the American Microscopical Society 113:15-21. , M. Balsamo, & P. Tongiorgi. 1992. Marine gastrotrichs from the Tuscan Archipelago (Tyr- VOLUME 108, NUMBER 4 rhenian Sea): I. Macrodasyida with description of three new species.—Bollettino di Zoologia 59:471-485. , W. A. Evans, & W. D. Hummon. 1991. Ma- rine Gastrotricha from Florida: status—Ameri- can Zoologist 31:107A. (Abstract). , J. W. Fleeger, & W. D. Hummon. 1995. Ma- pe | rine gastrotrichs from the sand beaches of the northern Gulf of Mexico.—Hydrobiologia (in press). Wieser, W. 1957. Gastrotricha Macrodasyoidea from the intertidal of Puget Sound.—Transactions of the American Microscopical Society 76:372-— 381. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(4):560—567. 1995 Rissoella ornata, a new species of Rissoellidae (Mollusca: Gastropoda: Rissoelloidea) from the southeastern coast of Brazil Luiz Ricardo L. Simone Secao de Moluscos, Museu de Zoologia da Universidade de Sao Paulo, Caixa Postal 7172, CEP 01064-970, Sao Paulo, SP, Brazil Abstract.—Rissoella ornata, a new species, is described from the northern coast of Sao Paulo State, Brazil. Characters of the shell and anatomy are de- scribed, demonstrating, among several interesting anatomical characters, that this species has eyes on small stalks with tentacles far removed from them, a special chamber in the right side of the mantle cavity delimited by the thickness of the mantle and head skin, simultaneous hermaphrodite gonad, the gonoducts free in the haemocoel of the body and two genital pores (a small penis and a female pore) behind the head. The analysis of these characters suggests a re- lationship of the Rissoellidae with the Euthyneura-Pulmonata. Only one species of the genus Rissoella J. E. Gray, 1847 (type species R. glaber J. E. Gray, 1847), Rissoelloidea, Rissoellidae, which comprises very small animals, is known from Brazil. This is Rissoella cari- baea Rehder, 1943 (Rios 1985:40, 1994: 181). In studies on associated fauna of the cor- al Mussismilia hispida (Verrill), carried on by Joao Miguel M. Nogueira (doctoral the- sis) on the Sao Paulo coast in the years 1993-1994, several specimens belonging to the genus Rissoella were found. Analysis of the shell characters suggested these speci- mens are a new species. Probably due to miniaturization (Fretter 1948), and consequent simplification of structures, the systematic placement of the Rissoellidae has been problematic. They were, for example, included in the Meso- gastropoda (Rios 1985), Heterogastropoda (Ponder & Yoo 1977), Neogastropoda (Vaught 1989) and Opisthobranchia (Fretter & Graham 1954) (for a discussion on this subject see Ponder & Yoo 1977). Haszpru- nar (1985:29) considered that the Rissoel- lidae appears to represent an intermediate level of organization between the Proso- branchia and the Heterobranchia. Haszpru- nar (1988) re-analyzed the position of the Rissoellidae, considering the group as part of a superfamily within Allogastropoda (Heterobranchia). This systematic problem has been aggravated by scanty knowledge of the anatomy of the Rissoellidae, for only two species have been described anatomi- cally (Fretter 1948). The family Rissoellidae includes only one genus, Rissoella, which was subdivided in four subgenera (Ponder & Yoo 1977), mainly characterized by the radula. Materials and Methods The specimens were collected by scuba diving, fixed in 4% formalin, preserved in 70% ethanol, and deposited in ““Museu de Zoologia da Universidade de Sao Paulo” (MZUSP) collection. Twenty specimens for anatomical studies were decalcified in Railliet-Henry fluid. Some of them were dehydrated in ethanol series, stained in carmine, cleared and fixed in creosote. Serial sections of three speci- mens were stained in haematoxylin and eo- VOLUME 108, NUMBER 4 sin. Shells, radulae, jaws and opercula were examined under SEM, in the ‘‘Laboratorio de Microscopia Eletrénica do Instituto de Biociéncias da Universidade de Sao Pau- lo’’, using the technique described by So- lem (1970, 1972). All drawings were made with the aid of a camera lucida. Systematic and shell terminologies were based on Pon- der & Yoo (1977), and the anatomical ter- minology on Fretter (1948). In the figures the following abbreviations are used: ag: albumen gland an: anus cg: capsule gland da: anterior lobe of the digestive gland dc: duct to capsule gland dd: duct of posterior lobe of the diges- tive gland dg: posterior lobe of the digestive gland ec: egg covering ey: eye fa: fold parallel to mantle border fc: phaecal chamber fh: flap of the head adjacent to the phaecal chamber fm: fold perpendicular to the mantle border fp: female genital opening ft: foot gi: gill vestiges hd: hermaphrodite duct hg: hypobranchial gland vestiges hs: head-foot skin in: intestine ja: jaw ki: kidney ll: left lobe of the columellar muscle Is: longitudinal posterior slit of the foot mantle border od: odontophore oe: oesophagus og: opercular peg op: operculum outer wall of egg capsule pe: posterior lobe of the capsule gland pe: penis 561 pt: prostate ra: radula rl: right lobe of the columellar muscle rt: rectum sn: snout Ss: style sac st: stomach te: tentacle vd: vas deferens vm: visceral mass yg: yolk granules of the egg Systematics Rissoella (Rissoella) ornata, new species Figs. 1-18 Types.—Holotype MZUSP 28004 (ength 1.00 mm by 0.60 mm). Paratypes: MZUSP 28005, 1 specimen (1.10 mm by 0.65 mm); MZUSP 28006, 1 specimen (1.10 mm by 0.68 mm); MZUSP 28007, 5 Shells (1.40 by 0.71 mm; 1.00 by 0.62 mm; 0.90 by 0.55 mm; 0.83 by 0.53 mm); MZUSP 28008, | shell (1.20 by 0.61 mm); MZUSP 28009, 13 specimens and anatom- ical slides; MZUSP 28010, 24 specimens; 28012, 7 specimens; 28013, 25 specimens; 28015, 24 specimens; all these from type locality. MZUSP 28014, 3 specimens, Sao Paulo, Sao Sebastiado, Buzios Island. MZUSP 28011, 1 specimen, Sao Paulo, Ubatuba, Palmas Island. Museu Oceano- grafico da Fundagao Universidade de Rio Grande: MORG 32289 (2 specimens from type locality). Museu Nacional da Univer- sidade Federal do Rio de Janeiro, MNRJ 6934 (2 specimens from type locality). Type locality.—Brazil, Sao Paulo State, Sao Sebastiao Municipality, Vitoria Island, 23°45'S 45°O1'W. Diagnosis.—South Atlantic species with tall spire, smooth surface of the shell, deep suture, umbilicus deep and bordered by thin walls. Description.—Shell: minute (to 1.5 mm), thin, tall, smooth and shiny (Figs. 1, 2, 14), transparent, coloriess. Protoconch with one smooth whorl, relatively large (Fig. 3). Spire length slightly longer than the length 562 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON lata) . 5 ‘ F . feo tey®, Noes ‘ Rey we eevee y *% ‘4 we + ase pve? eBente Fey are CER a ea Pabher,, ba pie ag Pp betes, 09a * She wv Se ao reeds eg gh bape Vee vi, bbe Pp PT vy, Preserve Hh, ge a wee ? y ete “ i TU Grdgense Miers a “*” i A ee 2 , “Ge voe es » * Recs, Stye - ‘ ° ‘¢ OS Pee ed : y @aore eo.” e. CU ar) Bae Rater soe? ; ‘ Secea@ens)/ os “\ Sesiigiee Py « es eae! CT ae . t ad ie “4°, ‘ ostoe i” ge . Rug > ” ( Figs. 1-6. Rissoella ornata. 1, frontal view of a paratype shell, scale = 200 wm; 2, the same for other specimen, scale = 200 pm; 3, detail of the specimen of Fig. 2, showing a profile of the protoconch and first teleoconch whorl, scale = 50 wm; 4, inner view of the operculum, scale = 100 wm; 5, outer view of the operculum, scale = 50 pm; 6, detail of a jaw plate, scale 5 wm. of the aperture. Teleoconch to three convex whorls. Suture deep. Surface with distinct growth lines, slightly undulated axially. Periostracum thin, velvet-like (Fig. 3). Ap- erture oval, of moderate size (discretely smaller than a half of total shell length). Inner lip slightly thickened. Umbilicus con- spicuous, externally bordered by a sharp edge (Figs. 1, 2, 14). Outer lip rounded and simple. Operculum: somewhat rounded, occupies entire aperture, pale-yellow, transparent. Externally with concentric growth lines, nu- cleus near mid region of the inner edge (Fig. 5). Internally (Fig. 4), the slightly con- vex columellar edge is bordered on its inner side by a sharp ridge from which arises a short, blunt peg; a short, rounded ridge aris- es from the base of the peg and passes across less than half of the inner surface of the operculum at right angles to the colu- mellar edge. Outer edge simple, convex. Head-foot: yellowish-white with brown pigment on lateral region of the foot and around eyes (Fig. 15). Snout bifid (Figs. 11, 13), with two rounded lobes and two ten- tacles, which arise one at the base of each lobe; tentacles cylindrical, tapering slightly towards a blunt tip (Figs. 11, 13). Eyes well-developed, dark, on short stalks far back on the neck; lens massive and very- large. Foot slightly lanceolate, its opercular VOLUME 108, NUMBER 4 563 Figs. 7-8. lobe on each side projects beyond the lat- eral margins of the sole. Posterior pedal mucous gland opening in a median longi- tudinal slit sited in posterior half of the ped- al sole (Figs. 13, 16). Head-foot greatly compressed posteriorly by the developed capsule gland (Fig. 13); upon dissection, a large concavity appears separating the head-foot from visceral mass, which are united only by oesophagus. Columellar muscle distally bilobed (Fig. 16), oesopha- gus and the female duct running between both lobes; right lobe thick, left lobe thin (Fig. 16). Visceral mass: visible through the trans- parent colorless shell (Fig. 15); in the pos- terior half of each whorl a yellowish-white digestive gland is present, in the anterior half of each whorl a beige gonad, in the form of several arcs (with convexity pos- terior) (Fig. 15). Some specimens (MZUSP 28015) have a dark pigment in the mantle. In these specimens, the above cited struc- tures are difficult to see. Pallial cavity: shallow (about a half of whorl) (Figs. 9, 10), mostly occupied by the female glands (albumen and capsule glands). No special structures found, except the faecal chamber (described bellow) on right side, and gill vestiges on left extremity (Fig. 9). These gill vestiges, which have Rissoella ornata. Details of the radular teeth, 7, scale = 10 pm; 8, scale = 5 pm. functional cilia, present great variation in number and form of the leaflets. Mantle border rather thickened, without tentacles or siphon (Fig. 10). Digestive system: jaw in two _ lateral plates (Fig. 11), which have several scales turned backward, each scale has an aligned series of minute cusps on its cutting edge (Fig. 6). Odontophore very long, cylindrical (Fig. 11). Radula short—about 10—12 rows; rachidian tooth large, wide, with convex cutting edge bearing about 10 similar-sized cusps (Figs. 7, 8); lateral teeth large, con- vex, triangular, with 4-6 pairs of small sharp cusps and one large terminal cusp (Fig. 8); inner marginal! teeth similar to (but smaller than) the lateral teeth (Fig. 7); outer marginal teeth missing. Oesophagus origi- nates in ventral-right region of the odonto- phore, flattened and wide (Fig. 11), skirts the columellar muscle, inserting in mid- ventral region of the stomach (Figs. 12, 13, 17). Stomach (Figs. 12, 13, 17) large, rather flattened, with style sac differentiated. Di- gestive gland with two lobes (Fig. 12), one small and anterior, bound posteriorly by the stomach, ventrally by the style sac, dorsally by the kidney and anteriorly by the albu- men gland. The large and posterior lobe of the digestive gland, which opens ventrally in stomach (Fig. 12), spreads through the 564 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Figs. 9-13. Rissoella ornata. 9, inner view of the pallial cavity, mantle border in frontal view; 10, mantle border in inner view, showing the faecal chamber; 11, ventral view of the head, foot removed, showing the anterior region of the digestive and genital systems; 12, ventral view of the posterior half of the first whorl and the second whorl; 13, dorsal-right view of the head-foot and stomach. Scales = 100 wm (see Materials and Methods for abbreviations). posterior half of the coils of the visceral mass (Fig. 15). Intestine a wide tube, orig- inates near the insertion of the oesophagus in stomach (Figs. 13, 17) and lies ventrally to the style sac and the albumen gland in the right margin of the pallial cavity (Fig. 12), running in this cavity for a short dis- tance and opening into a special faecal chamber (Figs. 9, 10), which is located in the right angle of the mantle edge. This chamber is delimited (Fig. 10) posteriorly by a mantle foid from anus until a short VOLUME 108, NUMBER 4 565 Figs. 14-18. extracted from the shell; 16, ventral view of the posterior half of the foot and the columellar muscle, anterior half of the foot extracted; 17, ventral-right view of the stomach; 18, mature egg capsule. Scales = 100 pm (see Materials and Methods for abbreviations). distance to the left (fa); right and dorsally by the mantle wall; on the left by a mantle fold rather perpendicular to the mantle bor- der (fm); and ventrally by a flap of the head skin (fh). Reproductive system: gonad (Fig. 15) on the columellar side of the visceral mass (de- scribed above), hermaphrodite, with sperm and ova developed in the same tubules. Gonadial duct single (Fig. 12) and coiled with thick walls, running along the colu- mella until the posterior end of the body whorl, where it divides into two divergent branches. One of these branches leads to the pallial vas deferens (vd) and the other to the albumen gland (ag). Vas deferens, initially narrow and _ thin- walled (Fig. 12), anteriorly passes forward on the right side of the head. Posterior to the odontophore it becomes larger, with Rissoella ornata. 14, shell of a young specimen in frontal view; 15, outer view of a specimen thick walls and rather coiled (prostate) (Fig. 11), turns at an opened angle along the dor- sal surface of the body whorl to the right side, where it ends in a relatively short tu- bular penis, which has a rather pointed tip (Figs. 11, 13). Albumen gland amorphous (Fig. 12), of considerable size, flattened between the kidney and the capsule gland. Capsule gland (Figs. 11, 13) well-developed, at- tached to the head-foot complex and not to the mantle. Between the albumen and cap- sule glands a flattened posterior lobe of the capsule gland is generally present (Fig. 13: pc). Anteriorly the capsule gland leads for- ward above the oesophagus and to the right side of the head, projecting from the dorsal wall as an opaque white mass (Fig. 11). The capsule gland opens into the right side of the head by way of a small pore, which is 566 anterior-ventral to the root of the penis, be- hind the tentacles (Figs. 11, 13: fp). Development: the proportionally large egg capsules are manufactured apparently one at a time in the pallial oviduct, when developed, occupies most of the space with- in the capsule gland. The egg capsules (Fig. 18) are similar to that described by Fretter (1948) for Rissoella diaphana (Alder) and R. opalina (Jeffreys). The capsule contains only one egg. Outer wall thick and trans- parent, between this and the thin transparent membrane, which covers the egg, a trans- parent fluid (Fig. 18). About half of exam- ined specimens have a developed capsule within the capsule gland. Even small spec- imens have developed gonad and egg cap- sules. Animals of several sizes are found together, which probably indicate that R. or- nata is not annual like its European rela- tives (Fretter 1948). Habitat.—R. ornata was found moving on coral Mussismilia hispida, about 5 m depth. According to the literature, other species of Rissoella live on and feed upon intertidal algae. Range.—Brazil, sub tidal level, northern coast of Sao Paulo State. Etymology.—The contrast between the dark gonad and the clear digestive gland, visible through the shell, gives the effect of each whorl is ornamented (it. ornata) by several arches (Fig. 15). Discussion Rissoella (R.) ornata differs from R. (Jef- freysilla) caribaea in having a longer spire, deeper suture, smaller body whorl and the umbilicus bordered by thin walls. Using the diagnosis of the Rissoella sub- genera by Ponder & Yoo (1977) the sub- generic allocation of R. ornata is doubtful as this species has radular similarity to both Rissoella s.s. and Jeffreysiella Thiele, 1912. The shape of the lateral and inner marginal teeth, and the similarity between both, re- semble Jeffreysiella; but the absence of the PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON outer marginal teeth and the shell shape re- semble Rissoella s.s. Based on the anatomical knowledge of three species: Rissoella diaphana, R. opal- ina (apud Fretter 1948) and R. ornata de- scribed herein, some interesting comments on the relationship of the Rissoellidae may be done. On the one hand, the Rissoellidae have small size, palcispiral operculum, snout, small gill (although vestigial), ripi- doglossate-like radula, large stomach with style sac, oesophagus insertions near intes- tine origin, among other characters, which reveal some similarity with the Rissoidea (= Hydrobioidea, “‘basal’’ Cenogastropo- da). On the other hand, the Rissoellidae have eyes on small stalks with tentacles far removed from them, a special chamber in the right side of the mantle cavity delimited by the thickness of the mantle and head skin, jaws, a peculiar arrangement of the digestive glands, simultaneous hermaphro- dite gonad, the gonoducts free in the hae- mocoel of the body and two genital pores (a male and a female) behind the head, which show similarity with the Pulmonata. The gonoducts free in the haemocoel of the body has been considered as a character of the “higher” Heterobranchia (Haszprunar 1985) and synapomorphy of the Euthyneura (Haszprunar 1988). Based on this fact, the Rissoelloidea may be considered as Euthy- neura. Acknowledgments Special thanks to Joao Miguel M. No- gueira for collecting material. Literature Cited Fretter, V. 1948. The structure and life history of some minute prosobranchs of rock pools: Skeneopsis planorbis (Fabricius), Omalogyra atomus (Phi- lippi), Rissoella diaphana (Alder) and Rissoella opalina (Jeffreys).—Journal of the Marine Bi- ological Association, United Kingdom 27:597— 632. , & A. Graham. 1954. Observations on the op- istobranch mollusc Acteon tornatilis (L.).— Journal of the Marine Biological Association, United Kingdom 33:565—585. VOLUME 108, NUMBER 4 Haszprunar, G. 1985. The Heterobranchia—a new concept of the phylogeny of the higher Gas- tropoda.—Zeitschrift fiir Zoologische Systema- tik und Evolutionsforschung 23(1):15—37. . 1988. On the origin and evolution of major gastropod groups, with special reference to the Streptoneura.—Journal of Molluscan Studies 54:367-441. Ponder, W. F, & E. K. Yoo. 1977. A revision of the Australian species of the Rissoellidae (Mollus- ca: Gastropoda).—Records of the Australian Museum 31:133-185. Rehder, H. A. 1943. New marine mollusks from the Antillean Region.—Proceedings of the United States National Museum 93(3161):187—203. 567 Rios, E. C. 1985. Sea shells of Brazil. Museu Ocean- ografico, FURG, Rio Grande, 329 pp. + 102 pls. 1994. Sea shells of Brazil, second edition. Museu Oceanografico, FURG, Rio Grande, 368 pp. + 113 pls. Solem, A. 1970. Malacological applications of scan- ning electron microscopy I, introduction and shell surface features.—Veliger 12:394—400. . 1972. Malacological applications of scanning electron microscopy II, radular structure and functioning.—Veliger 14:327—336. Vaught, K. C. 1989. A classification of the living Mollusca. In R. T. Abbott & K. J. Boss, eds. American Malacologists, Inc., Melbourne, 189 PP. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(4):568-576. 1995 Description of a new viviparous species of Dentatisyllis (Polychaeta: Syllidae) from Belize with an assessment of growth and variation, and emendation of the genus David E. Russell Department of Biology, Washington College, 300 Washington Avenue, Chestertown, Maryland 21620-1197, U.S.A. Abstract.—Dentatisyllis mangalis, a new viviparous species of syllid poly- chaete, is described from mangrove and adjacent shallow-water habitats of Twin Cays, Belize. The new species is distinguished by a combination of short dorsal cirri with only a few articles and compound falcigers with long tapering blades bearing bifid tips. Analysis of meristic and linear characters used in classical syllid taxonomy revealed that the mean number of articles per dorsal cirrus per individual increases only slightly with increasing body length (growth); proventricle length increases linearly with body length; and the ratio of proventricle length to width is relatively independent of body length. Fur- thermore, overall growth is accomplished more by the addition of segments than the elongation of existing ones. The genus diagnosis is emended to include the presence of nuchal organs. An investigation of the diversity and dis- tribution of syllid polychaetes inhabiting Shallow-water habitats at Twin Cays, Be- lize, has recorded a number of known and previously unknown species of the subfam- ilies Exogoninae and Eusyllinae (Russell 1989a, 1989b, 1991). Species of the sub- family Syllinae collected from these habi- tats include the new viviparous species of Dentatisyllis Perkins, 1981, described here- in. The new species is illustrated and the variability of selected linear and meristic characters is evaluated. The viviparous mode of reproduction, exhibited by the new species and docu- mented herein, is particularly noteworthy. Brooding, however, is quite common in syl- lids, and internal brooding with direct de- velopment (viviparity) has been reported (including the present paper) for seven spe- cies (San Martin 1984, Wilson 1991). Materials and Methods Specimens were obtained from core sam- ples collected along a 40-m transect span- ning a continually submerged mangrove forest, a shallow unshaded algal flat, and a turtlegrass (Thalassia testudinum) meadow. Depth along the transect at time of sam- pling ranged from 9 to 130 cm. Further de- scription of the habitats sampled and expla- nation of the core sample labels (used be- low under “‘Type material’) are provided by Russell (1989a). Material was fixed in a seawater solution of approximately five percent formaldehyde and one percent methanol, stained with Rose Bengal, and preserved in 70% etha- nol. Specimens were examined by mount- ing them temporarily on slides in a medium of glycerin and 70% ethanol (1:1). Linear measurements were made using an ocular micrometer. Body length was measured from the posterior margin of the tentacular segment to the posterior tip of the pygidium; body width across the proventri- cle without the parapodia. The mean num- ber of articles per dorsal cirrus was deter- mined for each type specimen by counting the articles in one dorsal cirrus on each se- VOLUME 108, NUMBER 4 tiger, provided a dorsal cirrus was present and could be seen in its entirety. Occasion- ally both dorsal cirri of one setiger were counted and the number of articles com- prising each cirrus differed. In such in- stances the larger number was used in the calculation of the mean to ensure that the data was not biased toward the conclusion that the species is characterized by short (few articles) dorsal cirri, a possibility that was of particular interest. The dorsal cirri on setiger 1, which appeared to always be longer than other dorsal cirri, and the dorsal cirri on the last three setigers, which typi- cally were shorter than others, were not in- cluded in the mean. It was assumed that the number of articles comprising these cirri was strongly determined by their position on the body, and that, in individuals with only a few dorsal cirri (e.g., small individ- uals or individuals that have lost dorsal cir- ri), including these cirri in the mean would result in a value that was not representative. Consequently, they were omitted from counts on all individuals. Summary statistics and the product-mo- ment correlation coefficient (7) were cal- culated using Microsoft Excel (version 4.0). Prior to the calculation of r for the relation- ship between body length and the ratio of proventricle length to width, the latter was transformed using a log, transformation due to the tendency of ratios not to be nor- mally distributed (Sokal & Rohlf 1981). Type material has been deposited in the National Museum of Natural History (USNM), Smithsonian Institution, Wash- ington, D.C., U.S.A. Comparative material was borrowed from the Museo Nacional de Ciencias Naturales de Madrid (MNCNM), Madrid, Spain. Systematics Family Syllidae Grube, 1850 Subfamily Syllinae Rioja, 1925 Genus Dentatisyllis Perkins, 1981, emended Diagnosis.—Holotype of the type spe- cies, Dentatisyllis carolinae (Day, 1973) 569 with two ciliated slits along the posterior margin of the prostomium, one on either side of the dorsal midline, each indicating the presence of a nuchal organ. Remarks.—Nuchal organs were reported absent in the original generic diagnosis (Perkins 1981), which is otherwise com- plete and accurate. Dentatisyllis mangalis, new species Figs. 1-3; Tables 1, 2 Dentatisyllis brevicirra—Wilson, 1991:506 nomen nudum. Type material.—Holotype: T-4LB (USNM_ 102495). Paratypes: M-3, 1 (USNM 102496); C-3, 2 (USNM 102499); C-5, 1 (USNM 102500); C-8, 1 (USNM 102502); T-2LE 3 (USNM_ 102510); T- 2LB, 2 (USNM 102511); T-4UB, 1 (USNM 102512); T-4LB, 4 (USNM 102513); T- SUB, 1 (USNM 102514); T-SLE 1, (USNM 102515); T-7E 3 (USNM 102516); T-7B, 3 (USNM 102517); T-9F 2 (USNM 102518); T-9B, 2 (USNM 102519); T-11K 3 (USNM 102520); T-11B, 4 (USNM 102521). All types from West Bay, Twin Cays, Belize (16°50'N, 88°05’W); 9-130 cm depth; col- lected Nov 1983. Additional material examined.—Denta- tisyllis carolinae (Day, 1973), holotype (USNM 43146); 4 paratypes (USNM 43147) off Beaufort, North Carolina, North Atlantic: 20 m, May 1965. Dentatisyllis sp. A of Uebelacker, 1984, 1 specimen (USNM 65669), id. by Barry Vittor and Associates, off Crystal River, Florida, Gulf of Mexico: 38 m, Nov 1977. Dentatisyllis junoyi Lopez & San Martin, 1992, holotype (MNCNM, 16.01/802), off Curral Velho, Boavista Is- land, Cape Verde Islands, North Atlantic, 15 m, Aug 1985; paratype (MNCNM, 16.01/803), off Salamanza, Sao Vicente Is- land, Cape Verde Islands, North Atlantic. Description.—All type specimens com- plete. Body slender and approximately oval in cross-section; pale yellow without mark- ings; length without palps, prostomium, and tentacular segment, 0.8—3.9 mm; width 570 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Summary statistics for selected features of Dentatisyllis mangalis. Feature Range xX SD n Body length (mm) 0.83.9 yall 0.7 35 width (mm) 0.1—0.2 0.17 0.32 18 number of setigers 19-47 33 7.4 35 Proventricle length (wm) 200-430 308.6 61.1 35 width (wm) 70—150 103.6 ied, 33 length/width 2.1—3.9 3.0 0.4 33 Number of articles median antennae 7-13 10.6 137/ 28 lateral antennae 4-8 6.1 ID 378 dorsal tentacular cirri 4—10 V3 1.8 248 ventral tentacular cirri 2-5 3.0 2 SE dorasl cirri of setiger #1° 4-10 VS 1.6 28 anal cirri 4-10 6.4 le} 33 a Some on same individual. > Summary statistics for other dorsal cirri are presented in Table 2. across proventricle without parapodia 130— 230 wm; number of setigers 19-47. (Table 1 provides descriptive statistics for these and other quantitative features.) Prostomium oval, 1.5—3 times wider than long, with two pairs of lensed eyes on pos- terior half of prostomium in a flattened trap- ezoidal arrangement (Fig. 1A). A third pair of smaller eyes (eyespots) present on ante- rior portion of prostomium, one at the base of each lateral antenna. Median antenna with 7—13 articles, arising between poste- rior pair of eyes; lateral antennae with 4-8 articles, originating on anterior portion of prostomium. Palps triangular, free for most of length; appear strongly contracted in most specimens. A pair of nuchal organs, each in a narrow ciliated groove on either side of the dorsal midline along posterior margin of prostomium (Fig. 1A). Dorsal tentacular cirri with 4—10 articles; ventral tentacular cirri with 2—5 articles. Dorsal cirri on setiger 1 longest with 4-10 articles, all other dorsal cirri with 3-8 arti- cles, fewer in posteriormost 3-4 setigers. Mean number of articles per dorsal cirrus increases with body length (Fig. 3A). (Table 2 provides descriptive statistics on the num- ber of articles in dorsal cirri by individual.) First article of tentacular and dorsal cirri typically slender and cylindrical (Fig. 1B), arising from a short broader cirrophore; other articles of these cirri larger, rounded or bulbous. Ventral cirri short, not extend- ing beyond parapodia, larger and bulbous in anterior setigers (Fig. 1C), becoming dig- itiform and tapering slightly in middle and posterior setigers (Fig. 1B). Pygidium short, semicircular, with terminal anus (Fig. 1D) and two anal cirri each with 4—10 articles usually larger than articles of dorsal cirri; with or without a third short smooth anal cirrus arising midventrally between articu- lated anal cirri. Aciculae slightly enlarged distally, tips protruding from parapodia through cuticle; 2 slender aciculae per parapodium in ante- rior setigers, tip of one less oblique than the other (Fig. 1E); aciculae solitary and heavi- er in middle and posterior setigers (Fig. 1F). Simple setae present only in posterior setig- ers: Superior simple setae (Fig. 2A) solitary, about as long as shafts of compound falci- gers, blunt bilobed to bifid tips, fine distal serrations typically along superior surface, approaching plumose condition; inferior VOLUME 108, NUMBER 4 Table 2.—Summary statistics by specimen of Den- tatisyllis mangalis for number of articles in dorsal cirri. nas Percent- age of Holotype (H) Total or y: Number of Paratype (P) Range 4 SD n Setigers H 4-8 6.0 0.7 31 94 P 4-7 So) 0.8 30 70 Pp 5-6 Soll 0.4 15 58 P 4-6 4.9 0.9 21 76 Pp 6-8 6.6 0.7 8 78) P 3-7 De 0.9 V7) 74 Pp 4-7 5.4 0.9 21 55 P 4-7 SZ 0.9 15 50 P 6-7 6.3 0.5 15 35 P 3-5 4.1 0.5 13 WZ P 4-6 Se 0.5 18 62 P 5-7 De) 0.6 19 54 P 4-5 4.4 0.5 20 91 P 4-5 4.6 0.5 21 84 P 4-5 4.7 0.4 21 100 P 5-8 6.4 0.7 37) 88 P 4-6 De 0.6 24 Wi P 4—6 D5) 0.6 24 89 P 4-6 5.1 0.6 Di 78 Je 4-7 Se) 0.7 WS) 82 P 3-4 3.8 0.4 13 81 P 4-6 Deo) 0.6 21 75 P 4-6 5.1 0.7 17 S// P 3-6 5.4 0.8 28 76 P 3-4 39) 0.3 16 94 P 3-5 4.2 0.6 19 86 P 4-6 4.9 0.5 27 96 P 5-7 6.1 0.5 35 oy P 5-7 5.6 0.6 33 94 P 3-6 Sell 0.6 PES 89 P 4-6 5.0 0.5 18 64 P 4-7 5.8 0.7 30 94 P 4-5 4.2 0.4 13 62 P 5-7 Doli 0.5 34 100 |e 4-5 4.1 (0.3) 15 100 4 Number of dorsal cirri examined. Only one cirrus (when present and visible) from each setiger was in- cluded in count. simple setae shorter, solitary, slightly curved (Fig. 2B), with bifid tips and fine distal serrations along inferior surface. About 9 compound setae per anterior fas- cicle, number gradually decreasing posteri- orly to 3 or fewer per fascicle in last few posterior setigers. Compound falcigers with bifid serrated blades (Fig. 2C—E). Blades of superiormost falcigers long, narrow, taper- D/A ing to minutely bifid tips with knob-like ter- minal tooth; longest in middle and posterior setigers. Commonly 2 of these long-bladed falcigers per fascicle in anterior setigers, usually 1—2 per fascicle in other setigers, occasionally absent from a fascicle. In mature specimens, pharynx usually extending to setiger 7 when inverted, to se- tiger 5 when everted, about equal in length to proventricle, with an anterior mid-dorsal tooth; anterior end surrounded by 10 large papillae or lobes, tips ciliated; anterior mar- gin of pharynx denticulate (Fig. 2F), with 10 distinct teeth arranged in opposition to pharyngeal lobes, tooth shape sometimes rough and irregular. Proventricle length 200—430 pm, width 70-150 wm, length to width ratio 2.1—3.9, with about 32 muscle rows. Proventricle length exhibiting a more or less linear relationship with increasing body length (Fig. 3B). Proventricle length to width ratio remaining relatively constant or increasing slightly with increasing body length (Fig. 3C). Reproduction through viviparity; holo- type and numerous paratypes with from 1 to 4 young present in coelom (Fig. 2G). Young not enclosed within a membrane; with up to 13 setigers bearing setae similar to those of adult. Etymology.—The species name refers to the mangrove community or mangal (sensu MacNae 1968; Tomlinson 1986) that is a conspicuous feature of the type locality. Discussion Synonymy.—Dentatisyllis brevicirra, as used by Wilson (1991:506, table 2), is a nomen nudum according to the Internation- al Code of Zoological Nomenclature (Third Edition 1985). That name had been, prior to Wilson’s paper, only informally proposed by me in a conference poster for the con- cept herein established as D. mangalis. Comparison to congeners.—Dentatisyllis mangalis agrees with the generic diagnosis (as emended above) and resembles Denta- tisyllis carolinae (Day, 1973), Perkins 72 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Dentatisyllis mangalis, new species. A. Anterior end, contracted (setae not drawn). B. Parapodium from middle setiger, posterior view. C. Ventral cirrus on anterior setiger, posterior view. Scale same for B and C. D. posterior end, dorsal view (setae not drawn). E. Aciculae and tip of parapodium from anterior setiger. FE Acicula and tip of parapodium from middle setiger. Scale same for E and E A from paratype USNM 102515; B, C, E, F from paratype USNM 102516; D from paratype USNM 102511. (1981:1166, fig. 38a—h) with respect to the bifid simple setae and superior compound falcigers, but differs by having antennae, tentacular cirri, and especially dorsal cirri comprised of relatively few articles, with the first article of dorsal cirri consistently more slender and cylindrical than others; only two aciculae per anterior parapodium; superiormost compound falcigers with long narrow blades; blades of other compound VOLUME 108, NUMBER 4 S13 U MMU MOOT > 0.01 mm Fig. 2. Dentatisyllis mangalis, new species. A. Superior simple seta from posterior setiger. B. Inferior simple seta from posterior setiger. C. Superior compound falciger from middle setiger. D. Inferior compound falciger from anterior setiger. E. Compound spiniger-like seta from anterior setiger. Scale same for A-E. E Anterior end of everted pharynx, ventral view. G. Two offspring prior to release, dorsal view of one, ventral view of other, inside setigers 21—32 of adult in ventral view, setiger 21 at top of figure (some external structures of offspring obscured by adult’s internal structures, especially musculature associated with aciculae in each parapodium; setae of offspring included where visible; setae of adult not drawn). A, B from paratype USNM 102510; C—E from paratype USNM 102516; F from paratype USNM 102511; G from holotype USNM 102495. 574 falcigers with sharply pointed terminal and subterminal teeth; blades of inferior com- pound falcigers without enlarged subtermi- nal tooth; and a typically smaller body size. Dentatisyllis mangalis differs in the same ways from Dentatisyllis sp. A of Uebelack- er (1984:30.115, fig. 30.110 a—d). The new species also resembles Denta- tisyllis junoyi Lopez & San Martin, (1992: 219, fig. la—l) particularly with regard to the superiormost compound falcigers with long narrow blades bearing bifid tips (called ““spiniger-like setae’? by Lopez and San Martin, 1992), but differs by having anten- nae, tentacular cirri, and, most notably, dor- sal cirri comprised of relatively few articles, with the first article distinctly cylindrical. Dentatisyllis mangalis differs further by having ventral cirri clearly shorter than par- apodial lobes, compound falcigers without an enlarged subterminal tooth or long slen- der serrations (or spines) arising from the distal margin of the blade, superiormost compound falcigers with long narrow blades along the entire body (sometimes two per fasicle in anterior setigers), and blades of these falcigers without long slen- der serrations arising from the distal blade margin. Viviparity, growth, and morphological variation.—One of the most interesting as- pects of the new species is its viviparous reproduction. Setigerous young are present in the coelom of a number of type speci- mens (Fig. 2G), including a paratype with as few as 30 setigers and a length of 1.6 mm. The brooded young were originally overlooked in some adult specimens. Their presence can be obscured by the alimentary tract, acicular and parapodial muscles, and other organs of the adult. The brooded young are most easily recognized by their proventricle, palps, and dark eyes. The ap- parent lack of a membrane enclosing each offspring, including those removed from brooding adults for closer study, suggests that Ovoviviparity is unlikely. The means by which offspring leave the adult is not known; however, once released, PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON individuals appear to grow more by the ad- dition of setigers than by the elongation of existing setigers, as indicated by a fairly lin- ear relationship between body length and the number of setigers (Fig. 3D). This close relationship indicates that setigers grow to full size fairly rapidly, soon after they are formed. Perhaps the most distinguishing and con- spicuous feature of Dentatisyllis mangalis is the dorsal cirri, each consisting of rela- tively few articles (from three to eight), ex- cept for those on setiger 1 which are typi- cally longer. In contrast, D. carolinae has 27 articles per dorsal cirrus on some ante- rior setigers (excluding setiger 1), thereafter about 30 on “short” dorsal cirri and about 55 on “‘long”’ dorsal cirri (Perkins 1981). Similarly, D. junoyi, has 7 to 9 articles on ‘short’? dorsal cirri and 14 to 21 articles on “long”? dorsal cirri (L6pez & San Martin 1992). (Except on a few anterior setigers, “‘short”’ and “‘long”’ dorsal cirri in the latter two species, as in many Syllinae, alternate throughout the body. Such alternation was observed on only a few specimens of D. mangalis; the difference between “‘short’’ and “‘long”’ cirri was only one or two arti- cles.) Observations I have made of numer- Ous specimens from Belize belonging to other genera of the subfamily Syllinae, as well as the observations of others (e.g., Per- kins, 1981:1169), have suggested that with- in a species the length of dorsal cirri and the number of articles comprising them probably increases with body size; small in- dividuals have fewer articles per dorsal cir- rus, larger individuals more. Although, such size-related variation does exist in D. man- galis and the relationship appears quite lin- ear, the mean number of articles per dorsal cirrus per individual does not change con- siderably with body length, even when the latter increases nearly five-fold (Fig. 3A). Furthermore, not only is the number of ar- ticles comprising a dorsal cirrus on D. man- galis small compared to D. carolinae and D. junoyi, but it also varies little along any one individual (Table 2). VOLUME 108, NUMBER 4 MEAN NO. OF ARTICLES PROVENTRICLE L/W 0 1 2 3 4 BODY LENGTH (mm) Bigs 3: S/5 PROVENTRICLE LENGTH 50 ie) ie) - (=) oO oO NUMBER OF SETIGERS S 0 1 2 3 4 BODY LENGTH (mm) Dentatisyllis mangalis type material. Relationship between body length and (A) mean number of articles in dorsal cirri for each individual (r = 0.858, n = 35, see Table 2 for variation associated with each mean), (B) proventricle length (r = 0.892, n = 35), (C) proventricle length to width ratio (r = 0.503, n = 33), and (D) total number of setigers (r = 0.886, n = 35). Other observations on the relationship between morphological variation (such as that in proventricle length or the proventri- cle length to width ratio) and growth have been noted in the species description. Acknowledgments I thank Dr. Kristian Fauchald, National Museum of Natural History (NMNH), Smithsonian Institution for use of a micro- scope equipped with a drawing tube, and for critically reading the manuscript; and, Dr. Jerry Kudenov, University of Alaska, Anchorage, Mr. Thomas Perkins, Florida Department of Natural Resources, and Dr. Guillermo San Martin, Universidad Auton- oma de Madrid, for critically reviewing the manuscript. Thanks are also due Mr. Oscar Soriano, Museo Nacional de Ciencias Na- turales de Madrid, for the loan of speci- mens; and Ms. Linda Ward, NMNH, for sharing office space and equipment and for cataloguing specimens. Financial support from the Caribbean Coral Reef Ecosystems (CCRE) Program, NMNH (Dr. Klaus Riit- Zler, Program Director), is gratefully ac- knowledged. This paper is CCRE Contri- bution No. 454. Literature Cited Day, J. 1973. New Polychaeta from Beaufort, with a key to all species recorded from North Caroli- na.—U.S. National Oceanographic and Atmo- spheric Administration Technical Report, Na- tional Marine Fisheries Service Circular 375:1— 140. Grube, A. E. 1850. Die Familien der Anneliden.— 576 Archiv fur Naturgeschichte (Berlin) 16:249- 364. Lépez, E., & G. San Martin. 1992. A new species of Dentatisyllis (Syllidae: Polychaeta) from Cape Verde Islands, West Africa.—Proceedings of the Biological Society of Washington 105:219-223. MacNae, W. 1968. A general account of the fauna and flora of mangrove swamps and forests in the Indo-West Pacific region.—Advances in Marine Biology 6:73-27. Perkins, T. H. 1981. Syllidae (Polychaeta), principally from Florida, with descriptions of a new genus and twenty-one new species.—Proceedings of the Biological Society of Washington 93:1080— 2” Rioja, E. 1925. Anélidos poliquetos de San Vicente de la Barquera (Cantabrico).—Trabajos de Mu- seo Nacional de Ciencias Naturales (Madrid), Serie Zoolégico 53:1—62. Russell, D. E. 1989a. Three new species of Sphae- rosyllis (Polychaeta: Syllidae) from mangrove habitats in Belize.—Zoologica Scripta 18:375— 380. . 1989b. A new species of Odontosyllis (Poly- chaeta: Syllidae) from Twin Cays, Belize.— PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Proceedings of the Biological Society of Wash- ington 102:768—771. 1991. Exogoninae (Polychaeta: Syllidae) from the Belizean barrier reef with a key to spe- cies of Sphaerosyllis—Journal of Natural His- tory 25:49-74. San Martin, G. 1984. Estudio Biogeografico, Faunis- tico y Sistematico de los Poliquetos de la Fam- ilia Silidos (Syllidae: Polychaeta) en Baleares. Tesis Doctoral, Editorial de la Universidad Complutense de Madrid, 529 pp. Sokal, R. R., & E J. Rohlf. 1981. Biometry. Second edition. W. H. Freeman & Company, New York, 859 pp. Tomlinson, P. B. 1986. The botany of mangroves. Cambridge University Press, Cambridge, 413 Pp. Uebelacker, J. M. 1984. Chapter 30, Family Syllidae. Pp. 30.1—30.151 in J. M. Uebelacker & P. G. Johnson, eds., Taxonomic guide to the poly- chaetes of the northern Gulf of Mexico. Final report to the Minerals Management Service, contract 14-12-001-29091. Volume 4. Barry Vittor and Associates, Inc., Mobile, Alabama. Wilson, H. W. 1991. Sexual reproductive modes in polychaetes: classification and diversity.—Bul- letin of Marine Science 48:500-516. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(4):577—582. 1995 New genera for two polychaetes of Lepidonotinae Marian H. Pettibone Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, U.S.A. Abstract.—Among the Lepidonotinae, two species of Lepidonotus are re- ferred to new genera: L. dictyolepis Haswell, from Australia, to Augenerilepi- donotus, and L. kumari Rullier, from Malaya, to Olgalepidonotus. As part of an on-going study on the po- lynoid polychaetes, two species that were described under Lepidonotus Leach, 1816, L. dictyolepis Haswell, 1883, from Aus- tralia, and L. kumari Rullier, 1970, from Malaya, show differences from Lepidon- otus and are referred to two new genera: Augenerilepidonotus, for the first species, and Olgalepidonotus, for the second spe- cies. The specimens examined in this report were received on loan from the Australian Museum, Sydney (AMS), through Eliza- beth Pope, Pat Hutchings, and Neville Coleman; from the Western Australian Museum, Perth, through R. W. George; and from Professor Francois Rullier, La- boratoire de Zoologie, Angers, France (LZA). Family Polynoidae Kinberg, 1856 Subfamily Lepidonotinae Willey, 1902 Augenerilepidonotus, new genus Type species.—Lepidonotus dictyolepis Haswell, 1883. Gender: masculine. Diagnosis.—Body short, flattened, sub- rectangular, with 26 segments (first achae- tous). Elytra and prominent elytrophores 12 pairs, on segments 2, 4, 5, 7, 9, 11, 13, 15, 17, 19, 21, and 23. Elytra large, oval, over- lapping, with surfaces mostly covered with chitinous irregular polygonal areas enclos- ing small secondary areolae, with some spi- nous microtubercles and lateral fringes of short, wide papillae. Dorsal cirri on non- elytrigerous segments, with bulbous cirro- phores and short styles; dorsal tubercles in- distinct. Prostomium bilobed, lepidonotoid, with 3 antennae, 2 palps, and 2 pairs of eyes; median antenna with bulbous cerato- phore in anterior notch of prostomium; lat- eral antennae inserted terminally on anterior extensions of prostomium. First or tentac- ular segment not visible dorsally; tentacu- lophores lateral to prostomium, achaetous, with 2 pairs of dorsal and ventral tentacular cirri, with bulbous upper lip and small in- distinct facial tubercle. Second segment with first pair of elytrophores, biramous parapodia, and long ventral buccal cirri. Bi- ramous parapodia with small, conical no- topodia on anterodorsal faces of larger neu- ropodia; neuropodia with subconical pre- setal lobe and shorter, rounded postsetal lobe, deeply cut dorsally and ventrally. No- tosetae numerous, slender, densely serrated, short, extending only slightly beyond tips of neuropodia, much more slender than neurosetae; few short, tapering to blunt tips and numerous, long, tapering to capillary tips. Neurosetae stout, relatively few (8— 12), with few spinous rows (4—5), and curved unidentate tips. Ventral cirri short, subulate. Pygidium with anal ridge and pair of anal cirri. Nephridial papillae small, bul- bous, beginning on segment 8. Pharynx (7?) not extended. Etymology.—The genus is named for Herman Augener (1927), whose descrip- tion and figures of a specimen of Lepidon- otus dictyolepis, from near the type local- 578 ity, supplemented Haswell’s original de- scription. Remarks.—Augenerilepidonotus dicty- olepis differs from other species of Lepi- donotus by the presence of chitinous polyg- onal areas on the elytra. Among the Poly- noidae, similar types of elytra with chitin- ous polygonal areas are found in the Iphioninae Baird, including [phione Kin- berg, Iphionides Hartmann-Schroder, and Iphionella McIntosh (See Pettibone 1986, Hanley & Burke 1991), and in Harmothoin- ae, including Gaudichaudius Pettibone, 1986. Augenerilepidonotus dictyolepis (Haswell, 1883), new combination Fig. 1 Lepidonotus dictyolepis Haswell, 1883, 287, pl. 9: figs. 7,8.—Augener, 1927:94, fig. 3a—c.—Not Fauvel, 1932:14 (Gulf of Manaar).—Rullier, 1972:29.—Day, 1975: 178, fig. Im—q.—Averincev, 1978:69.— Hanley & Burke, 1990:218, fig. 7A—J.— Hanley, 1993:314. ?Lepidonotus aeololepis Haswell, 1883: 286, pl. 9: figs. 3—5. Material examined.—Australia, New South Wales: Watson’s Bay, Fort Jackson, Sydney, dredged in shallow water, early June, 2 syntypes of L. dictyolepis (AMS G11274). Western Australia: Cockburn Sound, Harding Rock, east side of Garden Island, burrowed 5-7 inches into old coralline limestone, and Parmelia Bay, 1.6 km west of Woodman Point, washings from Pinna Shells with Caulerpa, 2—3 m, 6/13 Feb 1972, B. R. Wilson, coll., 3 specimens (WAM 43-72; ident. Day, 1975). Hall’s Bank, Fremantle, rubble, 8 m, 13 Apr 1972, N. Coleman, coll., | specimen (AMS W- 5491). Description.—Body with 26 segments, 7-12 mm long, 3-4.5 mm wide with setae. Elytra with most of surface covered with irregular polygonal areas, enclosing second- ary areolae, larger centrally, smaller periph- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON erally, with some smaller areolae, microtu- bercles, and short, stout papillae near exter- nal borders (Fig. 1G, H; Augener 1927: fig. 3a, b; Day 1975: fig. In, 0; Hanley & Burke 1990: fig. 7A—D). Elytrophores large, bul- bous (Fig. 1A, C). Dorsal cirri with cylin- drical cirrophores bulbous basally, with short subulate styles, shorter than neurose- tae; dorsal tubercles indistinct (Fig. 1A, D; Day 1975: fig. Im). Bilobed prostomium with bulbous cera- tophore of median antenna in anterior notch, style cylindrical, with tapered tip; slightly shorter lateral antennae inserted on anterior extensions of prostomium; 2 pairs of large eyes; tentaculophores lateral to stout palps and prostomium, with 2 pairs of dorsal and ventral tentacular cirri, similar to median antenna (Fig. 1A; Hanley & Burke 1990: fig. 7E). Segment II without nuchal fold, with first pair of large elytrophores, biramous parapodia and long ventral buccal cirri; neurosetae differing from following neurosetae, more slender, with more nu- merous rows of spines, tapering to sharp tips (Fig. 1A, B; Hanley & Burke 1990: fig. [8s Gy Ue Biramous parapodia with small, conical notopodium on anterodorsal side of large neuropodium, with subconical presetal lobe and shorter rounded postsetal lobe, deeply cut dorsally and ventrally (Fig. 1C, D). No- tosetae numerous, short, extending only slightly beyond neuropodium, slender, densely serrated, few short, tapering to blunt tips and numerous long ones tapering to capillary tips (Fig. 1C, E). Neurosetae relatively few (8—12), all similar, stout, with few spinous rows (4—5) and slightly curved unidentate tips (Fig. 1F; Haswell 1883: pl. 9: fig. 7; Augener 1927: fig. 3c, d; Day 1975: fig. 1q; Hanley & Burke 1990: fig. 7,1). Distribution.—Western Australia: Rott- nest Island, under limestone rubble, 6—7 m (Hanley 1993); South-West Australia: off Albany, low water under rocks (Hanley & Burke 1990; Eastern Australia: Port Jack- son, New South Wales, shallow water and VOLUME 108, NUMBER 4 Fig. 1. SY) Augenerilepidonotus dictyolopis, paratype of Lepidonotus dictyolepis (AMS 11274): A, Dorsal view of anterior end, prostomium partially withdrawn in segment II, with posterior eyes hidden from view; B, Neu- roseta from segment II; C, Right middle elytrigerous parapodium, anterior view, acicula dotted; D, Right middle cirrigerous parapodium, posterior view; E, Short and long notosetae from same; EK Two neurosetae from same; G. Right first elytron, with detail of lateral border; H, Right middle elytron, with detail of different areas. Scales = 0.5 mm for A; 0.1 mm for B, E, F; 0.3 mm for C, D; 1.0 mm for G, H. under stones and algae (Haswell 1883, Au- gener 1927); Tasman Sea: Norfolk Island, 50 m, coralline sand (Averincev 1978, Vin- ogradova et al. 1978); and Coral Sea: New Caledonia (Rullier 1972). Remarks.—Lepidonotus aeololepis Has- well, 1883, from Thursday Island, Austra- lia, from under rocks at low tide, is ques- tionably referred to A. dictyolepis. The type specimen in the Queensland Museum is lost (Day & Hutchings 1979:89) and not avail- able for study. The description and figures are deficient but the elytra, showing polyg- onal areas, and the stout neurosetae with few spinous rows and entire tips, agree with A. dictyolepis. 580 Olgalepidonotus, new genus Type species.—Lepidonotus kumari Rul- lier, 1970. Gender: masculine. Diagnosis.—Body short, flattened, sub- rectangular, with 26 segments (first achae- tous). Elytra and prominent elytrophores 12 pairs, on segments 2, 4, 5, 7, 9, 11, 13, 15, 17, 19, 21, and 23. Elytra large, overlap- ping, covering dorsum, without long fringe of papillae, densely covered with rounded microtubercles and micropapillae. Dorsal cirri on non-elytrigerous segments, with long cylindrical cirrophores and tapered styles extending to tip of neurosetae. Pro- stomium bilobed, with 3 antennae, 2 palps, and 2 pairs of eyes; median antenna with oval ceratophore in anterior notch of pro- stomium; lateral antennae inserted subter- minally on anterior continuations of prosto- mium, on same level as median antennae but with line of separation from prostomi- um. First or tentacular segment not visible dorsally; tentaculophores lateral to prosto- mium, achaetous, with 2 pairs of dorsal and ventral tentacular cirri, without distinct fa- cial tubercle. Second segment with first pair of elytrophores, biramous parapodia, and long ventral buccal cirri. Biramous para- podia with small conical notopodia on an- terodorsal faces of larger neuropodia; neu- ropodia with subconical presetal acicular lobe and shorter rounded postsetal lobe. Notosetae all one type, much more slender than neurosetae, delicate, finely spinous, ta- pering to capillary tips. Neurosetae not usu- al Lepidonotus type, rather slender, wider basally, tapering to sharp tips, mostly bare, upper ones with few spines (4-5), Ventral cirri short, subulate. Pygidium with anal ridge and pair of long anal cirri. Nephridial papillae rather long, cylindrical, beginning on segment 8. Pharynx (?) not extended. Etymology.—The genus is named for Olga Hartman, who contributed so much to the study of the Polychaeta. Remarks.—Olgalepidonotus kumari dif- fers from the usual Lepidonotus types of se- tae, having all long capillary notosetae and PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON stouter neurosetae mostly bare, with subter- minal enlargements, tapering to sharp tips. Olgalepidonotus kumari (Rullier, 1970), new combination Fig. 2 Lepidonotus kumari Rullier, 1970:221, Figs. A-I. Material examined.—Malaya, Port Swet- tenham, in mangrove, 30 Nov 1968, A. S. Kumar, collector, holotype (LZA). Description.—Body with 26 segments, 9 mm long, 4.5 mm wide with setae. Elytra large, overlapping, covering dorsum, last twelfth pair extra wide, covering posterior five segments (Rullier 1970: fig. A). Elytra oval to subreniform, densely covered with rounded, colored microtubercles and short micropapillae on surface and posterior and lateral borders (Fig. 2H; Rullier 1970: figs. H, I). Elytrophores large, bulbous (Fig. 2A, D). Dorsal cirri with long cylindrical cir- rophores on dorsoposterior faces of para- podia, with styles extending about to tips of neurosetae; dorsal tubercles slightly inflated (Fig. 2E). Bilobed prostomium with deep red pig- mentation on lateral margins and bases of antennae; median antennae with rounded ceratophore in anterior notch of prostomi- um, with style about as long as stout ta- pered palps; lateral antennae inserted sub- terminally on anterior extensions of prosto- mium, on same level as median antenna but with line of separation, appearing as distinct ceratophores, with styles shorter than me- dian antenna; eyes rather large, anterolateral pair and slightly smaller posterolateral pair; tentaculophores lateral to palps and prosto- mium, with dorsal tentacular cirri similar to median antenna and slightly shorter ventral tentacular cirri (Fig. 2A; Rullier 1970: fig. B). Segment II without nuchal fold, with first pair of large elytrophores, biramous parapodia, and long ventral buccal cirri, similar to tentacular cirri; notosetae similar to following; neurosetae differing from fol- lowing, more slender, with spinous rows, VOLUME 108, NUMBER 4 | Siifuss 2: 581 Olgalepidonotus kumari, holotype of Lepidonotus kumari (LZA): A, Dorsal view of anterior end, missing: styles of median antenna, right and left dorsal and ventral tentacular cirri, elytra of segment II and styles of dorsal cirri of segment III; B, Right parapodium of segment II, elytrophore and ventral buccal cirrus not shown, acicula dotted; C, Neuroseta from same; D, Right middle elytrigerous parapodium, anterior view, acicula dotted, some neurosetae broken or missing; E, Right middle cirrigerous parapodium, posterior view, some neurosetae broken or missing: EK Notoseta from same; G, Lower, middle and upper neurosetae from same; H, Right middle elytron, with detail of microtubercles and micropapillae. Scales = 0.5 mm for A; 0.3 mm for B, D, E; 0.1 mm for C, E G; 0.5 for H. and tapering to delicate tips (Fig. 2A—C; Rullier 1970: fig. B). Biramous parapodia with short, conical notopodia on anterodorsal sides of large neuropodia, with subconical presetal acic- ular lobe and slightly shorter, rounded pos- tsetal lobe, notched dorsally (Fig. 2D, E). Notosetae much more slender than neuro- setae, single type, delicate, finely spinous, tapering to long capillary tips (Fig. 2D, F; Rullier 1970: fig. D). Neurosetae rather slender, wider subdistally, tapering to sharp tips, few (4—5) supraacicular ones, with few (4-5) spinous rows, and more numerous (15—20) subacicular neurosetae, smooth, without spines, lower ones with slightly 582 curved tips (Fig. 2G; Rullier 1970: figs. E— G). Ventral cirri short, subulate (Fig. 2D, E). Distribution.—Indian Ocean, Port Swet- tenhan, west coast of Malaya, in mangrove swamp. Literature Cited Augener, H. 1927. Polychaeten von Sud6st- und Sud- Australien (Papers from Dr. Th. Mortensen’s Pacific Expedition 1914-16, no. 38).—Viden- sksbelige Meddelelser fra den Dansk Naturhis- toriska Forening 1 Kjg@benhavn 83:71—275. Averincev, V. C. 1978. [The Polychaetous annelids of the Aphroditiformia of the shelf and upper bathyal of Australian and New Zealand region and of Macquarie Island (on the base data of 16th Cruise of R/V “Dmitry Mendeleev” ].— Trudy Instituta Okeanologii im PP. Shirshova 113:51—72 [In Russian, English summary]. Day, J. H. 1975. On a collection of Polychaeta from intertidal and shallow reefs near Perth, Western Australia.—Records of Western Australian Mu- seum 3:167—208. , & P A. Hutchings. 1979. An annotated check-list of Australian and New Zealand Poly- chaeta, Archiannelida and Myzostomida.—Re- cords of the Australian Museum 32:80—-161. Fauvel, P. 1932. Annelida Polychaeta of the Indian Museum, Calcutta——Memoirs of the Indian Museum, Calcutta 12:1—262. Hanley, J. R. 1993. Scaleworms (Polychaeta: Poly- noidae) of Rottnest Island, Western Australia. Pp. 305-320 in FE E. Wells, D. I. Walker, H. Kirkman, & R. Lethbridge, eds., Proceedings of the fifth international marine biological work- shop: the marine flora and fauna of Rottnest Is- land, Western Australia. Western Australian Museum, Perth. , & M. Burke. 1990. Scaleworms (Polychaeta: Polynoidae) of Albany, Western Australia, Pp. 203-236 in E E. Wells, D. I. Walker, H. Kirk- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON man, & R. Lethbridge, eds., Proceedings of the third international marine biological workshop: the marine flora and fauna of Albany, Western Australia, Western Australian Museum, Perth, volume 1. 5 es 1991. Polychaeta Polynoidae: scaleworms of the Chesterfield Islands and Fair- way Reefs, Coral Sea.—Resultats des Campag- nes Musorstom, Volume 8.—Memoires du Mu- seum National d’Histoire Naturelle, serie A: Zoologie 151:9-82, Paris. Haswell, W. A. 1883. A Monograph of the Australian Aphroditea.—The Proceedings of the Linnean Society of New South Wales 7:250—299. Leach, W. E. 1816. Annulosa. Vermes Polychaeta.— Encyclopedia Britannica. Supplement Editions 4—6, 1(2):401-453. Kinberg, J. G. H. 1856. Nye slagten och arter af An- nelider.—Ofversigt af Konglia Vetenskaps-Ada- demiens Foérhandlingar, Stockholm 12:381—388. Pettibone, M. H. 1986. Review of the Iphioninae (Polychaeta: Polynoidae) and Revision of [phi- one cimex Quatrefages, Gattyana deludens Fau- vel, and Harmothoe iphionelloides Johnson (Harmothoinae).—Smithsonian Contributions to Zoology 428:1-43. Rullier, EF 1970. Lepidonotus kumari, une nouvelle es- pece d’Aphroditidae (Annelide Polychete) de Malaisie.—Bulletin de la Societe zoologique de France 95:221—223. . 1972. Annelides polychetes de Nouvelle-Ca- ledonie. Expedition Francaise sur les recifs cor- alliens de la Nouvelle Caledonie 6:1—169, Edi- tions de la Foundaton Singer-Polignac, Paris. Vinogradova, N. G., R. J. Levenstein, & E. P. Turpa- jeva. 1978. [Quantitative Distribution of Bot- tom Fauna in the Region Investigated on the 16th Cruise of R/V “Dmitry Mendeleev’ ].— Trudy Instituta Okeanologii im P. P. Shirshova 113:7—21. [In Russian, English summary]. Willey, A. 1902. Polychaeta. Jn, Report on the col- lections of natural history made in the Antarctic regions, during the voyage of the Southern Cross.—London 12:262—283. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(4):583-595. 1995 Two new species of Opisthotrochopodus (Polychaeta: Polynoidae: Branchinotogluminae) from the Lau and the North Fiji Bac-arc Basins, southwestern Pacific Ocean Tomoyuki Miura and Daniel Desbruyéres (TM) Faculty of Fisheries, Kagoshima University, 4-50-20, Shimoarata, Kagoshima 890, Japan; (DD) DRO-EP, IFREMER, Centre de Brest, B.P.70, 29280, Plouzane, France Abstract.—Opisthotrochopodus segonzaci, a new species of branchiate po- lynoid polychaete from the Lau and the North-Fiji Bac-arc Basins, southwest- ern Pacific Ocean, differs from the other species of the genus in having short dorsal cirri with rounded tips on the anterior segments and achaetous modified parapodia on segment 21. Opisthotrochopodus trifurcus, a new species from the North-Fiji Basin, is unique in the genus in having trifurcate neurosetae on the posterior modified segment. Polynoid polychaetes of the subfamily Branchinotogluminae Pettibone, 1985 in- clude three genera (Branchinotogluma Pet- tibone, 1985; Opisthotrochopodus Petti- bone, 1985; and Peinaleopolynoe Desbruy- eres & Laubier, 1988) and ten species to date. Most of them are known from deep- sea chemosynthetic communities. The sub- family was established by Pettibone (1985) when B. hessleri, B. sandersi, B. grasslei, and O. alvinus were described from the Ga- lapagos Spreading Center, and the East Pa- cific Rise at 21°N. Some of these species were later reported from the Explorer and the Juan de Fuca Ridges, Northeastern Pa- cific, as well as O. tunnicliffeae (Pettibone 1988). From the hydrothermal rift-area of the Mariana Bac-arc Basin in the western central Pacific, two additional polynoids, B. burkensis and O. marianus, were described (Pettibone 1989). About some hundreds ki- lometers North of this basin, O. japonicus was found associated with the hydrothermal vents of Kaikata seamount by Miura & Hashimoto (1991). A branchiate polynoid species, Peinaleopolynoe sillardi, was also described from a non-hydrothermal envi- ronment but attracted to enriched sediments in experimental trays disposed on the deep- sea floor in the North Atlantic Ocean to study the influence of organic matter on colonization processes (Desbruyéres & Laubier 1988). The genus was newly in- cluded in the subfamily by Pettibone (1993), along with the description of the second species, P. santacatalina, associated with a whale skeleton in the bathyal Santa Catalina Basin. Polynoid polychaetes collected by the French submersible Nautile from the hy- drothermal community of the central part of the North-Fiji and the Lau Basins were studied and Branchipolynoe pettiboneae Miura & Hashimoto, 1991 in Branchipo- lynoinae was reported associated with my- tilid mollusks by Desbruyeres et al. (1994). Two other polynoids, Thermopolynoe branchiata in Lepidonotopodinae and Thermiphione fijiensis in Iphioninae, have been described previously (Miura 1994). In this study, two new species of the genus Opisthotrochopodus are described. The types are deposited in the Museum Nation- al d’ Histoire Naturelle de Paris (MNHN), and the National Museum of Natural His- tory, Smithsonian Institution (USNM). 584 Some additional specimens were used for SEM or histological observation at Kago- shima University (KU). Subfamily Branchinotogluminae Pettibone, 1985 Opisthotrochopodus Pettibone, 1985 Opisthotrochopodus segonzaci, new species (Figs. 1-4) Material examined.—Vailili, Lau Basin, Nautile Dive 10, 22 May 1989, BIOLAU Station 2, 23°13’S 176°38'W, 1750 m, Ho- lotype (MNHN UD 852), 3 paratypes (in- cluding 1 modified, MNHN UD 853). White Lady, North Fiji Basin, Dive 20, 15 Jul 1989, STARMER II Station 4, 16°59.50’S, 173°55.47’E, 2000 m, 9 para- types (4 modified, MNHN UD 854, USNM 171052), 4 specimens (2 modified, KU). Description.—Holotype 29 mm long, 14 mm wide including parapodia, with 21 seg- ments, including first achaetous tentacular segment. Largest paratype 49 mm long, 19 mm wide, with 21 segments. Body short, spindle-shaped, slightly tapered anteriorly and posteriorly, flattened ventrally, arched dorsally (Figs. la, 2a—d). Integument smooth. Preserved specimens pale. Elytra 10 pairs, on segments 2, 4, 5, 7, 9, 11, 13, 15, 17, and 19, large, imbricated, oval, smooth, white or colorless (Fig. 2a); last pair on segment 19 much smaller on holotype and on specimens with modified posterior segments than on others. Dorsal cirri on non-elytrigerous segments with short cylindrical cirrophores, and short styles with rounded tips, extending to tips of neurosetae (Figs. le, g, 2c, 3d, e, h, i). Branchiae arborescent, separated into two groups: upper groups on lateral bases of elytrophores or dorsal tubercles; lower groups on bases of notopodia (Fig. 3d—1); present on segments 3—19 on holotype and on specimens with modified posterior seg- ments (Fig. Ic, e, f), or on segments 3—21 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON on specimens with non-modified segments (Figs. 2c, 3a, d—1). Prostomium bilobed. Anterior lobes prominent, cylindrical, extending anteriorly, with frontal filaments; median antenna in- serted in anterior notch, with short cylin- drical ceratophore and subulate style ex- tending to about tip of palp; palps thick, smooth, extending beyond prostomium, with rounded tips (Fig. 2c, d). Tentacular segment fused to prostomium, indistinct; tentaculophores lateral to prostomium, an- nulated, achaetous, with two pairs of ten- tacular cirri, stout, short, as long as tenta- culophores, with rounded or convex distal tips (Fig. 2c, d). Segment 2 with first pair of elytrophores, biramous parapodia, and ventral or buccal cirri attached basally on prominent cirro- phores lateral to mouth, with styles similar to tentacular cirri, longer than following ventral cirri (Figs. 2c, d, 3c). Mouth open- ing situated between segments 1 and 2 (Figs. la, 2b, d). Muscular pharynx encir- cled distally by four pairs of small papillae, subequal in size; bearing lateral lamellar ex- pansions with four distal papillae, and nu- merous small papillae forming proximal band; two pairs of jaws without denticula- tions on inner border (Fig. 2e). Segment 3 with first pair of arborescent branchiae, dorsal cirri, short ventral cirri and biramous parapodia similar to segment 2; dorsal cirri short, with rounded tips (Figs. 2c, d, 3d). Following biramous parapodia, with short notopodia on anterodorsal sides of large neuropodia (Fig. 3d—i). Notopodia subconical, with projecting acicular lobes hidden by numerous notosetae, and en- closed antero-dorsally by flaring bracts (Fig. 3d—1). Neuropodia subconical, deeply notched on upper part (Fig. 3d-1). Ventral segmental papillae long, attached to bases of neuropodia on segment 12, and extending to bases of ventral cirri; 5 pairs of flat semioval ventral lamellae on seg- ments 13-17 on holotype and on half of specimens (Fig. la, b), or lacking both pa- pillae and lamellae on other specimens (Fig. VOLUME 108, NUMBER 4 585 = ee Fig. 1. Opisthotrochopodus segonzaci, new species. Holotype (MNHN UD 852): a, Ventral view; b, Left halves of segments 12-14, ventral view; c, Posterior end, dorsal view; d, Same, ventral view; e, Right cirrigerous parapodium from segment 18, anterior view; f, Right elytrigerous parapodium from segment 19, elytron removed, anterior view; g, Right cirrigerous parapodium from segment 20, anterior view; h, Right cirrigerous parapodium from segment 21, anterior view. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 586 Y is } AW i! 4 x \ mh ; } ! r | ie \ NAS WANA)! Kak ( Al iY A\Q Opisthotrochopodus segonzaci, new species. Paratype (MNHN UD 853): a, Dorsal view; b, Ventral Fig. 2. view; c, Anterior end, dorsal view; d, Same, ventral view; e, Pharynx of another paratype (MNHN UD 854), ventral view. VOLUME 108, NUMBER 4 587 Fig. 3. Opisthotrochopodus segonzaci, new species. Paratype (MNHN UD 853): a, Posterior end, dorsal view; b, Same, ventral view; c, Right elytrigerous parapodium from segment 2, elytron removed, anterior view; d, Right cirrigerous parapodium from segment 3, anterior view; e-1, Right elytrigerous parapodium from segment 10, elytron removed, anterior view; e-2, Same, posterior view; f, Right elytrigerous parapodium from segment 11, elytron removed, anterior view; g, Right elytrigerous parapodium from segment 19, elytron removed, anterior view; h, Right cirrigerous parapodium from segment 20, anterior view; i, Right cirrigerous parapodium from segment 21, anterior view. 588 2b). Segments 18-21 modified, with re- duced parapodia on holotype and on spec- imens with ventral segmental papillae (Fig. lc—h), or not modified on other specimens (Figs. 2b, 3g—i1). Parapodia of segments 18 and 19 biramous, with small notopodia and slender neuropodia; ventral cirri slightly elongated on holotype and on modified specimens (Fig. 1c—f); notopodia developed with numerous notosetae on other speci- mens (Fig. 3g). Parapodium of segment 20 modified with reduced notopodium and elongated neuropodium, lacking notosetae; notopodial acicular lobe fused to cirrophore of long dorsal cirrus; cirrophore with ven- tral lamellar expansion and embedded no- toacicula; neuropodium with lamellar ex- pansion folding small bundle of neurosetae on holotype and on specimens with ventral segmental papillae (Fig. 1g); notopodia with numerous notosetae on other speci- mens (Fig. 3h). Parapodium of segment 21 strongly modified, lacking noto- and neu- rosetae; notopodial acicular lobe fused to cirrophore of short conical dorsal cirrus, ventral lamellar expansion enlarged; ventral cirrus as long as dorsal cirrus on holotype and on specimens with ventral segmental papillae (Fig. 1h); parapodia with noto- and neurosetal bundles on other specimens (Fig. 31). Pygidium visible dorsally as bulbous lobe, wedged between parapodia of poste- rior segments, with pair of long ventral anal cirri (Figs. lc, 3a) Notosetae numerous, forming radiating bundles, much stouter than neurosetae, ser- rated on distal margins; tips bare, tapered (Fig. 4a-1, -2). Neurosetae numerous, form- ing fan-shaped bundles. Supraacicular neu- rosetae with numerous prominent spines in two rows; tips bare, tapered (Fig. 4b-1, -2). Subacicular neurosetae serrated on distal margins; tips bare, tapered, hooked (Fig. 4c-1, -2). Variation in paratypes.—Half of the 16 complete specimens have modified para- podia on segments 18—21, with long ventral papillae on segment 12, and flat semi-oval ventral lamellae on segments 13-17 (Fig. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON la), while remaining specimens have no modified segments, papillae, or ventral la- mellae (Fig. 2b). This dimorphism is inde- pendent of the specimen size. Etymology.—The species is named for Michel Segonzac, who helped to prepare the materials examined in this study. Remarks.—Opisthotrochopodus segon- zaci differs from the other congeneric spe- cies in having short dorsal cirri with round- ed tips, and achaetous parapodia on seg- ment 21 on specimens with modified pos- terior segments. Opisthotrochopodus trifurcus, new species (Figs. 5—8) Material examined.—White Lady, North Fiji Basin, Nautile Dive 12, 7 Jul 1989, STARMER II Station 4, 16°59.50’S, 173°55.47'E, 2000 m, Holotype (MNHN UD 855), 7 paratypes (including 3 modi- fied, MNHM UD 856); Dive 10, 5 Jul 1989, 2 paratypes (MNHM UD 857); Dive 11, 6 Jul 1989, 26 paratypes (15 modified, USNM 171053); Dive 13, 8 Jul 1989, 36 specimens (21 modified, KU); Dive 14, 9 Jul 1989, 14 paratypes (7 modified, MNHM UD 858); Dive 16, 11 Jul 1989, 50 para- types (23 modified, MNHM UD 859); Dive 20, 15 Jul 1989, 23 paratypes (12 modified, USNM 171054). Description.—Holotype 15 mm long, 7 mm wide including parapodia, with 21 seg- ments, including first achaetous tentacular segment. Body short, spindle-shaped, slightly tapered anteriorly and posteriorly, flattened ventrally, slightly arched dorsally (Figs. 5a, b, 6a, b). Preserved specimens pale. Elytra 10 pairs, on segments 2, 4, 5, 7, 9, 11, 13, 15, 17, and 19; large, imbricated, oval, smooth, white or colorless; first pair subreniform (Figs. 5a, 6a). Dorsal cirri on non-elytrigerous segments with short cylin- drical cirrophores, and very long cirriform styles with tapered tips (Fig. 7b, c, e). Bran- chiae arborescent, separated into two VOLUME 108, NUMBER 4 eda seen SS Ree = star sber LULEe eed casas bond ican Sivedd auiaea, b-1 589 lya-c) 0.4mm 2(a-c) 0.1 mm Fig. 4. Opisthotrochopodus segonzaci, new species. Paratype (MNHN UD 853): a-1, Notoseta from segment 11; a-2, Same, enlarged; b-1, Supraacicular neuroseta; b-2, Same, enlarged; c-1, Subacicular hooked neuroseta; c-2, Same, enlarged. groups: upper groups on lateral bases of elytrophores and dorsal tubercles; lower groups on bases of notopodia; present on segments 3—18 on specimens with modified posterior segments (Figs. 5b, 7b—d), or on segments 3—21 on specimens with non- modified parapodia (Fig. 6a—d); dorsal tu- bercles nodular, projecting (Fig. 7b, c). Prostomium bilobed. Anterior lobes prominent, cylindrical, extending anteriorly, without frontal filaments; median antenna inserted in widely incised anterior notch, with very short ceratophore and subulate style, half as long as palp; palps thick, smooth, twice as long as prostomium, with slender tips (Figs. 5c, 6c). Tentacular seg- ment fused to prostomium, not distinct dor- sally; tentaculophores lateral to prostomi- um, smooth, achaetous, with two pairs of tentacular cirri, slender, longer than tenta- culophores (Figs. 5a—c, 6a—c). Segment 2 with first pair of elytrophores, biramous parapodia, and ventral or buccal cirri with short cirrophores and styles sim- ilar to tentacular cirri, longer than following ventral cirri; notopodium long, conical, with projecting acicular lobe and several notosetae; neuropodium long, conical, with numerous neurosetae (Fig. 7a). Muscular pharynx encircled distally by four pairs of 590 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON a,b 5 mm c-e 1 mm Fig. 5. Opisthotrochopodus trifurcus, new species. Holotype (MNHN UD 855): a, Dorsal view; b, Ventral view; c, Anterior end, dorsal view; d, Posterior end, dorsal view; e, Same, ventral view. VOLUME 108, NUMBER 4 a,b 5 mm c,d 2 mm 591 Fig. 6. Opisthotrochopodus trifurcus, new species. Paratype (KU): a, Dorsal view; b, Ventral view; c, An- terior end, dorsal view; d, Posterior end, dorsal view. large papillae, subequal in size; two pairs of jaws with denticulations on inner borders (observation on dissected paratypes). Segment 3 with first pair of arborescent branchiae, dorsal cirri, ventral cirri, bira- mous parapodia similar to segment 2, and triangular dorsal tubercles; dorsal cirri long, with tapered tips (Fig. 7b). Following bi- ramous parapodia, with short notopodia on anterodorsal sides of large neuropodia on segments 4—17 (Fig. 7c—d). Notopodia long, conical, with several notosetae; neuropodia long, conical, deeply notched on upper part (Fig. 7c—d). Ventral segmental papillae on segment 12, as long as segment; 5 pairs of flat semi- oval ventral segmental lamellae present on segments 13—17 on holotype and on half of specimens (Fig. 5b), or lacking semi-oval segmental lamellae and bearing short ven- tral segmental papillae on segment 11 on other specimens (Fig. 6b). Segments 18—21 592 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Za Fig. 7. Opisthotrochopodus trifurcus, new species. Holotype (MNHN UD 855): a-1, Left elytrigerous par- apodium from segment 2, elytron removed, anterior view; a-2; Same, dorsal view; b-1, Left cirrigerous para- podium from segment 3, anterior view; b-2, Same, dorsal view; c-1, Left elytrigerous parapodium from segment 10, elytron removed, anterior view; c-2, Same, dorsal view; d-1, Left elytrigerous parapodium from segment 11, elytron removed, anterior view; d-2, Same, dorsal view; e, Left cirrigerous parapodium from segment 18, anterior view; f-1, Left elytrigerous parapodium from segment 19, elytron removed, anterior view; f-2, Same, dorsal view; g-1, Left cirrigerous parapodium from segment 20, anterior view; g-2, Same, dorsal view; h, Left cirri- gerous parapodium from segment 21, anterior view. VOLUME 108, NUMBER 4 et ee: OT a et Fig. 8. 10; a-3, Same, enlarged; b-1, Supraacicular neuroseta; b-2, Same, enlarged; c-2, Subacicular hooked neuroseta; d-2, Trifurcate neuroseta from segment 20. modified, with reduced parapodia (Figs. 5b, d, e, 7c—h) on holotype and on specimens with semi-oval segmental lamellae, or not modified on specimens without segmental lamellae (Fig. 6b, d). Segment 18 modified slightly, with parapodia bearing very long dorsal cirri (Fig. 7e). Segment 19 with achaetous notopodia bearing broad dorsal lamellar expansion (Fig. 7f). Segment 20 modified strongly, with long achaetous no- topodia bearing bifurcate distal end; dorsal 8) x x N x NX x “ r S x NX NX N » x b-2 1: 0.1 mm 2: 0.1 mm 3: 0.03 mm Opisthotrochopodus trifurcus, new species. Holotype (MNHN UD 855): a-2, Notoseta from segment digitiform expansion and ventral fringed la- mella; club-shaped neuropodia fringed by trifurcate neurosetae found only on segment 20 (Fig. 7g). Segment 21 reduced in size, achaetous, with uniramous parapodia; no- topodia digitiform, with embedded acicula and very short cirri; ventral cirri, with basal lamellar expansion and digitiform short cir- riform distal end (Fig. 7h). Pygidium with pair of anal cirri; anal cirri short on speci- mens with modified posterior segments and 594 semi-oval segmental lamellae (Figs. 5b, d, e), or long on others with non-modified posterior segments (Fig. 6b, d). Notosetae as stout as neurosetae, with short spines on distal margins; tips bare, blunt (Fig. 8a-2, -3). Neurosetae numerous, forming fan-shaped bundles. Supraacicular neurosetae with numerous short spines on distal margins; tips bare, tapered, hooked (Fig. 8b-2). Subacicular neurosetae with short spines on distal margins; tips bare, ta- pered, hooked (Fig. 8c-2). Trifurcate neu- roseta on segment 20 with long blunt cen- tral rod-like projection and paired fine lat- eral extensions half as long as central pro- jection (Fig. 8d-2). Variation in paratypes.—Half of speci- mens (83/165) have modified parapodia on segments 18—21, ventral papillae on seg- ment 12, and flat semi-oval ventral lamellae on segments 13-17, while remaining spec- imens lack posetrior modified parapodia and ventral lamellae, but bear ventral pa- pillae on segment 11. Etymology.—The species name is de- rived from the characteristic trifurcate neu- rosetae. Remarks.—Opisthotrochopodus trifurcus differs from other species in having trifur- cate neurosetae on segment 20 in specimens with modified posterior segments. Opisthotrochopodus segonzaci and O. trifurcus show characteristic dimorphism. One type has modified posterior segments. In this type, ventral segmental papillae and lamellae are fully developed. The other type has no modified posterior segments nor ventral segmental lamellae. The dimor- phism is independent of the size of speci- men in these species. As none of specimens had matured gametes, it is not clear if this is sexual dimorphism. The dimorphism was reported also in Themopolynoe branchiata by Miura (1994). A couple of species, Opisthotrochopodus alvinus and Branchinotogluma hessleri, which were recorded from the Galapagos Spreading Center and the East Pacific Rise at 21° (Pettibone 1985, 1988) show almost PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON the same dimorphic state as mentioned above in each species from southwestern Pacific areas. In these species, the former has strongly modified posterior segments and long ventral papillae, and the latter lacks posterior modified segments and long papillae. We think these species have the possibility to be synonymized, but further study is necessary. Acknowledgments We thank the captain and the crews of the French research vessel Le Nadir and of the submersible Nautile. The STARMER project (North Fiji Basin study) was funded by the Science and Technology Agency of Japan and by the Institut Francais de Re- cherche pour |’exploitation de la Mer (IFREMER), France; the BIOLAU (Lau Basin study) was funded by IFREMER, France. Anne-Marie Alayse-Danet was chief scientist of the BIOLAU cruise and Suguru Ohta (with the second author) was co-chief scientist of the STARMER II cruise. The first author thanks Michel Se- gonzac and Patrick Briand of the Centre National de Tri d’Océanographie Biolo- gique for their kind communication of the materials examined in this study. The manuscript was improved by the careful re- view and corrections by Marian H. Petti- bone and Jon L. Norenburg of Smithsonian Institution. This work was supported in part by grants-in-aid for the first author from the Ministry of Education, Culture and Science, Japan (No. 03640630) and Kato Foundation for the Promotion of the Bioscience Re- searches. Literature Cited Desbruyéres, D., & L. Laubier. 1988. Exploitation d’une source de matiére organique concentrée dans |’océan profond: intervention d’une anné- lide polychéte nouvelle.-—Comptes Rendus Hébdomadaires des Séances de |’ Academie des Sciences, Série III 307:329-—335. , A.-M. Alayse-Danet, S. Ohta, & the Scientific Parties of Biolau and Starmer Cruises. 1994. Deep-sea hydrothermal communities in south- VOLUME 108, NUMBER 4 western Pacific back-ark Basins (the North Fiji and Lau Basins): composition, microdistribu- tion and food web.—Marine Geology 116:227— 242. Miura, T. 1994. Two new scale-worms (Polynoidae: Polychaeta) from the Lau Bac-are and North Fiji Basins, South Pacific Ocean.—Proceedings of the Biological Society of Washington 107: 532-543. , & J. Hashimoto. 1991. Two new branchiate scale-worms (Polynoidae: Polychaeta) from the hydrothermal vent of the Okinawa Trough and the volucanic seamount off Chichijima Is- land.—Proceedings of the Biological Society of Washington 104:166—174. Pettibone, M. H. 1985. Additional branchiate scale- worms (Polychaeta: Polynoidae) from Galapa- 95) gos hydrothermal vent and rift-area off western Mexico at 21°N.—Proceedings of the Biologi- cal Society of Washington 98:447-469. . 1988. New species and new records of scaled polychaetes (Polychaeta: Polynoidae) from hy- drothermal vents of the Northeast Pacific Ex- plorer and Juan de Fuca Ridges.—Proceedings of the Biological Society of Washington 101: 192-208. 1989. New species of scale-worms (Poly- chaeta: Polynoidae) from the hydrothermal rift- area of the Mariana back-arc basin in the west- ern central Pacific_—Proceedings of the Biolog- ical Society of Washington 102:137—153. . 1993. Polynoid polychaetes associated with a whale skeleton in the bathyal Santa Catalina basin.—Proceedings of the Biological Society of Washington 106:678—688. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(4):596—612. 1995 Two species of Oxyurostylis (Crustacea: Cumacea: Diastylidae), O. smithi Calman, 1912 and O. lecroyae, a new species from the Gulf of Mexico Daniel Roccatagliata and Richard W. Heard (DR) Departamento de Biologia, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Ciudad Universitaria, 1428-Buenos Aires, Argentina; (RWH) Gulf Coast Research Laboratory, PO. Box 7000, Ocean Springs, Mississippi 39566-7000, U.S.A. Abstract.—A form of O. smithi Calman, 1912 and Oxyurostylis lecroyae, new species, are described from the northeastern Gulf of Mexico. The speci- mens of O. smithi described herein and those collected near the type locality, Woods Hole, Massachusetts, differ in the development and arrangement of the carapace carinae, and in the shape and relative length of the female fifth tho- racic segment. Oxyurostylis lecroyae is distinguished from all the other de- scribed species of the genus by the presence of a horizontal carina on the carapace of the adult female. Calman (1912) created the genus Oxyu- rostylis to receive O. smithi, a species he described based on a few specimens from the Woods Hole, Massachusetts, area, but in the material examined two specimens from the Gulf of Mexico were also includ- ed. Subsequently, numerous authors report- ed the occurrence of O. smithi from the Bay of Fundy to Louisiana (e.g., Bousfield & Leim 1960, Bowen et al. 1979, Farrel 1979, Zimmer 1980, Corey 1984, Modlin & Dar- deau 1987, Cahoon & Tronzo 1990, Baces- cu 1992). In addition to O. smithi, five other spe- cies of the genus Oxyurostylis have been described: O. pacifica Zimmer, 1936 and O. tertia Zimmer, 1943 from California and Baja California; O. atlantica Radhadevi & Kurian, 1981 from Florida and Yucatan; O. antipai Petrescu et al., 1993 from Jamaica; and O. salinoi Brum, 1966 from Brazil. In the present paper two species of Ox- yurostylis from the northeastern Gulf of Mexico are described, one designated as O. lecroyae, new species, and the other pro- visionally considered as a form of O. smithi Calman, 1912. Although these descriptions clarify some taxonomic aspects of the Ox- yurostylis from the Gulf of Mexico, com- parisons among the species of this uniform and problematic genus still remain incom- plete. Material and Methods Total body length was measured from the tip of the pseudorostrum to the tip of the telson. Cephalothoracic and _ thoracic lengths include the distal process of the fifth thoracic segment. Small juveniles of O. smithi could not be identified with certainty; therefore, these specimens were omitted from the list of ma- terial examined. Since this limitation also applies to most of the immature instars of O. lecroyae, new species, only preparatory female and adults from both sexes are in- cluded in the material examined for this species. Specimens examined under scanning electron microscope (SEM) were dehydrat- ed, critical-point dried, mounted on stubs and coated with 15nm of gold-palladium. The materials used in this study were VOLUME 108, NUMBER 4 borrowed from museums, donated by col- leagues, and collected by the authors. The following museums loaned material to the authors: Museu Nacional do Rio de Janeiro (MNRJ); National Museum of Natural His- tory, Smithsonian Institution (USNM); and Gray Museum, Marine Biological Labora- tory, Woods Hole. In addition to the specimens from the Gulf of Mexico described herein, the fol- lowing material from the North American Atlantic coast and Southern Brazil was ex- amined for comparative purposes: Oxyurostylis smithi Calman, 1912.— Woods Hole, Massachusetts, surface: 16 Sep 1882, 1 preparatory 2, 1 adult 3, 1 preparatory ¢ (USNM 44145). Date un- known, 1 adult 6 (USNM 44160). 6 May 1906, 3 adult od, 1 juvenile (USNM 44163). Date unknown, | preparatory 2, 1 preparatory d, 2 juveniles (USNM 44164). Date unknown, | preparatory @, 2 juveniles (USNM 44165). 2 May 1888, 11 adult od, 1 juvenile (USNM 63646). Vineyard Sound, Massachusetts: depth unknown, 3 Sep -?, 1 juvenile (USNM 34897). Surface, 29 Jul 1881, 1 adult d, 1 juvenile (USNM 44152). Surface, 22 Aug 1881, 1 prepara- tory 2, 1 adult 6, 1 preparatory 6, 2 ju- veniles (USNM 44154). Barnstable Harbor, Massachusetts: depth unknown, 7 May 1968, 1 marsupial 2 (Gray Museum 1125). Off Falmouth Harbor, Massachusetts: 41°31.5'N, 70°36.5’W, depth unknown, 30 Mar 1965, 5 preparatory 2 2, 1 preparatory 3, 1 juvenile (Gray Museum 2337). Off New Jersey: 39°21’00"N, 74°05'18"W, 26 m, 7 Nov 1976, 10 marsupial °°, 5 pre- paratory 22, 2 adult dd, 5 juveniles (USNM 179608). 39°15'18"N, 74°08'00"W, 36 m, 14 Feb 1977, 6 preparatory °°, 1 adult 5, 1 preparatory 6d, 11 juveniles (USNM 179612). West end of Skull Creek, South Carolina: depth unknown, date un- known, 3 marsupial 2 2, 9 preparatory ¢ @, 2 preparatory 36d, 5 juveniles (USNM 92000). One mile inside May River, South Carolina: depth unknown, 17 Jan 1891, 2 597 preparatory °°, 1 adult ¢, 2 preparatory 36, 7 juveniles (USNM 92021). Oxyurostylis salinoi Brum, 1966.—Uba- tuba, Praia do Flamengo, Sao Paulo: 12 m, 18 Oct 1961, Sta. 3(6)II, 5 marsupial °°, 2 preparatory °°, 2 adult dd, 1 preparatory d (Paratypes, MNRJ 4213). Ubatuba, La- gosteiro, Sao Paulo 6 m, 19 Jan 1962, Sta. 4(6)II: 4 marsupial 2 2, 2 preparatory ¢ &, 2 preparatory ¢ d (Paratypes, MNRJ 4215). Oxyurostylis smithi Calman, 1912 (Figs. 1-22, 35, 36) Material examined.—Galveston Island, West Bay, Texas, 0.5-1.5m: Feb 1971, 1 marsupial 2, 5 preparatory 2°, 3 adult 36, 2 preparatory 6d. Mar 1971, 4 mar- supial 22, 1 adult 5, 2 preparatory do. Apr 1971, 7 marsupial 22, 1 preparatory 2, 1 adult 3, 3 preparatory 36. Jan 1972, 2 marsupial 2°, | preparatory 2, 2 adult 363, 5 preparatory 66. Feb 1973, 3 pre- paratory 22, 3 adult dd, 3 preparatory 33. Feb 1974, 5 marsupial 2 2, 9 prepa- ratory 22, 2 preparatory 6d, 3 juveniles. 2 Apr 1987, 2 marsupial 2 2. 31 Mar 1988; 1 marsupial 2, 3 preparatory ¢ 2. Biloxi, Mississippi, approx. | m depth: 22 Feb 1992, 2 marsupial 22, 1 adult d, 1 juve- nile. 2 May 1992, 3 marsupial °°, 3 pre- paratory 22, 2 adult 366d, 6 preparatory 36, 2 juveniles. 29 May 1992, 7 marsupial 22, 8 preparatory 2 ¢, 8 adult dd, 1 pre- paratory d (selected reference material: 6 marsupial 22, 6 adult ¢6d¢6, USNM 274182). 9 Jul 1992, 2 marsupial °°, 1 preparatory 2, 4 adult dd, 3 preparatory 364, 2 juveniles. 7 Nov 1992, 1 marsupial 2, 18 preparatory °°, 7 adult 56, 3 pre- paratory 66, 13 juveniles. Horn Island, Mississippi: depth unknown, 1|1 Jul 1973, 2 marsupial ¢ 2, 1 preparatory 2, 2 adult dd (USNM 150190). Ship Island, Mississippi: 4 ft; 2 Jul 1959; 10 marsupial 2? °, 4 pre- paratory 22, 1 adult ¢ (USNM Acc. #239395). Perdido Key, Florida: lagoon, 0.1—1 m, 1—5 Dec 1991, 46 marsupial ¢ @, 55 preparatory 22, 19 adult 2°, 33 pre- 598 paratory ¢66, 55 juveniles. 8-9 Mar 1992, 53 marsupial 22, 49 preparatory °°, 36 adult 6 6, 54 preparatory 6d, 30 juveniles (selected reference material: 6 marsupial 22, 6 adult dd, USNM 274183). 1 Jun 1992, 1 marsupial 2, 1 preparatory °, 3 juveniles. 21 Sep 1992, 6 marsupial ¢ 2, 3 preparatory 2°, 1 preparatory d, 1 juve- nile. St. Andrew Bay, Carl Gray Park, Flor- ida: 0O-0.5 m, 11 Dec 1991, 15 marsupial 22, 21 preparatory 22, 6 adult dd, 15 preparatory 6 d, 36 juveniles (selected ref- erence material: 5 marsupial 22, 5 adult 3636, USNM 274185). Tampa Bay, Florida (precise location not available): 0.5—1 m, grass bed; 13 Dec 1992, 59 marsupial ¢ &, 72 preparatory °°, 93 adult 6d, 46 pre- paratory ¢ i) oD aE = stay Sin S Sy A 97 © Fig. 3. Discorsopagurus maclaughlinae, new species. Holotype, male (CBM-ZC 603, SL 5.6 mm). A, right chela and carpus, dorsal; B, right cheliped, mesial; C, left chela and carpus, dorsal; D, left cheliped. mesial. VOLUME 108, NUMBER 4 row of moderately strong conical spines and tufts of moderately long stiff setae; dor- sal surface with several irregular rows of small or moderately strong spines extending to fixed finger; dorsolateral margin of fixed finger clearly delimited with row of mod- erately strong spines and dorsolateral mar- gin of palm weakly delineated; lateral, me- sial and ventral surfaces with scattered tufts of long stiff setae. Carpus slightly broad- ened distally, 1.4—1.5 times as long as distal breadth (excluding distolateral spines), al- most as long as merus, somewhat inflated ventrally; dorsomesial margin with single or double row of moderately strong spines, increasing in size distally, and tufts of mod- erately strong stiff setae, dorsal surface sometimes with small spines medially or short transverse rows of moderately long stiff setae, distal margin with few small spines medially; dorsolateral margin not sharply delimited, lateral face with scattered tufts of short setae; ventrolateral margin with row of small tubercles and tufts of short stiff setae, lateral face with scattered tufts of short or moderately short setae; ventromesial margin with row of low tu- bercles and tufts of short to moderately long setae, mesial face with scattered tufts of short to moderately long setae; ventral sur- face with small spinulose tubercles. Merus subtriangular in cross section; dorsal sur- face with short transverse ridges bearing stiff setae, distal margin with 3 or 4 acute Spines and stiff setae; mesial face some- times with vertical ridge with stiff setae and scattered tufts of short stiff setae, ventro- mesial margin with single row of moder- ately strong spines and tufts of long stiff setae; lateral face with scattered tufts of short setae, ventrolateral margin with small spines or spinulose tubercles and tufts of long stiff setae. Ischium with row of small denticles on ventromesial margin. Coxa with tufts of long setae on distal margin. Left cheliped (Fig. 3C, D) slightly over- reaching base of dactyl of right. Chela dis- tinctly longer than carpus; without torsion. Dactyl considerably elongated, approxi- 623 mately twice length of palm, slightly curved; with hiatus between fixed finger when closed; cutting edge with row of small corneous teeth, terminating in small corneous claw; dorsomesial margin with only few low tubercles and tufts of mod- erately long stiff setae; dorsal surface with few spinulose tubercles and row of tufts of moderately long stiff setae; mesial surface faintly tuberculate, with row of tufts of moderately long stiff setae; ventral face with few tufts of long stiff setae. Palm slightly inflated ventrally; dorsal and lateral surfaces with irregular rows of small spines and spinulose tubercles, and scattered tufts of stiff setae, dorsolateral margin not delim- ited; dorsomesial margin abruptly conver- gent posteriorly, weakly delineated, mesial face slightly tuberculate, with tufts of long stiff setae; ventral surface with tufts of long stiff setae. Carpus slightly shorter than me- rus; dorsal surface with 2 rows of moder- ately strong spines, distal margin with 2 or 3 acute spines and stiff setae laterally; me- sial face with short obliquely vertical ridges bearing long stiff setae; lateral face with scattered tufts of short stiff setae; ventro- distal margin not denticulated; ventral face with tufts of long stiff setae. Merus sub- triangular in cross section; dorsal surface with ridges bearing long tiff setae; mesial and lateral faces with few tufts of short se- tae; ventral surface with row of moderately strong spines medially, and tufts of long stiff setae. Ischium with row of small tu- bercles on ventromesial margin. Coxa sim- ilar to that of right cheliped. Second and third pereopods (Fig. 4A, C) differing slightly in armature of dactyls and shape of ischia; moderately long, second pair slightly exceeding tip of right cheliped, third pair slightly shorter. Dactyls (Fig. 4B, D) relatively long and moderately slender, 1.2—1.3 times as long as propodi, not twist- ed; apex terminating in strong corneous claw; dorsal surfaces each with single or double row of long stiff setae; lateral faces each with scattered tufts of short setae, and with shallow longitudinal sulcus; mesial 624 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 4. Discorsopagurus maclaughlinae, new species. Holotype, male (CBM-ZC 603, SL 5.6 mm). Left pereopods: A, second, lateral; B, dactyl of same, mesial; C, third, lateral; D, dactyl of same, mesial; E, fourth pereopod, lateral; EK propodus and dactyl of same. faces each with shallow longitudinal sulcus flanked by rows of tufts of short setae and rows of corneous spines in third, with row of corneous spines only along dorsal mar- gin in second; ventral margins each with 10-12 strong corneous spines over entire length. Propodi distinctly longer than carpi; dorsal surfaces each with tufts of long stiff setae; lateral and mesial faces each with scattered tufts of short to moderately long VOLUME 108, NUMBER 4 setae; ventral margins each with row of tufts of long stiff setae and with few cor- neous spines distally. Carpi moderately short; dorsal surfaces unarmed in left sec- ond and third pair, with few small spines in right second, each with tufts of long stiff setae; lateral, mesial and ventral surfaces with scattered tufts of stiff setae. Meri lat- erally compressed; dorsal surfaces each with row of low spinulose protuberances and tufts of long setae; lateral and mesial faces nearly naked; ventral margins slightly protuberant, each with tufts of long stiff se- tae. Ischia with dorsal and ventral margin bearing tufts of stiff setae; ischium of third pair somewhat longer than that of second, with slightly sinuous ventral margin. Coxae of third pereopods with gonopore in fe- males. Sternite of third pereopods (Fig. 1D) semicircular, not skewed, with numerous long setae on anterior surface. Fourth pereopods (Fig. 4E) subchelate, relatively stout, dorsal and ventral surfaces with tufts of long setae. Dactyls (Fig. 4F) strongly curved, without preungual process; cutting edge with relatively long corneous teeth laterally. Propodus (Fig. 4F) strongly inflated ventrally, rasp composed of 5 or 6 irregular rows of numerous scales. Fifth pereopods chelate. Male coxae (Fig. 1E) each with gonopore obscured by stiff setae or pointed scales. Sternite (Fig. 1E) developed as paired rounded protuberances, obscured by long stiff setae. Abdomen (Fig. 1B) well developed, Straight or slightly flexed, not twisted. Fourth somite with tergite in form of paired, incompletely fused, chitinous plates. Fifth somite with tergite strongly calcified, possessing median suture, thickly covered with stiff setae except in midline. Sixth so- mite (Fig. 2H) with tergite strongly calci- fied, subrectangular in shape, separated into 2 sections by transverse groove at level of midlength; dorsal surface with dense stiff setae, margins sloping; posterior margin slightly concave, with moderately deep si- 625 nus followed by shallow sulcus either side of median concavity. Pleopods of males (Fig. 1B) unpaired, third to fifth pleopods with exopod moder- ately well developed, endopod reduced. Pleopods in females unpaired, second to fourth pleopods with both rami moderately well developed; fifth pleopod with exopod moderately well developed, endopod re- duced. Uropods (Fig. 2G) stout, symmettri- cal. Telson (Fig. 2G) with weak median con- striction; posterior lobes separated by shal- low median notch, each terminal margin with 10 or more corneous spines increasing in size toward posterolateral angle. Coloration in life-—Chelipeds and am- bulatory pereopods generally brown; spines or tubercles on palm of right cheliped white. Shield mottled brown. Ocular pedun- cles and antennal flagella uniformly brown. In preservative, color fading to straw. Habitat.—The specimens from Usujiri, Pacific coast of southern Hokkaido, were dredged from rubble bottom at depths of 10-40 m. This species was never encoun- tered in the intertidal zone in the surveyed area. All of the specimens examined were found using detached tubes of serpulid po- lychaetes. Distribution.—Northern part of the west- ern Pacific: Russian coast of the Sea of Ja- pan, Saghalien, southern Kurile Islands, Hokkaido; 6—220 m. Etymology.—This interesting species is named in honor of Dr. Patsy A. Mc- Laughlin, for her great contributions to sys- tematics of the Paguridea. Relationship.—Although this new spe- cies has been confounded with Discorso- pagurus schmitti, the differences between them are numerous: Each article of the an- tennal flagellum possesses some short bris- tles in D. maclaughlinae, while in D. schmitti it bears long setae, giving a cast net-like appearance to the entire flagellum; in D. maclaughlinae, the third segment of the antennal peduncle is armed with a spine at the ventromesial distal angle. Such a 626 spine is absent in D. schmitti; in the new species, the tergite of the fifth abdominal somite is not particularly sculptured; its dorsal surface is obscured by dense stiff se- tae. However, in D. schmitti, it is somewhat sculptured with upturned margins and ele- vated central region; its dorsal surface bears scattered long setae; the shape of the telson is remarkably different in the two species. In D. maclaughlinae, the lateral margins of the telson is slightly constricted, and the posterior margins are concave, each with a row of 10 or more spines. In D. schmitti, the lateral margins are not constricted, and the posterior margins are nearly straight or slightly convex, each with 3 or 4 spines; the merus of the third maxilliped is armed with the dorsodistal and ventromesial spines in D. maclaughlinae, but in D. schmitti it is unarmed; the articulation between the palm and the carpus of the left cheliped does not show a slight degree of clockwise torsion in D. maclaughlinae; the dactyl of the left chela is much longer in D. maclaughlinae than in D. schmitti, and; the dactyls of the second and third pereopods are relatively longer and slenderer in D. maclaughlinae than in D. schmitti (1.2—1.3 times as long as propodus in the former, 0.9—1.1 times as long in the latter). In the new species, the mesial surface of each dactyl bears a single or double row of corneous spines, which is absent in the American species. Remarks.—Derjugin & Kobjakova (1935) included Discorsopagurus schmitti (as Orthopagurus) in their list of Decapoda from the Russian coast of the Sea of Japan. Makarov (1937, 1938b) first presented a somewhat detailed account of Discorsopa- gurus schmitti (as Orthopagurus) from Vla- dimir and Olga Bays. His description agrees in general with the present new species, though it lacks some important information, such as morphology of the telson. Vinogra- dov’s (1950:307, pl. 22, fig. 129A, B) figure of Orthopagurus schmitti, showing the dor- sal views of whole body of both sexes, is rather diagrammatic and uninformative. However, it is very likely that his illustra- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON tion represents D. maclaughlinae, since the figure shows smooth antennal flagella. Al- though Kobjakova (1956, 1958a, 1958b) did not present detailed accounts of her ma- terials, it is likely that her records from southern Saghalien and Kurile Islands by Kobjakova (1956, 1958a, 1958b) were ac- tually this new species. As McLaughlin’s (1974) distribution of Discorsopagurus schmitti was based on Russian literature, her report is considered partially to refer to the new species. In a recent report of ano- muran Crustacea of Soya Strait, northern Hokkaido, Takeda & Miyauchi (1992) gave a brief account under the name of Discor- sopagurus schmitti, together with photo- graphs of a specimen living in a polychaete tube. I have reexamined their specimens (NSMT-Cr 1811; 5 males, 3 ovig. females) and confirmed its identity as D. maclaugh- linae. At present, there is no evidence in- dicating the existence of the true D. schmitti in the western Pacific. The minor emendations made to the ge- neric diagnosis pertain to such intraspecifi- cally variable characters as the presence or absence of a spine at the ventromesial distal angle of the third antennal segment, the sculpture of the sixth abdominal somite, and the shape of the telson. Although McLaughlin (1974) suggested some rela- tionship between Discorsopagurus and the Parapaguridae in reference to the partially fused abdominal tergites and the telson with entire lateral and posterior margins, the presence of distinct posteromedian concav- ity of the telson in the new species seems to link D. schmitti to other pagurid genera. It is very likely that the similarity displayed by Discorsopagurus and parapagurids is su- perficial and the character states found in the former may be related to adaptation to worm tube usage. The difference in devel- opment of the posteromedian concavity of the telson may reflect degree of adaptation to the specialized habitats. The adaptation to worm tube usage seems to be more high- ly advanced in D. schmitti than in D. ma- claughlinae. Discorsopagurus — schmitti VOLUME 108, NUMBER 4 nearly exclusively inhabits attached tubes of the honeycomb worm Sabellaria cemen- tarium Moore (Gherardi & Cassidy 1994); the antennal flagellum, of which each arti- cle carries long setae, suggests that feeding is accomplished, at least partially, by anten- nal filtering in D. schmitti, as is also the case in members of Paguritta (McLaughlin & Lemaitre, 1993). As previously men- tioned, D. maclaughlinae uses detached polychaete tubes. Observation of specimens kept alive in aquarium has shown that the new species is a deposit feeder, as in usual pagurid. Acknowledgment My thanks are due to Mr. K. Nomura, the technical staff of the Usujiri Marine Bio- logical Laboratory, Hokkaido University, for helping to collect material at Usuyiri. I wish to express my sincere gratitude to Drs. R. B. Manning and R. Lemaitre of the Na- tional Museum of Natural History, Smith- sonian Institution, for making available specimens of Discorsopagurus schmitti and Orthopagurus minimus for comparison. I am deeply indebted to Drs. P A. Mc- Laughlin, R. Lemaitre, K. Baba of Kuma- moto University, and one anonymous re- viewer for critically reviewing the manu- Script and offering many helpful comments. Prof. Dr. L. B. Holthuis of the National Na- tuurhistorisch Museum, Leiden, kindly pro- vided me with copies of several Russian lit- erature. Literature Cited Derjugin, K. M., & Z. I. Kobjakova. 1935. Zur De- kapodenfauna des japonischen Meeres.—Zool- ogischer Anzeiger 112(5/6):141—147. Gherardi, EF, & P. M. Cassidy. 1994. Sabellarian tubes as the housing of the hermit crab Discorsopa- gurus schmitti.mCanadian Journal of Zoology 72:526—532. Holmes, S. J. 1900. Synopsis of the California stalk- eyed Crustacea.—Occasional Paper of the Cal- ifornia Academy of Science 7:1—262, pls. 1-4. Kobjakova, Z. I. 1956. Zakonomermosty raspreden- leniya desyatinigiskh rakov (Decapoda) v ray- one yuzhonogo Sakhalin. [The natural distri- 627 bution of decapods in the region of southern Sakhalin].—Trudy problemiikh tematicheskikh soveshchanii zoologischeskii institute, Akade- mie Nauk SSSR 6:47-—64, fig. 1. 1958a. Desjatinogi raki (Decapoda) rayona yuzhykl Kurilskikh Ostrovov. [Systematic re- view of the Decapoda of the southern Kurile Islands].—Issledovanija dalinevostochnykh morei SSSR 5:220-248. . 1958b. Sostav i raspredeleniey desjatinogikh rakov (Decapoda) v pribrezhiykh vodax ostro- vov Shikotan i Kunashir. [Composition and dis- tribution of Decapoda in shore waters of the Shikotan and Kunashir Islands].—Issledovanija dalinevostochnykh morei SSSR 5:249-259. Makarov, V. V. 1937. K faune rakov-otshelinikov (Pa- guridae) dalinevostochnikh morei. [Contribu- tion to the Paguridae fauna of the far eastern seas].—Issledovanija morei SSSR 23:55—-67, figs. 1-21 (with English summary). . 1938a. K faune rakov-otshelonikov (Paguri- dae) okrest-nostei ostrova petrova (japonskoe more). [A contribution to the Paguridae fauna in the vicinity of Petrov Island (Japan Sea).] Trudy gidrobiologicheskoi ekspedishii zmm an 1934 na japonskoe more, 1:405—423, figs. 1-5. . 1938b. Rakoobraznyey. Anomura. [Crustaces Decapodes anomures]. in A. A. Shtakelsberg, ed., Fauna SSSR, (n. ser.), 16(10)(3): 1i—x, 1- 324, text figs. 1-113, pls. 1-5. Moscow & Len- ingrad. Akademi Nauk SSSR. . 1962. Crustacea Anomura. Fauna of USSR, 10(3). Israel Program for Scientific Translation, Jerusalem, 278 pp., 5 pls. McLaughlin, P. A. 1974. The hermit crabs (Crustacea Decapoda, Paguridae) of northwestern North America.—Zoologische Verhandelingen 130:1-— 396) Iepl , & R. Lemaitre. 1993. A review of the hermit crab genus Paguritta (Decapoda: Anomura: Pa- guridae) with descriptions of three new spe- cies.—Raffles Bulletin of Zoology 41:1—29. Morgan, G. J., & J. Forest. 1991. A new genus and species of hermit crab (Crustacea, Anomura, Diogenidae) from the Timor Sea, north Austra- lia.—Bulletin du Museum d’ Histoire naturelle, Paris, 4e série, 13, section A nos. 1—2:182—202. Stevens, B. A. 1925. Hermit crabs of Friday Harbor, Washington.—Publications of the Puget Sound Marine Biological Station 3:273—309. . 1927. Orthopagurus, a new genus of Pagur- idae from the Pacific coast.—Publications of the Puget Sound Marine Biological Station 5:245— Dap Takeda, M., & T. Miyauchi. 1992. Anomura and brachyuran crustaceans from the Soya Strait, northern Hokkaido. Natural History Researches of northern Hokkaido (II])—Memoirs of the 628 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON National Science Museum 25:141—153 Gn Jap- shrimps, prawns and crabs from Far East.]— anese with English summary). Izvestija Tikhookeanskogo Nauchno-Issledova- Vinogradov, L. G. 1950. Opredeliteli krevetok, rakov teliskogo Instituta Riibnogo Khozjaistva i i krabov dalinego vostoka. [Classification of Okeanographii 33:179-—358, pls. 1-53. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(4):629—636. 1995 Redescription of Sudanonautes faradjensis (Rathbun, 1921), a fresh-water crab from Central Africa (Brachyura: Potamoidea: Potamonautidae) Neil Cumberlidge Department of Biology, Northern Michigan University, Marquette, Michigan 49855, U.S.A. Abstract.—The fresh-water crab Sudanonautes faradjensis (Rathbun, 1921), previously considered to be a subspecies of S. africanus (A. Milne Edwards, 1869), is recognized here as a valid species. The species is redescribed from the holotype and from a large series of specimens from four countries in Central Africa. Sudanonautes faradjensis is identified by a combination of characters of the carapace, chelipeds, and gonopod |. The distribution of S. faradjensis includes the northern part of the rain forest region of Central Africa, from south Cameroon to Zaire. The Central African fresh-water crab Su- danonautes faradjensis (Rathbun, 1921) was first described from material collected by the American Museum of Natural His- tory Congo expedition of 1911-1915, led by H. Lang and R. Chapin. Although S. far- adjensis was subsequently recognized as a valid species by a number of authors (Balss 1929, 1936; Chace 1942; Capart 1954), more recently others (Bott 1955, Monod 1977) have considered this taxon to be a synonym of S. (S.) africanus chaperi (A. Milne Edwards, 1887). The latter taxon was redescribed by Cumberlidge (1985), who assigned it to the genus Liberonautes, and also indicated that Potamon (Potamonau- tes) faradjensis Rathbun, 1921 was most likely a valid species. Based on the exam- ination of the types of both L. chaperi and S. faradjensis the two species have indeed been found not only to be distinct, but also to belong to different genera. Rathbun’s (1921) original description did not include adequate descriptions of gonopods | and 2, and of a number of other characters which are now thought to be necessary for the proper identification of the species in this genus, and so the species has been rede- scribed here from the holotype from Far- adje, Zaire. Materials and Methods The left mandible and left gonopods 1 and 2 were illustrated following removal from the specimen (Fig. 2a—g). Carapace length (CL), carapace width (CW), cara- pace height (CH), and front width (FW), were recorded from each specimen using digital calipers. Carapace proportions were calculated according to carapace length, and the resulting data pooled and used for descriptions of growth (Fig. 3a, b). Statis- tical comparisons between species were made between sexually mature adults only. Since many literature records are not relia- ble, the distribution of S. faradjensis de- scribed here is based on data from the direct examination of specimens from 20 different localities in four countries. The following abbreviations are used: AMNH, American Museum of Natural His- tory, New York; MNHN, Muséum National d’ Histoire Naturelle, Paris; NHM, The Nat- ural History Museum, London, UK; NNH, National Natuurhistorisch Museum, Leiden, The Netherlands; NMU, Northern Michigan University, Marquette; RCM, Royal Congo Museum, Tervuren, Belgium; USNM, Na- tional Museum of Natural History, Smith- sonian Institution, Washington, DC; ZMB, 630 Museum fiir Naturkunde der Humboldt- Universitat, Berlin, Germany; CW = cara- pace width at widest point; CL = carapace length, measured along median line; CH = cephalothorax height, maximum height of cephalothorax; FW = front width, width of front measured along anterior margin, M = male, F = female, juv = juvenile. Family Potamonautidae Bott, 1970 Sudanonautes faradjensis (Rathbun, 1921) Figs. 1-3 Potamon (Acanthothelphusa) faradjensis Rathbun, 1921:428—430, pl. 31, fig. 13 (type locality: Faradje, Zaire). Potamon (Potamonautes) faradjensis.— Balss, 1929:126, fig. 8; 1936:166—7, fig. 1. Potamon faradjensis.—Chace, 1942:211.— Capart, 1954:833, figs. 8, 20. Sudanonautes (Sudanonautes) africanus chaperi.—Bott, 1955:298—299 (part), pl. XXVII, figs. la—c, 2, 62, 96a, b—Mo- nod, 1977:1216 (part). Not Parathelphu- sa chaperi A. Milne Edwards, 1887. Types.—AMNH 3346, Holotype, Farad- je, Zaire, adult M (CW 71.5, CL 51.0), Dec 1912, coll. H. Lang & R. Chapin. AMNH 3351, paratypes, Faradje, 1 M, 2 E coll. H. Lang & R. Chapin (photographed and il- lustrated by Rathbun, 1921). RCM 836, paratype, van Kerkhovenville, coll. Apr 1912, H. Lang & R. Chapin, AMNH. Material examined.—Cameroon: NHM 25.1.1937/8, Papita section, dense forest country, Lomie District, 1 M, 2 E coll. 1936, Mr. Merfield (from Powell-Cotton Museum). RCM 53.283, Dokwa, 1 M, 3 Jul 1970, coll. Thijs van den Audenaerde. RCM 54.171, Dja Posten, River Dja, 1 M, CW 75.9 mm, 8 Mar 1975, coll. Thijs van den Audenaerde. RCM 54.197, Meri, 26 Jan 1976, 2 M, CW 58.5, 59.7 mm, coll. F Puylaert. GABON.-NNH 24761, Tussen Turindo en Makakou, 1 adult, 7 Mar 1962, coll. B. Condé (donated by J. Gery). SMF 7296, Ivindo u. Mariyots, Makokon, 1 EF CW 66.5 mm, 9 Dec 1975, coll. unknown. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Central African Republic: MNHN B5078, Bangui. NMU 1.1V.1991, river Oubangi, at Bangui, 1 M, 1 K (found with 1 Potamon- autes langi), 1 Apr 1991, coll. L. M. Bour- gault. RCM 55.400, Landjia river Ouban- cul, | M, 1 3 Feb 1982) coll} Exdegvos & Kempeneers. RCM 56.349, Kembé, above the falls on the river Kotto, 29 Mar 1984, 2 adults, coll. J. P Marquet. Zaire: NMU VII.1993, Kinshasa, Zaire river, 4 M, 8 Jul 1993, coll. L. M. Bourgault. USNM 54300, Faradje, 1 M, 1 K 7 Mar 1915. SMF 2382, Lisala, 1 M, CW 48.7 mm, coll. S. Deheyn (RCM 32071). RCM 71, river Ubangi, Banzyville, 1 juv, 1901, coll. G. T. Royaux. RCM 202, Equaterville, 1 F juv, 1921, coll. Verlaine. RCM 258, Leopold- ville (= Kinshasa), | specimen, 1933, coll. M. A. Cinant. RCM 50.379, Stanleypool (= Kisangani), 1 E 17 Jul 1957, coll. P. Brien, M. Poll, & J. Bouillion. RCM 50.380, Stan- leypool (= Kisangani), 1 M, Apr 1957, coll. P. Brien, M. Poll, & J. Bouillion. RCM 54.043, Stanleypool (= Kisangani), 2 spec- imens, 1 Mar 1953, coll. J. Mandeville. ZMB 13494, Dumé, 1 EF coll. 29 Jun 1909, O. Freyer. Diagnosis.—Postfrontal crest spanning entire carapace, crest deeply notched before meeting sharp, forward-pointing epibran- chial tooth; shallow notch in crest behind exo-orbital tooth; series of 7 to 8 sharp teeth on anterolateral margin posterior to epibranchial tooth. Posterior lateral surface of carapace with fields of raised short lines; cervical, semi-circular, cardiac, urogastric grooves very deep. Epibranchial, interme- diate and exorbital teeth all large, sharp, pointed forward. Terminal segment of gon- opod | long, thin and needle-like, subter- minal segment of gonopod | slender. Car- apace very flat (CH/CL = 0.41). Redescription.—Carapace (Fig. la-—c) ovoid, widest on anterior third (CW/CL 1.36), very flat (CH/CL 0.41), cervical, semi-circular, urogastric, cardiac grooves very deep; transverse branchial grooves present but weak. Front bilobed, indented, anterior margin curving down, relatively VOLUME 108, NUMBER 4 631 y Fig. 1. Sudanonautes faradjensis (Rathbun, 1921), adult male holotype, (CW 71.5 mm) from Faradje, Zaire (AMNH 3346). a, whole animal, dorsal aspect; b, carapace, frontal aspect; c, carapace showing detail of epi- branchial corner; d, sternum; e, left third maxilliped; f, abdomen; g, right cheliped, frontal view; h, left cheliped, frontal view; i, carpus, and merus of right cheliped, superior view; j, carpus, and merus of right cheliped, inferior view; k, left pereiopod 2. Scale bar equals 30 mm (a, b, c, d, f), 15 mm (g—k), 10 mm (e). 632 narrow, less than % carapace width (F'W/ CW = 0.30). Posterior lateral surface of carapace with fields of raised short lines, anterolateral surface with faint raised gran- ules. Postfrontal crest spanning entire car- apace, consisting of fused epigastric, post- orbital crests, almost straight in epigastric regions, curving forward behind orbits to meet anterolateral margins at the epibran- chial teeth; crest smooth in middle, slight crenulations at lateral ends. Postfrontal crest deeply notched before meeting sharp, for- ward-pointing epibranchial tooth; shallow notches in crest behind exo-orbital tooth and behind front; mid-groove on postfrontal crest short, forked at posterior end. Series of 7 to 8 small sharp teeth on anterolateral margin posterior to epibranchial tooth; an- terolateral margin continuous with postero- lateral margin. Posterior margin about % as wide as carapace width. Fields of conspicuous granules in sub- orbital regions. Suborbital, subhepatic, and pterygostomial regions with 2 sutures, | longitudinal, 1 vertical, dividing area into 3 parts (Fig. 1b, c). Longitudinal suture di- viding suborbital, subhepatic regions from pterygostomial region, beginning medially at lower margin of orbit, curving backward across entire region. Vertical suture short, v-shaped, dividing suborbital region from subhepatic region marked by row of small rounded granules (Fig. 1b, c); suture origins between exo-orbital and intermediate teeth, curving down to meet longitudinal suture. Third maxillipeds filling entire oral field, except for transversely oval efferent respi- ratory Openings at superior lateral corners; long, plumose flagellum on exopod of third maxilliped (Fig. le); ischium of third max- illiped smooth, with clear vertical groove (Fig. le). Mandibular palp 2-segmented; terminal segment single, undivided, fringed with hairs, longest at junction between seg- ments (Fig. 2a—c). First transverse groove on sternum (between sternal segments 2 and 3) complete; second groove (between sternal segments 3 and 4) consisting of 2 small notches at sides of sternum (Fig. 1d). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Segments 1—6 of abdomen four sided, last segment triangular, sides indented, rounded at distal margin (Fig. 1f); segment 3 broad- est, segments 4—7 tapering inwards (Fig. If). Gonopod | with very slender terminal segment, long (7% as long as subterminal segment), almost straight continuation of subterminal segment, only gently curved outward, tapering to pointed tip, longitudi- nal groove visible from caudal and superior views (Fig. 2d, f), not visible from cephalic view (Fig. 2e). Subterminal segment of gonopod 1 very slim (Fig. 2d, e), with raised flap extending halfway across seg- ment, margin vertical, edged with small spines, flap forming roof of chamber for gonopod 2; subterminal segment beneath flap forming lower floor of chamber for gonopod 2 (Fig. 2d). Gonopod 2 (Fig. 2g) shorter than gonopod | (reaching only to junction between last 2 segments of gono- pod 1). Terminal segment of gonopod 2 ex- tremely short, only % as long as subterminal segment, sides folded inwards and spoon- shaped, tip rounded. Subterminal segment gonopod 2 widest at base, tapering sharply inward, forming long, thin, pointed, upright process which supports short terminal seg- ment; rounded collar at junction between terminal segment and subterminal segment. Chelipeds (Fig. 1g—j) unequal, right lon- ger, higher than left. Dactylus of right che- liped long, slender, grooved, not arched; palm of propodus swollen; proximal region of fingers of digits of right cheliped each with 2 large teeth, opposing teeth meeting, forming small space at base, cut off from the longer interspace distal to these teeth. Both fingers with 3 intermediate sized pointed teeth, interspersed with a series of smaller pointed teeth along their lengths (Fig. 1g, h). Teeth of smaller cheliped all very small, fingers almost meeting when closed. Inferior margins of merus with rows of small teeth, cluster of granules surround- ing larger pointed tooth at distal end. Inner margin of carpus of cheliped with 2 large, slender, pointed teeth, second half size of VOLUME 108, NUMBER 4 633 Fig. 2. Sudanonautes faradjensis (Rathbun, 1921). a, left mandible anterior view; b, left mandible posterior view; c, left mandibular palp, superior view; d, left gonopod 1, caudal view; e, left gonopod 1, cephalic view; f, terminal segment of left gonopod 1, superior view; g, left gonopod 2, caudal view. Scale bar equals 3 mm (a—c), 2 mm (d—g). (a-f, adult male holotype (CW 71.5 mm) from Faradje, Zaire, AMNH 3346; g, adult male (CW 66 mm) from Kinshasa, Zaire, NMU VII.1993.) 634 CW = -0.22+1.36CL r=0.99 CH= -1.1 +0.43CL r=0.97 FW = 0.64 +0.36CL r=0.91 Carapace Dimension (mm) 20 2 30 35 40 45 50 55 60 65 70 Carapace Length (mm) igus: PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON CWI/CL = 1.35 +0.0001CL r= 0.031 1.0 CH/CL = 0.37 + 0.0008CL r= 0.291 0.9 FW/CL = 0.38 - 0.0003CL r= 0.089 Carapace Dimension 0.2 20 25 30 35 40 45 50 55 60 65 70 Carapace Length (mm) Comparisons of 24 specimens of Sudanonautes faradjensis (Rathbun, 1921). a, dimensions of the carapace (CW, CH, FW) compared to body size (CL), r values (all at df = 23) indicate a highly significant correlation (P < 0.001) between size classes; b, relative proportions of carapace width (CW/CL), carapace height (CH/CL), and front width (FW/CL), compared to body size (CL), 7 values (all at df = 23) indicate no significant correlation (P > 0.05) between size ciasses. first (Fig. li, j). Left cheliped similar to right, but smaller (Fig. 1h). Walking legs (pereiopods 2-5) slender (Fig. 1k), fourth longest, fifth shortest. Posterior margin of propodi serrated, dactyli tapering to point, each bearing rows of downward-pointing sharp bristles; dactylus of fifth shortest of the walking legs. Stridulating apparatus as- sociated with coxae of pereiopod 2. Adult female.—Right and left chelipeds similar proportions to male of similar size, unequal in both length and height. Mature female abdomen very wide, reaching coxae of pereiopods 2—5. Segments of female ab- domen 4-sided, becoming gradually longer distally, segments 1-5 becoming gradually wider (widest at groove separating seg- ments 4, 5); segment 6 and telson together forming near semicircle. Growth (Fig. 3a, b).—Carapace dimen- sions and relative proportions as shown in Fig. 3a, b. Sexual maturity judged by de- velopment of female abdomen: abdomen of mature females overlapping bases of coxae of walking legs, pleopods broad, hair- fringed. Pubertal molt occurring between CW 35-45 mm. Largest known specimen (male from Cameroon) CW 90 mm. Rela- tive width of carapace (in relation to cara- pace length, CW/CL) not changing with age: CW/CL of adults not significantly wid- er (P > 0.05) than that of juvenile and pu- bescent animals (Fig. 3b). Width of frontal margin (FW/CL) not changing with age: VOLUME 108, NUMBER 4 FW/CL of adults not significantly wider (P > 0.05) than that of juvenile and pubescent animals (Fig. 3b). Relative height of cara- pace (CH/CL) not changing with age: that of adult S. faradjensis not significantly dif- ferent (P < 0.05) than CH/CL of juvenile and pubescent animals. Shape of gonopod 1 and chelipeds changing as crabs grow older. Terminal segment of gonopod | of juveniles almost straight; right and left che- lipeds of juveniles (CW 25—40 mm) are even sized, palms not inflated, cutting edges meeting, not enclosing a space. Color.—(Based on living adult from Kinshasa, Zaire). Dorsal carapace and post- frontal crest dark brown-black, flanks paler brown; eyestalk brown, cornea black; ster- num, abdomen light brown. Arthrodial membranes between joints of chelipeds and pereiopods light brown; chelipeds, pereio- pods light brown. Distribution.—Rain forest regions of Cameroon, Central African Republic, Ga- bon, Zaire. Middle reaches of the Zaire riv- er basin, and in the rivers Ubangui and Uele, tributaries of the Zaire river. Distri- bution in Zaire described by Rathbun (1921) and Balss (1936), and in Cameroon by Balss (1929). Ecological Notes.—Sudanonautes far- adjensis is restricted to the more humid ar- eas of the rain forest from south Cameroon to Zaire. This species occurs in permanent aquatic habitats from large rivers to small streams. The following notes are based on Herbert Lang’s observations (leader of U.S.A. Congo Expedition) as reported by Rathbun (1921). Sudanonautes faradjensis is one of the largest river-living crabs of the Uele district in the Zaire basin. The species is found under stones in quiet portions of the river near rapids, where it is reportedly preyed upon by small crocodiles and car- nivorous fish. Sudanonautes faradjensis moves quickly and can usually escape from its predators. Remarks.—The short terminal segment of gonopod 2 (Fig. 2g), the long, slender, outward-curving terminal segment of gon- 635 opod | (Fig. 2d), and the intermediate tooth on the anterolateral margin between the exo-orbital and epibranchial teeth (Fig. la— c) place this species in Sudanonautes. Iden- tification of S. faradjensis to species can be made by noting the following characters: a very flat carapace (CH/CL = 0.41) with fields of raised short lines on the posterior lateral surface; large, sharp, forward-point- ing epibranchial, intermediate and exorbital teeth; a series of 7 to 8 sharp teeth on the anterolateral margin posterior to the epi- branchial tooth; a deep notch on the post- frontal crest before it meets the epibranchial tooth; and a long, thin and needle-like, ter- minal segment of gonopod 1. Comparisons.—Sudanonautes faradjen- sis is most likely to be confused with other large species of fresh-water crabs occurring in the rain forest zones of Central Africa, such as S. chavanesii (A. Milne Edwards, 1886) and S. africanus (A. Milne Edwards, 1869). Since the terminal segment of gon- opod 1 of all three species is similar (but not identical), other characters should be used to distinguish between species. For ex- ample, S. faradjensis can be distinguished from both S. africanus and S. chavanesii by examination of the anterolateral margin: that of S. faradjensis possesses rows of sharp teeth behind the epibranchial tooth (Fig. la), whereas the anterolateral margins of S. africanus and S. chavanesii are both smooth. In addition, the epibranchial tooth of S. faradjensis is sharp and pointed (Fig. la), whereas that of S. africanus is small and low. Furthermore, the epibranchial tooth of S. faradjensis is in line with the mid-point of the postfrontal crest (Fig. 1a), whereas that of S. chavanesii is set back posteriorly, well behind the line of the mid- point of the postfrontal crest. In addition to the above differences, there are a number of characters that distinguish S. faradjensis from the other species in the genus. For example, the carapace of S. far- adjensis is significantly flatter (P < 0.001) than most other species in the genus, and the carapace of S. faradjensis is rough-tex- 636 tured with patches of raised blisters and ridges in the posterior region, and deep uro- gastric and cardiac grooves (Fig. la). The spiny river crab, Liberonautes chap- eri (A. Milne Edwards, 1887), redescribed by Cumberlidge (1985) and Cumberlidge & Sachs (1989), is superficially similar to S. faradjensis. However, the two species can be easily distinguished by the terminal seg- ment of gonopod 2: that of L. chaperi is long, like a flagellum, whereas that of S. faradjensis is short and stubby (Fig. 2g). In addition, the terminal segment of gonopod 1 of L. chaperi curves inward, not outward as in S. faradjensis (Fig. 2d). Finally, L. chaperi is found only in West Africa from Cote d’Ivoire to Guinea, whereas S. farad- jensis is restricted to the forests of Central Africa. Acknowledgments Christopher B. Boyko, Department of In- vertebrates (AMNH) is thanked for the loan of the holotype and paratypes of S. farad- jJensis. The following are thanked for host- ing visits: H. Feinberg (AMNH); Paul Clark (NHM); D. Guinot and J. Forest (MNHN); L. Holthuis and C. Fransen, NNH, Leiden, The Netherlands; R. Joqué (RCM); R. Man- ning (USNM); and H. Gruner (ZMB). I es- pecially thank artist Jon C. Bedick of Northern Michigan University, for all of the illustrations used in this paper. Part of this work was supported by a Faculty Grant from NMU, Marquette. Literature Cited Balss, H. 1929. Potamonidae au Cameroon. Jn Con- tribution a l’étude de la faune du Cameroun.— Faune des Colonies Frangaises 3:115—129. . 1936. Beitrage zur Kenntnis der Potamidae PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON (Stisswasserkrabben) des Kongogebeites.—Re- vue du Zoologie et Botanie d’ Afrique 28:65— 204. Bott, R. 1955. Die Sitisswasserkrabben von Afrika (Crust., Decap.) und ihre Stammesgeschichte.— Annales du Musée Royal du Congo Belge, C- Zoologie Série III, HI 1(3):209-352. . 1970. Betrachtungen tiber die Entwicklungs- geschichte und Verbreitung der Siisswasser- Krabben nach der Sammlung des Naturhisto- rischen Museums in Genf/Schweiz.—Revue Suisse de Zoologie 77(2), 24:327—344. Capart, A. 1954. Révision des types des espéces de Potamonidae de |’ Afrique Tropicale conservés au Muséum d’Histoire Naturelle de Paris.— Volume Jubilaire Victor Van Strallen, Director de I’Institut Royale des Sciences Naturelles de Belgique, 1925-1934 II:819-847. Chace, F A. 1942. Scientific results of a fourth ex- pedition to forested areas in eastern Africa, III: Decapod Crustacea.—Bulletin of the Museum of Comparative Zoology 91:185—233. Cumberlidge, N. 1985. Redescription of Liberonautes chaperi (A. Milne Edwards, 1887) n. comb., a fresh-water crab from Ivory Coast (Brachyura, Potamonautidae).—Canadian Journal of Zoolo- gy 63:2704—-2707. , & R. Sachs. 1989. A key to the crabs of Liberian freshwaters.—Zeitschrift fiir Ange- wandte Zoologie 76:221—229. Milne Edwards, A. 1869. Révision du genre Thelphu- sa et description de quelques especes nouvelles faisant partie de la collection du Muséum.— Nouvelles Archives du Muséum d’ Histoire na- turelle, Paris 5:161—191. . 1886. La description de quelques Crustacés du genre Thelphusa recueillis par M. de Brazza dans les régions du Congo.—Bulletin de la So- ciété Philomathique de Paris, Série 7 10:148— IS). . 1887. Observations sur les crabes des eaux douces de 1’ Afrique-—Annales des Sciences Naturelles, Zoologie, Paris (7)4:121—-149. Monod, T. 1977. Sur quelques crustacés Décapodes africaines (Sahel, Soudan).—Bulletin de Mu- séum national d’ Histoire naturelle, Paris 3, 500: 1201-1236. Rathbun, M. J. 1921. The Brachyuran Crabs collected by the American Museum Congo expedition 1909-1915.—Bulletin of the American Muse- um of Natural History 43:379-474. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(4):637-642. 1995 Lithoscaptus pardalotus, a new species of coral-dwelling gall crab (Crustacea: Brachyura: Cryptochiridae) from Belau Roy K. Kropp Battelle Ocean Sciences, 397 Washington Street, Duxbury, Massachusetts 02332, U.S.A. Abstract.—A new species of coral-dwelling gall crab, Lithoscaptus pardal- otus, was found inhabiting the faviid coral Favia cf. helianthoides at Ngeme- lachel, Belau. The new species can be distinguished from others in the genus by the combination of having the internal orbital angle exceeding the external orbital angle and having very slender carapace that is nearly twice as long as wide. Among collections of gall crabs (Cryp- tochiridae) I made in Belau [Palau] in 1984, were several specimens that could not be assigned to any described species. These specimens, collected from colonies of Fa- via cf. helianthoides, are described as a new species of the genus Lithoscaptus Milne Edwards, 1862. Type material has been deposited in the National Museum of Natural History, Smithsonian Institution, Washington, D.C. (USNM), the Bernice P. Bishop Museum, Honolulu, Hawaii (BPBM), the Queensland Museum, Brisbane, Australia (QM), and the Muséum National D’ Histoire Naturelle, Paris (MNHN). Abbreviations used are: cl, carapace length; cw, carapace width; coll., collector; ft, feet; m, meters; mm, millimeters; MXP, maxilliped, ovig., ovigerous; P, pereopod; and PLP, pleopod. PHAP numbers refer to the author’s collection numbers. Place names in Belau follow orthography pro- posed by Motteler (1986). The former place name is presented in brackets. Lithoscaptus pardalotus, new species Figs. 1, 2 Type material.—Holotype: Belau, Nge- melachel [Malakal] Is., off causeway by park S of Micronesian Mariculture Dem- onstration Center; 1 m; 18 Jul 1984; on Fa- via cf. helianthoides (Wells); PHAP 131; °, USNM 274140. Paratypes: Belau; all same locality and host as holotype, 1—3.5 m; PHAP 058; 4 Jul 1984; 1 2, BPBM S$11274; PHAP 109; 14 Jul 1984; 2 2, 2 6, USNM 274142; PHAP 114; 14 Jul 1984; 2 2 (ovig.), MNHN B. 25100; PHAP 127; 18 Jul 1984; 3 & (2 ovig), MNHN B. 25101; PHAP 128; 18 Jul 1984; 1 4, USNM 274141; PHAP 220; 27 Jul 1984: 2 2 Cd ovig), QM W20647. Measurements.—Holotype: cl, 4.0 mm, cw, 2.2 mm. Paratypes: largest male: cl, 2.9 mm, cw, 1.6 mm; largest female: cl, 4.4 mm, cw, 2.5 mm; smallest ovigerous fe- male: cl, 3.0 mm, cw, 1.8 mm. Description.—(Female) Carapace convex in both directions, about 1.8 times longer than wide, widest just anterior to midlength, narrowing slightly anteriorly; in lateral view anterior portion sloping from mid- length toward front; dorsal surface with W- shaped depression, with many scattered conical and round tubercles. Lateral mar- gins of carapace subparallel; anterolateral margins lined with erect conical tubercles. Intraorbital margin concave, with scattered simple setae, irregularly denticulate; inter- nal orbital angle extending beyond external, each orbital angle marked with prominent spine; orbit broadly V-shaped, margin gran- ular. Posterior carapace convex, with scat- 638 tered round tubercles decreasing in size posteriorly; cardiac region marked by in- verted U-shaped furrow; posterolateral margins rounded, with rounded tubercles and granules. Carapace bearing scattered simple setae on entire surface. Pterygosto- mial region fused to carapace. Basal segment of antennule extending slightly beyond eye, subtriangular in dorsal view (excluding apical spine), mostly smooth but with few granules and scattered setae dorsally, margins with 8—9 spines, dis- tal apex marked with prominent spine; ven- trally bearing scattered granules, no setae, subtriangular in shape; outer margins sub- equal, inner margin concave, edge rounded. Eyestalks fully exposed dorsally, slightly swollen proximally, with scattered setae, mesial margin bearing prominent spine dis- tally, 2 smaller spines subdistally; smooth ventrally; eyestalk length about 1.9 times cornea length. Ischium of MXP-3 about 1.5 times lon- ger than wide, surface mostly smooth, gran- ular near distal margin, inner margin Straight, distal half crenulate, lined with simple setae; outer margin denticulate, bearing few simple setae; exopod about half length of outer margin of ischium, with few pappose setae on outer margin. Merus gran- ulate toward outer margin, latter finely den- ticulate, bearing few simple setae along en- tire length, distal angle produced, tubercu- late. Palp segments bearing few simple se- tae along outer margins. Cheliped (P-1) slightly shorter than first walking leg (P-2). Merus compressed, length about 2.2 times height; inner surface concave, smooth; outer surface convex, smooth; anterior margin smooth, with few simple setae; outer posterior margin with few granules, inner posterior margin serrat- ed. Carpus length about 1.6 times height; outer surface smooth, but with granules near anterior margin; anterior margin with few low conical tubercles directed distally, with few scattered setae. Manus length about 1.4 times height, smooth, but with few granules proximally near anterior mar- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON gin. Cutting edges of fingers entire, without setae; dactyl length about 0.6 times total chela length. Merus of P-2 compressed, length about 2.6 times height; inner surface concave, and outer surface slightly convex, both surfaces smooth; anterior margin straight, curving slightly downward distally, with few gran- ules along length and erect conical tuber- cles distally, with scattered simple setae along most of length; posterior margin straight, lined with angled tubercles giving serrated appearance, with proximally-pap- pose setae; posterodistal angle with 3 blunt tubercles. Carpus length about 2.0 times height; subtriangular in cross-section; an- terior surface slightly rounded, sloping me- sially, with erect conical tubercles, few sim- ple setae; inner surface smooth, with simple setae at anterior margin; outer surface flat, mostly smooth, with granules on upper half, some tubercles at anterior margin, latter with simple setae. Propodus about 2.8 times longer than high, subtriangular in cross-sec- tion; outer surface flat, with scattered gran- ules decreasing in size toward posterior margin; anterior surface flat, with erect con- ical tubercles and scattered setae; inner sur- face smooth, margin with 4 erect conical tubercles. Dactyl curved distally, with 2 tu- bercles proximally on upper surface, with few simple setae. Second walking leg (P-3) smaller than P-2; merus length about 2.0 times height; outer surface slightly convex, smooth ex- cept for few tubercles on upper distal cor- ner; inner surface concave, smooth; anterior margin straight, distal third curves sharply downward; margin bearing scattered simple setae, with tubercles on distal third; poste- rior margin relatively straight, lined with proximally-pappose setae, with | prominent and 2 smaller tubercles subdistally; postero- distal angle rounded, entire. Carpus sub- triangular in cross-section, length about 1.9 times height; inner surface flat, smooth; outer surface mostly smooth, with row of tubercles at upper third; anterior margin with row of erect conical tubercles, with VOLUME 108, NUMBER 4 639 Fig. 1. Lithoscaptus pardalotus, new species, female holotype (carapace length 4.0 mm). a, Carapace (dorsal view); b, Carapace (lateral view); c, Left antennule, antenna, and eyestalk (ventral view); d—h, Right P1—P5; i, MXP-3. Scale: 1.0 mm (a—h); 0.46 mm (i). few setae. Propodus slightly shorter than carpus, length about 2.3 times height; an- terior margin with row of erect conical tu- bercles; outer surface mostly smooth, with few rounded tubercles toward anterior mar- gin; inner surface smooth, with few setae. Dactyl curved distally, entire. Third walking leg (P-4) slightly smaller than P-3; merus length about 1.7 times height; outer surface slightly convex, smooth, but with few rounded tubercles near anterodistal angle; inner surface con- vex, smooth, bearing few setae; posterior margin with | larger, 1 smaller tubercle subdistally; anterior margin slightly convex, with few simple setae, with angled tuber- cles giving serrated appearance on distal third. Carpus length 2.7 times height; outer surface convex, smooth, with few setae; posterior margin entire; anterior surface 640 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Figa2: Lithoscaptus pardalotus, new species, male paratype (carapace length 2.9 mm. a, Carapace and right P2—P5 (dorsal view); b—f, Right P1—P5 (lateral view); g, Abdomen; h, PLP-1 (with tip enlarged). Scale: 1.0 mm (a—g); 0.46 mm (h). rounded, sloping toward outer surface, out- er margin with 3 rounded tubercles, with few simple setae; inner surface slightly con- vex, smooth. Propodus rotated slightly me- sially, length about 3.2 times height; outer surface slightly convex, smooth, with few simple setae; inner surface slightly convex, smooth, with few setae; posterior margin entire, with few setae; anterior margin straight, with 6 erect rounded tubercles, with scattered setae. Dactyl curved distally, rotated mesially, entire, with few scattered simple setae. Fourth walking leg (P-5) more slender, longer than P-4; merus not compressed, length 1.9 times height; outer, inner sur- faces convex, smooth, with scattered setae; posterior margin rounded, with tubercle near posterodistal angle; anterior margin rounded, convex, having few rounded tu- bercles distally, bearing few scattered sim- ple setae. Carpus length 2.9 times height; VOLUME 108, NUMBER 4 outer, inner surfaces slightly rounded, smooth, with scattered simple setae; poste- rior margin rounded, entire, having few simple setae; anterior margin rounded, with tubercles near distal margin, with few sim- ple setae. Propodus rotated about 90° me- sially, length 4.2 times height; outer, inner surfaces rounded, smooth, outer with scat- tered simple setae, inner having few round- ed tubercles and few pappose setae. Dactyl curved distally, rotated slightly mesially, smooth, with many simple setae on inner surface, posterior margin. Male smaller than female. Carapace pear- shaped in dorsal view, mostly smooth; an- terolateral margin with many spines; ante- rior depression wider, smoother than that of female. Chelipeds more robust than in fe- male; merus, carpus, and propodus with acute, angular tubercles along anterior mar- gins. Dactyl with angular tubercles proxi- mally; cutting surface of dactyl with tooth subproximally; fixed finger with low ridge running length of cutting surface. First pleopod as figured. Coloration.—Carapace of female off- white, anterior part with many regularly- spaced black spots. Eyestalks mostly white with black wedge dorsally, red-orange stripe laterally; cornea dark red with white spot. Basal segment of antennule white with scattered black dots and larger black area proximally; palp clear, with white mark- ings. Ventrum opaque, with scattered irreg- ularly-shaped black spots; P-1 and P-2 bright white dorsally, last 3 segments of P-1 and dactyl of P-2 with rose hue. Male carapace clear, covered with reticulated black spots except on anterior depressions. Eyestalk as carapace, corneas dark red with silver spot. Ventrum opaque; MXP-3 with black spots. P-1 through P-5 off-white with scattered irregular black and blue-white spots. Etymology.—From the Greek “‘pardalo- tos’, spotted like a leopard, in reference to the black spots on the anterior part of the carapace. Remarks.—Lithoscaptus pardalotus, new 641 species, L. tri (Fize & Seréne, 1956), and L. nami (Fize & Seréne, 1957) can be dis- tinguished from the other species presently included in the genus Lithoscaptus (Kropp 1990, 1994) by having the internal orbital angle extending beyond the external orbital angle. The slenderness of the new species, with a carapace length almost twice its width, distinguishes it from L. tri and L. nami in which the carapace length is about 1.3 times its width. The black-spotted pat- tern on the anterior carapace also differs from patterns found on other species. Hosts.—Faviidae: Lithoscaptus pardalo- tus was collected on Favia cf. helianthoi- des. I can find no other references to gall crabs occurring on this species of coral. Distribution.—Micronesia: Belau. Acknowledgments I thank the University of Guam Marine Laboratory, the University of Maryland, and the USNM for support during field work in Micronesia. Funding for field work was provided in part by a grant from the Biological Oceanography Section of the National Science Foundation to G. J. Ver- meij. Field work in Belau was supported by the Micronesian Mariculture Demonstration Center. My laboratory work at the USNM was kindly supported by R. B. Manning. I thank R. H. Randall for identifying the host coral. This is University of Guam Marine Laboratory Contribution Number 365. Literature Cited Fize, A., & R. Serene. 1956. Note préliminaire sur quatre espéces nouvelles d’Hapalocarcinidés avec quelques remarques au sujet du Crypto- chirus rugosus Edmonson [sic].—Bulletin de la Société Zoologique de France 80(5—6):379— 382, fig. 1 [issue date 1955]. DIOS 1957. Les Hapalocarcinidés du Viet-Nam. Mémoires de [’nstitut Océanograp- hique de NHATRANG, 10° Mémoire. in Ar- chives du Muséum National d’ Histoire Naturel- le, Paris (7) 5:1—202, figs. 1-43, pls. 1-18. Kropp, R. K. 1990. Revision of the genera of gall crabs (Crustacea: Cryptochiridae) occurring in 642 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON the Pacific Ocean.—Pacific Science 44:417— de la Réunion. Pp. F1—F16 in L. Maillard, Notes 448. sur l’ile de la Réunion (Bourbon), Paris. 1994. Gall crabs of the Rumphius Expedi- Motteler, L. S. 1986. Pacific island names: a map and tions with the descriptions of three new spe- name guide to the new Pacific_—B. P. Bishop cies.—Raffles Bulletin of Zoology 42:521—-538. Museum Miscellaneous Publications 34:1—91. Milne Edwards, A. 1862. Faune carcinologique l’ile PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(4):643—648. 1995 Charybdis hellerii (Milne Edwards, 1867), a nonindigenous portunid crab (Crustacea: Decapoda: Brachyura) discovered in the Indian River lagoon system of Florida Rafael Lemaitre Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, U.S.A. Abstract.—The portunid crab Charybdis hellerii (Milne Edwards, 1867), a species of Indo-Pacific origin, is reported from the Indian River lagoon system of Florida. Although this species was reported in 1987-1988 from Cuba, Co- lombia, and Venezuela, this is the first subsequent record in the western At- lantic, and the first in the continental United States. The specimens found in the Indian River include adults of both sexes (one female ovigerous), and juveniles, indicating that the population is reproducing and that this nonindig- enous species has become established in the region. This species must now be considered present in the tropical western Atlantic at least across the entire Caribbean region. Recognition characters of C. hellerii are provided, as well aS a Summary of what is known about its distribution, biology, and probable method of arrival. Comments on other marine decapods introduced in eastern North America are included. On 20 April 1995, two adult specimens, a male and an ovigerous female of Charyb- dis hellerii (Milne Edwards, 1867), a species of Indo-Pacific origin, were caught in baited traps in the Indian River lagoon system of Florida in the vicinity of the Ft. Pierce Inlet. The male, 74 X 49 mm of carapace width X carapace length, remained alive in an aquarium for over a month until it was sac- rificed. The ovigerous female (54 39 mm), died the same day it was captured. Follow- ing an announcement at The Crustacean So- ciety’s summer meeting (Lemaitre 1995) of the discovery of these specimens, several re- searchers (see names in acknowledgments) working in the Indian River during the pe- riod May—August 1995, informed me that they had found additional adults and juve- niles in the Sebastian and Fort Pierce Inlets. Charybdis hellerii was first collected in the western Atlantic in 1987 (Fig. 1), in Cuba by Gomez & Martinez-Iglesias (1990), in Venezuela by Hernandez & Bolafios (1995), and in Colombia by Campos & Tir- kay (1989) where it was also collected in 1988. Although there have been no subse- quent published reports of this species, it has been found repeatedly since 1988 in the Santa Marta area of Colombia (N. H. Cam- pos, pers. comm.). The present report of C. hellerii in eastern North America extends the range of this species in the western Atlantic across the Caribbean region, to include sub- tropical Florida. Keys for the identification of species of Charybdis as well as illustrations of C. hel- lerii, are available in a number of publica- tions (e.g., Chopra 1935, Leene 1938, Ste- phenson et al. 1957, Stephenson 1972, Go- mez & Martinez-Iglesias 1990). However, for the benefit of researchers working in the New World who may not have easy access to the earlier, more detailed taxonomic references, it seems appropriate to include here the most important recognition characters of this new- ly introduced species. Such information may facilitate future discovery and monitoring of populations of this portunid crab. 644 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Rigas in chronological order: 1, 2, Cuba: NE coast, Bahia Gibara (Provincia Holguin), and S coast, Bahia Cienfuegos (Provincia Cienfuegos) (G6mez & Martinez-Iglesias 1990); 3, NE Venezuela: Ensenada de Turpialito, Golfo de Cariaco (Estado Sucre) (Hernandez & Bolafios 1995); 4, 5, Colombia: Bahia de Portete (Guajira), and Bahia Chengue (Magdalena) (Campos & Tiirkay 1989); 6, eastern USA: Indian River, Ft. Pierce, Florida (this report). Method of capture.—The traps used were cylindrical, and consisted of 50 cm-long sections of PVC pipe of 20 cm diameter. The ends of each cylinder trap were fitted with cone entrances made of galvanized hardware cloth with a | cm mesh size. The cones were placed at each end with the nar- row end facing the inside of the pipe. A small plastic bait basket was tied in the midsection of each trap. Bait used was mul- let, purchased locally. Arrays of four traps were fixed to a long line, with a 20 kg lead weight on each end of the line. Two such arrays were left overnight in the vicinity of the Ft. Pierce Inlet: one in a channel south of Little Jim Island (in 1.5 m depth); and another in the Ft. Pierce turning basin just Map showing collecting dates and locations of capture of Charybdis hellerii (Milne Edwards, 1867) north of South Bridge (about 20 m from the mainland shoreline, in 4 m depth). After capture, the specimens were transported to the laboratories of the Smithsonian Marine Station at Link Port, Florida. The adult male and ovigerous female specimens have been deposited in the collections of the Na- tional Museum of Natural History, Smith- sonian Institution, Washington, D.C., under catalogue number USNM 275907. Family Portunidae Charybdis hellerii (Milne Edwards, 1867) JEG Me Z Recognition characters.—Carapace with dorsal surface naked; anterolateral margin VOLUME 108, NUMBER 4 645 Riga: Ft. Pierce, Florida, USNM 275907. with 6 sharp, black-tipped teeth (including outer orbital). Frontal region with 6 prom- inent teeth: 2 inner orbitals plus 4 blunt submedians, latter reaching slightly in ad- vance of orbitals. Chela stout; palm with 5 strong black-tipped spines on dorsal face. Swimming leg with merus and carpus each armed with strong black-tipped spine on posterior margin distally; propodus with row of spines on posterior margin. Sixth ab- dominal segment of male about as broad as long, posterodistal margins rounded; telson bluntly triangular, basal width less than dis- tal width of sixth abdominal segment. Male first gonopods, when viewed in situ, reach- ing approximately to suture between fifth and sixth thoracic somite, diverging in dis- tal third; each with row of about 25 slender spines on lateral margin distally (spines di- minishing in length distally). Coloration.—Overall dark green except Charybdis hellerii (Milne Edwards, 1867), male (74 X 49 mm), channel south of Little Jim Island, for light purple area on upper, inner surface of palm, and dark purple on dorsal surfaces of distal 4 segments of walking and swim- ming legs. Carapace with pale green or whit- ish areas on frontal, hepatic and epibranchial regions. Fingers of chelipeds dark purple. Distribution.—Indo-Pacific: Japan, Phil- ippines, New Caledonia, Australia, Hawaii, and throughout the Indian Ocean, including the Red Sea (Stephenson 1972, Kathirvel & Gopalakrishnan 1974, Vannini 1976, Javed & Mustaquim 1994). Eastern Mediterranean: Israel, and Egypt (Steinitz 1929, Galil 1992). Western Atlantic: eastern Florida, Cuba, Co- lombia, and Venezuela (Campos & Ttrkay 1989, Gédmez & Martinez-Iglesias 1990, Hernandez & Bolanos 1995, this report). Biology.—This species is considered commercial in southeast Asia (Moosa 1981). Not much is known about its biol- ogy. The larvae of C. hellerii, for example, 646 have not been described. Other species of Charybdis for which the complete larval development is known, have five to six zoeal stages and a megalopa (Greenwood & Fielder 1980). Fecundity of C. hellerii, as expressed by number of eggs, has been re- ported to range from 22,517 to 292,050 eggs per female (Siddiqui & Ahmed 1992). Other congeneric species (e.g., C. cruciata, C. hoplites pusilla, C. natator, C. feriatus) with biology probably similar to that of C. hellerii, are known to spawn year round but with peaks in early spring and fall, and have fecundities ranging from 181,230 to 3,200,000 million eggs per female (Pillai & Nair 1970, 1976; Campbell 1984; Sumpton 1990). Charybdis hellerii prefers soft bot- tom, although it is also found in rocky bot- tom and among live coral, and ranges from the intertidal to 51 m in depth (Stephenson et al. 1957, Galil 1992). Javed & Musta- quim (1994) recently reported specimens of C. hellerii from Pakistan that were carrying acorn barnacles (Chelonibia patula) on the carapace and chelipeds. Timing and method of introduction.— Reports on the presence of C. hellerii in the Caribbean and Florida suggest that the ar- rival of this species in the region occurred in relatively recent times, probably in the late 1980’s (see Fig. 1). Campos & Tiirkay (1989), and Gomez & Martinez-Iglesias (1990), observed that the presence of C. hellerii in Colombia and Cuba was corre- lated with increased ship traffic coming from the eastern Mediterranean where this species migrated via the Suez Canal and is now firmly established (Steinitz 1929, Galil 1992). Campos & Tiirkay (1989:122) indi- cated the possibility that the species arrived in ballast water or “‘clinging to the ships trunk”. Transport in ballast water is the most probable explanation for the arrival of this species to the Caribbean and eastern Florida. It is unlikely that crabs could cling to the hull of any modern ship, although crabs could be transported by ships in fouled sea chests or fouled seawater intake pipes. As suggested by G6mez & Martinez- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Iglesias (1990:71), it is also possible that this species was introduced in one or more sites and subsequently dispersed via larval stages to various other coastal environments in the region. Presence in the Indian River of adults, in- cluding at least one ovigerous female, and juveniles, leaves little doubt that a reproduc- ing population exists in this area and that the species is already established. The scant in- formation available on the biology of C. hel- lerii indicates the potential of this species to successfully colonize marine ecosystems of South Florida and the Gulf of Mexico. Marine decapods introduced in eastern North America.—Charybdis hellerii is the third marine decapod to be introduced and become established in the North American Atlantic coast. The other two being the west- ern Pacific grapsid crab Hemigrapsus san- guineus (see Williams & McDermott 1990, McDermott 1991), and the European portun- id crab Carcinus maenas (see Hedgpeth 1993, Carlton & Geller 1993, Cohen et al. 1995, Carlton & Cohen 1996). Hemigrapsus sanguineus 1s now known from Cape Cod to Chesapeake Bay (J. T. Carlton, G. Ruiz, A. H. Hines, Jr., pers. comms.). Carcinus maenas is the oldest decapod introduction to eastern North America. This species is known to have been present in eastern North America at least since the early 1800’s, al- though it was then known under the name Cancer granulatus Say, 1817 [see Rathbun 1930:15, under Carcinides maenas (Linnae- us)]. By the late 1800’s, C. maenas was weil established from Cape Cod to New Jersey (Smith 1879), and currently is known to be established from Nova Scotia to New Jersey (Cohen et al. 1995). It has also been intro- duced to and is established in Australia, South Africa, Japan, and Pacific North America (Le Roux et al. 1990, Carlton & Cohen 1996). Curiously, the older published information on this species does not mention how it reached eastern North America. Re- cently, Cohen et al. (1995) have indicated that the global dispersal of C. maenas ap- VOLUME 108, NUMBER 4 pears to be linked to the movement of fouled and bored wooden ships. At least five other marine decapods have been considered in the literature as “‘intro- duced”’ in eastern North America but have not become established. Of these, two Indo- West Pacific penaeids (Penaeus japonicus, P. monodon) and two eastern Pacific pen- aeids (P. stylirostris, P. vannamei) are cul- tured extensively in South Carolina, Flori- da, and Texas, and while specimens of the latter three have been found in the wild as escapees from the shrimp farms, no repro- ducing populations have been reported (Wenner & Knott 1992). The Asian Erioch- eir sinensis or ““Chinese mitten crab’’, is believed to have arrived to eastern North America in ship ballast water (Nepszy & Leach 1973). Adults of E. sinensis have been steadily collected in the Great Lakes, most in Lake Erie, since the early 1970’s to 1994 (J. T. Carlton, pers. comm.). However, since E. sinensis is catadromous, it cannot reproduce in the Great Lakes. One transient specimen of Eriocheir sinensis was found in Louisiana in 1987 but has not been found there since (D. L. Felder, pers. comm.). This species apparently has also been transported as a live food item sold in Asian-American markets (Horwath 1988, 1989). Acknowledgments I thank Woody Lee, Smithsonian Marine Station Link Port, Ft. Pierce, Florida, for his invaluable help in trap building, navigation- al guidance through the Indian River, and assistance during trips to the Station. For providing me with information on addition- al specimens of Charybdis hellerii found in the Indian River, I thank D. L. Felder (Uni- versity of Southwestern Louisiana, Lafa- yette), and A. H. Hines, Jr., and R. B. Man- ning (both from the Smithsonian Institution, Washington, D.C.). I am particularly grate- ful to J. T. Carlton (Williams College-Mys- tic Seaport, Connecticut), and an anony- mous reviewer for extensive comments and up-to-date information that considerably 647 improved the manuscript. The following in- dividuals also supplied useful information: Dave K. Camp (Florida Marine Research Institute, St. Petersburg, Florida), A. J. Provenzano, Jr. (Old Dominion University, Norfolk, Virginia), M. Tavares (Universi- dade Sta. Ursula, Rio de Janeiro, Brazil), E. L. Wenner (Marine Resources Research In- stitute, Charleston, South Carolina), A. B. Williams (National Marine Fisheries Ser- vice, Washington, D.C.), and J. D. Williams (National Biological Survey, Gainesville, Florida). This study was supported by the Smithsonian Marine Station at Link Port, Ft. Pierce, Florida, and is contribution no. 387 from that Station. Literature Cited Campbell, G. R. 1984. A comparative study of adult sexual behaviour and larval ecology of three commercially important portunid crabs from the Moreton Bay region of Queensland, Australia. Unpublished Ph.D. Thesis, University of Queensland, 253 pp. Campos, N. H., & M. Tiirkay. 1989. On a record of Charybdis helleri from the Caribbean coast of Colombia.—Senckenbergiana Maritima 20(3/ 4):119-123. Carlton, J. T., & A. N. Cohen. 1996. Episodic global dispersal in shallow water marine organisms: the case history of the European green crab Carcinus maenas.—Journal of Biogeography (in press). , & J. B. Geller. 1993. Ecological roulette: the global transport of nonindigenous marine or- ganisms.—Science 261:78-—82. Chopra, B. 1935. Further notes on Crustacea Decap- oda in the Indian Museum. 8. On the decapod Crustacea collected by the Bengal Pilot Survey off the mouth of the River Hooghly. Brachyg- natha (Oxyrhyncha and Brachyrhyncha).—Rec- ord of the Indian Museum 37:463-514. Cohen, A. N., J. T. Carlton, & M. C. Fountain. 1995. Introduction, dispersal and potential impacts of the green crab Carcinus maenas in San Fran- cisco Bay, California—Marine Biology 122: 225-237. Galil, B. S. 1992. Eritrean decapods in the Levant. Biogeography in motion.—Bulletin de 1’ Institut océanographique, Monaco, no. special 9:115— WB), Gomez, O., & J. C. Martinez-Iglesias. 1990. Reciente hallazgo de la especie indopacifica Charybdis helleri (A. Milne Edwards, 1867) (Crustacea: 648 Decapoda: Portunidae:) en aguas cubanas.—Ca- ribbean Journal of Science 26 (1—2):70-72. Greenwood, J. G., & D. R. Fielder. 1980. The zoeal stages and megalopa of Charybdis callianassa (Herbst) (Decapoda: Portunidae:), reared in the laboratory.—Proceedings of the Royal Society of Queensland 91:61—76. Hedgpeth, J. W. 1993. Foreign invaders.—Science 261:34—35. Hernandez, G., & J. Bolafios. 1995. Additions to the anomuran and brachyuran fauna of northeastern Venezuela. The Crustacean Society Summer Meeting, May 25-27, 1995 [abstract]. Horwath, J. L. 1988. Injurious wildlife: mitten crabs. Proposed rule.—Federal Register 53(219): 45784-45787. 1989. Importation or shipment of injurious wildlife: mitten crabs. Final rule-—Federal Reg- ister 54(98):22285—22289. Javed, M., & J. Mustaquim. 1994. New record of an acorn barnacle, Chelonibia patula (Cirripedia, Thoracica) from Pakistan.—Crustaceana 66: 124-126. Kathirvel, M., & K. N. Gopalakrishnan. 1974. On the occurrence of Charybdis (Charybdis) hellerii (A. Milne Edwards) (Decapoda: Portunidae) along the west coast of India.—Journal of the Marine Biological Association of India 16:286—287. Leene, J. E. 1938. The Decapoda Brachyura of the Siboga Expedition. 7. Brachygnatha: Portuni- dae.—Monograph of the Siboga Expeditie 39C?, Livre 131:1—156. Lemaitre, R. 1995. A nonindigenous portunid crab discovered in the Indian River lagoon system of Florida. The Crustacean Society Summer Meet- ing, May 25—27, 1995 [abstract]. Le Roux, P. J.. G. M. Branch, & M. A. P. Joska. 1990. On the distribution, diet and possible impact of the invasive European shore crab Carcinus maenas (L.) along the south African coast.— South African Journal of Marine Science 9:85— OP, McDermott, J. J. 1991. A breeding population of the western Pacific crab Hemigrapsus sanguineus (Crustacea: Decapoda: Grapsidae) established on the Atlantic coast of North America.—Bio- logical Bulletin 181:195—-198. Milne Edwards, A. 1867. Description de quelques es- peces nouvelles de Crustacés Brachyures.—An- nales de la Societé Entomologique de France, series 4, 7:263—288. Moosa, K. 1981. Crustacés décapodes: Portunidae. Résultats des Campagnes MUSORSTOM I— Philippines (18-28 Mars 1976).—Collection Mémoires ORSTOM 91:141—150. Nepszy, S. J., & J. H. Leach. 1973. First record of the Chinese mitten crab, Eriocheir sinensis, (Crus- tacea: Brachyura) from North America.—Jour- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON nal of the Fisheries Research Board of Canada 30(12)1:1909-1910. Pillai, K. K., & N. B. Nair. 1970. Observations on the reproductive cycles of some crabs from the south-west coast of India.—Journal of the Ma- rine Biology Association of India 10:384—386. ,& . 1976. Observations on the breed- ing biology of some crabs from the southwest coast of India.—Journal of the Marine Biology Association of India 15:754—770. Rathbun, M. J. 1930. The cancroid crabs of America of the families Euryalidae, Portunidae, Atele- cyclidae, Cancridae and Xanthidae.—United States National Museum Bulletin 152:1—593. Say, T. S. 1817. On a new genus of the Crustacea, and the species on which it is established. Journal of the Academy of Natural Sciences of Philadelphia 1(4):49-64. Siddiqui, G., & M. Ahmed. 1992. Fecundities of some marine brachyuran crabs from Karachi (Paki- stan).—Pakistan Journal of Zoology 24:43-45. Smith, S. I. 1879. The stalked-eyed crustaceans of the Atlantic coast of North America north of Cape Cod.—Transactions of the Connecticut Acade- my of Arts and Sciences 5:27—136, pls. 8-12. Steinitz, W. 1929. Die Wanderung indopazifischer Ar- ten ins Mittelmeer seit Beginn der Quartarper- iode.—Internationale Revue der gesamten Hy- drobiologie Hydrographie 22:1—90, figs. 1-9. Stephenson, W. 1972. An annotated check list and key to the Indo-West-Pacific swimming crabs (Crus- tacea: Decapoda: Portunidae).—Royal Society of New Zealand Bulletin 10:1—64. Stephenson, W., J. J. Hudson, & B. Campbell. 1957. The Australian portunids (Crustacea; Portuni- dae). Il. The genus Charybdis.—Australian Journal of Marine and Freshwater Research 8: 491-507. Sumpton, W. 1990. Biology of the rock crab Charyb- dis natator (Herbst) (Brachyura: Portunidae).— Bulletin of Marine Science 46:425—431. Vannini, M. 1976. Researches on the coast of Soma- lia. The shore and the dune of Sar Uanle. 8. Notes on Atelecyclidae and Portunidae (Decap- oda Brachyura).—Monitore Zoologico Italiano, Italian Journal of Zoology, suppl. 8(2):119—127. Wenner, E. L., & D. M. Knott. 1992. Occurrence of Pacific White Shrimp, Penaeus vannamei, in coastal waters of South Carolina. Pp. 173-181 in M. Richard DeVoe, ed., Proceedings of the Conference and Workshop—introductions and transfers of marine species, achieving a balance between economic development and resource protection. Hilton Head, South Carolina, 198 PP. Williams, A. B., & J. J. McDermott. 1990. An eastern United States record for the western Indo-Pa- cific crab, Hemigrapsus sanguineus (Crustacea: Decapoda: Grapsidae).—Proceedings of the Bi- ological Society of Washington 103:108—109. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(4):649-655. 1995 Two new species of freshwater crabs of the genus Hypolobocera from Colombia (Crustacea: Decapoda: Pseudothelphusidae) Martha R. Campos and Gilberto Rodriguez (MRC) Universidad Nacional de Colombia, Instituto de Ciencias Naturales, ~ Apartado Aéreo 53416, 114 Bogota 2, Colombia, S.A.; (GR) Centro de Ecologia, Instituto Venezolano de Investigaciones Cientificas, Apartado 21827, Caracas 1020-A, Venezuela Abstract.—Two new species of the genus Hypolobocera Ortmann, 1897, H. kamsarum and H. emberarum, from the Colombian Andes mountains, are de- scribed. The presence of a mesial lobe on the first gonopod of H. kamsarum, and a large triangular lateral lobe on the first gonopod of H. emberarum, dis- tinguish these species from all others in the genus. The genus Hypolobocera Ortmann, 1897, comprises a group of 28 species of fresh- water crabs that inhabit a vast territory in Colombia, Venezuela, Ecuador and Peru. This is by far the most extended distribution in all the pseudothelphusid genera. The sys- tematics and biogeography of the genus has been reviewed by Rodriguez (1982, 1992), and corrections to the diagnosis of several species have been published recently by Rodriguez (1994). The morphology of the first male gono- pod, a basic character for the diagnosis of the species, displays considerable variabil- ity in this group. To facilitate species iden- tification, Rodriguez (1982) proposed the division of the genus into six groups, based on morphological and biogeographical characters. However, several transitional or aberrant species are still difficult to assign with complete certainty to any group. This is the case with one of the two new species described herein, Hypolobocera kamsarum, obtained during recent explorations of the Andean mountains of Colombia. In the description of these two new spe- cies we have used the terminology estab- lished by Smalley (1964) for the different processes of the male first gonopods. The material is deposited at the Museo de His- toria Natural, Instituto de Ciencias Natura- les, Universidad Nacional de Colombia, Bogota (ICN-MHN). The abbreviations cb and cl are used for carapace breadth and carapace length, respectively. Family Pseudothelphusidae Rathbun, 1893 Tribe Hypolobocerini Pretzmann, 1971 Genus Hypolobocera Ortmann, 1897 Hypolobocera kamsarum, new species icsan lez Material examined.—Vereda Alto Cam- pucana, Municipio Mocoa, Putumayo De- partment, Colombia, 1350 m alt., 2 Jun 1994, leg. O. V. Castafio: 1 male holotype, cl 14.0 mm, cb 23.7 mm (ICN-MHN-CR 1349); 1 male paratype, cl 13.4 mm, cb 21.8 mm, 1 female paratype, cl 14.6 mm, cb 24.8 mm (ICN-MHN-CR 1350). Diagnosis.—First gonopod apex oval, with field of spines, mesial border strongly projected proximally, forming strong trian- gular mesial lobe; caudal ridge straight, strongly concave proximally in lateral view; lateral lobe small, transverse, evenly round- ed in lateral view, ending far from apex of gonopod. Exognath of third maxilliped rel- atively long, approximately 0.6 length of is- chium of third maxilliped. Description of holotype.—Carapace nar- row (cb/cl = 1.7). Cervical groove almost 650 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Cc Fig. 1. Hypolobocera kamsarum, new species, holotype: a, dorsal view of carapace and pereiopods; b, chela of largest cheliped, external view; c, frontal view of carapace. VOLUME 108, NUMBER 4 651 Fig. 2. Hypolobocera kamsarum, new species, holotype, left first gonopod: a, whole gonopod, caudal view; b, whole gonopod, mesial view; c, whole gonopod, cephalic view; d, distal portion of lateral view; e, detail of apex, cephalic view; f, left third maxilliped; g, aperture of left efferent channel, frontal view. 652 straight and deep, ending far from lateral margin. Anterolateral margin with papillae not well defined; posterior half smooth. Postfrontal lobes small, ovally shaped, low. Median groove absent. Surface of carapace between front and postfrontal lobes inclined anteriorly and towards mid-line. Upper bor- der of front bilobed in dorsal view, without tubercles; lower margin strongly sinuous in frontal view; front between upper and lower margin low, vertical. Lower orbital margins each with row of tubercles. Surface of car- apace smooth, covered by small papillae; limit between regions indistinct (Fig. la, c). Palm of larger chela (right) strongly swollen, without proximal tubercle on fin- gers, fingers not gaping when closed (Fig. 1b). Palm of smaller chela moderately swollen, fingers not gaping when closed. Walking legs (pereiopods 2—5) slender, but not unusually elongated; pereiopods 2 and 3 longest of similar in length (length 1.16 times breadth of carapace). Dactylus of per- e1opods 2—5 each with 5 rows of spines di- minishing in size proximally; arrangement of spines on dactylus of left third pereiopod as follows: anterolateral row with 4 spines, anteroventral row with 5 spines plus 1 prox- imal papilla, external row with 4 spines plus 2 intercalated papillae and 1 pair of proximal papillae, posteroventral row with 3 spines, and posterolateral row with 3 spines plus 1 proximal papilla. Exognath of third maxil- liped relatively long, approximately 0.6 length of ischium of third maxilliped; merus of third maxilliped with acute angle on distal half of external margin (Fig. 2f). Efferent branchial channel open (Fig. 2g). First gonopod with caudal ridge straight, strongly concave proximally (lateral view); lateral lobe small, transverse, evenly round- ed (lateral view), ending far from apex of gonopod (Fig. 2a, b, d); apex oval, with me- sial border strongly projected proximally, forming strong triangular mesial lobe (Fig. 2e). Color.—Specimens preserved in alcohol: carapace overall chestnut mottled with or- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ange. Walking legs lighter than carapace with scattered orange specks. Etymology.—The species is named for the Kamsa Indians, in whose territory the species was found. Remarks.—Although the hypoloboceran traits are quite distorted in this species by the development of a triangular apical me- sial lobe of the first gonopod, the lateral lobe is still characteristic of the genus. Based on Rodriguez (1982), Hypolobocera kamsarum can be considered a transitional species as mentioned in the introduction. Hypolobocera emberarum, new species Figs. 3, 4 Material examined.—Vereda El Veinte, Municipio El Carmen, near the Atrato Riv- er, Choc6é Department, Colombia, 2800 m alt., 30 May 1994, leg. I. D. Vélez: 1 male holotype, cl 14.2 mm, cb 23.8 mm (ICN- MHN CR-1358); 2 male paratypes, cl 13.1 and 13.0 mm, cb 22.0 and 21.8 mm, 3 fe- male paratypes cl 13.4, 13.2 and 12.7 mm, cb 22.6, 22.3 and 21.4 mm (ICN-MHN CR- 1359). Valle de Pérdidas, Municipio Urrao, Antioquia Department, Colombia, 1800 m alt., 3 Sep 1994, leg P. Duque: 3 males, cl 14.1, 11.4 and 10.9 mm, cb 23.8, 19.1 and 18.3 mm, 3 females, cl 14.0, 12.8 and 12.7 mm, cb 23.5, 21.5 and 21.4 mm (ICN- MHN CR-1383). Diagnosis.—First gonopod with caudal ridge long, straight; lateral lobe prominent, subtriangular, hatchet-shaped, wider proxi- mally, spinulated, extending near apex of gonopod and forming deep notch distally, with longitudinal ridge at base; apex with oval field of spines, with cephalic side rounded, projected cephalad. Exognath of third maxilliped strongly reduced, approxi- mately 0.20 length of ischium of endog- nath. Lateral sides of carapace conspicu- ously hairy in largest specimens. Description of holotype.—Carapace nar- row (cb/cl = 1.7). Cervical groove deep, wide, slightly sinuous, ending far from lat- eral margin. Anterolateral margin smooth. VOLUME 108, NUMBER 4 653 1 cm 1 cm sah ek , Mbyte uy Cc Fig. 3. Hypolobocera emberarum, new species, holotype: a, dorsal view of carapace and pereiopods; b, chela of largest cheliped, external view; c, frontal view of carapace. 654 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON : ICPRurvey d Fig. 4. Hypolobocera emberarum, new species, holotype, left first gonopod: a, whole gonopod, caudal view; b, whole gonopod, lateral view; c, whole gonopod, cephalic view; d, detail of apex, caudal view; e, detail of apex, superior view; f, left third maxilliped; g, aperture of left efferent channel, frontal view. VOLUME 108, NUMBER 4 Postfrontal lobes ovally shaped, low. Median groove deep, narrow. Surface of carapace between front and postfrontal lobes inclined anteriorly. Upper border of front bilobed in dorsal view; lower margin sinuous in frontal view, with small tubercles. Surface of cara- pace smooth, covered by small papillae. Lat- eral sides of carapace conspicuously hairy in largest specimens (Fig. 3a, c). Palm of larger chela (left) elongated, swollen, without proximal tubercle on fin- gers; fingers slender, slightly gaping when closed (Fig. 3b). Palm of smaller chela mod- erately swollen, fingers not gaping when closed, with tips crossing. Walking legs (per- e1opods 2—5) slender, dactylus of pereiopods each with 5 rows of spines diminishing in size proximally; arrangement of spines on dactylus of right third pereiopod as follows: anterolateral row with 6 spines plus 3 prox- imal papillae, anteroventral row with 5 spines plus 3 proximal papillae, external row with 5 spines plus 2 papillae intercalated and 1 pair of proximal papillae, posteroventral and posterolateral rows with 4 spines. Ex- ognath of third maxilliped 0.2 as long as is- chium of third maxilliped; merus of third maxilliped with acute angle on distal half of external margin (Fig. 4f). Efferent branchial channel open (Fig. 4g). First gonopod strongly bent in middle, caudal ridge long, straight, reaching to apex of gonopod; lateral lobe prominent, sub- triangular, hatchet-shaped, wider proximal- ly, extending near apex of gonopod and forming deep notch distally, caudal face covered with spinules (Fig. 4a, c, d); apex oval, with cephalic side rounded, projected cephalad (Fig. 4e). Remarks.—This species resembles Hy- polobocera chocoensis Rodriguez, 1980, in the shape of the lateral lobe of the first gon- opod. However, the lateral lobe of H. cho- coensis does not extend near the apex, and lacks a distal notch. Based on Rodriguez (1982), and on the general morphology and biogeography of H. emberarum, this spe- cies can be considered to belong to group IV. 655 Color.—Specimens preserved in alcohol: carapace and chelipeds overall dark brown- olive. Walking legs slightly lighter than car- apace. Etymology.—The species is named for the Embera Indians, in whose territory the Species was found. Acknowledgments We thank Ivan Dario Vélez, Patricia Du- que and Olga Victoria Castafio, for collect- ing the specimens. We are also very grate- ful to Dr. Rafael Lemaitre and the referees for their valuable comments. The illustra- tions were prepared by Juan Carlos Pinzon. Literature Cited Ortmann, A. 1897. Carcinologische Studien.—Zool- ogische. Jarbiicher, Abtheilung fiir Systematik, Geography and Biology der Tiere 10:258—372, plaalive Pretzmann, G. 1971. Fortschritte in der Klassifizi- erung der Pseudothelphusidae.—Anzeiger der Osterreichischen Akademie der Wissenschaften Mathematische Naturwissenschaftliche Klasse 179(1/4):14—24. Rathbun, M. 1893. Descriptions of new species of American freshwater crabs.—Proceedings of the United States National Museum 16(959): 649-661, pl. 73-77. Rodriguez, G. 1980. Description préliminaire de quel- ques espéces et genres nouveaux de crabes d’eau douce de |’Amérique tropicale (Crusta- cea, Decapoda, Pseudothelphusidae).—Bulletin du Muséum Nationale d’ Histoire Naturelle, Par- is (4) 2, section A (3):889-894. . 1982. Les crabes d’eau douce d’ Amérique. Famille des Pseudothelphusidae.—Faune Trop- icale 22:1—223. . 1992. The freswater crabs of America. Fam- ily Trichodactylidae and supplement to the fam- ily Pseudothelphusidae.—Faune Tropicale 31: 1-189. . 1994. A revision of the type material of some species of Hypolobocera and Ptycophallus (Crustacea, Decapoda, Pseudothelphusidae) in the National Museum of Natural History, Wash- ington, D.C., with descriptions of a new species and a new subspecies.—Proceedings of the Bi- ological Society of Washington 107:296—307. Smalley, A. 1964. A terminology for the gonopods of the American river crabs.—Systematic Zoology 13:28-31. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(4):656—661. 1995 Redescription of a unique feather star (Echinodermata: Crinoidea: Comatulida: Comasteridae) with the diagnosis of a new genus Charles G. Messing Nova Southeastern University Oceanographic Center, 8000 North Ocean Drive, Dania, Florida 33004, U.S.A. Abstract.—A new genus, Aphanocomaster, is erected for Comaster pulcher A. H. Clark, which is known only from the holotype. The genus is similar to Comaster in having an apparent cryptosynarthry in the primibrach series, the first brachial syzygy chiefly between the first and second brachials on arms arising from the tertibrach series, and a central mouth. It differs in lacking a transverse initial comb tooth. The new genus differs from all other comasterid genera in having secundibrach series of four ossicles series in which both the first and second ossicles and the third and fourth are united by synarthrial articulations. In the course of revising the Indo-West- ern Pacific genus Comaster L. Agassiz, a re-examination of Comaster pulcher A. H. Clark revealed that the single known spec- imen lacks a critical feature of the genus as currently diagnosed and exhibits another, previously unnoticed characteristic unique to the family. The combination requires re- moval of the species from Comaster and establishment of a new generic name for it. In addition, published descriptions are uni- formly incomplete (A. H. Clark 1912, 1918, 1931). Terms, abbreviations, measurements and symbols are as follows: Centrodorsal: cen- tral aboral plate. Cirri: aboral, segmented hooks attached to centrodorsal; Roman and Arabic numerals indicate numbers of cirri/ individual and segments (cirrals)/cirrus, re- spectively (a range of values is usually giv- en); LW of cirral: length to median width ratio when viewed laterally. Ray: one of five branched series of ossicles radiating from center of specimen. Radial: (n.) first ossicle of a ray or (adj.) a structure asso- ciated or oriented with a ray. Axil: ossicle at which a ray branches. Brachitaxis: series of ossicles following radial or axil and in- cluding the next axil; I-IVBr: first through fourth brachitaxes (written out primi-, se- cundi-, terti-, and tetrabrachitaxis, or tetra- brach series); Arabic numeral immediately following indicates number of ossicles in that brachitaxis (e.g., I[Br2). Arm: unbran- ched series of ossicles following distalmost axil; brachial (br; plural, brr): arm ossicle; subscript number indicates specific ray os- sicle (brachitaxis or arm) counting from first ossicle after preceding axil or radial (e.g., IIBr,, br,); WL of ray ossicle: median width to midaboral length ratio when viewed aborally. Synarthry (—) (See *“Note”’ below): articulation typically be- tween first two ossicles of a brachitaxis or arm consisting of two ligament bundles separated by an aboral-oral fulcral ridge, sometimes with midaboral swelling. Cryp- tosynarthry: synarthry modified as a tight junction, visible externally as a fine line; the articular faces either smooth or with round or elongate, often concentric raised areas (Hoggett & Rowe 1986, fig. 2c). Syz- ygy (+): articulation between two succes- Sive ray ossicles consisting of radiating VOLUME 108, NUMBER 4 ridges and grooves and appearing externally as a perforated line (e.g., br3,,); intersyzy- gial interval: number of articulations be- tween successive syzygies. Pinnules (P): unbranched segmented appendages arising from alternate sides of successive brachials, and from exterior side of brachitaxes of more than two ossicles; on brachitaxes, sub- script Roman numerals indicate brachitaxis from which pinnule arises (e.g., P;, on IIBr); on arms, subscripts count pinnules from the most proximal; Arabic numbers and letters refer to pinnules along exterior and interior side of an arm, respectively (that is, the sides away from and toward the extrapolat- ed axis of the preceding axil) (e.g., P,, P.); LW of pinnule ossicles (pinnulars): length to median width ratio. Comb: modification of distal pinnulars of proximal (oral) pin- nules producing comblike profile. Disk: central visceral mass or, specifically, its oral surface; anal interambulacral area: area on disk surrounded by food grooves and bear- ing anal papilla. For further discussions and examples of comatulid morphometrics, meristics, abbreviations and symbology, see A. M. Clark & Rowe (1971), Breimer (1978), Hoggett & Rowe (1986) and Mess- ing & Dearborn (1990). In all illustrations, Sparse uniform stippling indicates articula- tions between successive ossicles. Note: The following descriptions include an exception to standard comatulid sym- bology. Synarthries are designated by one- em dashes (—), following the usage of Gis- lén (1934), so that, for example, I[Br4(1— 2, 3—4) indicates a secundibrach series of four ossicles in which both the first and sec- ond and the third and fourth ossicles are united by synarthry. In comatulid crinoids, the first two ossicles following a radial or axil are joined either by synarthry (some- times modified as a cryptosynarthry) or syzygy. Typically, only the latter is shown (as a + sign) in written symbology. When a syzygy is not present, a synarthry is as- sumed in this position (e.g., in HIBr2, the two ossicles are joined by synarthry). Be- cause the species discussed herein exhibits 657 an unusual distribution of synarthries, they must be clearly indicated. However, even if additional specimens eventually demon- Strate that the distribution of synarthries de- scribed below is an abnormality, the unique combination of other features exhibited by this specimen requires diagnosis of a new genus, as follows. Aphanocomaster, new genus Diagnosis.—A genus of Comasteridae with IBr2 ossicles apparently united by cryptosynarthry; I[Br2(1—2) and 4(1—2, 3—4); MIBr2(1_—2), 4(1—2, 3—4) and 4(1—2, 3 + 4) [possibly also 2(1 + 2)]; in brachitaxes of four ossicles, the second os- sicle (Br, and HIBr,) laterally enlarged ex- teriorly to accommodate robust base of Py, and Py, respectively; first brachial syzygies chiefly br,,53,, on arms arising from IIIBr; br,,, alone on exterior arms arising from IIBr; comb teeth tall, triangular or spade- shaped, confluent with exterior lateral mar- gin of pinnular; initial tooth sometimes slightly twisted, but not oriented transverse- ly; mouth central; anal papilla midway be- tween mouth and disk margin. Type species.—Comaster pulcher A. H. Clark (1912), by monotypy. Etymology.—From aphanes (adaved), unseen, invisible, secret, obscure (Brown 1978), and the genus name Comaster. Gen- der is masculine. Distribution.—Known only from a single station in the Kei Islands, Indonesia, O—52 m (A. H. Clark 1931). Remarks.—A. H. Clark (1921) and Gis- lén (1934) considered the few records in co- matulids of four-ossicle brachitaxes con- sisting of two synarthrial pairs as abnormal. They recorded this pattern in only three species: Cyllometra manca (Carpenter) (Colobometridae) (originally as C. anomala A. H. Clark), Heterometra quinduplicava (Carpenter) (Himerometridae) and Adelo- metra angustiradia (Carpenter) (Antedoni- dae). A. H. Clark (in A. H. Clark & A. M. Clark 1967) suggested that the single 658 known specimen of A. angustiradia might be a juvenile Himerometra or related genus, but placed it in the Antedonidae on the ba- sis of its cirri and pinnules. Messing (1975) noted [[Br4 without II[Br,,, in several spec- imens of Crinometra brevipinna (Pourtaleés) (Charitometridae), but did not otherwise identify the articulation between the third and fourth ossicles. If, in specimens collected in the future, brachitaxes with two pairs of synarthries prove to be abnormally doubled forms of the much more widespread (among comat- ulids) brachitaxes of two ossicles united by synarthry, a new genus still must be erected for the type specimen of Comaster pulcher. No genus currently exists in the Comaster- idae with the combination of features ex- hibited by this specimen: first brachial syz- ygy chiefly br, ,,3,,4, both I1[Br2(1—2), and IBr4(1—2, 3 + 4) present; no initial transverse comb tooth; mouth central. Only if the IBr4(1—2, 3 + 4) are also highly unusual and all brachitaxes are normally of two ossicles in future specimens would pulcher fall within a currently recognized genus: Comissia (Hoggett & Rowe 1986). Aphanocomaster pulcher (A. H. Clark, 1912) Fig. | Comaster pulcher A. H. Clark, 1912:22.— 1918:37, 40, 275, pl. 14, figs. 14, 15.— 1931:443-444, pl. 50.—Gislén, 1919:14. Holotype.—University of Amsterdam, Zoological Museum, U.Cri.-2114. Siboga Expedition Station 257, Du-roa Strait, Kei Islands, Indonesia, 0-52 m, 11 Dec 1899. 1 specimen. Description.—Centrodorsal a_ thick, rounded pentagonal disk, 4.1 mm across (Fig. la). Aboral pole slightly depressed, with cirrus sockets encroaching on margin. Cirri XXXIV, 15-17, up to 11.2 mm long, in crowded double marginal row; first seg- ment short, second with LW = 1.5-2.0; third segment longest, with LW 2.8; fourth and following segments decreasing in PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON length, becoming compressed and slightly expanded distally; eighth segment with LW = 1.1—1.3; one or two segments preceding penultimate with LW = 1.0-1.1 (Fig. 1b, c). Fourth and following segments with low, aboral, subdistal, transverse swelling or ridge, sometimes finely spinulose, nar- rower on more distal segments, becoming low blunt and triangular on the four to five segments preceding penultimate (Fig. 1d, e). Fourth and following segments shiny. Opposing spine low and wide, sharp or blunt. Arms 35, all in single plane, none com- plete. Radials very short, just visible. IBr2 ossicles joined by tight articulation, proba- bly cryptosynarthry (see remarks). IBr, very short, WL = 4.5—5.0, completely sep- arated or just touching at proximal corners, and slightly narrower distally so that adja- cent IBr, are separated by V-shaped gap (Fig. la). One I[Br2(.1—2) bearing I[Br4- (1—2, 3—4) exteriorly and I[I[Br4(i—2, 3+4) interiorly; nine I[Br4(1—2, 3—4) bear- ing nine IIJBr2(1—2) [one of these with a close articulation; possibly II[Br2(1 + 2)] and three I[Br4(1—2, 3 + 4) [two of the latter arising from one IIBr4]. IIBr developed in pairs or interiorly on IIBr. One interior IVBr2(1 + 2)?; the articulation uniting its os- sicles a tight line similar to syzygy on prox- imal brr, but perforations not clearly visible. First ossicles of adjacent brachitaxes beyond the first (I-IVBr,) joined interiorly for most or all of their length; adjacent IIBr and fol- lowing brachitaxes separated interiorly by narrow U-shaped gaps; gaps becoming nar- rower between more distal brachitaxes, but remaining distinct between adjacent arm bases. Width at IIBr, axil 1.6 mm (1.3 at IBr axil). Br, slightly longer exteriorly; adjacent br, joined interiorly; in arm bases with br, unit- ed by synarthry, br, cuneate and projecting exteriorly to accommodate base of large P). Proximal brachials otherwise oblong as far as bre,, WL = 1.7; triangular or almost so by about br.,9, WL = 1.9. Middle brachials tri- angular, WL = 1.3—1.4. Distal brachials elon- VOLUME 108, NUMBER 4 659 Fig. 1. a-i. Aphanocomaster pulcher (A. H. Clark), holotype. a. Aboral view showing centrodorsal, bases of cirri, bases of three rays, and basal few ossicles of proximal pinnules. b. Cirrus with weak aboral ornamen- tation. c. Cirrus with strong aboral ornamentation. d. Oblique view of fourth (left) and fifth cirrals showing finely spinulose transverse subdistal ridge. e. Oblique view of tenth (left) to thirteenth cirrals showing transverse aboral swelling narrowing to triangular process on more distal segments. f. P,, reconstructed from fragments; most of comb lost. g. P,. h. Lateral view of proximal part of comb from a detached oral pinnule showing slightly twisted proximal tooth (arrow). i. Same, oblique view. j. Comaster distinctus (Carpenter), lateral view of proximal part of comb from a detached oral pinnule showing well-developed transversely-oriented proximal tooth (arrow). Scale bar: a, 2 mm; b, c, f, g, 1 mm; d, e, h-j, 0.5 mm. 660 gated, shaped like bent hourglasses, WL = 0.5 (LW = 1.9). Aboral surface of rays smooth, somewhat flattened. On II and I]{Br4(i—2, 3—4), Il and IIIBr, cuneate with exterior lateral mar- gin projecting for articulation of very ro- bust P,, and P,,; P;, and Py, so large that the ossicle bearing them almost appears as an asymmetric axil (Fig. la). Synarthries on brachitaxes sometimes weakly swollen midaborally. Distal margins of mid-arm brachials finely dentate but not flared or thickened. Distal brachials smooth. Epi- dermis fairly thick, obscuring several artic- ulations. First brachial syzygies chiefly br,,.3,, on arms arising from IIIBr; br,,, alone on ex- terior arms arising directly from IIBr4 (one with br3,,,,5, another with br,,, instead) (Fig. la). On one ray, the exterior arms aris- ing from a pair of adjacent IIIBr2 both lack br,,, (one has br,,,, the other is broken at br,); interiorly, one of these IIIBr2 bears an arm with br, ,,53,,4; the other bears a [VBr2(1 + 2) which, in turn, bears an interior arm with br,,53,, and an exterior arm with br;,, alone. Next syzygy brj,,,,. Distal intersy- zygial interval chiefly 4 (occasionally 3 or 5))). P,, of 35+ segments, comb lost; estimat- ed length ~14 mm; very robust at base, ta- pering to slender and flagellate distally; middle segments longer than broad (to LW = 1.5); no spines visible (Fig. If). One de- tached oral comb of 15 teeth possibly be- longs to P, or Py. P, of 37 segments with 11 teeth, L ~ 10.0 mm; similar to P,, but slightly less robust; apparently shorter with fewer segments. P, of 31-33 segments, 12 teeth, L = 8.2 mm; much less robust than P, (Fig. 1g). P. of at least 26 segments with 6 teeth (tip possibly lost), remaining length = 7.6 mm. P, of at least 22 segments with 7 teeth (tip possibly lost), remaining length = 5.7 mm. Comb teeth tall, triangular or spade-shaped and at least slightly incurved, confluent with exterior lateral margin of pinnular. Initial tooth sometimes slightly twisted (Fig. lh, i); no initial transverse PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON tooth. [Compare with initial transverse tooth of Comaster distinctus (Carpenter), Fig. 1j.] Comb either rising abruptly with tall first tooth, or developing gradually over 1-3 increasingly tall teeth. Distal teeth smaller & shorter but still well developed. Combs coiling planospirally; present to at least Px. P, of 18 segments, no teeth, L = 5.4 mm; most segments (except at base and tip) lon- ger than wide, cylindrical and constricted in middle, LW to 2.0; distal segments with cluster of lateral spines; hooks on last 3 segments. P, and P, first genital pinnules; spherical gonads developed on fourth—sixth segments. Middle pinnules of 17—18 seg- ments, 6.0 mm long; first two segments short, third and fourth squarish; these first four segments stout; following segments narrower so pinnule tapers to slender tip; most segments little longer than broad with bundle of strong spines along lateral mar- gin. Distal pinnules of 17 segments, 7.2 mm long; first two segments short; following segments elongate (except near tip), and of about equal length but becoming more slen- der distally; LW increasing from about 3.3 proximally to 5.2 distally (not including basal or distalmost segments). Disk completely naked; oral integument somewhat transparent with tiny sclerites visible in tissue. Mouth central; anal papilla large, about midway between mouth and disk margin; anal interambulacral area only Slightly larger than other interambulacral areas. Color.—White in alcohol. Distribution.—Kei Islands, Indonesia. Known only from the type specimen. Remarks.—tThe identities of the articu- lations in several brachitaxes are uncertain. The two ossicles of each IBr2 are joined by a tight articulation that may be either a syn- arthry or cryptosynarthry. No trace of the external perforations exists that would iden- tify these articulations as syzygies and place the specimen in the genus Comatula. The articulation has not been dissociated for de- tailed examination because the specimen is VOLUME 108, NUMBER 4 unique. The articulations between the two ossicles of the single [IVBr2, and the IIIBr2 that precedes it, are tight, similar to syzy- gies on the proximal brachials, but with perforations not clearly visible. The specimen is in somewhat poorer condition than when last photographed (A. H. Clark 1931). However, the current re- description is substantially more detailed and differs in several important respects from those previously published (A. H. Clark 1912, 1918, 1931). I found XXXIV rather than XX VII cirrus sockets with the longest cirrus segment having LW = 2.8 rather than 3 to 4. The previous descriptions refer to a single IIBr2(1—2), nine IIBr4(1—2)(3 + 4), and all WIBr2(1 + 2). This re-examination clearly shows that syz- ygies are not present in the IIBr, and that the IIIBr includes 4(1—2, 3—4), 4(1—2, 3 + 4), 2(1—2) and, possibly, 201 + 2). In addition to the diagnostic characteris- tics of brachitaxes and pinnule combs, sev- eral other features also distinguish A. pulch- er from the several nominal cirrus-bearing species of Comaster of similar size: C. fru- ticosus A. H. Clark, C. brevicirrus (Bell), C. schoenovi A. H. Clark, C. sibogae, A. H. Clark, C. distinctus (Carpenter) and C. serratus (A. H. Clark). In these Comaster spp., the longest cirral is usually the fourth, not the third; cirrals bear well developed aboral transverse ridges with a stronger op- posing spine; middle brachials are shorter and usually cuneate with everted spinose distal margins; the first syzygy following br... (or br;,,, when present) occurs at brio,; to br;>,,3; the distal intersyzygial in- terval is usually 3, not 4, and the disks usu- ally bear at least some slender conical nod- ules. Acknowledgments This research was carried out as part of a grant awarded to the author by the Mu- seum national d’ Histoire naturelle, Paris for July 1993. I thank Alain Crosnier for mak- ing the grant available to me, and for his 661 hospitality, and J. Bleeker of the Instituut voor Taxonomische Zoologie, University of Amsterdam, for making the type specimen of Comaster pulcher available to me. I also thank Nadia Cominardi (curator of echino- derms, MNHN) for bringing the availability of funding to my attention, and for provid- ing research space, great assistance and ex- traordinary generosity during my stay in Paris. This is Contribution no. 8 from the Deep Ocean Society, Inc., Hollywood, Flor- ida. Literature Cited Brown, R. W. 1978. Composition of Scientific Words. Smithsonian Institution Press, Washington, D.C. 882 pp. Clark, A. H. 1912. Seventeen new East Indian cri- noids belonging to the families Comasteridae and Zygometridae.—Proceedings of the Biolog- ical Society of Washington 25:17—28. . 1918. Unstalked crinoids of the Siboga Ex- pedition.—Siboga Expedition 42b:1—300, 28 pls. . 1921. A monograph of the existing crinoids. 1(2).—Bulletin of the United States National Museum 82. xxvi + 795 pp., 57 pls. . 1931. A monograph of the existing crinoids. 1(3) Superfamily Comasterida.—Bulletin of the United States National Museum 82. viii + 816 pp., 82 pls. , & A. M. Clark. 1967. A monograph of the existing crinoids 1(5).—Bulletin of the United States National Museum 82:1—860. Gislén, T. 1919. Results of Dr. Mjoberg’s Swedish Scientific Expedition to Australia, 1910-1913. XXIII. Crinoids.—Kungl. svenska Vetenskap- sakademiens Handlingar 59(4):3—37, 1 pl. . 1934. A reconstruction problem; analysis of fossil comatulids from N. America with a sur- vey of all known types of comatulid arm-ram- ifications.—Kungl. Fysiografiska Sallskapets Handlingar N.E 45(11):1—59. Hoggett, A. K., & E W. E. Rowe. 1986. A reappraisal of the family Comasteridae A. H. Clark, 1908 (Echinodermata: Crinoidea), with the descrip- tion of a new subfamily and a new genus.— Zoological Journal of the Linnean Society 88: 103-142. Messing, C. G. of the Crinoidea Comatulida (exclusive of the Macrophreatina) collected by the R/V Gerda in the Straits of Florida and adjacent waters. Un- published MS Thesis, University of Miami, Coral Gables, Florida. x + 284 pp. 1975. The systematics and distribution PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(4):662-—678. 1995 Gymnotus maculosus, a new species of electric fish (Chordata: Teleostei: Gymnotoidei) from Middle America, with a key to species of Gymnotus James Spurling Albert and Robert Rush Miller Museum of Zoology, University of Michigan, Ann Arbor, Michigan 48109-1079, U.S.A. Abstract.—A new species of gymnotoid knifefish, Gymnotus maculosus, is described from Pacific drainages of Middle America, from Chiapas, Mexico, to Costa Rica, and an Atlantic drainage in Nicaragua. Morphological, meristic, and pigmentation features are compared with congeners. Variation among the species of Gymnotus from many localities in South America remains to be analyzed for a robust hypothesis about numbers of species and their interre- lationships. A phylogenetic diagnosis of Gymnotus, and a key to the recognized species is provided. Resumen.—Se describe una nueva especie de anguila electrica, Gymnotus maculosus. Esta especie se encuentra en las vertientes Pacificas de Mesoam- érica desde Chiapas, México, hasta Costa Rica, y en una cuenca Atlantica de Nicaragua. Se discuten y comparan caracteristicas morfol6gicas, meristicas, y de pigmentacié6n con otros miembros del género. Para poder establecer el nu- mero de especies de Gymnotus y sus relaciones, es necesario un andalisis de la variabilidad entre especies de diversas localidades en Sudamérica. Se provee una diagnosis filogenética del género Gymnotus y una clave de identificaciOn de las especies reconocidas. Gymnotus 1s the most widely distributed genus of American electric knifefishes (Te- leoste1: Gymnotoidei). Members of the ge- nus range from the Rio San Nicolas of southeastern Chiapas, Mexico, south to the Rio de la Plata, Argentina (Eigenmann & Ward 1905, Eigenmann & Fischer 1914, Ei- genmann & Allen 1942, Miller 1966, Buss- ing 1985). Members of this lineage are ag- gressive, nocturnal predators who prefer small tributaries, lagoons, and backwater swamps, and in Middle America sometimes inhabit highland lakes. The systematic position and organization of Gymnotus are problematic. Gymnotus was proposed by Peter Artedi in 1738 to include the type species G. carapo Linnae- us. According to Ellis (1913) the original specimens were collected in the Guyanas. Ellis (1913) was the first to propose a for- mal hypothesis of interrelationships among the major gymnotoid lineages, placing Gymnotus and Electrophorus as the sister lineage to all other gymnotoids (Ellis 1913, Eigenmann & Fischer 1914). Ellis synony- mized all specimens of Gymnotus into the type species, G. carapo. There are presently thirteen described species of Gymnotus, of which five have been synonymized with the type species (Mago-Leccia 1994). Several additional species are recognized in museum collec- tions, however, so that the actual number of species may be substantially higher. De- scriptions of several of these species await the acquisition of sufficient samples from populations across their geographical range (Mago-Leccia 1994). Most specimens of Gymnotus in museum collections are iden- tified as G. carapo, regardless of their ap- pearance or origin. This is due, in part, to the wide geographical extent of this puta- VOLUME 108, NUMBER 4 663 Table 1.—Number of precaudal vertebrae in Gymnotus species. Counts from a total of 75 specimens are reported. 30 32 34 36 38 40 42 44 46 48 50 n Mode anguillaris 0) 0) O O 5 0) 0) O 0) 0) 0 5 38 carapo O 0) 0 0) 6 O 0) 0) 0) O O 6 38 catanaiapo 0) O 0) 0) 0) 0) 0) O 0) 0) 2 2 50 coatesi 0) 0) O 1) 0) 3 1 O 0) 0) 0) 4 40 cylindricus Z 15 6 0) 0) 0) 0) 0) 0) 0) 0 728 32 maculosus 1 11 9 0) 0) 0) 0) 0) 0 0) 0) Di 32 pedanopterus O 0) O 7 0 0) O 0) 0) 0) 0) dl 36 stenoleucus 0) 0) 0) 0 0) 0) 0) 0) 4 3 0) of 46 Total 3} 26 15 7 7 3 1 0) 4 3 2 YS tive species, which is known from Guyana to the Rio La Plata basin and many inter- vening sites. Furthermore, color patterns, and body shapes of specimens referred to G. carapo vary widely across the range. The species diversity of Gymnotus is greatest in the Amazon basin, from where SIX Species are recognized. Five of these species also inhabit the Upper Rio Orinoco basin (Mago-Leccia 1994). Gymnotus an- guillaris Hoedeman, originally described from Guyana, has also been collected from the Orinoco, Amazon, and La Plata basins. The other species have more restricted dis- tributions. Gymnotus coatesi LaMonte is known only from the western Amazonian lowlands of Brazil and Pert, G. inaequila- biatus (Valenciennes) is from the Rio de La Plata drainage, and G. pantherinus (Stein- dachner) from Atlantic drainages of south- ern Brazil. Gymnotus also ranges into Middle Amer- ica. Understanding the biogeographic his- tory of Gymnotus in northwestern South America and Middle America is confound- ed by the complex geological history of the region (Bussing 1987, Pittman et al. 1993). Although the species diversity of Gymnotus has not been recently studied west of the Andes, there appear to be several unrecog- nized species inhabiting small coastal streams and rivers along the Pacific coast from Ecuador, north to Chiapas, Mexico. The purpose of this paper is to describe one of these species, discovered one of us (RRM). Methods.—Methods used herein are modified or elaborated from Lundberg & Mago-Leccia (1986), Fink (1989), and Fink & Machado-Allison (1992). Specimens ex- amined are listed in the Appendix. Museum acronyms are as presented by Leviton et al. (1985). Osteological data were taken from cleared and stained specimens using a mod- ified version of the enzyme technique of Taylor & Van Dyke (1985). Standardized dissection methods of small teleosts were used (Weitzman 1974). We follow Fink & Fink (1981) for bone nomenclature. All drawings were made with the aid of a Zeiss dissecting microscope, equipped with a camera lucida, and edited by means of a PC graphics editor. Counts of precaudal vertebrae (Table 1) and anal-fin rays (Table 2) were taken from radiographs. The number of precaudal ver- tebrae includes those of the Weberian ap- paratus; number of precaudal vertebrae is used here as a proxy for body-cavity length (Albert & Fink, in press). Body proportions reported include head length (HL), from posterior margin of bony operculum to tip of snout; postorbital head length (PO), from posterior margin of bony opercle to poste- rior margin of eye; preorbital head length (PR), from anterior margin of eye to tip of snout; body depth (BD), vertical distance from origin of anal fin to dorsal body bor- 664 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 2.—Number of anal-fin rays in Gymnotus species. Counts from a total of 58 specimens are reported. 155 165 175 185 195 205 215 225 235 245 255 n anguillaris 0) 0) 0) 0) 0) 1 0) 2 1 1 0) 5) carapo 0) 0) 0) 0) 2 1 2 1 0) 0) 0) 6 catanaiapo 0) 0) 0) 0 0) 0) 0) 2 3 0) 1 7 coatesi 0) 0) 0) 0) 0) DD 1 1 0 0) 0 4 cylindricus 0) 1 1 5 0) 3 3 1 0) 0) 0) 14 maculosus 2 4 D 0) O 0) 0) O 0) 0) 0) 8 pedanopterus 0) 0 0) 0) 0) 0) 1 1 1 1 0) 7 stenoleucus 0) 0) 0) 0) DD 8 1 1 O 0) 0) 7 Total 2 5 3 5) 4 10 8 9 5 2) 1 58 der, with lateral line held horizontal; pec- toral-fin length (P1), from dorsal border of fin base where it contacts cleithrum to tip of longest ray; interorbital distance (IO), between dorsomedial margins of eyes; size of branchial opening (BO), from postero- dorsal to anteroventral extent of fold along anterior margin; pre-anal distance (PA), from anterior insertion of anal fin and pos- terior margin of anus. Body size is repre- sented by head length (HL) in millimeters. The use of head length as a rough measure of overall body length is discussed by AIl- bert & Fink (in press). Specimens in which the caudal filament was obviously damaged and unrepaired were excluded from mea- surements of total length. Some measurements used in this report differ from those of earlier works on gym- notoids (Mago-Leccia 1978, 1994). Head length, for example, is defined as the dis- tance between tip of snout and posterior ex- tent of the bony opercle. Head length mea- sured from the posterodorsal termination of the branchial opening (Mago-Leccia 1978) is subject to greater preservation-induced artifact and was not used. Measures of pec- toral-fin length, postorbital length, and body depth reported here may also differ from results of previous studies. For these measurements, we attempted to use the un- ambiguous landmarks noted above to define endpoints. This report is part of a larger research effort by the first author on systematics and evolution of the Gymnotoidei. A total of more than 2400 specimens representing at least 80 gymnotoid species have been ex- amined (Albert 1995). Character polarities were determined from the hypothesis of re- lationships proposed by Albert (1995; see Discussion). Monophyly and Nomenclature of Gymnotus Gymnotus Linnaeus 1758 Gymnotus Linnaeus 1758:246 (type spe- cies, by original monotypy, G. carapo Linnaeus 1758). Carapus Cuvier 1816 (nec Rafinesque): 2 Giton Kaup in Duméril 1856:201 (type spe- cies, by original monotypy, G. fasciatus Pallas 1767:35). Diagnosis.—Species of Gymnotus have multiple (13-50) pale oblique bands of re- duced melanophore density along lateral surface of body (obscured in some spe- cies); anterior narial opening located with- in gape such that pore opening faces an- teroventrally (also present in many rham- phichthyoids); anterior pharyngobranchial (associated with gill arch 3) unossified; all five basibranchials unossified; capacity to regenerate postcoelomic neural spines (Gayet & Meunier 1991); cylindrical or barrel shaped electrocytes, without a stalk, their long axis oriented longitudinally (Bass 1986, fig. 1). Description.—The polarity of several VOLUME 108, NUMBER 4 characters useful in identifying specimens of the genus Gymnotus are ambiguous. Mesethmoid short and broad. Frontals broad. Extrascapular firmly articulated with cranium. Lateral ethmoids absent. Meso- pterygoid edentulous, bearing an ascending process. Infraorbital series represented by superficial bony tubes. Urohyal well devel- oped. Four broad branchiostegal rays, two anterior branchiostegals narrower than the others. Basihyal elongate and hollow. Basi- branchials not ossified. Three ossified phar- yngobranchials. Five epibranchials, fifth one cartilaginous. Gill-rakers small, dentig- erous plates, located on both margins of fourth branchial arch and on anterior mar- gin of fifth arch, and covered by connective tissue. Branchial openings large, branchios- tegal membrane united across isthmus. Posttemporal not fused with supracleith- rum. Mesocoracoid present. Coracoid not bearing ventral process. Four pectoral ra- dials. Lateral line complete. Presence of 30-51 precaudal vertebrae. No displaced hemal spines. Anus not growing anteriorly during ontogeny. Anterior insertion of anal fin not extending anterior to vertical through posterior margin of pectoral fin. Anal fin confluent with tip of tail (rays of the posterior portion of anal fin are tightly aggregated and lie horizontally caudally, making it difficult to count the exact num- ber of rays other than in stained specimens or radiographs). Species of Gymnotus mature at moderate body sizes (approximately 150 to 400 mm TL), possess a cylindrical or subcylindrical body, and are completely covered by cy- cloid scales on the postcranial part of their bodies. In many species, color pigmentation patterns are formed by alternating bands of dark and pale oblique bands along the body axis. The band-interband contrast increases ventrally, and is generally more pronounced in juveniles (less than 100 mm TL). Al- though the number, width, and shape of these bands are variable, banding patterns have diagnostic value for several species (Mago-Leccia 1994). 665 We follow Mago-Leccia (1994) in rec- ognizing the following species of the genus and their synonyms: Gymnotus carapo Linnaeus G. carapo Linnaeus, 1758:246 (type lo- cality: America). G. fasciatus Pallas, 1767:35 (type local- ity: Brazil). | G. albus Pallas, 1767:35 (type locality: Suriname). G. brachiurus Bloch, 1787:61, plate 157, fig. 1 (type locality: Brazil). G. putaol Lacépéde, 1800:176 (type lo- cality: Brazil). G. cingulatus Brind, 1935 (type locality: Brazil). Gymnotus inaequilabiatus (Valenciennes) Carapus inaequilabiatus Valenciennes in D’Orbigny 1847:11, plate 14 (type lo- cality: Rio de La Plata, Argentina). Gymnotus pantherinus (Steindachner) Giton fasciatus var. pantherinus Stein- dachner, 1908:129 (type locality: San- tos, Brazil). Gymnotus coatesi LaMonte G. coatesi LaMonte, 1935:1, fig. 1 (type locality: Rio Amazonas, Brazil). Gymnotus cylindricus LaMonte G. cylindricus LaMonte, 1935:2 (type lo- cality: Los Amates, Rio Motagua Ba- sin, Guatemala). Gymnotus anguillaris Hoedeman G. anguillaris Hoedeman, 1962a:55, fig. 2 (type locality: Coropina Creek, Su- riname). G. coropinae Hoedeman 1962a:55, fig. Ic (type locality: Coropina Creek, Su- riname). Gymnotus cataniapo Mago-Leccia Gymnotus cataniapo Mago-Leccia 1994: 90, fig. 100, table 9 (type locality: Rio Cataniapo, Amazonas, Venezuela). Gymnotus pedanopterus Mago-Leccia Gymnotus pedanopterus Mago-Leccia 1994:92, fig. 98, table 10 (type locali- ty: Cafio La Esmeralda, Rio Orinoco, Amazonas, Venezuela). 666 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Vista 8 km W of Pajal, Departamento Santa Rosa, Guatemala. Gymnotus stenoleucus Mago-Leccia Gymnotus stenoleucus Mago-Leccia 1994:94, figs. 99-101, table 11 (type locality: Rio Cataniapo, Amazonas, Venezuela). Gymnotus maculosus, new species Fig. 1-6 Gymnotus carapo in part, Eigenmann & Fisher, 1914:235. Rivers and lakes on both slopes of Guatemala. Gymnotus carapus Carr & Giovannoli, 1950:11—12. Rio Choluteca, Honduras. Gymnotus cylindricus Bussing, 1985, fig. 36. Gymnotus sp. Miller, 1966:786 (Pacific Slope of Middle America from Guate- mala to the Gulf of Fonseca). Gymnotus sp. Miller, 1986:129 (Pacific Slope of Middle America between Te- huantepec and Guatemala). Gymnotus sp. Espinosa Perez et al., 1993: 29 (Rio Usumacinta). Holotype.—UMMZ 230830, 191 mm TL. Diversion of channel from Maria Lin- da, 14°04’N; 90°37’'W, c. 20 km. East of Escuintla, Departamento Santa Rosa, Gua- temala; collected by D. E. Rosen and R. M. Bailey, 27 March 1971. Paratypes—_UMMZ 190783, 3. speci- Lateral view of Gymnotus maculosus, paratypes, USNM 134700, 185 and 195 mm TL, Rio Buena mens, 176-203 mm TL, type locality. UMMZ 224128, 5 specimens, 158-222 mm, Rio Higuerdn, Canas, Costa Rica. UMMZ 230354, 20 specimens, Rio Buena Vista, trib. of Rio Tigre, on road between Escuintla and Chiquimulilla, 8 km. West of Pajal, Departamento Santa Rosa, Guate- mala, 128—230 mm, 1946.1V.18. USNM 134700, 49 specimens, Rio Buena Vista, 8 km. West of Pajal, Departamento Santa Rosa, Guatemala, 67-195 mm TL, 1946.1V.18. TU 24965, 36 specimens, 60— 260 mm TL, Boca del Rio Sapoa at Sapoa, Nicaragua. Diagnosis.—A gymnotid unique in pos- sessing numerous small brown spots, vary- ing in size from two to four times eye di- ameter, distributed over most of the body surface (Figs. 1, 2). The body lacks distinct oblique bands in both juveniles and adults. Scales above the lateral line large, 6—8 rows to the dorsal midline at about midbody. Description.—Figure 3 illustrates body Shape. Body subcylindrical (its greatest width 0.61—0.84% BD); BD = 10-13% TL; dorsal profile of body almost straight, ven- tral profile slightly convex; head small, de- pressed, HL = 9.8-12.7% TL; PO = 58- 64.5% HL; anterior nares tubular, partially concealed in labial groove; posterior nares round, and positioned close to eye; mouth VOLUME 108, NUMBER 4 667 Fig. 2. Lateral view of Gymnotus maculosus, paratype, TU 24965, 238 mm TL, Rio Sapoa at Sapoda, Nic- aragua. broad, gape large, rictus of mouth not reaching vertical through eyes; lower jaw protruding, teeth small, conical, in a single row in each jaw, with irregular second row; snout blunt, PR = 33.1-—37.5% HL; IO = 37.0-41.0% HL; origin of anal fin posterior to tip of pectoral fin, PA = 7.4—-11.1% TL; branchial opening large, BO = 33.0-44.3% HL; nape without pale band; lateral line complete, paralleling main body axis; pec- toral fin small, rounded, Pl = 43-48% HL; pectoral-fin rays 11+15—-16 (n = 8); 30—34 A precaudal vertebrae (Table 1, mode = 32, n = 21); anus near vertical through branchial opening; anal fin short, 155-182 anal-fin rays (Table 2, mode = 169, n = 8). Color in preservative.—Ground color yellow to pale brown laterally on preserved specimens. Dorsum dark brown due to densely concentrated melanophores. Dark brown dorsally and laterally. Ground color overlaid by numerous brown spots on head and body, ranging in size from about 2 to 4 eye diameters, and varying somewhat in Fig. 3. Arrow indicates position of the anus. Scale bar equals 10 mm. Diagrammatic representation of Gymnotus maculosus. (A) dorsal, (B) lateral, and (C) ventral view. 668 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON parietal prootic sphenotic frontal Se pterotic supraoccipital exoccipital epioccipital extrascapular mesethmoid SOO bs ses aj ge vomer orbitosphenoid ventral ethmoid pterosphenoid F.V, Flv FeMipague waa ixGX F. int. carotid basioccipital parasphenoid Fig. 4. Lateral view of neurocranium of Gymnotus maculosus, UMMZ 190531, 185 mm TL. Cartilage indicated by stippling. Abbreviations: FE V,, foramen of cranial nerve V, (profundus); EK H—VI, foramen of optic tract and cranial nerves III (oculomotor), [TV (trochlear), and VI (abducens); E int. carotid, foramen of internal carotid artery; E V,,;, foramen of V, (superficial ophthalmic and maxillary) and V, (internal mandibular) rami of trigeminal nerve; E VII, foramen of cranial nerve VII (facial); KE [X—X, foramen of cranial nerves IX (glossopharyngeal) and X (vagus). Scale bar equals 1 mm. size and arrangement. Some specimens with spots arrayed into approximately 13-— 21 oblique lines, corresponding to oblique lateral bands of other species of Gymnotus; other specimens with smaller spots, arrayed into more than 40 oblique lines, or with spots arrayed in a somewhat random pat- tern. The organization of spot and pigment density is not a function of body size. Anal and pectoral fins without spots. Anal- and pectoral-fin rays and interradial membranes hyaline. Etymology.—From the Latin maculosus, meaning spotted, in reference to the con- Spicuous color pattern possessed by this Species. Distribution.—Gymnotus maculosus lives in freshwater streams of Middle America (Fig. 7). On the Pacific slope it ranges from the Rio San Nicolas (15°20'N, 92°50'W), Chiapas, Mexico, south to trib- utary streams to the Gulf of Nicoya in northwestern Costa Rica (9°47’N, 84°48'W). On the Atlantic slope it appears to be restricted to streams on the western slope of the Lake Nicaragua basin in Nic- aragua, the Rio Sapoa of northern Costa Rica, and the San Carlos and Sarapiqui drainages of northeastern Costa Rica. The seemingly disjunct distribution of G. ma- culosus in Fig. 7 is presumed to result from the lack of collections from Pacific slope drainages of El Salvador and Nicaragua. Remarks.—Gymnotus maculosus shares two characters of ambiguous polarity with G. cylindricus; absence of discrete oblique bands along the body, and short body cav- ity, with 31—35 precaudal vertebrae. Comparisons with other species.—Gym- notus maculosus can be separated from par- apatric members of G. cylindricus in the Lake Nicaragua basin by its unique color pattern, by its smaller interorbital distance, and by the larger size of scales above the lateral line. Specimens of G. cylindricus possess few if any small brown spots on the VOLUME 108, NUMBER 4 A Fig. 5. indicated by stippling. Scale bar equals | mm. body. Although some specimens of G. cy- lindricus possess a mottled appearance, with low contrast, pale spots on the ventral portion of the body, only G. maculosus pos- sesses high contrast spots over the entire body. The interorbital distance of G. ma- culosus 1s 37-41% HL, whereas that of G. cylindricus 1s 41-46% HL. Scales above the lateral line are larger in G. maculosus, with 6-8 rows to the dorsal midline at about midbody, whereas those of G. cylindricus are smaller, with 10—12 rows above the lat- eral line at about midbody. Gymnotus maculosus can be separated from the other species of Gymnotus by its color pattern and meristic, features. Gym- notus maculosus has fewer precaudal ver- tebrae (30-34, n = 21; vs. 38, n = 6) and fewer anal-fin rays (155-182, n = 8; vs. 190-222, n = 6) than does G. carapo. Un- like G. carapo, G. maculosus is spotted and does not have conspicuous oblique bands. Gymnotus maculosus can be separated from other spotted species of Gymnotus, G. in- aequilabiatus and G. pantherinus, by the supracleithrum 669 posttemporal cleithrum mesocoracoid scapula coracoid Pectoral girdle of Gymnotus maculosus, UMMZ 190531. (A) lateral view. (B) medial view. Cartilage Same meristic characters used to separate it from G. carapo. Discussion Variation in characters examined.—Sev- eral body measurements vary among and between species of Gymnotus. Uses of body depth, head length, interorbital distance, and pectoral-fin length to identify species are noted in the key. Despite this variation, the characters used in the diagnoses do prove useful when making certain compar- isons. The deep body of G. inaequilabiatus (BD greater than HL), for example, can be used to separate it from all other South American species of Gymnotus (except some specimens of G. cataniapo) in which the BD is less than HL. Specimens of G. carapo and G. pedanopterus possess a long head (HL greater than 11.8% TL for spec- imens 100—250 mm TL) when compared to other species of Gymnotus. Although the snout length of most Gymnotus species ranges from PR = 32.5—38.0% HL, speci- 670 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON hyomandibula opercle metapterygoid mesopterygoid palatine maxilla p a yds 7 )\ ANS *S ee N y ae dentary anguloarticular quadrate preopercle subopercle Seis oa retroarticular symplectic interopercle Fig. 6. Lateral view of suspensorium of Gymnotus maculosus, UMMZ 190531. Cartilage indicated by stip- pling. Scale bar equals 1 mm. mens of G. anguillaris tend to have a longer snout (PR = 36.0-41.0% HL), and speci- mens of G. coatesi, G. cataniapo, G. pe- danopterus, and G. stenoleucus tend to have shorter snouts (PR = 28.5—36.5% HL). The anal fin of G. carapo and G. in- aequilabiatus tends to insert more anteri- orly than in other Gymnotus species (PA = 55-86% HL vs. 72—112% HL). These mea- surements exclude two specimens of G. cy- lindricus with abnormally short PA distanc- es (31% and 46% HL). PA data for G. pe- danopterus and G. stenoleucus were not available for this study. The two species of Gymnotus from Middle America share a short pectoral fin (Pl = 38.0-47.5% HL) with the Suriname specimens of G. carapo, and with the slender species G. cataniapo, G. pedanopterus, and G. stenoleucus. Body proportions of Gymnotus carapo in South America vary considerably. Speci- mens examined from Suriname can be dis- tinguished from those in the Amazon basin by having a deeper body (BD = 11.5- 12.5% TL vs. 10-11% TL) and a shorter pectoral fin (P1 = 38-45% HL vs. 46-57% HL). The anal-fin origin also tends to be more remote from the anus in the Suriname specimens (PA = 9.5—-11.0% HL vs. 6.5— 10.0% TL). Specimens collected west of the Andean cordillera in Colombia, from the Rio Atrato basin and the Pacific drain- age at Pizarro, possess the two diagnostic traits of G. carapo: a long head (HL = 11.8-12.5% TL) and a clear anal-fin mem- brane at the posterior end of the fin. Mea- surements of these specimens (NV = 5) re- veal that they have the more slender body (BD = 10.0-10.2% TL) and longer pectoral fin (Pl = 45.2-58.3% HL) of G. carapo from the Amazon, and the posterior origin of the anal fin (PA = 9.0-11.2% TL) of G. carapo from Suriname, and of the other Gymnotus species. VOLUME 108, NUMBER 4 671 Fig. 7. Geographic distribution of the two Middle American species of Gymnotus. Type localities indicated by hollow symbols. The absence of G. maculosus from the Pacific slope of El Salvador and Nicaragua is due to sampling bias. Although color patterns are of use in identifying species of Gymnotus, certain as- pects are variable (Nijssen & Isbriicker 1968, Mago-Leccia 1994). The trans-An- dean specimens of G. carapo, for example, differ from other populations G. carapo in possessing narrow pale bands. This aspect of coloration is otherwise only observed in elongate, slender species of Gymnotus (see below). Another example of variable color pattern is G. anguillaris from the Lawa River, Suriname, which resemble G. carapo in their dusky brown ground color and in possessing more than 23 alternating pale and dark bands. The pale bands are narrow- er than the dark bands only on the posterior end of the body. Specimens of G. anguil- laris from other localities, however, ranging from Guyana to Paraguay, more closely re- semble the description of the type specimen in which the pale bands are more narrow along the entire length of the body (Hoede- man 1962a). Comments on the interrelationships of Gymnotus.—Species diversity and system- atics of the Gymnotoidei have received considerable attention in recent years (Mago-Leccia 1978, 1994; Lundberg and Mago-Leccia 1986; Triques 1993; Gayet et al. 1994: Alves-Gomes et al. 1995; Albert 1995). The results of Albert’s (1995) study support Ellis’s (1913) hypothesis that Gym- notus and Electrophorus form a monophy- letic group, the Gymnotoidea (Gymnotoidei of Mago-Leccia 1978), which is itself the sister lineage to other gymnotoids (Albert 1995, fig. 1; Albert & Fink 1995, fig. 2). The Gymnotoidea is diagnosed by 14 char- acters. Gape large, extending to vertical through posterior nares; premaxilla large, 672 articulating head of maxilla oriented ante- riorly; lateral margin of premaxilla pro- duced into a posterior process; ventral mar- gin of descending maxillary blade with a sharp angle about two-thirds distance to its tip; base of lateral ethmoid narrow; cranial fontanels closed in adults; m. adductor mandibulae undivided bundle at origin; lat- eral portion of valvula cerebelli larger in cross sectional area than medial portion at isthmus; depression on dorsal surface of ba- sihyal; ventral process of coracoid absent; anal-fin pterygiophores long, more than one-third total body depth; caudal fin or fil- ament absent, caudal anal-fin rays extend to tip of tail; body cavity long, more than 30 precaudal vertebrae; posterior chamber of gas bladder elongate, passing between he- mal arches of postcoelomic axial skeleton and musculature. Although a formal hypothesis of relation- ships among species within Gymnotus is beyond the scope of this paper, the data used in this study may be used to divide Gymnotus into three species-groups, on the basis of color pattern and general body pro- portions. Because the characters used to recognize these species-groups are not po- larized, the monophyly of these groups and their interrelationships are ambiguous. One species-group is composed of elon- gate, slender species, with a BD less than 11% TL. Members of this group also pos- sess a color pattern consisting of alternating dark and pale bands, in which the pale bands are narrower than the dark bands, and the dense brown or grey dark bands extend onto the dorsum. Gymnotus anguillaris is the most widely distributed member of this group and exhibits the greatest variation in these characters. Members of the second species-group have deeper bodies, with a BD greater than 11% TL, and the alternat- ing bands of heavy and light pigmentation are of about equal width, grading to a dusky brown on the dorsum. This group includes the type species G. carapo. The third spe- cies-group is composed of the two exclu- sively Middle American species of Gym- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON notus. These species lack the oblique non- pigmented bands of South American Gym- notus species, and have a shorter body cavity (31-35 precaudal vertebrae), char- acteristics regarded to be relatively ple- siomorphic among gymnotoids (Albert 1995). Based on both phylogenetic and biogeo- graphic considerations, the Gymnotidae can be presumed to have originated in South America (Lundberg 1993). There is no ev- idence, however, that the Middle American species of Gymnotus are derived from an extant South American congener. If the Middle American species-group were to be identified as the sister taxon to other species of Gymnotus, the presence of an elongate body cavity with more than 35 precaudal vertebrae, would be evidence for the mono- phyly of the G. carapo and G. anguillaris species-groups. Unfortunately, the polarity of body cavity length in Gymnotus is am- biguous. The immediate outgroup taxon, Electrophorus, does not possess caudal ver- tebrae at all; the body cavity and gas blad- der extend to the caudal end of the animal, hemal spines do not form, and the hemal arches do not meet along the midline. The next outgroup is not elongate; species of Sternopygus, which retain a relatively ple- siomorphic gymnotoid morphology, pos- sess fewer precaudal vertebrae than any species of Gymnotus. In fact, the phyloge- netic distribution of the number of precau- dal vertebrae among extant gymnotoid taxa indicates that the plestomorphic condition for the group is 17—19 (Albert 1995). The polarity of oblique depigmented bands in Gymnotus is also ambiguous. Many rhamphichthyoid species possess al- ternating color bands, but these bands are vertical regions of pigment investing a non- pigmented background. Narrow bands are not present in Electrophorus, Sternopygus, or the plesiomorphic hypopomid Hypopo- mus (Albert 1995). Furthermore, the pre- sumed absence of oblique bands in the Mid- dle American species of Gymnotus is not entirely clear; the pattern of spots in some VOLUME 108, NUMBER 4 specimens of G. maculosus is somewhat similar to the oblique bands of other Gym- notus, 1n position, orientation, and number. Biogeography of Middle American Gym- notus.—The geographic distribution of Gymnotus is discussed by Meek and Hil- debrand (1913), Eigenmann & Fischer (1914), Martin (1972), Miller & Carr (1974), Villa (1982), and Bussing (1987). Eigenmann & Allen (1942) claimed a range from Costa Rica to the Rio de la Plata. The northernmost record of a South American species of Gymnotus, as recognized here, is G. cf. anguillaris from Almirante on the At- lantic slope of Panama (Behre 1928, Hil- debrand 1938, see Appendix), located about 170 km south of the southern limit of G. cylindricus at Tortuguero, Sixaloa, Costa Rica (Gilbert & Kelso 1971). The northern limit of G. carapo on the pacific slope is from Pizarro, Choco, Colombia. No refer- ences were found for the presence of G. carapo from the Rio Tuira or elsewhere in Panama. The two species of Gymnotus limited to Middle America come into closest contact in small streams on opposite sides of Lake Nicaragua, and in several tributaries of the Rio San Juan in Northern Costa Rica, an Atlantic drainage. The presence of G. ma- culosus in these localities may have resulted from stream capture due to tectonic uplift of the region in the past several million years (Durham 1944; Miller 1950, 1986). Geologic evidence for the uplift of the Nic- araguan graben is discussed by Riedel (1972). The parapatric distribution of species of Gymnotus is similar to that of some other fishes in the lake Nicaragua basin. The shad Dorosoma chavesi, the poeciliid Poeciliop- sis cf. turrubarensis, and the gar Atractos- teus tropicus found in this basin are other- wise only known from Pacific-slope drain- ages. A second species of poeciliid record- ed from the Lake Nicaragua basin, P. cf. gracilis, also occurs on the Atlantic slope in southern Mexico and in the Rio Motagua of Guatemala. The sawfish Pristis pectina- 673 ta, and the tarpon Tarpon atlanticus, both found in Lake Nicaragua, are confined to the Atlantic basin. Key to adults of species of Gymnotus (HL greater than 10 mm) la. No oblique bands on body; body cavity with 31—35 (mode = 32) precaudal ver- teDrac eee ee ee ee ee nee 2 Ib. Alternating bands of dark and pale pig- mentation arranged at an oblique angle along length of body; body cavity with 36—51 precaudal vertebrae ......... 3 2a. Numerous small brown spots on body; eyes set close together, interorbital dis- tance 37-41% HL; scales above lateral line large, 6—8 rows at midbody Gymnotus maculosus, n. sp. 2b. Few if any spots on body; eyes set apart, interorbital distance 41-46% HL; scales above lateral line small, 10—12 rows at midbody Sede ae Te Gymnotus cylindricus LaMonte 3a. Body deep, depth at anal-fin origin more than about 9% TL (except G. pantherinus); alternating dark and pale oblique bands of equal width at mid- body, or pale bands broader; ground color dusky brown on dorsum; anal fin hyaline posteriorly 3b. Body slender, depth at anal-fin origin less than about 9% TL; dark oblique bands broader than pale bands at mid- body; ground color dark brown or dark grey on dorsum; anal fin pigmented posteriorly 4a. Head long, HL greater than about 11% TL for specimens 100—250 mm TL); no spots on body, dark oblique bands un- interrupted; pigment density in middle part of dark bands less dense than near margins; bands present on anterior por- tion of body .. Gymnotus carapo (Linneus) 4b. Head short, HL less than about 11% TL; spots on head and back, dorsal part of dark oblique bands interrupted; dark oblique bands of uniform pigment den- sity; bands not present on anterior half of body in specimens larger than 220 mm TL oe) (0 (ee) es e © © © © © © © © eo OOF 0100 OOO O00 0 080 0.0 OF 010) 674 5a. Body deep, BD greater than 9% TL Gymnotus inaequilabiatus (Valenciennes) 5b. Body shallow, BD less than 9% TL : Gymnotus pantherinus (Steindachner) 6a. Eyes set further apart, IO 39-44% HL; nape dark, pale bands not extending above lateral line on anterior half of the body Gymnotus anguillaris Hoedeman 6b. Eyes set closer together, IO 25-38% HL (except some G. cataniapo); white band at nape, pale bands extending to dorsal midline on anterior half of body EE eo oe Oe ee ee me MCS ce 7 7a. 13-23 oblique bands; pectoral fin lon- PCTS HOA om blll ae acs sue eaieke wank. 8 7b. 23-50 oblique bands; pectoral fin short- Cra 8 Sion Ae oy ie ck ovo cic seme 9 8a. Body cavity shorter, 40-42 precaudal vertebrae; alternate pale bands incom- plete Gymnotus coatesi LaMonte 8b. Body cavity longer, 47-48 precaudal vertebrae; all pale bands complete Le ee Gymnotus stenoleucus Mago-Leccia 9a. Head shorter, HL 8—10% TL; body cav- ity shorter, 50-51 precaudal vertebrae; posterior end of anal fin dark Gymnotus cataniapo Mago-Leccia 9b. Head longer, HL 12-14% TL; body cavity shorter, 36—37 precaudal verte- brae; posterior end of anal fin pale Gymnotus pedanopterus Mago-Leccia ee © ee Acknowledgments William Bussing, David Catania, Mary Anne Rogers, Bill Saul, and Richard Vari provided many specimens used in this study. Sara Fink helped interpret the mor- phology of the ethmoid region. Tamaki Yuri helped with construction of the key. Ricar- do Campos-da-Paz and John Sullivan pro- vided useful comments on the characters. William Bussing, John Lundberg, and Rich- ard Vari made many useful suggestions on earlier drafts of the manuscript. We thank the Smithsonian Institution photographic laboratory for the photograph reproduced in Fig. 1, and Margaret Van Bolt who re- touched the background. Brian Dyer pro- vided the Spanish abstract. We acknowl- edge the NEODAT project (NSF/AID DEB grant 90-24797) for collection information. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON We thank Doug Nelson for curatorial assis- tance, and the Secretaria, Departamento de Pesca, for permission to collect in Mexico. Part of the research reported was conducted in facilities provided by NSF grant BSR 9015158 to Willim Fink. Literature Cited Albert, J. S. 1995. Phylogenetic systematics of the American Knifefishes (Teleostei: Gymnotoidei). Unpublished Ph.D. Thesis, University of Mich- igan, 528 pp. , & W.L. Fink. 1995. Sternopygus xingu, a new species of electric fish from South America (Gymnotoidei, Teleostei), with comments on the phylogenetic position of Sternopygus.—Copeia (in press). Alves-Gomes, J. A., G. Orti, M. Haygood, A. Meyer, & W. Heiligenberg. 1995. Phylogenetic analy- sis of the South American electric fishes (Order Gymnotiformes) and the evolution of their elec- trogenic system: a synthesis based on morphol- ogy, electrophysiology, and mitochondrial se- quence data.—Molecular Biology and Evolu- tion 12(2):298-318. Artedi, P. 1738. Ichthyologia sive Opera Omnia de Piscibus, in Carolus Linneaus, ed., Lugdini Ba- tavorum. Behre, E. H. 1928. A list of the freshwater fishes of western Panama, between long. 81°45’ and 83°15’ W.—Annals of the Carnegie Museum 18(2):305-328. Bloch, M. E. 1787. Naturgeschichte der Auslandisch- en Fische, Berlin. Brind, W. L. 1935. Hunting the ““Tiger Knife Fish” in the Amazons.—Aquarium News, New York 2(2):5, 8-10. Breder, C. M. Jr. 1927. The fishes of the Rio Chu- cunaque Drainage, Eastern Panama.—Bulletin of the American Museum of Natural History 52(3):91-176. Bussing, W. A. 1985. Patterns of distribution of the Central American ichthyofauna. Pp. 453-473 in E G. Stehli and S. D. Webb, eds., The Great American Biotic Interchange. Plenum, New York. 1987. Peces de las aguas continentales de Costa Rica. Editorial de la Universidad de Costa Rica. San José, 271 pp. Carr, A., & L. Giovannoli. 1950. The fishes of the Choluteca drainage of southern Honduras.— Occasional Papers of the Museum of Zoology, University of Michigan 523:1-—37. Cuvier, G. 1816. Le Regne Animal distribué d’apres son organisation pour servir de base a I’ histoire VOLUME 108, NUMBER 4 naturelle des animaux et d’introduction a |’an- atomie compareé, Edition 1. v.2i—xviii+ 1—532. Dumeéril, M. C. 1856. Ichthyologie Analytique ou Es- sai d’une classification naturelle des Poissons a Vaide de Tableaux Synoptiques.—Memiores Academie France 27, Paris. Durham, H. W. 1944. New volcanoes and new moun- tain ranges.—Science 100:49—50. Eigenmann, C. H., & W. R. Allen. 1942. Fishes of Western South America, University of Ken- tucky, Lexington, 494 pp. , & H. G. Fischer. 1914. The Gymnotidae of Trans-Andean Colombia and Ecuador.—Indiana University studies no. 25, 141:235—237. , & D. P. Ward. 1905. The Gymnotidae.—Pro- ceedings of the Washington Academy of Sci- ence 7:157-—86. Ellis, M. M. 1913. The gymnotid eels of South Amer- ica.—Memiores of the Carnegie Museum 6(3): 109-195. Espinosa-Pérez, H., M. T. Gaspar-Dillanes, & P. Fu- entes-Mata. 1993. Listados faunisticos de Méx- ico. III. Los peces dulceacuicolas Mexicanos.— Universidad Nacional Aut6énoma de México 1993:1-98. Fink, S. V., & W. L. Fink. 1981. Interrelationships of the ostariophysan fishes (Teleostei).—Zoologi- cal Journal Linnean Society 72(4):297-353. Fink, W. L. 1989. Ontogeny and phylogeny of shape change and diet in the South American fishes called piranhas. Jn B. David et al., eds., Onto- genése et évolution.—Geobios, mémiore spécial 12:167-172. , & A. Machado-Allison. 1993. Three new species of piranhas from Brazil and Venezuela (Teleostei: Characiformes).—Ichthyological Ex- plorations of Freshwaters 3(1):55-—71. Gayet, M., & E J. Meunier. 1991. Premiére décou- verte de Gymnotiformes fossiles (Miocéne su- périeur, Bolivie).—C. R. Academie Sciences Paris, Série II:471—476. , & E Kirschbaum. 1994. Ellisella kirschbaumi Gayet & Meunier, 1991, gymnoti- forme fossile de Bolivie et ses relations phylo- génétiques au sein des formes actuelles.—Cy- bium 18(3):273—306. Gilbert, C. R., & D. P. Kelso. 1971. Fishes of the Tortuguero area, Caribbean Costa Rica.—Bul- letin of the Florida State Museum of Biological Science 16:1—54. Hildebrand, S. EF 1938. A new catalogue of the fresh- water fishes of Panama.—Field Museum of Natural History, Zoological Series 22(4), Pub- lication 425:219-359. Hoedeman, J. J. 1962a. Notes on the ichthyology of Suriname and other Guyanas, 9. New records of gymnotid fishes.—Bulletin Aquatic Biology, Amsterdam 3(26):53—60. 675 . 1962b. Notes on the ichthyology of Suriname and other Guyanas, 11. New gymnotoid fishes from Suriname and French Guyana, with addi- tional records and a key to the groups and spe- cies from Guyana.—Bulletin Aquatic Biology, Amsterdam 3(30):97—107. Lacépéde, B. G. E. 1800. Histoire Naturelle des Pois- sons. Vol. 2, Chez Plassan, Imprimeur-Libraire, Paris. LaMonte, EF 1935. Two new species of Gymnotus.— American Museum Novitates No. 781:1-3. Leviton, A. E., R. H. Gibbs, Jr, E. Heal, & C. E. Dawson. 1985. Standards in herpetology and ichthyology: Part I. Standard symbolic codes for institutional resource collections in herpe- tology and ichthyology.—Copeia 1985(3):802— 832. Linnaeus, C. 1758. Systema Naturae, 10th edition, Vol. 1. Laurentii Salvii, Holmiae, 824 pp. Lundberg, J. G. 1993. African-South American fresh- water fish clades and continental drift: problems with a paradigm. Pp. 156-199 in P. Goldblatt, ed., Biological relationships between Africa and South America. Yale University Press, New Ha- ven. , & E Mago-Leccia. 1986. A review of Rhab- dolichops (Gymnotiformes, Sternopygidae), a genus of South American freshwater fishes, with descriptions of four new species.—Pro- ceedings of the Academy Natural Sciences Phi- lalephia 138:53-85. Mago-Leccia, EF 1978. Los peces de la familia Ster- nopygidae de Venezuela.—Acta Cientifica Ve- nezolana 29:1—51. . 1994. Electric fishes of the continental waters of America. Biblioteca de la Academia de Cien- cias Fisicas, Matematicas, y Naturales, Vol. 29, Caracas, Venezuela, 206 pp. Martin, M. 1972. A biogeographic analysis of the freshwater fishes of Honduras. Unpublished Ph.D. Thesis, University of Southern California, 456 pp. Meek, S. E., & S. E Hildebrand. 1913. New species of fishes from Panama. Family Gymnotidae.— Field Museum of Natural History, Zoological series, 10(8), Publication 166:305—313. Miller, R. R. 1950. A review of the American clupeid fishes of the genus Dorosoma.—Proceedings of the U.S. National Museum 100:387—410. 1966. Geographical distribution of Central American freshwater fishes.—Copeia 1966: 773-802. 1986. Composition and derivation of the freshwater fish fauna of Mexico.—Anales de la Escuela Nacional de Ciencias Biol6gicas, Méx- ico 30(1-4):121-153. Miller, R. R., & A. Carr. 1974. Systematics and dis- 676 tribution of some freshwater fishes from Hon- duras and Nicaragua.—Copeia 1974:120-125. Nijssen, H., & I. J. H. Isbriicker. 1968. Gymnotus car- apo and G. anguillaris (syn. G. coropinae), two often confused species of gymnotid fishes. Pi- sces, Cypriniformes.—Beaufortia 15(203):161— 168. Pallas, P S. 1767. Spicilegia Zoologica, Petrop., 1769-79. Pittman, W. C. III., S. Cande, J. LaBrecque, & J. Pin- dell. 1993. Fragmentation of Gondwana: The separation of Africa from South America. Pp. 15-61 In P. Goldblatt, ed., Biological relation- ships between Africa and South America. Yale University Press, New Haven. Putzer, H. 1984. The geological evolution of the Am- azon Basin and its mineral resources. Pp. 14— 46 in H. Sioli, ed., The Amazon. Limnology and landscape ecology of a mighty tropical river and its basin. Dr. W. Junk Publisher, The Neth- erlands, ix + 763 pp. Riedel, D. 1972. Die Genesis der nicaraguensischen Grabenseen (Teil I) und des mesoamerickan- ischen (Teil II) aus der Sicht des Fischereibiol- ogen.—Archives Hydrobiologia 70:82—107. Steindachner, EF 1908. Brachyplatystoma -Art aus dem Rio Parnahyba un iiber eine ditch gefleckte und gestrichelte Varietaet von Giton fasciatus aus dem Gewassern von Santos. Staat Sao Paulo.— Anzeiger der Akademie der Wissenschafften, Wien 45(9):123—130. Taylor, W. R., & G. C. Van Dyke. 1985. Revised pro- ceedures for staining and clearing small fishes and other vertebrates for bone and cartilage study.—Cybium 9:107-119. Triques, M. L. 1993. Filogenia dos géneros de Gym- notiformes (Actinopterygii, Cstariophysi), com base em characters esqueleticos.—Comunica- ¢des do Museu Ciéncias da PURCS, série Zool- ogia, Porto Alegre, 6:85—130. Valenciennes, A. 1847. Poissons. Pp. 5—11 in D’Or- bigny, Voyage dans |l’Amérique Meridionale Vol. 5, 2nd part, Paris. Villa, J. 1982. Peces nicaraguenses de agua dulce.— Banco de América, serie geografia y naturaleza 3:i—xiv, 1-253. Weitzman, S. H. 1974. Osteology and evolutionary relationships of the Sternoptychidae, with a new classification of stomiatoid families.—Bulletin of the American Museum of Natural History 153:327-478. Appendix 1 List of 125 lots containing 698 specimens of gym- notids examined. Data are arranged by country, then alphabetically by museum acronym and number. Cat- alogue numbers are followed in parentheses by number of specimens, and then by locality, size range in mil- limeters total length (mm), and date of capture. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Gymnotus anguillaris—Guyana: FAU uncat. (2), Madewini River, 132-165 mm, 1994.IV.20. Paraguay: UMMZ 206080 (21), Arroyo in Parque Nacional Yby- cui, 82—260 mm, 1979.V1.20. Suriname: UMMZ 190413 (6), Maka Creek, Lawa River, Morowyne, 1967.1V.21. ZMA 100338 (2), Coropina Creek, 228— 236 mm, 1956.V.20. Gymnotus cf. anguillaris—Panama: CAS 72209 (2), small creek into Rio Cricamola, behind Konkintu, Boca del Toro Province, 225—239 mm, 1923.11.25. Gymnotus carapo.—Bolivia: UMMZ 066433 (2), Lake Rogoagua, Rio Beni, 214-338 mm, 1921.IX. UMMZ 066462 (2), Reyes, Rio Beni, 205-245 mm, 1921.1X. UMMZ 204299 (1), Costa Marquez, Rio It- enez (Guapore), Madeira, 164 mm, 1964.1X.02. UMMZ 204771 (5), Costa Marquez, Rio Itenez, 44— 51 mm, 1964.X.1. UMMZ 204886 (3), Rio Baures, Rio Itenez, 145-345 mm, 1964.X.7. Brazil: UMMZ 143282 (2), Lagoa dos Quadros, Rio Grande do Sul, Tramandai, 55-129 mm, 1941.VHI. UMMZ 230732 (2), Aggassiz collections, presumably Brazil, 174—202 mm, Acc. 1876.V.9. UMMZ 230734 (2), Rio Cayari near Benjamin Constant, Amazonas, 190-210 mm, 1993.V.11. USNM 199215 (1), Upper Yuruena, Mato Grosso, 85 mm, 1964.VII.20. USNM 199219 (2), Up- per Yuruena, Mato Grosso, 1965.IV.9. Colombia: CAS 072192 (2), Rio Sucio, Rio Truando, Rio Atrato, 150— 179 mm, 1913. FMNH 56793 (2), mouth of Rio Cal- ima, Rio Atrato, 1913. FMNH 56794 (2), Rio Sucio, Rio Truando, Rio Atrato, 174-175 mm, 1913. FMNH 70511 (5), Pizarro, Choco, Pacific drainage, 124—247 mm, 1945. IX. Peri: UMMZ 072636 (2), Yurimaguas, Rio Maranon, 148—214 mm, 1920.1X. UMMZ 230733 (1), Quebrada near Buen Sucesso, Rio Javari, Loreto, 250 mm, 1993.V.15. UMMZ 228998 (2), Rio Momon, near Iquitos, Rio Nanay, Loreto, 38-172 mm, 1993.V.5. UMMZ 228999 (1), Rio Tapira, Rio Tahu- ayo, Loreto, 161 mm, 1993.V.7. Suriname: UMMZ 187499 (1), Brokopondo, 195 mm, 1966.IX.15. UMMZ 190414 (6), Brokopondo, 71-225 mm, 1966.1X.15. ZMA 100409 (1), Aloike, Litany River, 71 mm, 1957.XI.27. ZMA 100430 (1), Ile de Cayenne, 305 mm, 1957.X.6. ZMA 100434 (2), Degrad Cacao, 190-255 mm, 1957.XI.10. ZMA 100439 (1), 256 mm, 1957.X.5. Trinidad: UMMZ 169080 (2), Gunupia, Mt. Plaisance, 46-74 mm, 1953.11.12. Venezuela: VMMZ 212345 (1), Esteros de Camaguan, Guarico, Apure, 45 mm, 1981.VIII. UMMZ 214766 (1), Cafio Falc6n, Por- tuguesa, 241 mm, 1987.]1.19. Gymnotus cataniapo.—Venezuela: MBUCV-V- 14154 (3), camo near Salto Nieve, Rio Orinoco, 75— 194 mm, 1983.XII.22. MBUCV-V-14300 (1), Cafio Merete, Rio Casiquiare, 213 mm, 1984.VIII.20. MBUCV-V-14736 (1), San Carlos de Rio Negro, Rio Orinoco, 253 mm, 1984.X.4. MCUCV-V-14757 (1), Cano Guayabal, near Puerto Ayacuchu, Rio Orinoco, 114 mm, 1984.XTI.12. MBUCV-V-14781 (15), Cafio Las Pevas, Rio Casiquiare, 143—316 mm, 1984.VIII.2. VOLUME 108, NUMBER 4 Gymnotus coatesi.—Bolivia: UMMZ 205149 (7), near Guayaramerin, Arroyo Grande, 45-175 mm, 1964.X.23. Paraguay: UMMZ 206155 (2), near San Rafael, Arroyo Tembley, 194-225 mm, 1979.VI.26. Peru: UMMZ 224596 (10), Buen Sucesso, Rio Javari, Loreto, 45-167 mm, 1993.V.15. UMMZ 224599 (2), Quebrada near Buen Sucesso, Rio Javari, Loreto, 1993.V.15. UMMZ 224601 (1), cafio near Santa Ana, Rio Tahwayo, Loreto, 1993.V.4. UMMZ 224607 (6), near Iquitos, Rio Nanay, 30-144 mm, 1993.V.4. Gymnotus cylindricus.—Costa Rica: UCR 1014-1 (12), Rio Escondido, Limon, Rio Sixaola, 123—249 mm, 1976.1X..17. UCR 280-2 (3), Rio Escondido, Li- mon, Rio Sixaola, 1968.X.3. UMMZ 224129 (10), Rio Escondido, Limon, Rio Sixaola, 131-205 mm, 1968.X.3. Guatemala: UMMZ 193873 (2), el Progres- so, Morazan, Rio Yeguare, 192—210 mm, 1973.]I1.24. UMMZ 193986 (14), Quebrada de Vegega, Los Amates, Rio Izabal, 26—183 mm, 1973.IV.5. Honduras: UMMZ 155831 (1), bridge on Tegucigalpa-Danli road, Rio Yeguare, 174 mm, 1948.11.30. UMMZ 155832 (2), Quebrada near Los Flores, Rio Yeguare, 161—204 mm, 1947.VII.23. UMMZ 188110 (2), Pito Solo, Cortez, Rio Jaitique, 93-191 mm, 1948.V.23. UMMZ 188273 (1), Lago Yoyoa, Santa Barbara, Rio Yeguare, 135 mm, 1949.VIII.1. UMMZ 188274 (5), Lago Yojoa, Santa Barbara, Rio Yeguare, 97-175 mm, 1948.IV.14. UMMZ 188275 (16), Lago Yojoa, Rio Yeguare, 166— 224 mm, 1949. VIII.10. UMMZ 188296 (2), Quebrada Lagunita, Morazan, Rio Yeguare, 120-138 mm, 1947.VII.29. UMMZ 188297 (5) Quebrada behind fin- ca de Rudolfo Rosales, Rio Yeguare, 69-129 mm, 1947.XI. UMMZ 199598 (1), Laguna Sikalanka, Rio Sucre, 215 mm, 1975.V.11. Nicaragua: CAS 161383 (3), Rio Frio, about 1.5 mi. above San Carlos, in bayou with sluggish current, Lago Nicuaragua drainage, Rio San Juan state, 1963.11.12. TU uncat. (1), San Carlos, Lago Nicuaragua drainage, Rio San Juan state. UMMZ 199622 (13), SE of Bilwaskarma, Rio Kurnog, 1975.V.13. UMMZ 199633 (2), Rio Putkrukira near Waspan, Rio Coco, 121-194 mm, 1975.V.27. Gymnotus inaequilabeatus.—Brazil: USNM 1643 (1), Rio Paraguay, 791 mm. USNM 1644 (1), Rio Par- aguay. USNM 1645 (1), Rio Paraguay. Paraguay: UMMZ 206939 (1), near Pto. Stroessner, Arroyo Ve- necia, 154 mm, 1979. VIII.5. UMMZ 206703 (4), Pe- dro Juan Caballero, Parana, 113—280 mm, 1979.VI.24. UMMZ 206971 (2), Estancia la Golondrina, Presidente Hayes, Confuso, 255-261 mm, 1979.VIIIl. UMMZ 207025 (17), marsh 34 km N.W. Pt. Remaro bridge, Rio Confuso, 485 mm, 1979.VIII.9. UMMZ 207096 (3), Estancia la Golondrina, Presidente Hayes, Confu- so, 132-210 mm, 1979. VIII. UMMZ 207564 (2), Rio Pilcomayo near Peurto Falc6n, Rio Paraguay, 220—242 mm, 1979. VIII.29. UMMZ 207619 (1), Riachtelo Pil- co, Presidente Hayes, 144 mm, 1979.VIII.31. UMMZ 207760 (2), Arroyo Peguajho, Ypane, 77 mm, 1979. 1X.04. UMMZ 207893 (6), Rio Aequidaban near 677 Paso MHasqueta, Rio Parand, 16-390 mm, 1979. 1X.6.UMMZ 215183 (1), Estancia la Golondrina, Presidente Hayes, Confuso, 170 mm, 1981.X.02. UMMZ 216576 (1), Estancia la Golondrina, Presidente Hayes, Rio Confuso, 322 mm, 1981.X.02. UMMZ 216576 (1), near Trans Chaco-Villa, Presedente Hayes, 322 mm, 1981.X.2. Gymnotus maculosus.—Costa Rica: TU 24935 (16), La Virgin, Rio Sarapiqui, 1960.11.22. TU 25063 (3), Rio Tempisque, near Liberia, Guanacaste, 193-218 mm, 1961.1.21. UCR 969-9 (20), Santa Cruz, Rio Gar- zon, Guanacaste, 123-195 mm, 1976.1.25. UCR 980- 9 (5), Arenal, Rio Dos Bocas, Guanacaste, Rio San Carlos drainage, 1968.VI.10. UMMZ 158451 (1), Fin- ca la Trinidad, Guanacaste, 179 mm, 1949.X.21. UMMZ 224128 (5), Rio Higuer6n, near Canas, Guan- acaste, 158—222 mm, 1968.VI.10. Guatemala: UMMZ 188072 (1), Rio Bravo, Rio Nahualate, 1968.IV.3. UMMZ 190531 (7), El Obraje, Jutiapa, Rio Grande de Pasaco, 146-200 mm, 1971.111.9. UMMZ 190783 (3, paratypes), diversion of channel from Rio Maria Lin- da, 20 km East of Escuintla, Departamento Santa Rosa, 176—203 mm, 1971.111.27. UMMZ 194122 (6), 2.6 km ESE Eca Cocales, Rio La Primavera, 105-135 mm, 1973.1V.20. UMMZ 194150 (2), Rio Siguacan, near Escuintla, Santa Rosa, 141-172 mm, 1973.1V.24. UMMZ 197103 (17), near Taxisco, Santa Rosa, 1974.1V.26. UMMZ 197103 (20), Taxisco, Santa Rosa, 78—227 mm, 1974.1V.6. UMMZ 230354 (20, para- types), Rio Buena Vista, trib. of Rio Tigre, on road between Escuintla and Chiquimulilla, 8 km West of Pajal, Departamento Santa Rosa, Guatemala, 128—230 mm, 1946.1V.18. UMMZ 230830, (1, holotype), diver- sion of channel from Rio Maria Linda, 20 km East of Escuintla, Departamento Santa Rosa, 191 mm, 1971.111.27. USNM 114235 (2), Rio Colojate, 1947. USNM 114539 (22), near Malacatan, Rio Gramal, 1956. VII.6. USNM 134700 (49, paratypes), near Payal, Rio Buen Vista, 138-226 mm, 1946.1V.18. USNM 225435 (1), Rio Chiquimulilla, 1946. USNM uncat. Rio Hondo, 1946. USNM uncat. (8), Rio Lato, 1946. USNM uncat. (23), Rio Grammal, 1947. Mexico: UMMZ 191702 (2), Tapachula, Rio San Nicolas, Chia- pas, 163-189 mm, 1971.11.13. UMMZ 191712 (3), Tapachula, Rfo San Nicolas, Chiapas, 184-228, 1971.11.27. Nicaragua: TU 24965 (38, paratypes), Boca de Rio Sapoa, Sapoa, Lago Nicaragua drainage, Rivas Province, 1960.IV.23. TU 25032 (1), Boca de Rio Sapoa, Sapoa, Lago Nicaragua drainage, Rivas Province, 43 mm, 1961.1.21. Gymnotus n. sp. N.—Ecuador: FMNH uncat. (46), Aguarico, Rio Napo, 1983.X1.21. Gymnotus pantherinus.—Brazil: USMN 297933 (20), Naneia, Sao Paulo, 1988.11.21. USMN 297939 (12), Guaratuba, Rio da Praia, 1988.X.3. Gymnotus pedanopterus.—Venezuela: MBUCV-V- 14737 (2), Cahlo Chola, near San Carlos de Rio Negro, Rio Casiquiare, 198-228 mm, 1984.XI. MBUCY-V- 678 14738 (1), Caflo Temblador, Rio Casiquiare, 215 mm, 1984.XII.5. MBUCV-V-4860 (1), Cafio Esmeralda, Rio Orinoco, 281 mm, 1966.XI.6. MBUCV-V-7135 (2), El Pozo de Lucas, near San Fernando de Atabapo, Rio Orinoco, 109-119 mm, 1973. MBUCV-V-7310 (3), cafio near San Fernando de Atabapo, Rio Orinoco, 92-185 mm, 1972.IV.14. Gymnotus stenoleucus.—Venezuela: MBUCV-V- 14028 (15), Rio Cataniapo, Rio Orinoco, 80-171 mm, 1983.11.30. MBUCV-V-14747 (1), Cafio Guayabal, near Peurto Ayacuchu, Rio Orinoco, 140 mm, 1984. XII.12. MBUCV-V-4644 (2), Rio Quiritare, Rio PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Orinoco, 91-95 mm, 1966.X.30. MBUCV-V-6218 (1), Cano Caripo, Rio Casiquiare, 140 mm, 1969.1.26. MBUCV-V-9417 (4), Cafio Caripo, Rio Casiquiare, 91-142 mm, 1969.1.26. Electrophorus electricus.—Aquarium: UMMZ 183710-S (1), 100 mm+, 1961.VII.1. Bolivia: UMMZ 204426-S (2), Costa Marques, Rio Itenez, Rio Madei- ra, 940 mm, 1964.1X.12. Guyana: USNM 228883 (3), Essequibo River. UMMZ uncat. (1), 1000+ mm, 1992.X11.12. Suriname: USNM 225669 (1), Amotopa landing, Nickerie District, Rio Corantijn, 511 mm, 1980.VII.19. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(4):679-686. 1995 A new species of the genus Bryconamericus Eigenmann, 1907 from southern Brazil (Ostariophysi: Characidae) Luiz R. Malabarba and Andreas Kindel (LRM) Departamento de Zoologia, Universidade Federal do Rio Grande do Sul, Av. Paulo Gama s/n, 90046-900—Porto Alegre-—RS—Brasil; and Museu de Ciéncias e Tecnologia, PUC-RS, Av. Ipiranga 6681, Cx.P. 1429, CEP 90619-900 Porto Alegre-—RS—Brazil (AK) Curso de Pé6s-Graduagao em Ecologia, UFRGS Abstract.—A new characid species, Bryconamericus lambari, is described from Southern Brazil. This fish is known only from the arroio Feitoria, a small and rocky mountain tributary of the rio Cai, laguna dos Patos drainage, from municipio de Dois Irmaos, Rio Grande do Sul, Brazil. The identities of other Bryconamericus species described from southern South America are briefly discussed. Bryconamericus boops is considered a junior synonym of B. ther- ingit and a lectotype is designated for Tetragonopterus theringii. Resumo.—Bryconamericus lambari, espécie nova, é descrita para o sul do Brasil. A espécie é conhecida somente do arroio Feitoria, um pequeno arroio tributario do rio Cai, sistema da laguna dos Patos, municipio de Dois Irmaos. A identidade de outras espécies do género Bryconamericus descritas para 0 Sul da América do Sul é discutida. Bryconamericus boops é considerado sinédnimo de B. iheringii. E designado o lect6tipo de Tetragonopterus iheringiti. The characid genus Bryconamericus E1i- genmann (in Eigenmann, McAtee & Ward, 1907) comprises 30—40 species distributed through freshwater basins of South and Central America (Géry 1977). The current definition of the genus follows Eigenmann (1927), who included in Bryconamericus all characid fishes that possesses a single row of teeth on the dentary, two rows of teeth on the premaxilla with four teeth in the in- ner series, a low number of teeth along the anterior margin of the maxilla, a lack of scales on the caudal fin, a large third infra- orbital contacting the preopercle along its posterior and ventral margins, setiform gill- rakers, a complete laterosensory canal sys- tem on the body, and the absence of a glan- dular pouch on the caudal fin in males, as recently summarized by Vari & Siebert (1990:517). The naturalness of Eigen- mann’s genera have been long discussed by several authors and for Bryconamericus most recently by Fink (1976), Vari & Sie- bert (1990) and Malabarba & Malabarba (1994). We here describe a new species, Brycon- americus lambari, from the laguna dos Pa- tos drainage, Rio Grande do Sul, Brazil. We assign the new species to Bryconamericus because it has the combination of characters used by Eigenmann to define the genus. A phylogenetic study of the relationships of Bryconamericus species and a reconsidera- tion of its monophyly are beyond the scope of this paper. Type specimens of other southern South American species at present assigned to Bryconamericus were examined during this study. Comments about the identity of these species are provided. Methods Specimens examined belong to ANSP— Academy of Natural Sciences, Philadelphia; 680 BMNH—Natural History Museum, Lon- don; CAS—California Academy of Sci- ences, San Francisco; MAPA—Museu An- chieta, Porto Alegre; MCP—Museu de Ciéncias, now Museu de Ciéncias e Tec- nologia, Pontificia Universidade Catdlica do Rio Grande do Sul, Porto Alegre; UFRGS—Departamento de Zoologia, Universidade Federal do Rio Grande do Sul, Porto Alegre; and USNM—National Museum of Natural History, Smithsonian Institution, Washington. Counts and measurements follow Fink & Weitzman (1974), except for number of scale rows between lateral line and pelvic fin origin, counted from the scale row im- mediately ventral to lateral line to the scale row closest to the first pelvic-fin ray. Counts of vertebrae and procurrent rays of caudal fin were taken from radiographed and cleared and stained specimens. C&S in- dicates specimens cleared and stained for cartilage and bone according to the method of Taylor & Van Dyke (1985). Bryconamericus lambari, new species Fig. 1, Table 1 Bryconamericus sp.—Malabarba, 1989:131 (undescribed species from laguna dos Pa- tos drainage). Holotype.-—MCP. 15448 (55.9 mm SL, male), small creek flowing into arroio Fei- toria, under bridge of avenida Ipiranga, municipio de Dois Irmaos, Rio Grande do Sul, Brazil, 51°07'S, 29°36’W, 4 December 1991, Eunice A. Isaia and Andreas Kindel. Paratypes.—MCP 15449 (14 specimens, 41.4-56.3 mm SL), MCP 15450 (2 speci- mens C&S, 44.9-52.5 mm SL), USNM 324628 (3 specimens, 42.2—-52.9 mm SL), collected with holotype. UFRGS 4539, (1 specimen, 38.5 mm SL), UFRGS 4513 (3 specimens, 50.5—63.5 mm SL), same local- ity as the holotype, 4 September 1991, Eu- nice A. Isaia and Andreas Kindel. UFRGS 1784 (2 specimens C&S and 5 specimens colored with alizarin only, 27.3-46.3 mm SL), arroio Feitoria, municipio de Dois Ir- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON maos, Rio Grande do Sul, Brazil, Luiz R. Malabarba. Diagnosis.—Bryconamericus lambari, new species, is distinguished from all other species assigned to Bryconamericus by a combination of the following characters: the total number of anal-fin rays (19-22); number of perforated scales along the lat- eral line (38-40); its elongated body with greatest depth 24.6—29.8% of standard length; lack of dark pigmentation on the distal tips of the caudal fin; 3 or 4 maxillary teeth; and 5 or 6 scale rows between the lateral line and dorsal-fin origin. Brycon- americus lambari is further distinguished from B. iheringii (Boulenger 1887), the sec- ond species of the genus found in the la- guna dos Patos drainage (Malabarba 1989), by the shallower body depth and shorter pelvic fins (Tables. 1 & 2). Bryconamericus lambari is further distinguished from B. stramineus Eigenmann, 1908 described from the rio Uruguai drainage in its absence of a conspicuous and wide midlateral stripe (present in B. stramineus). Description.—Body elongate, fusiform, laterally compressed. Greatest body depth anterior to or at dorsal-fin origin. Dorsal body profile convex from tip of supraoccip- ital to dorsal-fin origin; nearly straight from base of last dorsal-fin ray to caudal pedun- cle. Ventral body profile convex from tip of lower jaw to anal-fin origin; nearly straight or slightly concave along anal-fin base. Dorsal and ventral profiles of caudal pe- duncle slightly concave. Snout deeply rounded from margin of upper lip to vertical line through anterior nostrils. Mouth slightly inferior. Maxilla short, reaching vertical line through anterior border of eye. Ventral profile of head gently convex. Dentary with 4 large teeth, each with 5 cusps, median cusp distinctly larger; fol- lowed by smaller teeth (3 or 4 teeth counted in cleared and stained specimens) either with 3—5 cusps or conical. Premaxilla with two tooth rows. Outer row with 3 or 4 (usu- ally 3) teeth, each tooth with 3—5 cusps. VOLUME 108, NUMBER 4 681 Table 1.—Morphometric data of Bryconamericus lambari, new species. Standard length in mm, measurements numbered 2—11 as percents of standard length and 12-14 as percents of head length. Data of the smallest paratype (UFRGS 1784—27.3 mm SL) and of UFRGS 4539 paratype not included. Character Holotype n Range X 1 Standard length Sa) 28 30.9-63.5 45.6 2 Predorsal distance 52.6 28 49.1-55.0 51.4 3 Prepelvic distance 45.4 28 43.8-48.4 45.5 4 Anal-fin base Ws) 28 21.0—26.0 Ma) 5 Caudal peduncle length e2 Jol) 16.1-18.3 17.1 6 Caudal peduncle depth 10.9 28 10.2-11.4 10.8 7 Body depth at dorsal-fin 27.6 28 24.6—29.8 26.9 8 Dorsal—fin length 20.8 24 20.0—24.2 21.8 9 Pelvic—fin length W227 28 12.1-15.1 13.4 10 Pectoral fin length ew 28 16.6—21.1 18.6 11 Head length D2) 28 21.2—26.3 Mp jct 12 Upper jaw length 36.2 28 29.4—-37.4 399) 13 Orbital diameter 37.8 28 36.0—-44.2 Shoal) 14 Interorbital width Bo 28 27.1-35.5 oie) Table 2.—Morphometric (1—14) and meristic (15—25) data of the type specimens of Tetragonopterus iheringii (A—lectotype—BMNH 1886.3.15.30, and 4 paralectotypes—BMNH 1886.3.15.31-34), Tetragonopterus pliodus (B) Astyanax eigenmanni (C) and Bryconamericus boops (D). Standard length in mm; measurements numbered 2-11 as percents of standard length and 12-14 as percents of head length. Lecto- A Character type range xX B C D 1 Standard length 64.7 48 .8-66.8 58.1 Se) 61.1 59.2 2 Predorsal distance 54.6 55.5—56.8 Se) 49.0 49.8 50.1 3 Prepelvic distance 47.0 46.8—-48.0 47.4 46.9 44.3 45.7 4 Anal-fin base 24.4 20.1—23.4 22.1 DOG, Dw) DEES 5 Caudal peduncle length 14.5 14.6—16.0 S)o33 14.5 16.7 16.3 6 Caudal peduncle depth 12.9 11.3-12.6 11.7 11.6 13.2 9.8 7 Body depth at dorsal-fin 39.1 32.7-38.1 35.6 34.0 372 DAES 8 Dorsal-fin length 25.0 22.7-25.8 24.2 236 DON 2353 9 Pelvic-fin length 16.5 15.4—-18.0 NN o72 16.5 16.5 16.6 10 Pectoral-fin length pbk 19.8—21.7 20.8 20.8 Py 22.6 11 Head length 24.1 23.1—25.4 23.8 Doral 2346 BSI) 12 Upper jaw length 32.4 33.8—34.7 34.2 35of/ 38} 293 13 Orbital diameter 3556 34.0-38.3 36.3 38.0 30.6 38.7 14 Interorbital width B3el 31.3-34.5 324i 3335 3h8\53) 33.8 15 Unbranched anal fin rays 3 3 3.0 s) 3) 3 16 Branched anal—fin rays 17 15-17 16.0 i19/ 15 19 17 Dorsal—fin rays 10 10 10.0 10 9 fal 18 Pelvic—fin rays 8 8 SiO 8/9 8 8 19 Pectoral—fin rays V2 12 12.0 13 12 13 20 Caudal—fin rays 19 19 19.0 19 19 NC 21 Perforated lateral line scales 87 37-38 37hc3) 38 38 38 22 Scale rows between lateral line and dorsal fin 6 5-6 5:3 5 5 5 23 Scale rows between lateral line and anal fin 3) 8) 3110) 3 3 4 24 Predorsal scales 12 11-13 12.0 12 12 i2 25 Scale rows around caudal peduncle 14 14 14.0 14 14 14 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Four teeth in the inner row, all with 5 cusps, except medial tooth with 4 cusps. Maxilla with 3 or 4 teeth with 3—5 cusps; teeth grad- ually becoming smaller posteriorly. All maxillary teeth with median cusp distinctly larger. Dorsal-fin rays 11, 8; tip not reaching ad- ipose fin when fin depressed. Posterior bor- der of dorsal fin straight, perpendicular to body margin when fin erect. Adipose fin present. Caudal fin forked, lobes equal in size, rounded; 8—11 procurrent rays dorsally and ventrally. Anal-fin rays i1i—iv, 14-19 (usually 15- 18). Distal margin of anal fin concave with the 6 anterior branched fin-rays longest and gradually decreasing in size posteriorly. Anal-fin origin along vertical line through base of last dorsal-fin ray. Anal fin of males bearing bony hooks from last unbranched ray to 11th branched ray in some individ- uals, usually reaching the 5th branched ray. Anal-fin rays with | pair of hooks per seg- ment, hooks present in posterolateral sur- face of ray and only caudal branch of each anal-fin branched ray that bears hooks. Hooks retrorse, curved towards ray base. Hooks typically present on up to 10 seg- ments of anterior anal-fin rays; number pro- Holotype of Bryconamericus lambari, new species (MCP 15448, male, 55.9 mm SL). gressively reduced posteriorly, with hooks on only one segment on last anal-fin ray that bears hooks. Pectoral-fin rays i, 11-12 G, 14 in one specimen). Pectoral fin rounded; tip of fin not reaching pelvic-fin origin. Pelvic-fin rays 1, 6—7 (i, 6 in only one specimen). Pel- vic-fin margin rounded distally, not reach- ing anal-fin origin. Pelvic-fin rays of males with ventromedial retrorse bony hooks. Hooks present on all branched rays but ab- sent on unbranched ray. Usually 1 pair of slender, long hooks per segment along me- dial branch and most of their lengths. Hooks also present on other branches of third to fifth rays. Scales cycloid. Caudal fin without scales. Single row of 4—7 scales on base of anterior 4—7 branched anal-fin rays. Lateral line complete with 38-40 perforated scales. Scale rows between dorsal-fin origin and lateral line 5—6; scale rows between lateral line and pelvic-fin origin 4—5. Predorsal scales 11-14, usually in regular series. Scale rows around caudal peduncle 14. Pre- caudal vertebrae 16—18; caudal vertebrae 20-22. Color in alcohol.—Ground color yellow- ish. Dark midlateral horizontal stripe indis- VOLUME 108, NUMBER 4 683 Figs. 2-4. 2, Lectotype of Tetragonopterus iheringii Boulenger, 1887 (BMNH 1886.3.15.30, male, 64.7 mm SL); 3, Holotype of Astyanax eigenmanni Evermann & Kendall, 1906 (USNM 55570, male, 61.1 mm SL); 4, Holotype of Bryconamericus boops Eigenmann, 1927 (MCZ 20700, male, 59.2 mm SL). tinct anteriorly but becoming well defined posterior to vertical line through dorsal-fin origin. Caudal peduncle without a well de- fined spot. Scattered chromatophores on posterior border of scales. Pigmentation most intense on dorsal portions of head and above midlateral stripe on body. Humeral spot roundish and well defined, centered on 684 fourth to sixth scales of scale row just dor- sal to lateral line. Sexual dimorphism.—Males of B. lam- bari are easily recognized by presence of bony hooks on anal and pelvic fins. Etymology.—The species name, lambari, refers to the common name of small char- acids in southern Brazil. Distribution.—Despite extensive collec- tions that have been made in several tribu- taries of laguna dos Patos drainage, B. lam- bari have been found only in the type basin: the arroio Feitoria and tributaries, rio Cai drainage, laguna dos Patos drainage, Rio Grande do Sul, Brazil. Additional collec- tions were also made in other tributaries of rio Cai to determine whether the species also ocurred elsewhere in the basin, but no additional specimens were collected. These other localities are inhabited by B. theringii, which occurrs in the same drainage, but was never found syntopic with B. lambari. We cannot explain this peculiar distribution. Ecology.—All specimens of B. lambari were collected in a small sandy and rock bot- tomed river in the Serra Geral formation of Rio Grande do Sul, Brazil. Although the river consists mainly of rapids, the specimens were caught in areas of low current flow. Discussion We examined and compared the type specimens of four species of Bryconameri- cus described from South Brazil, Uruguay and Argentina. Tetragonopterus iheringii Boulenger (1887:172—-173), with Tetragon- opterus pliodus Cope (1894:90-91) as a ju- nior synonym, were both described based on specimens originating in laguna dos Pa- tos tributaries, Rio Grande do Sul, Brazil. Astyanax eigenmanni Evermann & Kendall (1906:83) was described from the lower rio Parana, Argentina, and Bryconamericus boops Eigenmann (1927:371) was de- scribed from a small coastal drainage at Maldonado, Uruguay. All these species, now assigned to Bry- conamericus, are deep bodied fishes, and PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON clearly differ from B. lambari in body shape and greatest body depth (Figs. 1-4, Tables 1 & 2). However, we found no differences in examined features that clearly permit the recognition of multiple deep bodied Brycon- americus species in the region (Table 2). The differences found among the type specimens fall within the range of what we identify as B. theringii, when comparing different sam- ples of that species from several laguna dos Patos tributaries. | Eigenmann (1927:377) previously placed Tetragonopterus pliodus as a junior syn- onym of B. theringii, both species having been described from the same drainage. We follow Eigenmann in maintaining Tetra- gonopterus pliodus as a junior synonym of B. theringii. Eigenmann (1927:380) also pointed out the close similarity of B. eigen- manni and B. iheringii, but retained both as valid species. The holotype of B. eigen- manni presents no clear differences in counts and measurements to B. theringii (Table 2), but we also keep both as valid species. We suggest that a useful compari- son needs an in depth statistical study of good population samples from the areas of distribution of both nominal species and the only material we examined from lower rio Parana was the holotype of B. eigenmanni. The type specimen of B. boops (Fig. 4) is an abnormal specimen as was already noted by Géry (1977:390), who nonetheless retained it as a valid species. The measure- ments presented in Table 2 for B. boops are not comparable to the other species. We have found no specimens at the type local- ity of this species with such a profile pre- sented by the holotype, but rather found only specimens refereable to B. iheringii. We propose B. boops as a junior synonym of B. iheringii, hypothesizing that the ho- lotype of B. boops is a deformed specimen of B. iheringii. Comparative material: Bryconamericus iheringii: BMNH 1886.3.15.30 (Tetragonopterus Theringii VOLUME 108, NUMBER 4 Boulenger, 1887—Lectotype by present designation—64.7 mm SL, male), Sado Lourengo, Rio Grande do Sul state, Brazil, H. von fhering. BMNH 1886.3.15.31-34 (Tetragonopterus Iheringii Boulenger, 1887—paralectotypes, 11 ex., 48.8-66.8 mm SL, five males and six females), col- lected with the lectotype. ANSP 21578 (Te- tragonopterus pliodus Cope, 1894—holo- type, 55.9 mm SL, female?), Rio Grande do Sul, Brazil, H. H. Smith (type locality re- stricted to laguna dos Patos drainage by Malabarba, 1989:120—-121, 131). MCZ 20700 (Bryconamericus boops Eigenmann, 1927—holotype, 59.2 mm SL), Maldonado, Uruguai, T. G. Cary. Rio Grande do Sul, Brazil: MCP 10074 (57 ex.) Barragem Er- nestina, rio Jacui, 16 November 1983, C. A. S. Lucena & L. R. Malabarba. MCP 8430 (33 ex.), agude Garcia, km 56 of road BR 116, Barra do Ribeiro, 18 June 1985, C. A. S. Lucena & R. E. Reis. MCP 11446 (133 ex.), rio Camaqua e pogas laterais, Ca- maqua, 5 May 1987, C. A. S. Lucena, L. R. Malabarba & E. Pereira. MCP 11492 (19 ex.), arroio Chasqueiro, BR 116, between Pelotas and Jaguarao, Arroio Grande, 9 Jan- uary 1987, C. A. S. Lucena, A. Bergmann & P. Azevedo. UFRGS 2874 (©4 ex.), UFRGS 710 (2 ex.), Estagaéo Ecolégica do Taim, Rio Grande, 7-8 May 1981, R. E. Reis & J. R. Stehmann. MCP 11264 (176 ex.), arroio Jaguarao, Passo do Centuriao, Herval, 8—9 Jan 1987, R. E. Reis, PR Azev- edo & I. Costa. Maldonado, Uruguai: MAPA 2056 (16 ex.), arroio Salso, Ruta 9, km 122, 12 Jan 1982, P A. Buckup & R. E. Reis. Bryconamericus eigenmanni: USNM 55570 (Holotype of Astyanax eigenmanni Everman & Kendal, 1906—61.1 mm SL, male), rio Primero, Cordoba, Argentina, 1903-1904, J. W. Jiiteomb. Bryconamericus stramineus: CAS 40833 (holotype, 39.5 mm SL, female?), Piraci- caba, Sao Paulo, Brazil, R. von Ihering. Acknowledgments We are indebted to G. Howes, J. Cham- bers (BMNH), FE Meyer (MAPA) and K. 685 Hartel (MCZ) for the loan of specimens; R. Vari and M. Weitzman (USNM) for per- mission to examine type specimens under their care and on loan (B. stramineus and Tetragonopterus pliodus, respectively). The photograph of the holotype of B. boops was taken by S. Weitzman. This work was sup- ported by grants from Conselho Nacional de Desenvolvimento Cientifico e Tecnol6- gico (CNPq—proc. 414067/90-0) and by a scholarship (CNPq—proc. 806.041-88.9) to Andreas Kindel. This paper was improved with comments from C. Lucena, R. Reis, Z. M. Lucena, R. Vari, S. Weitzman and an anonynous reviewer. Literature Cited Boulenger, G. A. 1887. Description of new South American Characinoid fishes.—Annals and Magazine of Natural History 5(19):172-174. Cope, E. W. 1894. On the fishes obtained by the Nat- uralist Expedition in Rio Grande do Sul.—Pro- ceedings of the American Philosophical Society 33(144):84—-108. Eigenmann, C. H. 1908. Zoological results of the Thayer Brazilian Expedition—Preliminary de- scriptions of new genera and species of tetra- gonopterid characins.—Bulletin of the Museum of Comparative Zoology 52(6):91—106. . 1927. The American Characidae. IV.mMem- oirs of the Museum of Comparative Zoology 43:311-428. , W. L. McAtee, D. P. Ward. 1907. On further collections of fishes from Paraguay.—Annals of the Carnegie Museum 4(7):110—157. Evermann, B. W., & W. C. Kendall. 1906. Notes on a collection of fishes from Argentina, South America; with description of three new spe- cies.—Proceedings of the United States Nation- al Museum 31(1482):67—108. Fink, W. 1976. A new genus and species of characid fish from the Bayano River basin, Panama (Pi- sces: Cypriniformes).—Proceedings of the Bi- ological Society of Washington 88(3):331-444. Fink, W. L., & S. H. Weitzman. 1974. The so-called Cheirodontin fishes of Central America with de- scriptions of two new species (Pisces: Characi- dae).—Smithsonian Contributions to Zoology 172:1—-46. Géry, J. 1977. Characoids of the World. Neptune City, T.EH. Publication. 672 pp. Malabarba, L. R. 1989. Hist6rico sistematico e lista comentada das espécies de peixes de agua doce do sistema da laguna dos Patos, Rio Grande do 686 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Sul, Brasil—Comunicagoes do Museu de Cién- cedures for staining and clearing small fishes cias da PUCRS, Série Zoologia, 2:107—179. and other vertebrates for bone and cartilage Malabarba, M. C. S. L., & L. R. Malabarba. 1994. study.—Cybium 9:107-119. Hypobrycon maromba, anew genus and species’ Vari, R. P., & D. J. Siebert. 1990. A new, unusually of characiform fish from the upper rio Uruguai, sexually dimorphic species of Bryconamericus Brazil (Ostariophysi: Characidae).—Ichthyo- (Pisces: Ostariophysi: Characidae) from the pe- logical Exploration of Freshwaters 5:19—24. ruvian Amazon.—Proceedings of the Biological Taylor, W. R., & G. C. Van Dyke. 1985. Revised pro- Society of Washington 103:516—524. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(4):687—-694. 1995 Gobiodon acicularis, a new species of gobioid fish (Teleostei: Gobiidae) from Belau, Micronesia Antony S. Harold and Richard Winterbottom (ASH & RW) Department of Ichthyology and Herpetology, Royal Ontario Museum, 100 Queen’s Park Crescent, Toronto, Ontario MSS 2C6, Canada; (RW) Department of Zoology, University of Toronto, Toronto, Ontario M5S 1A1, Canada Abstract.—A new species of gobiid, Gobiodon acicularis, bearing an un- usual, highly elongate first dorsal spine, is described. Other distinguishing char- acters, in combination, include lack of post-symphysial canine teeth in the lower jaw, high number of papillae on the dorsum, overall uniform pigmen- tation of head, body and fins, lack of an interopercle-isthmus groove and a narrow gill opening. The material forming the basis of the description was collected in Belau (Palau) in the 1950’s. Available habitat data indicate that the species occurs in reef margin or lagoonal habitats. The Indo-Pacific genus Gobiodon Bleek- er, 1856, is a group of gobies that live as adults as obligate associates of corals of the genus Acropora (Patton 1994). The species, which reach up to about 40 mm SL, are characterized by a scaleless body covered by a thick mucus layer, a reduced head pa- pillae pattern, and generally small jaw teeth with the exception of one or two pairs of well-developed canine teeth posterior to the dentary symphysis. Gobiodon appears to share with another hermaphroditic goby ge- nus, Paragobiodon, an unusual and proba- bly derived gonad structure (Cole 1990). Pseudogobiodon Bleeker, 1874, is similar to Gobiodon but the post-symphysial ca- nine teeth are lacking in the only known species, P. macrochir Bleeker, 1875 (Gob- ius citrinus Riuippell, 1838, was erroneously ascribed to Pseudogobiodon, as the type species, by Bleeker 1874:309). Whether or not species lacking these canine teeth should be considered generically distinct from Gobiodon is a phylogenetic problem we hope to solve in the near future. Many of the described species are bright- ly colored, often with distinctive body bars, spots and/or fin striping (see, for example, Akihito 1984:265—266, plate 246, F —P). Gobiodon has attracted considerable atten- tion from systematists, resulting in more than 30 nominal species that can be attrib- uted to it. Of these described species we currently recognize 14 as valid: G. albofas- clatus Sawada & Arai, 1972; G. axillaris De Viz, 1884; G. ceramensis (Bleeker, 1852); G. citrinus (Riippell, 1838); G. ful- vus Herre, 1927; G. heterospilos Bleeker, 1856; G. histrio (Valenciennes, 1837); G. micropus Giinther, 1861; G. oculolineatus Wu, 1979; G. okinawae Sawada et al., 1972; G. quinquestrigatus (Valenciennes, 1837); G. reticulatus Playfair, 1867; G. ri- vulatus (Riippell, 1830) (mot of Suzuki et al. 1995); and G. unicolor (Castelnau, 1873). During our survey of collections of Go- biodon from throughout the Indo-Pacific re- gion we found the present undescribed spe- cies in the collection of the California Academy of Sciences. Since this species so clearly differs from all other congeners, we have decided to describe it now, making its name available for other, planned papers on the genus. Materials and Methods Determination of meristic character val- ues follows Hubbs & Lagler (1947), with 688 exceptions given below. Osteological char- acters determined from three paratypes (ROM 1603CS, 27.2 mm SL o4 and 31.3 mm SL @; CAS 81515, 34.0 mm SL 2) cleared and stained following the method outlined by Potthoff (1984). Vertebral and caudal-fin ray counts were taken from ra- diographs. The last branched ray of the dor- sal and anal fins is divided to its base and is counted as a single element. Caudal-fin ray counts are given as the number of seg- mented and unsegmented rays, as opposed to procurrent and principal rays (see Hoese & Gill 1993:419), and the number of un- branched and branched rays. First dorsal-fin pterygiophore formula follows Birdsong et al. (1988:175). Terminology for lateralis pores follows Lachner & Karnella (1980): AI, anterior interorbital; AO, anterior otic; IT, intertemporal; NA, nasal; PI, posterior interorbital; POP, preopercular; SO, superi- or otic. Meristic data are reported as the range with the value for the holotype un- derlined, followed, in parentheses, by the mean and number of specimens counted. Institutional abbreviations follow Leviton et al. (1985). Morphometric measurements were made to the nearest 0.1 mm using digital calipers interfaced by a Smartcable (Gage Connec- tions, Inc.) with the software DATAQ, ver- sion 1.02 (D. L. Schultz). Data were ana- lyzed for univariate statistics using SAS (SAS Institute Inc.) for personal computers, version 6. Standard length (SL) was mea- sured from the median point of the premax- illary groove (anteriormost point of snout) to midlateral base of caudal fin. Head length (HL) was measured from the snout tip to posteriormost position on opercular membrane. The following measurements are reported as percentages of either HL (measurements of the head) or SL (all oth- ers): snout to origin of first dorsal fin; origin of first dorsal fin to origin of second dorsal fin; origin of second dorsal fin to origin of anal fin; anterior base of pelvic spine (pel- vic-fin origin) to origin of anal fin; snout to pelvic-fin origin; origin of first dorsal fin to PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON pelvic-fin origin; origin of first dorsal fin to origin of anal fin; anterior base of pelvic spine to origin of second dorsal fin; pelvic- fin length, pelvic-fin origin to tip of longest ray; anal-fin length, anterior base of third branched ray to its tip; first dorsal-fin first spine length, anterior base of first spine of first dorsal fin to its tip; first dorsal-fin sixth spine length, anterior base of sixth spine of first dorsal fin to its tip; second dorsal-fin length, anterior base of first branched ray of second dorsal fin to its tip; pectoral-fin length, base of longest ray of pectoral fin to its tip; caudal peduncle length, anterior base of central caudal rays to posterior base of last ray (insertion) of anal fin; caudal- peduncle depth, insertion of last ray of sec- ond dorsal fin to insertion of last ray of anal fin; orbit diameter, the maximum diameter in horizontal plane; snout length, minimum distance from anterior margin of orbit to tip of snout; interorbit (bony), minimum dis- tance between orbits. Gobiodon acicularis, new species Figs. 1, 2, Table 1 Holotype.-—CAS 81525 (31.4 mm SL), western Pacific Ocean, Belau (Palau), off Babelthuap Island, coral-enclosed area north of Arakataoch Stream, collected by H. A. Fehlmann et al., 22 Sep 1957. Paratypes.—CAS 81515 (5, 30.1—34.3 mm SL), western Pacific Ocean, Belau (Pa- lau), north shore of Korer Island, reef bor- dering eel-grass flat east of Ebadel’s Pier (T-dock), collected by R. R. Harry et al., 4 Aug 1955; CAS 81522 (6, 31.0—39.1 mm SL), western Pacific Ocean, Belau (Palau), Auluptagel Island in. Ngarahelngael Pass, ca. 45 meters north of Ngarahelngael, col- lected by H. A. Fehlmann et al., 9 Oct 1957; CAS 82377 (5, 21.5—30.4 mm SL), ROM 1603CS (2, 27.2—31.3 mm SL, CS), collected with holotype; ROM 69038 (2, 33.1—36.6 mm SL), collected with CAS 81522. Diagnosis.—A species of Gobiodon that is distinguished by a derived, highly elon- VOLUME 108, NUMBER 4 689 Fig. 1. gate (18.7-50.7 % SL) spine I of the first dorsal fin and, in combination, by lack of post-symphysial canine teeth in the lower jaw, spines of first dorsal fin decreasing in length progressively from anterior to pos- terior, subequal upper jaw teeth (i.e., outer teeth not prominently enlarged), an absence of stripes or other markings, 16 or 17 pec- toral-fin rays, more than four papillae in the anterior predorsal group and lack of an in- teropercle-isthmus groove. Description.—Dorsal-fin rays VI + I, 10-11 (X = 10.2, n = 16), first spine fila- mentous and highly elongate (Fig. 1), its relative length variable but not varying by sex or body size, in some specimens length of the spine reaching one half of standard length (Table 1); spines of first dorsal fin decreasing sequentially in length posterior- ly; a shallow notch present between dorsal fins; first dorsal-fin pterygiophore formula 3-22110; second dorsal-fin posterior rays elongate, in some specimens reaching pos- teriorly as far as bases of dorsal segmented caudal-fin rays; anal-fin rays I, 9 in all spec- imens, posterior rays elongate, reaching bases of segmented caudal-fin rays as with second dorsal fin; pectoral-fin rays 16—17 (X = 16.1, n = 20), fin extending posteri- Gobiodon acicularis, lateral view, left side (holotype, CAS 81525, 31.4 mm SL). orly as far as second or third branched anal- fin ray in some specimens; pelvic-fin rays I,5 in all specimens; pelvic-fin fraenum and basal membrane complete, fin relatively large, usually reaching as far as anus and occasionally base of genital papilla; seg- mented caudal-fin rays 9 + 8, dorsal unseg- mented rays 5-6 (X = 5.6, n = 11), ventral unsegmented rays 5—6 (X = 5.7, n = 11); dorsal branched caudal-fin rays 8-9 (X = 8.3, n = 11); ventral branched caudal-fin rays 7-8 (X = 7.2, n = 11); scales absent; first gill slit open; gill opening restricted, ending ventrally opposite bases of pectoral- fin ray 11-14-16 (X = 13.5, n = 19); gill rakers, short, basally ossified 1 + 2, 1 + 4, 2 + 5 (n = 3); mouth small, terminal, gape extending posteriorly to below anterior one- half of eye; upper jaw teeth subequal, small, conical to slightly recurved, arranged in 2 to 3 irregular rows; lower jaw teeth similar to those of upper jaw, post-symphysial ca- nine teeth absent; anterior nasal opening at tip of long fleshy tube, posterior nasal open- ing with raised rim or short tube; anterior oculoscapular canal pores 6 (NA, AJ, PI, SO, AO, IT) (Fig. 2); preopercular canal pores 3; head papillae in a reduced trans- verse pattern (Fig. 2), with suborbital row 690 Table 1.—Morphometric characters of Gobiodon acicularis, based on the holotype, CAS 81525, and 19 paratypes, CAS 81515, CAS 81522, CAS 82377, ROM 69038 and ROM 1603CS, all from Belau. Stan- dard length in mm; interorbital width, horizontal eye diameter, snout length and upper jaw length as per- centages of head length; all other morphometric char- acters as percentages of standard length. Holotype Range xX Standard length Sie) Aloo SIO Snout to first dorsal—fin origin 34.4 33.1-39.2 35.7 First dorsal—fin origin to second dorsal—fin ori- gin Wo) DNO WO. 234 Second dorsal—fin origin to anal—fin origin 34.4 33.5-40.7 36.6 Pelvic—fin origin to anal— fin origin Pel] PSB) SX0)K0) Snout to pelvic—fin origin 33.4 33.0-40.1 35.4 First dorsal—fin origin to pelvic—fin origin 3 SoA sik BSS First dorsal—fin origin to anal—fin origin 43.6 43.6-51.7 46.7 Pelvic—fin origin to sec- ond dorsal—fin origin 42.0 40.1-49.1 44.4 Pelvic—fin length GS) S788) Its) 7/ Anal-—fin length 19.7 18.4-28.4 22.2 First dorsal—fin first spine length 36.6 18.6-50.7 38.3 First dorsal—fin sixth spine length 13.6 8.0-16.3 12.5 Second dorsal—fin length 19.4 15.2-24.5 18.8 Pectoral—fin length 31.8 29.4-38.6 34.4 Caudal—peduncle length 24.8 22.0-25.3 23.9 Caudal—peduncle depth 18.8 17.6-20.8 19.3 Head length 26.8 26.0-30.3 28.3 Interorbital width DDO la = 2326s Sey Horizontal eye diameter 10.2 7.9-11.4 9.6 Snout length 2924 233-30:3), 270 Upper jaw length 3059" 27-1345); 30:2 6-7-8 (X = 7.6, n = 14), dorsal preoper- cular row 6-7-9 (X = 7.1, n = 14) and ventral preopercular row 4-7-8 (X = 6.9, n = 14); two clusters of papillae on predorsal surface (Fig. 2), anterior cluster 1 + 1—2- 3, posterior cluster 1-2 + 1 with anterior papilla in cluster located anteriorly or an- terolaterally to posterior papilla; tongue rounded; no groove present between inter- opercle and isthmus; branchiostegal rays 5; vertebrae 10 + 15 + ural centrum = 26; PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON epurals 1; morphometric values given in Ta- ble 1. Color in alcohol (all material preserved in 55 % isopropyl alcohol).—Head and body covered nearly uniformly with small, pale brown chromatophores, except on pos- terior margin of gill membrane where they are slightly larger and darker; all fins with many scattered, small chromatophores sim- ilar to those of body; pectoral fin and distal margin of first dorsal fin distinctly darker than other fins and body; no stripes or other markings present. Etymology.—The name is based on the Latin adjective acicularis, meaning like a needle, in reference to the elongate spine I of the first dorsal fin. Remarks.—Gobiodon acicularis is dis- tinguished from all other described con- geners by its highly elongate first spine of the first dorsal fin. In other respects the new species resembles G. ceramensis (Bleeker, 1852) and the adults, according to a manuscript key (D. E Hoese, pers. comm.), of G. albofasciatus Sawada & Arai, 1972 with which it shares relatively uniform body pigmentation without mark- ings, lack of a groove between the intero- percle and isthmus, and the outer row of the upper jaw teeth not prominently en- larged. Both of these species share with G. acicularis the possibly derived first dorsal fin shape with the lengths of the spines de- creasing progressively from anterior to posterior. Unlike G. acicularis, G. albofas- ciatus has a relatively broad gill opening, occupying nearly the entire base of the pectoral fin. The gill opening of G. aci- cularis is quite narrow, terminating ven- trally at the level of the base of the 14th to 16th pectoral-fin ray. Gobiodon citrinus is also similar to the above species in the shape of the first dorsal fin, but has prom- inent transverse stripes on the head and body posterior to the pectoral-fin base and a dusky spot at the dorsal margin of the gill opening. This species is further distin- guished by its relatively small pelvic fin and narrower gill opening. VOLUME 108, NUMBER 4 Pl Al NA 691 Fig. 2. Head of Gobiodon acicularis (paratype, CAS 82377, 26.3 mm SL) showing arrangement of papillae and lateralis pores in left lateral (above) and dorsal (below) views. Abbreviations for lateralis pores, following Lachner and Karnella (1980): AI, anterior interorbital; AO, anterior otic; IT, intertemporal; NA, nasal; PI, pos- terior interorbital; POP, preopercular; SO, superior otic. Scale bar = 3 mm. The morphology of Pseudogobiodon ma- crochir Bleeker, 1875 also bears some, pos- sibly important, similarities to G. acicular- is. Bleeker’s erection of Pseudogobiodon for P. macrochir distinct from Gobiodon, was based on its slightly enlarged outer teeth of the upper jaw and, more impor- tantly, the absence of post-symphysial ca- nine teeth in the lower jaw. Neither of these characters are likely to be of use in diag- nosing Pseudogobiodon based on observed variation within and between Gobiodon 692 species and the very small sample size upon which Bleeker’s description was based (we only know of the holotype). Pseudogobio- don is almost never used in regional faunal works on Indo-Pacific fishes (see Koumans 1953:13, for example). Our examination of the holotype of P. macrochir (RMNH 4463, 26.7 mm SL) re- veals that the upper jaw teeth are subequal, as they are in G. ceramensis and G. citri- nus, for example, a condition which is prob- ably derived within Gobiodon. We confirm that the post-symphysial dentary canine teeth are absent in the holotype of P. ma- crochir, as described by Bleeker (1875). Post-symphysial canine teeth are absent in G. acicularis, variously present or absent in adults of another, probably undescribed, Gobiodon species and present, but reduced in size, in adults of G. okinawae (Sawada et al. 1972:59, fig. 2). Reduction of these canine teeth is probably derived and may support the monophyly of a clade. Pending results of our phylogenetic analysis, if such a clade, or P. macrochir alone, were found to be the sister group of a clade comprising all other Gobiodon species then the former could be referred to Pseudogobiodon. If, however, P. macrochir is found to be non- basal in the Gobiodon clade then Pseudo- gobiodon should be synonymized with Go- biodon to preserve the monophyly of the latter genus. Harrison (1989:348) indicated that some features of the skull of a species of Gobio- don (not identified to species) he examined were unique among gobioids and therefore likely derived. These characters are: (1) an enlarged supraoccipital crest with addition- al, paired crests anteriorly formed by dorsal extensions of the frontals along their medial surfaces, and (2) a deepened sphenotic and pterotic. These features are common to all species of Gobiodon we have examined, in- cluding G. acicularis, and to Pseudogobio- don macrochir. Given these shared, derived characters, Gobiodon can be provisionally diagnosed, whereas Pseudogobiodon can- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON not, and the former is therefore the appro- priate genus for the new species. Specific characters shared by P. macro- chir and G. acicularis include a uniformly pigmented body with some fins, especially the first dorsal, noticeably darker than the body, lengths of the spines of the first dor- sal fin decreasing progressively from ante- rior to posterior, and modally 16 pectoral- fin rays. Bleeker (1875:117) does not men- tion a horizontal stripe at the base of the second dorsal fin in P. macrochir, nor 1s one shown in the recently published illus- tration (Bleeker 1983: pi. 431, fig. 4). How- ever, we have found there to be a lightly pigmented stripe along the base of each dorsal fin in the holotype of P. macrochir, indicating an important difference in pig- mentation from G. acicularis. Gobiodon acicularis is known only from the collections reported here from Belau. That the species is recognized now for the first time in spite of there being many col- lections of Gobiodon from reefs throughout the Indo-Pacific, including Belau (Randall, pers. comm.), is possibly explained by a re- stricted geographic distribution and/or what we know of its ecology so far. Collections from Belau were all made in back-reef ar- eas where there may have been less col- lecting effort than at other locations where Gobiodon species occur. We have no infor- mation on host corals that might have been present, but given the general water condi- tions it is likely that G. acicularis occupies an unusual, probably derived, habitat with respect to congeners (see, for example, Pat- ton 1994), which, as adults, are commensal primarily on Acropora species. New collec- tions with detailed habitat data would be invaluable, if not essential, in our attempt to trace the evolution of host-commensal relationships in the genus. Acknowledgments We are grateful to W. Eschmeyer, D. Ca- tania and J. Fong, CAS, who made the ma- terial available for this study, and to E. O. VOLUME 108, NUMBER 4 Murdy and J. T. Williams for their helpful comments on the manuscript. Thanks are also due M. Rouse and C. Gilmore, ROM, for their technical assistance with radiog- raphy and B. Boyle for photography. Fi- nancial support for this study was provided through Natural Sciences and Engineering Research council of Canada operating grant A 7619 to RW. Literature Cited Akihito, Prince. 1984. Suborder Gobioidei. Pp. 236— 289 in H. Masuda, K. Amaoka, C. Araga, T. Uyeno, & T. Yoshino, eds., The fishes of the Japanese Archipelago. Tokai University Press, Tokyo. Birdsong, R. S., E. O. Murdy, & E L. Pezold. 1988. A study of the vertebral column and median fin osteology in gobioid fishes with comments on gobioid relationships.—Bulletin of Marine Sci- ence 42:174—214. Bleeker, P. 1852. Nieuwe bijdrage tot de kennis der ichthijologische fauna van Ceram.—Natuurkun- dig Tijdschrift voor Nederlandsch Indié 3:689— 714. . 1856. Bijdrage tot de kennis der ichthyolo- gische fauna van het eiland Boeroe.—Natu- urkundig Tijdschrift voor Nederlandsch Indié 11:383-414. . 1874. Esquisse d’un systéme naturel des Go- bioides.—Archives Néerlandaises des Sciences exactes et naturelles 9:289-331. . 1875. Gobioideorum species insulindicae no- vae.—Archives Néerlandaises des Sciences ex- actes et naturelles 10:113—134. . 1983. Atlas Ichthyologique des Indes Orien- tales Néérlandaises. Vol. XI-XIV. Smithsonian Institution Press, Washington, D.C., pp. 1-22, pls. 421-447. Castelnau, EF L. 1873. Contribution to the ichthyology of Australia. Number V. Notes on fishes from northern Australia.—Proceedings of the Zoo- logical and Acclimatisation Society of Victoria 2:83-97. Cole, K. S. 1990. Patterns of gonad structure in her- maphroditic gobies (Teleostei: Gobiidae).—En- vironmental Biology of Fishes 28:125—142. De Viz, C. W. 1884. Fishes from South Sea islands.— Proceedings of the Linnaean Society of New South Wales 3:445—457. Ginther, A. 1861. Catalogue of the actinopterygian fishes in the collection of the British Museum. Volume 3. Gobiidae through Notacanthi. Taylor and Francis, London, 586 pp. Herre, A. W. 1927. Gobies of the Philippines and the 693 China Sea. Bureau of Science, Manila, 352 pp. Hoese, D. EF & A. C. Gill. 1993. Phylogenetic rela- tionships of eleotridid fishes (Perciformes: Go- bioidei).—Bulletin of Marine Science 52:415— 440. Hubbs, C. L., & K. E Lagler. 1947. Fishes of the Great Lakes Region. University of Michigan Press, Ann Arbor, Michigan, 186 pp. Koumans, E P. 1953. Gobioidea. In M. Weber & L. E De Beaufort, Fishes of the Indo-Australian Archipelago. E. J. Brill, Leiden 10:1—423. Lachner, E. A., & S. J. Karnella. 1980. Fishes of the Indo-Pacific genus Eviota with descriptions of eight new species (Teleostei: Gobiidae).— Smithsonian Contributions to Zoology 315:1- Wile Leviton, A. E., R. H. Gibbs, Jr, H. Heal, & C. E. Dawson. 1985. Standards in ichthyology and herpetology: Part I. Standard symbolic codes for institutional resource collections in herpe- tology and ichthyology.—Copeia 1985:802- 832. Patton, W. 1994. Distribution and ecology of animals associated with branching corals (Acropora spp.) from the Great Barrier Reef, Australia.— Bulletin of Marine Science 55:193-211. Playfair, R. L. 1867. Acanthopterygii. Pp. 1-80 in R. L. Playfair & A. Giinther, The fishes of Zanzi- bar. London, 154 pp., 21 pls. Potthoff, T. 1984. Clearing and staining techniques. Pp. 35-37 in H. G. Moser, W. J. Richards, D. M. Cohen, M. P. Fahay, A. W. Kendall, Jr, & S. L. Richardson, eds., Ontogeny and system- atics of fishes. American Society of Ichthyolo- gists and Herpetologists, Special Publication No. 1. Rtippell, E. 1830. Atlas zu der Reise im No6rdlichen Afrika. Fische des Rothen Meeres. Frankfurt- am-Main, part 3, pp. 95-141, pls. 25-35. 1838. Neue Wirbelthiere zu der Fauna von Abyssinien gehorig. Fisches des Rothen Meer- es. Frankfurt-am-Main, pp. 81-148, pls. 22-33. SAS Institute Inc. 1985. SAS procedures guide for personal computers, version 6 edition. SAS In- stitute Inc., Cary, North Carolina. Sawada, Y., & R. Arai. 1972. Gobiodon albofasciatus, a new coral-goby from the Ryukyu Islands, Ja- pan.—Bulletin of the National Science Muse- um, Tokyo 15:415—420. s , & T. Abe. 1972. Gobiodon okina- wae, a new coral-goby from the Ryukyu Is- lands, Japan.—Japanese Journal of Ichthyology 19:57-62. Schultz, D. L. Documentation for DataQ and DataQb. Unpublished software manual, available from the author, Department of Biology, Nicholls State University, Thibodaux, Louisiana, 20 pp. 694 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Suzuki, T., M. Aizawa, & H. Senou. 1995. A prelim- Cuvier & M. A. Valenciennes, Histoire naturelle inary review of the Gobiodon rivulatus complex des poissons. E G. Levrault, Paris. from Japan.—I.O.P. Diving News 6(7):2-7. Wu, H.-L. 1979. Description of two new species of Valenciennes, M. A. 1837. Tome Douziéme. Suite du Gobiodon Bleeker from China.—Oceanologia livre quatorziéme. Gobioides. Pp. 1—261 in G. et Limnologia Sinica 10:157—160. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(4):695-—716. 1995 South American rocky habitat Leptodactylus (Amphibia: Anura: Leptodactylidae) with description of two new species W. Ronald Heyer Department of Vertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, U.S.A. Abstract.—There are four groups of populations of Leptodactylus associated with rocky habitats in northern South America. The available data are most consistent with recognizing three species for these four units: L. rugosus and two new species, L. lithonaetes and L. myersi. A single specimen from Para- guay, although distinctive, is considered to be conspecific with L. syphax, pre- viously known from disjunct localities in eastern Brazil. Data are inadequate at present to determine whether the South American species of Leptodactylus associated with rocky habitats are a monophyletic group. Charles W. Myers brought my attention to a distinctive new species of the frog ge- nus Leptodactylus that he had collected from granitic outcrops in the State of Ro- raima, Brazil. As study of the material pro- gressed, comparisons were made between the new species from northern Brazil with Leptodactylus rugosus, a species from gra- nitic and sandstone habitats of the Guiana shield region. It became apparent that there was considerably more variation among populations of the Guiana shield frogs than recognized previously (Heyer 1979). From the other end of South America, a single specimen of Leptodactylus was collected several years ago from rocky outcrops in Paraguay; this specimen’s affinities are problematic. The purpose of this paper is to re-evaluate the species status of the Lepto- dactylus associated with granitic and sand- stone habitats in South America. Methods and Materials As aspects of variation in Leptodactylus syphax have recently been addressed (Car- doso & Heyer 1995), members of that spe- cies are not treated in detail here. As many specimens as possible were borrowed of all other granitic and sandstone habitat Lepto- dactylus. Data were taken on patterns of the dorsum, upper lip, posterior thigh, and up- per shank using the standards described in Heyer (1979). In addition, belly and ventral thigh surface patterns were recorded. Infor- mation was noted on dorsal folds, texture of the dorsum, upper shank, outer tarsus, and foot, as well as male secondary sexual characteristics. The snout—vent length (SVL) was recorded for all specimens. For all adults and specimens near adult size, the following measurements were also recorded (following Heyer et al. 1990): head length (HL), head width (HW), eye—mid-nostril distance (E—N), tympanum diameter (TD), thigh, shank, and foot. Statistics were ana- lyzed with SYSTAT for Windows, version 5 (1992). Museum abbreviations are those recommended by Leviton et al. (1985) with the addition of IND-AN = INDERENA, Ministerio de Agricultura, Bogota, Colom- bia. Variation in Northern South American Rocky Habitat Leptodactylus As data were being collected on speci- mens from northern South America, it be- 696 -& 7 7 4 7 1 ) vi ™, VENEZUELA ' Uae So uuns \ @ S ----- (4 A) wv COLOMBIA ve \) e ¢ x \ ~~ 2 RE \X N q Meallrye \a-y e., ae Poot ESO i 4 \ . r v \ ’ > ! gk eA i Ro iil SS oo Ke / I ‘ N 0. g ? if ews seco Lore > 80° 70° Fig. 1. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON oO ue ca Oo Oo o = to 2 y) Oo 8 i Sf 5° a 0 Ree 4 s re) fexe) , FG (e) Sear ee I] ¥, 9 S / ( GC: pea i) aA ' Sa Tos 7 Qy’ ry = ai b) “—N_N sy 0 \ t A BRAZIL 5° 60° 50° Localities for Group 1-4 specimens from northern South America. Group I = dots (L. lithonaetes), Group II = squares (L. rugosus), Group III = triangles (L. myersi), Group TV = circles (L. myersi). came apparent that there were four groups involved based on character states and geo- graphic distributions. Each group is dis- cussed in some detail and then decisions are drawn with respect to species limits. There are few tadpole samples; those that are available do not all contain Gosner (1960) Stage 25—32 specimens. Tadpole character- istics described in this section are based on Gosner stage 33-42 specimens. Because re- cordings of advertisement calls exist for only Group 2 individuals, those data are not discussed. Group 1.—Specimens in this group are from Colombia and Venezuela near the Co- lombian border (Fig. 1). Seventeen adult fe- males, 31 adult males, 176 total specimens. Dorsal patterns characteristically have a series of 3-4 (rarely 2) pairs of spots rang- ing from small to large in size and ranging from discrete to patterns of fusion with oth- er spots both across, as well as, lengthwise along the dorsum (Heyer 1979, fig. 1, pat- terns H through K). About 10% of the in- dividuals have a uniform dorsum. Upper lip patterns show a continuum among the following states. Thirteen per- cent of the specimens have little pattern on the upper lip (Heyer 1979, fig. 2, patterns C, E); 32% have some expression of alter- nating light and dark vertical bars (Heyer 1979, fig. 2, pattern J); 29% have some form of alternating light and dark oblique bars (Heyer 1979, fig. 2, pattern N); and 26% have some sort of irregularly defined light area in the loreal region to under the eye (Heyer 1979, fig. 2, pattern M). The posterior thigh surface pattern also shows a complete continuum among the following conditions. The posterior thigh surfaces are indistinctly mottled in 39% of the individuals (Heyer 1979, fig. 3, pattern P); distinctly mottled with small light irreg- ular marks in 17% of the individuals (Heyer 1979, fig. 3, patterns B, C, D); distinctly mottled with large light irregular spots and marks in 38% of the individuals (Heyer 1979, fig. 3, pattern A); 5% of the individ- uals have distinctly mottled thigh surfaces with some expression of light vertical marks or bars on the upper portions of the thigh surfaces (Heyer 1979, fig. 3, pattern I); one individual has a large light area on the lower thigh surface containing a few distinct dark spots; one individual has a large light area on the upper thigh surface. VOLUME 108, NUMBER 4 One juvenile has a distinct lengthwise light band in the middle of the ventral thigh surface, 8% of the juveniles have a notice- able light band, whereas no adults have any indication of such a band; 76% of juveniles and 44% of adults have very light ventral thigh surfaces with few or no melanophores (Fig. 2); 9% of juveniles and 33% of adults have almost uniformly dark ventral thigh surfaces; 3% of juveniles and 12% of adults have lightly to moderately mottled ventral thigh surface patterns; 3% of juveniles and 10% of adults have the upper-lateral sector boldly mottled and the rest of the ventral thigh surface uniformly light. Sixteen percent of the juveniles and 45% of the adults have relatively uniform gray/ brown bellies, although in some, the pos- terior belly is lighter than the anterior; 50% of the juveniles and 2% of the adults have uniformly light bellies with very few me- lanophores (Fig. 2); 9% of the juveniles and 22% of the adults have dark bellies with moderate to low contrast light spots and/or vermiculations; 1% of the juveniles and 16% of the adults have a mottled pattern of smaller irregular lighter areas on a darker ground; and 24% of the juveniles and 14% of the adults have a bold mottled pattern of large light spots/flecks on a darker ground color. Dr. Charles W. Myers (pers. comm.) pro- vided life color information based on spec- imens AMNH 100656—100667 from the southwest sector of Cerro Yapacana, Ama- zonas, Venezuela: “Small white markings on lip. Rear of thigh usually suffused with orange (dotted pale tan on black in one Specimen). Ventral surfaces grayish white. Juvenile (small specimen, 14875) [= AMNH 100666, 20.7 mm SVL] has pure white venter and bright orange under thighs, and a strong suffusion of orange on rear of thigh. Iris overall pale bronze, or pale bronze above and pale gray below— with overall dense black venation.’ Dr. John D. Lynch (pers. comm.) provided life color information based on specimens from Cueva Arévalo, Vichada, Colombia, es- 697 pecially ICNMNH 13972-13974: ‘“‘Dorsum brown with slightly darker brown and black spots on upper flanks. Patches on head and center of back rust. Pale brown (almost cream) interorbital bar and some marks on back. Face cream with black canthal stripe. Tympanum reddish-brown. Limb bands black. Yellow warts on flanks (glands). Venter and throat cream with brown spots and reticulation. Undersides of thighs pink. Posterior surfaces of thighs marbled black with faint rose spots. Iris bright copper above, gray below, flecked with black and bearing black horizontal streak. Other in- dividuals may have posterior thighs black with small yellow spots above, becoming more rose below. Dorsum varies from tan to nearly black. In males the lateral % of the gula is black (center white). Face gen- erally pale (some dark individuals do not have the pale labial patch). Venter scarcely to heavily spotted with brown. Some dorsal patches have olive cast. Variation based on series of 15 individuals. ... ”’ The degree of juvenile and adult pattern differences from individuals from the same localities, although noticeable in preserved specimens, is not striking. The most notice- able features are that the bellies and ventral thigh surfaces are lighter than in the adults, but these differences are more of degree than fundamentally different. The commonest conditions for dorsolat- eral fold development are either no indica- tion of dorsolateral folds or one short pair of ridges or elongate warts in the shoulder region. Some specimens have series of warts or ridges in the dorsolateral fold field, some extending the entire length of the dor- sum from behind the eye, others extending only to the sacrum. Most individuals have a shagreen togeth- er with black and/or white tubercles on the dorsum. When tubercles are present, they are more abundant on the posterior dorsum. Most juveniles (80%) either have a mod- erate to pronounced shagreen and/or small bumpy glands without any black and/or 698 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2: white tubercles; 31% of adults lack tuber- cles. The upper shank surface may have a sha- green or not. Most specimens (93%) have few to many black and/or white tubercles; few specimens (7%) have very few or no tubercles. Characteristic ventral patterns of juveniles. Upper left, Group I, ICNMNH 13980; upper right, Group II, USNM 291249; lower left, Group II], AMNH 128023; lower right, Group IV, RMNH 23921. The texture of the outer tarsus may be shagreened or not. Most specimens (81%) have few to many black and/or white tu- bercles; some specimens (19%) have very few or no tubercles. Texture of the foot is usually smooth, lacking shagreen or tubercles (86%); some VOLUME 108, NUMBER 4 individuals (12%) have no more than a few black and/or white tubercles; very few in- dividuals (2%) have a weakly developed shagreen. Adult males have one black thumb spine on each hand and a pair of chest spines. There is also a band of black tubercles across the chest in larger males. Males have a patch of black chin tubercles on the an- terior portion of the throat. The vocal sacs are laterally expanded and darkly pigment- ed. Adult females range between 54.8 and 78.4 mm SVL, adult males 45.3 and 71.4 mm SVL. A principal components analysis was run on the measurement data for adult specimens. No obvious outliers are evident on the resultant plot of individuals using the first two factors. Two series of tadpoles are available, one well-preserved series of specimens from Cerro Patava, Colombia, and two alcohol preserved dehydrated specimens from Can- aripo, Venezuela. It is difficult to determine to what degree the differences observed be- tween larvae from these two localities are due to preservation artifact. Tadpoles from both localities demonstrate the semiterres- trial ecomorph as described by Altig & Johnson (1989). The body length is 28— 31% of the total length in the Cerro Patava tadpoles, 31—32% in the Canaripo tadpoles. The anterior oral gap is 80—88% of the oral disk width in the Cerro Patava specimens, 64-77% in the Canaripo specimens. The upper beak is highly arched; its depth is 27— 37% of the upper beak width in the Cerro Patava sample, 36% in the Canaripo sam- ple. The body is flattened, with body depth 67—74% of body width in the Cerro Patava larvae, 46-56% in the Canaripo larvae. Lar- vae have series of glandular ridges on the body above the abdominal cavity. Habitat notes are available from AMNH 100656—100667 collected at Cerro Yapa- cana, Amazonas, Venezuela, 18—19 Febru- ary 1978 (C. W. Myers, pers. comm.): ‘“Mountain stream by night, sitting on the rock stream bed at or near edge of water. 699 ... They are timid and quick to take cover in crevices and under large boulders if one’s light is not kept on them while approaching; they seem to avoid diving into the water.” Group 2.—This group includes speci- mens from Guyana and southeastern Ven- ezuela (Fig. 1). Seventeen adult females, 35 adult males, 126 total specimens. Most specimens have a dorsal pattern of a series of 3-4 (rarely 2) pairs of large to small spots ranging from discretely defined to patterns of fusion with other spots both across the dorsum as well as lengthwise along the dorsum (Heyer 1979, fig. 1, pat- terns H through K). Twenty-seven percent of the individuals have uniform or almost uniform dorsal patterns. Upper lip patterns represent a continuum among the following states. Eleven percent of the specimens have little pattern on the upper lip (Heyer 1979, fig. 2, patterns C, E); 8% have some expression of alternating light and dark vertical bars (Heyer 1979, fig. 2, pattern J); 43% have some form of alternating light and dark oblique bars (Heyer 1979, fig. 2, pattern N); 26% have some sort of irregularly defined light area in the loreal region to under the eye (Heyer 1979, fig. 2, pattern M); 7% have an irreg- ularly defined light area in the loreal region, separated from light oblique bars behind the eye by irregular dark bars (Heyer 1979, fig. 2, pattern K); and 3% have extensive dark mottling on the upper lip, heaviest near the mouth. Posterior thigh surface patterns are quite variable, with a continuum including the following states. The posterior thigh sur- faces are indistinctly mottled in 8% of the individuals (Heyer 1979, fig. 3, pattern P); distinctly mottled with small light irregular marks in 4% of the individuals (Heyer 1979, fig. 3, patterns C, D); distinctly mot- tled with large light irregular spots and marks in 40% of the individuals (Heyer 1979, fig. 3, pattern A); the upper thigh sur- faces have some sort of light vertical marks or bars in 15% of the individuals (Heyer 1979, fig. 3, patterns E, F); the lower thigh 700 surface has one or more extensive light area, Sometimes with a few small dark spots in 22% of the specimens; and the upper thigh surface has a large light area in 10% of the individuals. The ventral thigh surfaces are very light with few or no melanophores in 72% of the juveniles and 31% of the adults (Fig. 2); the ventral thigh surfaces are almost uniformly dark in 13% of the adults (no juveniles); the ventral thigh surfaces are lightly to mod- erately mottled in 5% of the juveniles and 37% of the adults; the ventral thigh surfaces have the upper-lateral sector boldly mottled with the rest of the thigh surface uniformly light in 23% of the juveniles and 18% of the adults. Two percent of the juveniles and 21% of the adults have relatively uniform gray/ brown bellies; 44% of the juveniles and 13% of the adults have almost uniformly light bellies with very few melanophores (Fig. 2); 3% of the juveniles and 8% of the adults have dark bellies with moderate to low contrast light spots and/or vermicula- tions; 17% of the juveniles and 45% of the adults have a mottled pattern of smaller ir- regular lighter areas on a darker ground; and 33% of the juveniles and 13% of the adults have a boldly mottled pattern of large light spots/flecks on a darker ground. Donnelly & Myers (1991:22) provided color descriptions for specimens from Cerro Guaiquinima, Bolivar, Venezuela: “In life, some juveniles had white-edged dorsal blotches whereas others were uniformly blackish. The throat was heavily marked with gray mottling on white and there was less gray on the venter. The body glands of the groin and the ventral and posterior thigh surfaces had variable suffusion of orange that was bright in some individuals. The iris was bronze above, white on the medial ven- tral section, and brown between, with dense black venation overall.’’ Dr. Robert P. Reynolds (pers. comm.) provided copies of his color notes for specimens from Kaieteur Falls, Guyana. A series of adults and small juveniles (USNM 291245—291250) had the PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON following: “Brown dorsally with warty ru- gose skin.... Venter cream with darker mottling. Rose-orange groin and undersur- faces of thighs. Bronze-gold iris.”” A female (USNM 291251) was: “‘Dorsally greenish with dark speckling throughout. Rims of eyelids yellowish with yellow band be- tween eyelids. Three—4 light spots on lower eyelid. Yellowish with mottled brown be- tween upper lip and eyelid. Tympanum rust brown. Iris bronze. Venter brownish with white throughout. Rear of thigh gold and brown mottled.’ A series of adults of both sexes (USNM 291252—-291256) were: ““Dorsally greenish brown with yellow mot- tling. Three distinct yellow lines from eye to lip. Yellow line between eyes on top of head and a line from rear of each eye back to rear of head. Rust wash over neck and shoulder area. Mottled light and dark ven- ter. Undersides of legs rose tint with dark spots.”” A single male (USNM 342151) had: ‘“‘Dorsum reddish brown, limbs with reticulate dark markings, light line between orbits, bronze eye, dark canthal stripe; chin, throat, chest and belly with dark grey spot- ting, underside of thighs and calfs salmon neds: The degree of juvenile and adult pattern differences from individuals from the same localities are striking with respect to ventral patterns, especially when small juveniles are compared with adults. The throats and bellies of juveniles are bright white in re- cently preserved specimens (Kaieteur Na- tional Park, Guyana), whereas the throats and bellies of adults are densely mottled dark gray; the ventral thigh surfaces of the juveniles lack melanophores and stand in contrast to the dark pigmented ventral thigh surfaces of adults. No differences in dorsal pattern between juveniles and adults are ev- ident. The commonest conditions for dorsolat- eral fold development are either no indica- tion of dorsolateral folds or one short pair of ridges or elongate warts in the shoulder region. Three individuals were recorded as having two pair of short ridges and three VOLUME 108, NUMBER 4 individuals were recorded as having a series of short ridges in the dorsolateral fold field. Most individuals have a shagreen and black and/or white tubercles on the dorsum. When tubercles are present, they are more abundant on the posterior dorsum. Twenty- two percent (20 individuals) of the juve- niles have a pronounced dorsal texture of a strongly developed shagreen with glandular warts; only 5% (2 individuals) of the adults have the same texture. The upper shank surface may have a sha- green or not. Almost all specimens (122 of 124 recorded) have few to many black and/ or white tubercles; only 2 individuals have very few or no tubercles. The texture of the outer tarsus may be shagreened or not. Almost all specimens (98%) have few to many black and/or white tubercles; only 2 individuals (2%) have very few or no tubercles. The texture of the foot is smooth, lacking shagreen or tubercles in 50% of the indi- viduals; 49% of the specimens have no more than a few black and/or white tuber- cles; very few individuals (2%) have a weakly developed shagreen. Adult males have one or two black thumb spines on each hand and a pair of chest spines. There is no distinctive band of black tubercles across the chest in larger males. Males lack a patch of black chin tu- bercles on the anterior portion of the throat. The vocal sacs are laterally expanded and darkly pigmented. Adult females range between 53.6 and 73.5 mm SVL, adult males 50.9 and 71.6 mm SVL. A principal components analysis was run on the measurement data for adult specimens. There is a general cluster of points on the plot of individuals using the first two factors, but three individuals lie somewhat outside the general cluster. KU 166499 is an individual from a locality in which all other specimens lie in the general cluster of points. RMNH 23906 and USNM 258130 are the only adult individuals from each of two different localities. Re-exami- nation of these specimens does not indicate 701 that any change should be made in their as- signment to the geographically based four clusters. Data were taken from a single tadpole from Cerro Auyantepui, Venezuela, a single tadpole from La Escalera, Venezuela (both alcoholic) and seven larvae from Kartabo, Guyana (formalin). These tadpoles demon- Strate the semiterrestrial ecomorph as de- fined by Altig & Johnson (1989). The body length is 28—29% of the total length. The anterior oral gap is 75—81% of the oral disk width. The upper beak is highly arched; the upper beak depth is 33-40% of the upper beak width. The body is flattened; the depth is 64—70% the body width. Larvae have se- ries of glandular ridges on the body above the abdominal cavity. Habitat information specific for members of Group 2 has been published by Donnelly & Myers (1991:22) for specimens from Cerro Guaiquinima, Bolivar, Venezuela: **Adults were taken at night in the rocky stream bed at Camp 2 and in the north stream at Camp 1. At Camp 2, several adults were collected by night in water in the middle of small waterfalls. Juveniles were active both day and night. During the day, juveniles were commonly seen in ex- posed situations, sitting on rocks around small pools and jumping into the pools when disturbed.’’ Robert P. Reynolds (pers. comm.) collected a series of specimens (in- cluding USNM 291245—291256) at night at Kaieteur National Park, Guyana, 29-30 March 1989, on a trail from the airstrip to the guest house above Kaieteur Falls. Some individuals were calling and others were not. On the afternoon of 7 April 1994, Reynolds collected USNM 342151—342160 on bedrock puddles, near the airstrip at Kai- eteur Falls. Group 3.—Specimens from this group are from northern Brazil (Fig. 1). Three adult females, 13 adult males, 47 total specimens. The dorsal patterns characteristically have two large dark spots on the dorsum (Heyer 1979, fig. 1, pattern K). The spots may have fuzzy borders, or be well defined 702 with a black outline border, or be well de- fined by a contrasting light border. The spots may be fused with each other and the interocular spot. Nine percent of the indi- viduals have a dark dorsum with a few small, discrete, irregular, lighter spots. One individual has paired series of dark dorsal spots (Heyer 1979, fig. 1, pattern H). No individual has a completely uniform dor- sum. Upper lip patterns demonstrate a contin- uum among the following states. Nine per- cent of the specimens have little pattern on the upper lip (Heyer 1979, fig. 2, patterns C, E); 49% have some expression of dark vertical bars (Heyer 1979, fig. 2, patterns A, J); 16% have some form of alternating light and dark oblique bars (Heyer 1979, fig. 2, pattern N); 20% have some sort of irregularly to regularly defined light area in the loreal region to under the eye (Heyer 1979, fig. 2, patterns L, M); and 7% have a mottled upper lip with small dark spot- ting. The posterior thigh surface patterns grade among the following states. A single individual has an indistinctly mottled pos- terior thigh surface (Heyer 1979, fig. 3, pat- tern A); one specimen has an almost uni- formly dark thigh surface; 70% have large light spots on the upper portion of the pos- terior thigh surfaces with or without smaller light spots on the lower portion of the pos- terior thigh surfaces; 15% have narrow ver- tical light stripes on the upper portion of the posterior thigh surface; and 11% have few to several distinct small light spots on the posterior thigh surface (Heyer 1979, fig. 3, patterns N, O). Most juveniles (79%) have at least a no- ticeable lengthwise light band in the middle of the ventral thigh surfaces (very distinct in 38% (Fig. 2)), whereas only one adult has a faint indication of this band; one ju- venile and most adults (69%) have almost uniformly dark gray or brown ventral thigh surfaces; 17% of the juveniles and 25% of the adults have lightly to moderately mot- tled ventral thigh surfaces. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON No juveniles, but 29% of the adults, have relatively uniform gray or brown bellies; 63% of the juveniles, but no adults, have dark bellies with very contrasting distinct light spots (Fig. 2); 33% of the juveniles and 71% of the adults have dark bellies with moderate to low contrast distinct light spots or vermiculations. Ronald I. Crombie (pers. comm.) took the following notes on a 79.4 mm SVL ju- venile female (USNM 302192): “Light markings on back rich tan, especially on head, darker markings deep wood brown. Warts on side with reddish brown markings. Venter gray with lighter gray spots, lighter marks distinctly greenish in groin and under legs. Red-brown pustules under arms and near axilla (parasites?). Rear of thighs black with a few greenish blotches and some red- brown ones near anus. Soles of hands and feet dark gray, feet almost black. Iris gold with brassy vermiculations above, dark be- low.”’ Crombie noted that on two other large juveniles (USNM 302194, 72.3 mm SVL; USNM 302195, 69.8 mm): ““Consid- erable red spotting on anterior and posterior thighs.”’ Charles W. Myers (pers. comm.) took color notes for adults and juveniles: **Adults [AMNH 128021—128022 = CWM 18514, 18516]: Throats blackish gray; rest of ventral surfaces gray, with or without small white spots. Thighs above with black bands separated by brown interspaces that may have a faint reddish suffusion (not bright as in juveniles). Rear of thigh black with a line of silvery white dots. Iris pale bronze with reddish brown horizontal stri- pe. Juveniles [AMNH 128023-128031 = CWM_ 18526-18534]: Throat and other ventral surfaces gray. with irregular white spotting. Thighs with black bands and bright orange-red interspaces antero- and posterodorsally (red color not continuous across dorsal midline of thigh). Iris pale bronzy gray on upper half—above an ill- defined reddish brown horizontal stripe— and with a faint reddish suffusion on lower part of iris.” The degree of juvenile and adult pattern VOLUME 108, NUMBER 4 differences from the same locality is strik- ing in preserved specimens. The most no- ticeable differences occur on the thighs. The upper portion of the posterior thigh surfaces in juveniles have well defined large light spots (bright red in life), whereas such spots are not distinct or completely ab- sent in adults. In most juveniles, there is a distinct light longitudinal band on the mid- ventral thigh surfaces; no such distinct light band occurs in adults. The belly spotting differences between juveniles and adults are not as striking as the thigh surface pattern differences—in juveniles, the dark bellies have more distinct light spots than adults. No dorsolateral folds are evident in 16% of the individuals examined. In most spec- imens (77%) dorsolateral folds are distinct, but broken (not continuous) and range from short (extending only to the shoulder re- gion) to long (entire length of back from behind eye to leg). Only two individuals have almost continuous dorsolateral folds extending to the sacrum, and only one spec- imen has a continuous dorsolateral fold ex- tending to the leg. Most specimens have a shagreen and white tubercles on the dorsum. When tu- bercles are present, they are more abundant on the posterior dorsum. Only in juveniles (60%) is a strong warty shagreen devel- oped. The upper shank usually has a shagreen (93%). Most specimens (87%) have few to many black and/or white tubercles; some (13%) have very few or no tubercles. The texture of the outer tarsus is usually shagreened (85%). Most specimens (93%) have few to many black and/or white tu- bercles; a few (7%) have very few or no tubercles. The texture of the sole of the foot is usu- ally smooth (87%); 11% have a weak sha- green; only one individual has scattered white tubercles. Adult males have one black spine on each thumb. There is no indication of chest Spines, chest tubercles, or chin tubercles. 703 Vocal sacs are neither laterally expanded nor differentially patterned. Adult females range from 103.8 to 112.9 mm SVL, adult males 74.2 to 116.8 mm SVL. The relatively small sample size does not justify analyzing the measurement data with principal components to identify po- tential outlier individuals. No tadpoles are available for this group. Habitat notes are available from two sets of collectors from the same rock outcrop from Mucajai, Roraima, Brazil. Charles W. Myers (pers. comm.) indicated for speci- mens AMNH 128021-—128031, collected on 11 July 1987: “‘On granite inselberg sur- rounded by humid scrubby forest with many palms. Some juveniles by day under granite flakes microsympatric with Tropi- durus hispidus and Leptodactylus fuscus. Other juveniles by night, sitting on bare rock or in small pools of water (with sparse aquatic vegetation) on the bare rock. The four adults all in the small pools by night.”’ On 30 May-1 June 1988, Ronald I. Crom- bie (pers. comm.) collected USNM 302066—302068, 302190—302203, MZUSP 660889 at night from on or under rock. Ronald I. Crombie (pers. comm.) docu- mented that USNM 302267 from Col6énia Apiat, Roraima, Brazil, collected on the af- ternoon of 14 June 1988, was taken from under roofing tiles at an abandoned hunting camp in the forest, not near any rocky out- crop. Group 4.—Specimens in this group are from French Guiana and Surinam (Fig. 1). One adult female, 1 adult male (both from French Guiana), 54 total specimens. Most specimens have two large, dark, relatively well-defined blotches on the dor- sum posterior to the dark interorbital blotch (Heyer 1979, fig. 1, pattern K); these blotches may be outlined by a white ring or a dark border. A few specimens have some paired large or small dark spots on the dor- sum (Heyer 1979, fig. 1, patterns H, J). A few have fused blotches (Heyer 1979, fig. 1, patterns D, I). Uniform patterns (Heyer 704 1979, fig. 1, pattern C) occur in 9% of the sample. Upper lips demonstrate variability among the following states. Seven percent of the specimens have little pattern on the upper lip (Heyer 1979, fig. 2, patterns C, E); 54% have some expression of alternating light and dark vertical bars (Heyer 1979, fig. 2, patterns A, J); 13% have some form of al- ternating light and dark oblique bars (Heyer 1979, fig. 2, pattern N); and 26% have some sort of irregularly defined light area in the loreal region which may extend to under the eye (Heyer 1979, fig. 2, patterns K, M). _ The posterior thigh surface patterns are quite variable. They are distinctly mottled with small irregular light marks in 13% of the individuals (Heyer 1979, fig. 3, patterns C, D); distinctly mottled with large light ir- regular spots and marks in 13% (Heyer 1979, fig. 3, patterns A, EK K); mottled with some expression of light vertical marks or bars on the upper portions of the thigh sur- faces in 11% (Heyer 1979, fig. 3, pattern I); mottled with large distinct light spots or blotches on the upper portions of the thigh surfaces (light marks may be confluent with each other) in 51% of the individuals; one individual has distinct light spots on the posterior thigh surfaces (Heyer 1979, fig. 3, pattern N); three specimens have some large dark blotches on the lower portion of the thigh surfaces; two specimens have rel- atively uniform thigh surfaces. Juveniles do not have a distinct light band lengthwise in the middle of the ventral thigh surfaces, although 21% of the juve- niles have a noticeable light band (Fig. 2), neither adult has any indication of a light band; 21% of the juveniles have very light ventral thigh surfaces with few or no me- lanophores, neither adult has this pattern; no juveniles have almost uniformly dark ventral thigh surfaces, one adult has this pattern; 4% of the juveniles have lightly to moderately mottled ventral thigh surfaces, one adult has heavily mottled ventral thigh surfaces; 50% of the juveniles have the up- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON per-lateral sector of the ventral thigh sur- faces boldly mottled with the rest of the ventral thigh surface uniformly light, nei- ther adult has this pattern. No juveniles, but both adults, have a rel- atively uniform gray/brown belly; 23% of the juveniles have a uniform and light belly with very few melanophores; 27% of the juveniles have a dark belly with distinct light spots/vermiculations of moderate to low contrast; 25% of the juveniles have a mottled belly of smaller irregular light areas on a darker ground; 25% of the juveniles have a bold mottled belly pattern of large light spots/flecks on a darker ground (Fig. 2). Dr. Charles W. Myers (pers. comm.) took color notes on a series of juvenile speci- mens from Voltzberg, Surinam (AMNH 87705-87736): *“‘Dorsal surfaces brown with a remarkably constant pattern of darker brown markings that are outlined in light yellowish brown. Top of thigh with orange-red flash mark (which is concealed in frog’s normal position) that is broken by black bars; the flash mark is present in all but is quite faint on several specimens of different size. Rear of thigh black with pale blue or greenish blue dots. Venters white with variable amounts of gray mottling, turning light gray underneath limbs. Iris pale bronzy orange, in some individuals turning pale gray below pupil; dense black venation and a tendency for a brownish horizontal streak.” There is no locality for which a series of adults and juveniles are available. In com- paring adults and juveniles from different localities, the degree of adult and juvenile pattern differences in preserved specimens is noticeable but not striking. Most notice- able are the upper posterior thigh light spots/blotches (red in life) on juveniles, that are not present in either adult. The bellies and ventral thigh surfaces are much lighter than in the adults, with the smallest juve- niles having the lightest bellies, with a size gradient of darkening evident. Almost all individuals (94%) have some VOLUME 108, NUMBER 4 sort of dorsolateral fold development (the 4 individuals scored as having no indication of dorsolateral folds perhaps due to pres- ervation artifact?); 34% of the specimens have a pair of entire dorsolateral folds ex- tending from behind the eye to at least as far as the sacrum; 60% of the individuals have a pair of interrupted folds to a series of ridges or elongate warts to at least the sacral region. All individuals have at least a weakly- developed dorsal shagreen. Two juveniles have a pronounced shagreen with small bumpy glands. Thirty-one percent of the ju- veniles have at least a few tubercles on the dorsum, that are more abundant posteriorly. Both adults have a weakly developed dorsal shagreen; one lacks noticeable tubercles, the other has very few white dorsal tuber- cles. Most specimens have a shagreen on the dorsal shank surface (lacking in only two). Most specimens (89%) have few to many white tubercles; few specimens (11%) have very few or no white tubercles. The texture of the outer tarsus may be shagreened or not. Many specimens (70%) have few to many white tubercles; some specimens (30%) have very few to no white tubercles. Only two specimens have a weak sha- green on the sole of the foot; only one spec- imen has a few white tubercles; all others have a smooth sole that is lacking shagreen or tubercles. The single adult male has one white Spine on each thumb and one white bump on each prepollex. There is no indication of chest spines, chest tubercles, or a patch of tubercles on the chin. Vocal sacs are neither laterally expanded nor differentially pig- mented. The adult female is 111.2 mm SVL, the adult male 117.6 mm SVL. No tadpoles are available for this group. Habitat data are available from Saramac- ca, Surinam (Charles W. Myers, pers. comm.) for AMNH 87705-87736 collected 23 February 1972 from a granite outcrop: 705 “Three specimens of a considerably larger Leptodactylus [not collected] were seen on the granite at night, but when approached they backed under rock crevices in which lizards were sleeping. [Following notes for juveniles.] Several by day, under rock flakes around edges of bare granite sur- faces; a few in pools in circular depressions after afternoon rain, an hour or so before dark. Most however at night, when they proved to be exceedingly abundant out on bare rock far away from any cover or crev- ice. The majority were sitting in the little damp gullies that resulted from overflowing of water from the circular holes and other depressions on the granite.”’ Intergroup differentiation and evaluation of species limits.—The group characteriza- tions described in the preceding section are used to evaluate differentiation among the groups. In addition, some data are available for two or three of the groups that pertain to group differentiation. These additional data are presented below. The preceding group characterizations demonstrate that the greatest differentiation occurs between the combined Groups | and 2 on the one hand and Groups 3 and 4 on the other. The most distinctive differentiat- ing characteristics are size of adults, sec- ondary male features, and contrast of ju- venile and adult color patterns (Table 1). Discriminant function analyses using measurement data support this clustering, although there are not enough data to in- clude Group 4 in the analyses. Data run separately for males and females show the same results. Results of the male data show somewhat less differentiation among groups, so they are described as examples for both sets of analyses. Data for the Group 4 individual were not used in esti- mating the multivariate models. The pos- terior classifications based on group dis- criminant function means for Groups 1, 2, and 3 result in rather good group assign- ments overall, but most robust for Group 3, with only one male being posteriorly as- signed to the wrong group (Group 1, Table 706 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Distribution of certain characteristics among four groups of rocky habitat Leptodactylus from north- ern South America. Group 1 is from Colombia and nearby Venezuela, 2 from southeast Venezuela and Guyana, 3 from northern Brazil, 4 from Surinam and French Guiana. Group 1 Group 2 Group 3 Group 4 Female size in mm Male size in mm Number of spines/thumb in males Chest spines in males Band of chest tubercles in males Chin tubercles in males Vocal sacs laterally expanded and darkly pigmented Juvenile and adult color patterns strikingly different 2). The graphical results when the first ca- nonical factor is plotted against the second (Fig. 3) indicates that Group 3 individuals are separated from Group | and 2 individ- uals along the first axis, typically size re- lated, which is consistent with the adult size distributions of the three groups. The single Group 3 individual that was posteriorly classified as belonging to Group 1 rather than 3 (USNM 302202) is the left-most symbol for Group 3 individuals in Fig. 3 and is the smallest male in the Group 3 sample. Some micro-complement fixation tests, comparing albumins of some of these frogs included one sample from Group 2 and one sample from Group 3. Although materials were insufficient to raise antisera to either the Group 2 or Group 3 specimens, both samples were tested against antisera to al- bumins of other members of the L. penta- dactylus species group. Results (Table 3) indicate that the Group 2 and Group 3 sam- Table 2.—Discriminant function analyses for male data for Groups 1 (Colombia and nearby Venezuela), 2 (southeast Venezuela and Guyana), and 3 (northern Brazil). Number of observations classified into groups Group 1 Group 2 Group 3 Group 1 26 5) 0 Group 2 4 3) 0 Group 3 1 0) 1 55-78 54-74 104-113 111 45-71 51-72 74-117 118 1 1-2 1 1 + — = + — = +++ + ples are not identical and, in fact, suggest that they are quite different. Thus, the data are unequivocal that at the very least, Groups | and 2 combined rep- resent a distinct species from Groups 3 and 4 combined. The more difficult questions to resolve are whether Group 1 represents a species distinct from Group 2 and whether Group 3 represents a species distinct from Group 4. Each of these two situations is discussed in turn. The greatest differences between Group 1 and Group 2 individuals are with male secondary sexual characteristics (Table 1). Whereas males in Group 1 have chin tu- bercles and, in the larger adults, a band of chest tubercles, no males in Group 2 have such tubercles. Whereas some males in Group 2 have two spines per thumb, all males in Group | have but one spine per thumb. There are differences in terms of de- gree of expression of states in the upper lip patterns, posterior thigh surface patterns, ventral thigh surface patterns, belly pat- terns, outer tarsal texture, and foot texture. Whereas there is complete overlap of adult sizes between the two groups, there is some morphological differentiation demonstrated by measurement data (Fig. 3). There is clearly some separation of Group 1 from Group 2 along the second canonical factor, indicating some differentiation of shape. The dependent variables that have the high- est values on the second canonical factor VOLUME 108, NUMBER 4 4 3 oO O ; = oA Oo ; Oo Factor 2 > > sd Pt of > -1 A SO aa -2 A A A -3 A A -4 -5 Fig. 3. 707 A A 3 4 ts) 6 7 8 9 Factor 1 Graph of first and second canonical factors for Group I-IV males. Group I = open squares, Group II = open triangles, Group III = filled triangles, Group IV = filled square. for females are shank length (0.275), head length (0.253), thigh length (0.251), eye— nostril distance (0.250), and head width (0.236) and for males are thigh length (—0.512), shank length (—0.493), eye—nos- tril distance (—0.446), and foot length (—0.413), suggesting an overall shape dif- ference. Given the discrete differences of male secondary sexual characteristics, com- bined with the evidence of demonstrable Table 3.—Immunological distances between albu- mins of two samples of northern South American gra- nitic outcrop associated Leptodactylus. Immunological distance Anti- serum- Anti- serum- L. penta- L. Anti- serum- L. flavo- labyrin- Anti- serum-L. Samples dactlus fallax pictus thicus Group 2 (Bolivar, Venezuela) 67 24 43 SH Group 3 (Roraima, Brazil) Dap 30 5)7/ 43 differentiation in pattern and morphology between Groups | and 2, the evidence is most consistent with recognizing members of these two groups as distinct species. Evaluating the differentiation between Groups 3 and 4 is confounded by the fact that there is but a single adult male and a single adult female available for Group 4. In comparing the available data, there is less apparent differentiation between Groups 3 and 4 than between Groups | and 2 (also see Table 1). Group 3 juveniles dif- fer from Group 4 juveniles in terms of ven- tral coloration pattern. Group 3 juveniles often have a very distinct longitudinal light band on dark ventral thigh surfaces; no Group 4 juveniles have this expression. Group 3 juveniles often have very light pat- terns over the entire ventral thigh surface; no Group 4 juveniles have this expression. Most Group 3 juveniles have very contrast- ing belly patterns of distinct light spots on 708 a dark belly; no Group 4 juveniles have this pattern. As a consequence of these differ- ences in juveniles, the contrast between ju- venile and adult ventral patterns is more pronounced in Group 3 than in Group 4. The single adult male in Group 4 is 117.6 mm SVL, minimally larger than the largest Group 3 male, 116.8 mm SVL. The Group 4 male has a single white spine on each thumb and the merest indication of a second spine. White thumb spines in Leptodactylus occur due to two different reasons. They are characteristic of very young males, which have just become sexually mature, and some Leptodactylus seasonally shed the black keratin sheath of the spine. If the white spine condition of the Group 4 male is because it is a young male, then there is probably a size difference between Group 3 and 4 adults. When actually comparing the available adults from Group 3 and 4 side- by-side, I find no differences among them other than the male thumb condition noted. The data indicate that there has been some differentiation between Groups 3 and 4, but the data are not at all conclusive in deciding whether Groups 3 and 4 represent one or two species. Given the lack of advertise- ment calls and adequate samples of adults of Group 4, I prefer to take the nomencla- turally conservative (and very possibly the biologically incorrect) position of recogniz- ing a single species for Group 3 and Group 4 combined. : Nomenclature.—Only a single name has been proposed for any of the frogs involved in this portion of the study: Leptodactylus rugosus Noble, 1923. As Donnelly & My- ers (1991:23) pointed out, the holotype (AMNH 1169, 38 mm SVL) from near Kai- eteur Falls, Guyana, described as an adult by Noble (1923:297), is in fact a juvenile. Leptodactylus rugosus applies to the frogs of Group 2 in this paper. Without available names for the other two species recognized as distinct, they are described as the follow- ing new species. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Leptodactylus lithonaetes, new species Figs. 4A, 5 Holotype.-—AMNH 100656, an adult male from Venezuela; Amazonas, SW sec- tor Cerro Yapacana, 600 m, 3°57'N, 67°00'’W. Collected by Charles W. Myers and John Daly on 18—19 February 1978. Paratopotypes—AMNH 100657—100666, 100668, same data as holotype except col- lected from 18—22 February 1978. Diagnosis.—The species of Leptodacty- lus that share conditions of free toes (no lateral fringe) and indistinct dorsolateral folds (at least in some individuals) with L. lithonaetes are bufonius, labialis, labyrin- thicus, laticeps, latinasus, myersi, rugosus, syphax, and troglodytes. The upper shank and posterior tarsus of bufonius, labialis, latinasus, and troglodytes are covered with large prominent white tubercles and males lack thumb spines; the upper shank and posterior tarsus of lithonaetes has black- tipped tubercles (in some preserved speci- mens the black tips may be lost, leaving white tubercles which are noticeably small- er than those of bufonius, etc.), and the males have a spine on each thumb. Lepto- dactylus laticeps has a tile-like dorsal pat- tern (Heyer 1979, fig. 1, pattern F) and is larger (minimum adult SVL 91 mm) than lithonaetes (maximum adult SVL 78 mm), which does not have a distinct tile-like pat- tern. Leptodactylus labyrinthicus is larger (minimum adult SVL 117 mm) than Jith- onaetes; no L. labyrinthicus have light lo- real blotches, whereas several lithonaetes do. Leptodactylus myersi is larger (females 104-113 mm SVL, males 74-118 mm SVL) than lithonaetes and myersi males lack the chest spines found on Jithonaetes. Leptodactylus lithonaetes is most likely to be confused with L. rugosus and syphax; the most distinctive characteristics among these three species are male secondary sex- ual characteristics. Leptodactylus lithonae- tes has a single black spine on each thumb and a patch of brown/black tubercles on the chin/throat; all L. syphax and some rugosus VOLUME 108, NUMBER 4 have two spines per thumb and no syphax or rugosus males have a patch of chin tu- bercles. Description of holotype.—Snout rounded from above and in profile; canthus rostralis indistinct; loreal weakly obtuse-concave; tympanum distinct, greatest diameter about ¥, eye diameter; vomerine teeth in strongly arched series, between and posterior to cho- anae, separated medially by about ¥, length of single vomerine tooth series; vocal slits present; vocal sacs expanded laterally and darkened; finger lengths in increasing order II just < IV < I just < III; extensive finger ridging best developed on medial surfaces of fingers II and III; metacarpal tubercles large, about same size, inner triangular shaped, outer ovate and bifid; arms strongly hypertrophied; thumb with one large black medial spine, one pair of large bicuspid black chest spines, narrow band of brown- black tubercles across chest, field of brown- black tubercles on thumb on either side of spine, well developed field of brown-black tubercles on chin medially extending pos- teriorly to about mid-throat; dorsum rela- tively smooth, a few scattered black tuber- cles posteriorly; one short pair of interrupt- ed dark-outlined ridges behind eye and above and posterior to tympanum, well de- veloped supratympanic fold from eye to hu- merus; commissural gland well developed, flanks with three large irregular glands on each side, lower posterior thigh with elon- gate longitudinal gland; ventral texture smooth; belly disk fold well developed; toe tips bulbous, broader than toes immediately behind tips; toes free, lacking fringe or web; subarticular tubercles well developed, Ovoid; outer rounded metatarsal tubercle relatively well developed, about 'Y, size ovoid inner metatarsal tubercle; tarsal fold moderately developed, sinuous, extending about 7; length of tarsus; no metatarsal fold; outer tarsus with several brown-black tu- bercles; sole of foot smooth. SVL 71.4 mm, head length 27.2 mm, head width 27.4 mm, eye—nostril distance 6.4 mm, interorbital distance 5.1 mm, great- 709 est tympanum diameter (including annulus) 6.3 mm, thigh length 31.5 mm, shank length 31.5 mm, foot length 36.1 mm. Dorsum with a light interocular bar; rest of back with ill-defined pattern of series of fused blotches on each side of midline and series of darker brown flecks more numer- ous posteriorly; upper limbs with faint sug- gestion of transverse bands, most distinct on dorsal surfaces of tarsus and foot; irreg- ular dark canthal stripe, with an irregular light tan loreal blotch ventrally; upper lip with irregular dark border. Throat almost uniformly dark brown medially, laterally expanded vocal sacs dark brown, edges of jaws with black borders; chest, belly, and ventral limb surfaces with extensive, fine brown mottling. Posterior surface of thigh with large irregular-shaped dark and light blotches. Etymology.—From the Greek lithos, stone, rock and naetes, inhabitant, in ref- erence to its habitat. Variation.—Variation is described earlier in the paper under Group 1 individuals. Larval characteristics (based on sample from Cerro Patava, Colombia, UTA 23499, 23507-23510, n = 10).—Dorsal head and body relatively uniform brown, dorsal pat- tern ending somewhat abruptly on side of head-body; oral disk lacking melanophores; belly with or without white flecks; ventral region anterior to guts with melanophores visible under skin; tail dorsal fin and most of muscular region mottled, with or without darker flecks and/or weakly-developed light ocelli; ventral musculature and tail fin lack- ing melanophores entirely or with melano- phores on distal half, most heavy at tip re- gion; nostril nearer eye than tip of snout; internarial distance greater than interorbital distance; eye diameter 14—16% head-body length; oral disk width 25—26% head—body length for Gosner stages 28-29, 22-26% head—body length for Gosner stages 35—40; anterior oral papilla gap 80-88% oral disk width; 32—34 denticles on one side of split tooth row anterior to beak (row A-2) for Gosner stages 28-29, 42-51 denticles for 710 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 4. New species of Leptodactylus. A. Male paratopotype (AMNH 100668) of Leptodactylus lithonaetes. Photograph courtesy of C. W. Myers, American Museum of Natural History. B. Male paratopotype (AMNH 128021) of Leptodactylus myersi. Photograph courtesy of C. W. Myers, American Musuem of Natural History. VOLUME 108, NUMBER 4 EEE Pig. 3: Vy é a My A d MY, Sony, : My Ea My 7 7 i lM, sce 2 28s yyy Lateral view and mouthparts of larval Leptodactylus lithonaetes. Scale bars 5 and | mm respectively. Diagrammatic illustrations based on specimen from UTA-A 23509, Gosner stage 38. Gosner stages 35—40; upper beak width 1 1— 12% head—body length; upper beak depth 30-33% upper beak width for Gosner Stages 28-29, 27-37% for Gosner stages 35—40; single row of marginal papillae; tooth row formulae 2(2)/3 or 2(2)/3(1); field of glandular ridges on body above abdom- inal cavity; spiracle sinistral; vent median; fins 29-33% tail depth at maximum tail depth for Gosner stages 28-29, 23-36% for Gosner stages 35-40; body depth 38-42% head—body length for Gosner stages 28—29, 42-48% for Gosner stages 35-40; body width 55-56% head—body length for Gos- ner stages 28-29, 58-64% for Gosner stages 35-40; body depth 70-75% body width for Gosner stages 28-29, 67-74% for Gosner stages 35-40; head—body length 32-35% total length for Gosner stages 28— 29, 28-31% for Gosner stages 35—40; total length, stage 38, 36.1 mm (Fig. 5). Advertisement Call.—Not recorded. Karyotype.—Unknown. Referred specimens (specifically not des- ignated as types)—COLOMBIA. AMA- ZONAS: Rio Igara-Parana, 50 km arriba “La Chorrera,”” IND-AN 2822. GUAINIA: Puerto Inirida, Rio Atabapo, IND-AN 2808, 2811; Rio Negro, opposite Casiquare Canal and Guainia, AMNH 23160-23162. VAU- PES: Cerro Patava, IND-AN 3399, UTA-A 3724-3725, 3772-3785, 3857-3877, 23499— 23500 (larvae), 23503-23510 Carvae). VI- CHADA: Parque Nacional Natural El Tupar- ro, ICNMNH 14045-14047, IND-AN 2303-— 2304, 3736; Puerto Carreno, Cueva de Arévalo, ICNMNH 13970-13996. VENEZUELA. AMAZONAS: Canaripo, RMNH 23913-23916, 23931-23950, 23970- 23973 (plus 5 unnumbered juveniles and me- tamorphs and 3 larvae); Cerro Yapacana, AMNH 100667 (skinned specimen); Mision Coromoto-Atures, USNM_ 137186—137192; Puerto Ayacucho, AMNH 23209-23219, FMNH_ 175460—-175465, 176197—176201, 176203—-176222, KU 207531-—207533, USNM 80634, 80665-80673, 291081- 291082; Sanariapo, MCZ 27827, USNM 80635-80639. APURE: Hato Caribén, 46 km NE Puerto Paez, Rio Cinaruco, USNM 216795—216797. BOLIVAR: 3 km E Rio Orinoco just below Rio Horeda, 100 m, AMNH 62169; mouth of Rio Parguaza, AMNH 62168. 12 Leptodactylus myersi, new species Fig. 4B Holotype.—MZUSP 66089, an adult male from Brazil; Roraima; Mucajai, 2°25'N, 60°55'W. Collected by Celso Mor- ato de Carvalho and Ronald I. Crombie on 31 May 1988. Paratopotypes.—AMNH 128021—128031, MZUSP 70976-70986 collected by Celso Morato de Carvalho, Charles W. Myers, Thelma Schmidt, and P. E. Vanzolini on 11 July 1987, USNM 302190—302203, collected by Celso Morato de Carvalho and Ronald I. Crombie on 30 May-—1 June 1988, USNM 302204—302205, collected by Carvalho and Crombie on 20 June 1988. Diagnosis.—The species of Leptodacty- lus that share conditions of free toes (no lateral fringe) and indistinct dorsolateral folds (at least in some individuals) with L. myersi are bufonius, labialis, labyrinthicus, laticeps, latinasus, lithonaetes, rugosus, syphax, and troglodytes. The upper shank and posterior tarsus of bufonius, labialis, latinasus, and troglodytes are covered with large prominent white tubercles and males lack thumb spines; the upper shank and posterior tarsus of myersi has black-tipped tubercles (dn some preserved specimens the black tips may be lost, leaving white tuber- cles which are noticeably smaller than those of bufonius, etc.) and the males have a spine on each thumb. Leptodactylus lati- ceps has a very distinct tile-like dorsal pat- term (dleyern 97/9 ince les pattetnael)) le. myersi does not have a tile-like dorsal pat- tern. Leptodactylus labyrinthicus is. larger (minimum adult SVL 117 mm) than myersi; all large (>170 mm SVL) male labyrinthi- cus have chest spines, no myersi males have chest spines. Leptodactylus syphax is small- er (maximum adult SVL 78 mm) than myersi (maximum adult SVL 118 mm) and male syphax have two spines per thumb in contrast to the single thumb spine found in myersi. Leptodactylus lithonaetes and ru- gosus are smaller (maximum adult SVL 78 mm) than myersi. The males of L. lithon- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON aetes and rugosus have laterally expanded vocal sacs; the male vocal sacs of L. myersi are not expanded externally. Description of holotype.—Snout nearly rounded from above and rounded in profile; canthus rostralis indistinct; loreal obtuse- concave; tympanum distinct, greatest di- ameter about *%3 eye diameter; vomerine teeth in arched series, between and poste- rior to choanae, separated medially by about %4 length of single vomerine tooth se- ries; vocal slits present; vocal sac single, internal; finger lengths in increasing order II just < IV < I just < III; finger ridges well developed medially on fingers II and III; inner and outer metacarpal tubercles large, about same size, outer bifid; arms moderately hypertrophied; thumb with one large black medial spine; no chest spines; band of black tubercles aross chest, on ven- tral surfaces of upper arms, on flanks near axilla and along lateral edges of jaws ven- trally; dorsum relatively smooth, a few small black or white tubercles scattered posteriorly; extensively interrupted pair of dark-outlined dorsolateral folds from eye to beyond sacrum, distinct continuous supra- tymypanic fold from eye to humerus; flanks glandular; belly smooth; belly disk fold dis- cernible but weak; toe tips bulbous, broader than toe widths just behind tips; toes ridged laterally, especially on toes I, I, and III, not developed into fringe; subarticular tubercles well developed, ovoid; outer metatarsal tu- bercle round, flattened, about % size of ovoid inner tubercle; tarsal fold well devel- oped, extending about *% length of tarsus; no metatarsal fold; outer tarsus with a few small white or black tubercles; sole of foot smooth. ; SVL 109.4 mm, head length 47.2 mm, head width 44.1 mm, eye—nostril distance 9.9 mm, interorbital distance 8.4 mm, great- est tympanum diameter (including annulus) 7.8 mm, thigh length 47.2 mm, shank length 44.1 mm, foot length 46.8 mm. Dorsum dark gray brown with irregular darker brown interorbital blotch and two darker dorsal chevrons, one in shoulder re- VOLUME 108, NUMBER 4 gion, one in sacral region; dark well-defined canthal stripe; upper lip rather uniform gray-brown; upper limbs transversely band- ed. Venter dark and gray, chin and throat uniform, chest and belly with indistinct light, small vermiculations. Upper portion of posterior thigh surface with extension of dorsal band pattern, lower portion with black elongate oval region containing a few contrasting small light regular and irregular spots. Etymology.—Named for Dr. Charles W. Myers for his contributions to Neotropical herpetology in general and bringing this new species to my attention in particular. Variation.—Variation is described earlier in this paper under Group 3 and Group 4 individuals. Advertisement Call.—Unknown. Karyotype.—Unknown. Referred Specimens (specifically not des- ignated as types)—BRAZIL.' AMAZO- NAS: Rio Araca (Serrinha), MZUSP 59016, 59018, 59026-59028. RORAIMA: Colonia Apiat, MZUSP 65949, USNM 302267; Mucajai, USNM 302206 (gutted and partially deboned). FRENCH GUIANA. No further locality, MNHN 1982-81; Massif des Emerilions, MNHN 1982-82; Montagne St. Marcel, MNHN 1982-73—1982-80; Montagne des Trois Pitons, MNHN 1982-153; Peolaeu (Haut Oyapock), MNHN 1982-83; Trois Saut, MNHN 1982-84. SURINAM. BROKOPONDO/NICKER- IE: Tafel Berg, 1025 m, MCZ 97259- SI ZOMn OT 299", 97/3035 97306; 97308, RMNH 23912, 23919-23930, 23951- 23961. NICKERIE: Amotopo, , RMNH 23964—23968; Blanche Marie-Vall, RMNH 23910-23911. SARAMACCA: Raleighval- ' After the manuscript was submitted for publica- tion, another locality for L. myersi was discovered in the collections at MZUSP MZUSP 28405, 54110— 54114 are from Brazil, Para, Igarape Jaramacaru, Cam- pos do Ariramba, 1°09’S, 55°54’W. The frogs were collected from sandstone formations (P. E. Vanzolini, pers. comm.), a common feature within the Campos do Ariramba region (Egler 1960). YS len-Voltzberg Nature Preserve, MCZ 92363; Voltzberg, RMNH 23974-23976. Disposition of Specimen from Paraguay On 22 July 1973, the mammalogist Phil- ip Myers collected a single specimen of Leptodactylus, MVZ 111027, that resem- bles L. syphax from a rocky region with caves | km S of Tobati, La Cordillera, Par- aguay (25°15’S, 57°04'W). The same cave system has been re-collected by mammal- ogists, who have also collected whatever frogs they encountered (P. Myers, pers. comm.). Even though these newer collec- tions contain Leptodactylus, none of them are the same species as MVZ 111027. Description of the adult female from near Tobati follows: tympanum large, greatest diameter about % eye diameter; vomerine teeth in arched series posterior to the cho- anae, contacting medially; palatine teeth present; inner metacarpal tubercle ovoid, the outer (about same size as inner) some- what rounded, tubercles separated by less than width of inner tubercle; dorsum glan- dular and smooth; supratympanic fold dis- tinct, no other folds present; no distinct glands other than the general glandular Structure of skin on top of shanks and on back; venter smooth; ventral disk not well defined; tips of toes barely swollen; toes lacking fringe or web; no metatarsal fold; tarsal fold weakly developed extending about ¥% length of tarsus; dorsal shank and outer tarsus surfaces glandular with scat- tered white tubercles; sole of foot smooth; side of head uniform brown; dorsum brown with faint quadrangular ocellations; upper arm surfaces almost uniform brown; upper leg surfaces weakly cross banded with darker brown; venter cream, lacking mela- nophores; posterior surfaces of thighs dis- tinctly mottled, almost spotted; SVL 79.7 mm, head length 31.7 mm, head width 29.1 mm, eye—nostril distance 7.4 mm, tympa- num diameter 6.5 mm, thigh length 28.7 mm, shank length 30.0 mm, foot length 30.6 mm. 714 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 6. Known distribution of Leptodactylus syphax in Brazil and Paraguay. Political boundaries outlined for Brazil and Paraguay. Map truncated at 40°S. The Tobati specimen is very similar to L. syphax from Brazil. They are the same size and share the same features of pattern and texture. The shared pattern of a low-con- trast mosaic pattern of quadrangular ocel- lations is uncommon in the genus. The striking visual difference between the To- bati specimen and Brazilian L. syphax is the short leg of the Tobati specimen. The thigh/ SVL, shank/SVL, and foot/SVL ratios for the Tobati specimen are 36%, 38%, and 38% respectively. The mean female ratios for L. syphax are 42%, 42%, and 44% re- spectively. The only other feature that dif- fers between the Tobati specimen and Bra- Zilian L. syphax is that the subarticular tu- bercle on the thumb of the Tobati specimen is less pronounced than in Brazilian L. syp- hax. Without additional material, it is impos- sible to know whether the short leg of MVZ 111027 is characteristic of the population or is an individual aberration. The species identification of the Tobati deme is made more difficult by the lack of both advertise- ment call data and information on male sec- ondary sexual characteristics. Given the available data, I think the best taxonomic conclusion is to consider that the Tobati specimen is conspecific with L. syphax. This conclusion should only be taken as a working hypothesis until further data be- come available. The geographic extension into Paraguay of L. syphax, although con- siderable, might be expected given the dis- junct nature of the rest of the known pop- ulations of L. syphax (Fig. 6). Discussion The main conclusion of this study is that there is considerably more variation among populations of Leptodactylus associated with rocky habitats than previously sum- marized (Heyer 1979). Fully understanding and interpreting that variation is not possi- ble at this time. Two suites of questions are raised: those associated with distributions and those associated with relationships. The degree of habitat restriction to rocky habitats, either streambeds or outcrops, is not known for certain, but such habitat fi- delity is high, if not complete. The semi- terrestrial larval ecomorph of L. lithonates and rugosus is certainly an adaptation as- sociated with rocky substrates and probably VOLUME 108, NUMBER 4 flowing water. The larvae of L. myersi and syphax are unknown, but adults and juve- niles of both have been collected most fre- quently (entirely so for L. syphax) from rocky outcrops. The scanty data suggest a greater association of L. lithonaetes and ru- gosus with rocky streams (either in a gran- ite or sandstone setting) and L. myersi and syphax with granitic or sandstone outcrops. The patchy distributions of the taxa also in- dicate a high fidelity to rocky habitats, which themselves are patchily distributed. This habitat fidelity allows prediction of where the species should occur in areas where they have not been collected up to the present. For example, the easternmost locality in Colombia for L. lithonaetes (Amazonas, Rio Igara-Parana), coincides reasonably well with an isolated occurrence of Precambrian-aged rocks with granite be- ing among the most common rock types (Kroonenberg 1985, fig. 1, p. 58). There are additional isolated patches of Precambrian rocks of the same formation in Colombia (Kroonenberg 1985:58) from which L. lith- onaetes are currently unknown. Leptodac- tylus lithonaetes would be expected to oc- cur (at least historically, if not currently) on these Precambrian rock formations if run- ning water exists in the formations. There are several questions remaining re- garding relationships. Although L. lithon- aetes is considered distinct from rugosus in this paper, it is possible that the differences between the two taxa represent (pro- nounced) geographic differentiation in a single species. On the other hand, the dif- ferences herein interpreted to represent geo- graphic variation between the Surinam and French Guiana populations of L. myersi and the Roraima, Brazil, populations of myersi, may be too conservative an interpretation and two species should be recognized in- stead. The species allocation of the frogs from Brazil, Amazonas, Rio Aracé need verification. At present, only rather small- sized juveniles are available from the Rio Araca site. The characteristics they have are consistent with being conspecific with L. D'S myersi (my best guess), or L. lithonaetes, or representing yet another new species. The Paraguayan population needs resampling to verify whether it is conspecific with L. syp- hax. It is very likely that advertisement call data could resolve the outstanding species level problems. At another level, the avail- able data can not answer the question whether the rocky habitat associated species of Leptodactylus are a monophyletic cluster, having a common ancestor that became adapted to the habitat, or not. Larval infor- mation for L. myersi and syphax may rfe- solve this question, but it is more likely some appropriate molecular based analysis is needed to answer it. Given the many questions about relation- ships, any detailed discussion of zoogeog- raphy of these species is premature. Acknowledgments Charles W. Myers, AMNH, alerted me to the new species from Roraima, freely shared all the extensive data he had col- lected on frogs studied in this paper, and reviewed the manuscript. Linda R. Maxson, The University of Tennessee, kindly pro- vided the micro-complement fixation data reported herein. For providing color and habitat notes, I thank Ronald I. Crombie (USNM), John D. Lynch (University of Ne- braska), Charles W. Myers, and Robert P. Reynolds (National Biological Service). Kate Spencer drew the tadpole illustrations (Figure 5). George R. Zug (USNM) re- viewed the manuscript. For their patience in allowing me to study materials on loan over a several year period, I thank the following institutions and their curatorial staffs: AMNH—Charles W. Myers; BMNH—Barry T. Clarke; CM—Ellen J. Censky and the late Clarence J. McCoy; FMNH—Harold Voris and Alan Resetar; ICNMNH—Maaria Christina Ardi- la-R. and Pedro M. Ruiz-C.; IND-AN— José Vicente Rueda-A.; KU—wWilliam E. Duellman and John E. Simmons; MCZ— José P Rosado and Ernest E. Williams; 716 MNHN—Alain Dubois; MZUSP—P. E. Vanzolini; RMNH—Marinus S. Hoog- moed; UMMZ—Ronald A. Nussbaum and Gregory Schneider; UTA—Jonathan A. Campbell and William FE Pyburn. Research for this paper was supported by the Instituto Nacional de Pesquisas da Amazonia as part of the ““‘Herpetofauna of Roraima”’ Project under the direction of Celso Morato de Carvalho; the Museu de Zoologia da Universidade de Sao Paulo and retired Director P E. Vanzolini; and the Smithsonian Institution’s Neotropical Low- lands Research Program, funded by the In- ternational Sciences Program and the Di- rector’s Office (USNM). Literature Cited Altig, R., & G. E Johnston. 1989. Guilds of anuran larvae: relationships among developmental modes, morphologies and habitats.—Herpeto- logical Monographs 3:81—109. Cardoso, A. J., & W. R. Heyer. 1995. Advertisement, aggressive, and possible seismic signals of the frog Leptodactylus syphax (Amphibia: Lepto- dactylidae).—Alytes 13:67—76. Donnelly, M. A., & C. W. Myers. 1991. Herpetolog- ical results of the 1990 Venezuelan expedition to the summit of Cerro Guaiquinima, with new PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON tepui reptiles—American Museum Novitates 3017:1-54. Egler, W. A. 1960. Contribuigdes ao conhecimento dos campos da Amazonia. I—Os campos do Ariramba.—Boletim do Museu Paraense Emilio Goeldi, nova série, Botanica 4:1—36 + 4 plates. Gosner, K. L. 1960. A simplified table for staging anuran embryos and larvae with notes on iden- tification.—Herpetologica 16:183—190. Heyer, W. R. 1979. Systematics of the pentadactylus species group of the frog genus Leptodactylus (Amphibia: Leptodactylidae).—Smithsonian Contributions to Zoology 301:1-43. , A. S. Rand, C. A. G. Cruz, O. L. Peixoto, & C. E. Nelson. 1990. Frogs of Boracéia.—Ar- quivos de Zoologia, Museu de Zoologia da Universidade de Sao Paulo 31:231-410. Kroonenberg, S. B. 1985. El borde occidental del Es- cudo de Guayana en Colombia. Pp. 51-63 in M. I. Mufoz, ed., I Simposium Amazonico. Boletin de Geologia, Publicacién Especial No. 10. Caracas. xxii + 768 pp. Leviton, A. E., R. H. Gibbs, Jr., E. Heal, & C. E. Dawson. 1985. Standards in herpetology and ichthyology: Part I. Standard symbolic codes for institutional resource collections in herpe- tology and ichthyology.—Copeia 1985:802— 832. Noble, G. K. 1923. New batrachians from the Tropical Research Station, British Guiana.—Zoologica 3:289-299. SYSTAT for Windows: Statistics, Version 5 Edition. 1992. Evanston, Illinois. SYSTAT, Inc. 750 pp. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(4):717-—728. 1995 The taxonomic status of the shrew of St. Lawrence Island, Bering Sea (Mammalia: Soricidae) R. L. Rausch and V. R. Rausch (RLR) Department of Comparative Medicine, University of Washington, Box 357190, Seattle, Washington 98195-7190, U.S.A.; (VRR) Burke Museum, University of Washington, Box 353010, Seattle, Washington 98195-3010, U.S.A. Abstract.—Shrews of the subgenus Otisorex inhabit two islands in the Be- ring Sea, within the limits of the former Beringian Refugium. Whether those taxa represent independent species, or subspecies of the nearctic Sorex cinereus Kerr, has remained uncertain. The karyotype (2N = 66, FN autosomes = 70) of one of these, described as Sorex jacksoni Hall et Gilmore from St. Lawrence Island, has been defined and compared with that of S. c. cinereus on the Alas- kan mainland, from which it could not be distinguished. No differences could be discerned between the two taxa in structure of the glans penis or in rela- tionships of the medial tines of the incisors. The shrew on St. Lawrence Island is regarded as being a subspecies of S. cinereus, for which the designation S. c. jacksoni Hall et Gilmore is applicable. A review of published karyograms and other information supports the concept that no species of the subgenus Otisorex is holarctic. Sorex cinereus Kerr has the most exten- Sive geographic range of any species of shrew in the Nearctic, occurring in North America from approximately lat. 35°N to the northern shores of Alaska and Canada (Hall 1981, map 14). Records of Rancho- labrean age indicate that its distribution had been more extensive in the southern regions of the continent (Kurtén & Anderson 1980: 105). The fossil record of that species dates from deposits of late Mindel-Kansan time, and possibly earlier. Diverse interpretations concerning the taxonomy and the geographic ranges of some shrews of the Sorex cinereus-group (subgenus Otisorex) have been given in the recent literature, especially relating to the northern forms. Hall (1981) distinguished 12 nearctic subspecies of S. cinereus, in- cluding S. c. jacksoni Hall et Gilmore, on St. Lawrence Island (Bering Sea), and S. c. ugyunak Anderson et Rand, which inhabits mainly the treeless regions of the continent from northwestern Alaska eastward to the western shores of Hudson Bay and Foxe- Basin. Another of the group, S. pribilofensis Merriam, occurs on St. Paul Island, Pribilof Islands (Bering Sea) (Hoffmann & Peterson 1967). In Eurasia, Stroganov (1956) de- scribed S. c. portenkoi from the vicinity of Anadyr’ (Chukotka) (approx. 64°40’N, 177°20’E), providing the first indication that S. cinereus might have an holarctic dis- tribution. Two additional subspecies have been recognized in northeastern Eurasia: S. c. camtschaticus ludin, 1972 (type locality: Kambal’ Bay, Ust’-Bol’sheretsk region, Kamchatka) (approx. 52°45'N, 156°30’B), and S. c. leucogaster Kuroda, 1933 (syn. S. beringianus ludin, 1967) (type locality: Paramushir Island, Kurile Islands). On the basis of morphometric analysis, van Zyll de Jong (1982) tentatively consid- ered that S. cinereus, S. jacksoni, and S. pri- bilofensis are independent species. Accord- ing to his concept, S. jacksoni would be 718 holarctic in distribution, with S. 7. jacksoni on St. Lawrence Island, S. j. ugyunak in northern North America, and S. j. portenkoi and S. j. leucogaster in Eurasia. The taxon previously designated S. c. camtschaticus was considered also to represent a distinct species. The results of van Zyll de Jong’s (1991) further analyses of cranial characters were taken to be compatible with those tax- onomic conclusions, except that leucogas- ter also was regarded as being an indepen- dent species. Van Zyll de Jong pointed out that his conclusions were tentative, and that study of other taxonomic criteria, particu- larly cytogenetic and biochemical, was needed. He earlier (1982) noted that ‘‘clar- ification of phallic morphology is needed.”’ Hutterer (1993) acknowledged van Zyll de Jong’s concept that ugyunak may be specif- ically distinct from cinereus. Junge & Hoff- mann (1981) considered S. jacksoni and S. pribilofensis (designated by them as S. hy- drodromus Dobson) to be distinct from S. cinereus. Okhotina (1984) retained the Eur- asian taxa as subspecies of S. cinereus. A different concept was presented by Ivanit- Skaia & Kozlovskii (1985), who proposed on the basis of chromosomal criteria that S. cinereus does not occur west of Bering Strait. They regarded S. c. ugyunak (without karyological data) as being a distinct spe- cies with an holarctic distribution, repre- sented in Eurasia by S. u. portenkoi. Zaitsev (1988) pointed out that in such case the trivial name ugyunak would have to be re- placed by portenkoi on grounds of priority. In their review of the systematics of mam- mals of the USSR, Pavlinov & Rossolimo (1987) concluded that S. cinereus does not occur in Eurasia, and that no species of the subgenus Otisorex is holarctic. They distin- guished three palaearctic taxa in that sub- genus: S. leucogaster, S. ?portenkoi, and S. camtschaticus. To contribute towards a resolution of some of the taxonomic uncertainties involv- ing shrews of the cinereus-group, we de- scribe here the karyotype of S. c. cinereus on the basis of shrews collected on the PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Alaskan mainland, and compare that of S. c. jacksoni (Fig. 1) on St. Lawrence Island. The structure of the glans penis and other taxonomic characters are also compared. Our attempt to collect ugyunak in the vicin- ity of Barrow, Alaska, during August 1994 was unsuccessful, and its karyotype has not been defined. Indications of its intergrada- tion with S. c. cinereus along the northern front of the Brooks Range, arctic Alaska, are briefly discussed. Material and Methods For convenience, and with respect to our conclusion concerning the status of one of the taxa studied, we apply the nomenclature of Hall (1981) for subspecies of Sorex ci- nereus. The specimens examined consisted of 239 shrews of the S. cinereus-group, col- lected by us and coworkers in Alaska dur- ing the period 1949-1975 and by us during 1980-1994. Specimens of the taxa consid- ered here included S. c. cinereus, 90; S. c. Jacksoni, 77; S. c. hollisteri Jackson, 58; and S. c. ugyunak, 14, along with measure- ments of nine additional specimens for which skulls were not retained. Standard data were recorded for all specimens col- lected, as well as information about repro- ductive status; skulls or skeletons were rou- tinely prepared. Since many of the shrews were collected primarily for helmintholog- ical investigations, skins were prepared only as time permitted; nonetheless, small series were available for all of the taxa con- sidered. For examination of the glandes, en- tire penes were usually fixed in extended condition in 10% formalin solution. One each from S. c. cinereus and S. c. jacksoni was stained in acetic carmine, dehydrated in ethanol, and cleared in terpineol for ex- amination. Two additional male specimens of S. c. jacksoni (August 1957 and May 1993) were used only for the study of the genital organs. Measurements of larger cra- nial dimensions were made by means of a dial-caliper, graduated in tenths of milli- meters and provided with a fine adjustment. VOLUME 108, NUMBER 4 TAS: Fig. 1. Lawrence Island. In measuring tooth-rows and other features of small size, a stereoscopic microscope with a calibrated scale, graduated in tenths of millimeters, was used. Degrees of lati- tude and longitude given below are approx- imate. Chromosomes were examined from 7 specimens of S. c. cinereus collected in Alaska as follows: 1 female (August 1972), Chena Hot Springs Road (64°53’N, 147°W); 1 female (August 1993), 6 km NE of Palmer (61°39'N, 149°13'W); 1 female, 1 male (August 1991), 16 km N of Gakona Junction Village (62°24'’N, 145°22’W); 1 male (August 1993), 35 km S of Paxson (62°52'N, 145°29'W); 2 females (August 1993), 17 km S of Copper Center (61°51'N, 145°15'W). From S. c. jacksoni, prepara- tions were made from 5 adults and 2 em- bryos collected near Savoonga, St. Law- rence Island (63°42’N, 170°29'W): 3 males (June 1980, June 1984, June 1987); 1 fe- male (June 1987), and 1 female (June 1988), with one female and one male em- bryo. Cells from marrow and lymphatic tissue Sorex cinereus jacksoni. Photographed on 15 June 1992, approximately 1 km west of Savoonga, St. were treated with colchicine and hypotonic solution, centrifuged, fixed, and placed on slides in the field, where also testicular tu- bules were fixed and stained in acetic or- cein, employing standard karyological pro- cedures. Most preparations were stained in the laboratory at the University of Wash- ington; in some cases, only orcein-staining was feasible. Banding of chromosomes was produced by application, for G-banding, of the method of Seabright (1972), and for C- bands, that of Sumner (1972). Chromo- somes were counted and evaluated in intact cells in metaphase; at least 10 cells were photographed from each animal, from which, in non-banded complements (stan- dard Giemsa stain), measurements were made as recommended by Levan et al. (1964). In karyograms constructed for com- parisons, arm-ratios and size were the bases for assembling pairs of non-banded chro- mosomes. Those with G-bands were distin- guished according to banding pattern and size; C-banded chromosomes were identi- fied by size and by location of centromeric heterochromatin. The fundamental number 720 qe ON at a3 “> oe E - eb eh 98 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON TaCa\ Mime $¢ 58 ee Gé ge bb 06 68 ae €6 8&@ @t ee KY. Fig. 2. Karyogram of Sorex c. cinereus, male. Giemsa-banding. Scale-bar represents 5 micrometers. (FN) of major chromosomal arms was de- termined according to the method of Matth- ey (1945). Voucher-specimens of the two subspecies of Sorex cinereus have been deposited as follows: Burke Memorial Washington State Museum, University of Washington, Seat- tle, No. 39403 (Sorex cinereus jacksoni); and Museum of Southwestern Biology, University of New Mexico, Albuquerque, Nos. 69701, 69702, 69703 (Sorex cinereus jJacksoni), and Nos. 69704 and 69705 (So- rex cinereus cinereus). Results Karyograms.—Sorex cinereus cinereus: The diploid complement (66) consisted of the following: autosomes—centromere in subterminal to near-terminal region (pairs 1-31) (range of arm-ratio 3.8 to 12.0); cen- tromere in median region (pair 32) (arm- ratio 1.2 to 1.4); sex chromosomes—X-chro- mosome with centromere in submedian re- gion (range of arm-ratio 2.0 to 2.7), in metaphase usually approximately equal in total length to that of largest autosomes; Y- chromosome with centromere in subtermi- nal region (arm-ratio not determined). The Y-chromosome (in metaphase) was the smallest of the complement (Fig. 2). The FN was calculated to be 70. The distribu- tion of constitutive heterochromatin in the diploid complement (C-banded) is shown in Fig. 3. Meylan (1968) preliminarily reported a diploid number of 66 and FN of 70 for two female specimens of S. c. cinereus collected in the Province of Ontario, Canada; that publication appeared later than his descrip- tion and illustration of the chromosomal complement (Meylan 1967). Recently, Vo- lobouev & van Zyll de Jong (1994) provid- ed karyograms of a male shrew of that spe- cies from Pennsylvania. We conclude. that the aforementioned specimens, all collected near the eastern limits of the range of S. c. cinereus, were karyotypically identical with those obtained in Alaska. The karyotype of S. c. cinereus is probably uniform through- out its geographic range. VOLUME 108, NUMBER 4 G66 6A OA 46 Ah 06 Ah fe &G && 4h && Fig. 3: Sorex cinereus jacksoni: The diploid complement (66) consisted of the follow- ing: autosomes—centromere in subterminal to near-terminal region in 31 pairs (1-31) (range of arm-ratio 3.6 to 13.6); centromere in median region (pair 32) (arm-ratio of 1.1 to 1.6); sex chromosomes—X-chromosome with centromere in submedian region (range of arm-ratio 1.7 to 2.6); Y-chromo- some (metaphase) was the smallest in the complement (Fig. 4). As in S. c. cinereus, the FN of S. c. jacksoni was 70. The prep- arations from all but two were processed fhR OA AG OA Aa Q@a4 en 44 A@ aéa &4@ G66 & 4 & & dy & $@ 72 46 @8 68 64 @6@ AR GA 84 > ® @e Ge €&¢ Ke OY Karyogram of S. c. cinereus, male. C-banding. Scale-bar represents 5 micrometers. with acetic orcein; G-banded chromosomes were not of uniform quality, but homo- logues could be identified and compared by selecting the best individuals from each preparation. We were unable to discern karyological differences between S. c. ci- nereus and S. c. jacksoni. The glans penis.—The glans penis of S. c. cinereus, first illustrated by Iudin (1969), differs in form from that of any other spe- cies of Sorex for which information is avail- able. We compared the glandes of six spec- imens of S. c. jacksoni with four from S. c. 64 40 Qn nad AA AA G4 4A Ahn AA BA AG A & Oo & 4 & £ & ¥ Ue xX Y Fig. 4. Karyogram of S. c. jacksoni, male. Orcein. Scale-bar represents 5 micrometers. V2 = =_ e Fig? 5: PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Glans penis of S. c. cinereus (a—c) and S. c. jacksoni (d—e). a, lateral view of penis; b, dorsal view of glans; c, en face view of glans; d, semi-lateral view of penis; e, en face view of glans. Scale-bar represents 1 millimeter. cinereus. The penes of both were of ap- proximately uniform diameter distally, and were about | to 1.5 mm in diameter just posterior to the glans. When not fully ex- tended, the surface of the penis immediately posterior to the glans exhibited folds, some of which extended around the entire cir- cumference of the organ. In our specimens, such folding of the superficial tissue was not so strongly defined as was portrayed by Iudin (1969). In lateral view, the anterior surface of the glans appeared truncate to curved, with its distal portion projecting ventrad or ventroposteriad, depending on degree of contraction (Fig. 5a). In en face view, the glans was more or less cordate in shape, and was bisected by a medial groove extending anteriad and ventrad from the dorsal surface to the end of the ventral pro- trusion (Fig. 5c). The orifice of the urethra was situated in that groove near the middle of the anterior surface of the glans, opening at the apex of a small papilla, which in some cases protruded slightly above the surrounding surface of the glans (Fig. 5a, b). In one specimen, the end of the urethra protruded from the papilla, appearing as a processus urethrae; probably, however, it was a Structural artifact, produced in the process of removing the organ for fixation. It was evident that the glans is subject to muscular control, which may cause slight modification in its shape. The glandes of S. c. jacksoni (Fig. 5d, e) were indistinguish- able from those of S. c. cinereus. A fully relaxed penis of S. c. jacksoni was 19 mm long from its point of attachment to its dis- tal end. The animal (TL 91 mm) was in breeding condition, with testes measuring 5 x 3.5 mm. Also compared was the glans penis of one specimen of S. c. hollisteri col- lected at Hooper Bay (Kuskokwim- Yukon VOLUME 108, NUMBER 4 delta) (61°27'’N, 166°W) in May 1966 and preserved intact in formalin solution. The glans was indistinguishable from that of S. c. cinereus and S. c. jacksoni. Other morphological characters.—Cra- nial characters of shrews of the cinereus- group have been compared by van Zyll de Jong (1976, 1982, 1991). We found in S. c. jacksoni that the accessory medial tines of the incisors, also important taxonomically (Junge & Hoffmann 1981; Dannelid 1989, 1994) did not differ in position from those of S. c. cinereus, S. c. hollisteri, and S. c. streatori, from the Alaskan mainland, but were perhaps slightly greater in length. Like some other small mammals inhabiting is- lands in the Bering Sea, S. c. jacksoni 1s somewhat larger than shrews representing the various subspecies of S. cinereus on the Alaskan mainland. External measurements (Table 1) were based on a series of 38 males and 23 females, judged to be adults on the basis of the reproductive status of those col- lected during July and August, and on the large size of those obtained during Septem- ber through November. Another series of 23 specimens collected in June, considered to be over-wintered adults, averaged some- what smaller (Table 1). Compared with specimens of S. c. cinereus collected in east-central Alaska during August, mean to- tal lengths of S. c. jacksoni were greater and tail-length was less. Weights were not ob- tained for the series of S. c. cinereus. Youngman (1975) found weights of nine specimens from the southern Yukon Terri- tory to range from 3.1 to 5.1 g (X = 4.3 g). Mean total length for a series of 18 speci- mens of S. c. hollisteri (one collected in June, the remainder in August) from Na- paskiak (62°41’N, 161°54’W) on the lower Kuskokwim River was similar to that of males of S. c. jacksoni; S. c. jacksoni had a shorter tail, and the mean body-weight was greater (Table 1). For the series of S. c. Jacksoni collected in summer, weights of 13 non-pregnant females ranged from 4.8 to 7.3 g (X = 5.0 g). In 11 pregnant animals, Table 1—Body measurements (mm) and weights (g) of some subspecies of Sorex cinereus. Weight Hind foot Length tail Total length Range SD Range SD SD Range Subspecies 4.1-7.0 5.5 (0.81) 4.3-8.8 6.3 (1.20) s5) OS) I 22 12.4 (0.44) 12.3 (0.43) 12.5 (0.48) W255) (O52) 12-13 11.5-13 38 66 Beye 2 13 66 ie 2 32-38 35.2 (1.39) 38 3d De) OO 13 366 iS 2 92-103 97.8 (2.67) 89-105 99.4 (4.10) 91-100 94.9 (3.47) 91-104 95.7 (5.60) 3156S jJacksoni 32-37 34.5 (1.61) WE Ls? IB SS 1EONS (July—Nov) jJacksoni 12-13 30-38 34.8 (2.34) 12-13.5 31-36 34.3 (2.09) (June) cinereus 12.2 (1.06) 10-13 12 36-43 38.5 (2.21) 12 12 89-101 93.2 (3.78) COQ e2 cS) hollisteri 3.8-6.5 4.5 (0.75) 14 11-13 Ie (OFS5)) 18 32-41 37.2 (2.19) 18 91-101 97.7 (2.72) 18 @ 2 WSS BY) 723 724 the numbers of embryos ranged from 5 to 14, with a mean of about 10 (9.7). Discussion With respect to the relationships and dis- tributional status of shrews of the subgenus Otisorex, two differing hypotheses have been proposed: Sorex jacksoni may be an independent species, with three subspecies, S. j. jacksoni (St. Lawrence Island), S. /. portenkoi (NE Siberia), and S. j. ugyunak (tundra of NW North America); S. pribilo- fensis (St. Paul Island), S. leucogaster (Par- amushir Island), and S. camtschaticus (NE Siberia) would each be specifically distinct; or jacksoni and leucogaster may be closely related if not conspecific, and ugyunak would be an independent species, repre- sented in North America by S. u. ugyunak and in Eurasia by S. u. portenkoi. Since portenkoi (a prior name, see introductory remarks) would then replace ugyunak, the subgenus Otisorex would be represented in Eurasia by the species /eucogaster, cam- tschaticus, and portenkoi, of which the last would be holarctic. Karyological investi- gations, although incomplete, have been helpful in resolving some of the questions about conspecificities. As described and illustrated by Ivanit- Skaia & Kozlovsku (1985), the karyotype of portenkoi (2N = 60, FN = 60) distin- guishes that taxon from S. c. cinereus and S. c. jacksoni (2N = 66, FN = 70). The karyogram of leucogaster, however, is like that of the latter two taxa with respect to diploid number, FN, and perhaps morpho- logical characteristics of the chromosomes. Nonetheless, /eucogaster is considered to be specifically distinct from S. cinereus (Ivanitskaia & Kozlovskii 1985, Pavlinov & Rossolimo 1987, van Zyll de Jong 1991). A significant distinguishing character, as shown by Iudin (1969, 1971), is the glans penis of S. beringianus ludin (= S. leuco- gaster), which differs from that of S. ciner- eus in being rounded and smooth, described by Iudin (1969:29) as having “*... a mas- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON sive, globular, rounded crown,”’ with an ap- parent corona, and lacking the ventral ex- tension typical of the glans of S. cinereus. Definition of karyotypes can be expected to clarify the affinities of pribilofensis and ugyunak. We examined the glans penis of only one specimen of S. pribilofensis, dis- sected from an animal preserved in forma- lin, and it was not suitable for detailed com- parison. Some indications exist that ugyu- nak will be found to represent a subspecies of cinereus. A converse possibility was in- dicated in the vicinity of the Mackenzie Delta, Northwest Territories (Canada), where cinereus and ugyunak appeared to be associated with taiga and tundra, respec- tively, with no reported evidence of inter- gradation (Martell & Pearson 1978). With reference to S. c. cinereus, Youngman (1975:41) remarked that “‘Shrews of this subspecies become smaller in a cline from the central Yukon to the northern part of the Territory, where they intergrade with the smaller S. c. ugyunak,’’ and evidence for intergradation of ugyunak with cinereus and possibly with hollisteri was reported by Bee & Hall (1956) in northern Alaska. Twenty- three specimens collected by us along the northern margin of the Brooks Range, be- tween about lat. 67°57'N and 68°20'N (en- compassing taiga as well as tundra) were designated S. c. ugyunak; body-measure- ments and cranial dimensions of them were interpreted as indicating intergradation with S. c. cinereus. The medial tines of the in- cisors of those specimens were like those of the latter taxon. (The data concerning ug- yunak will be reported elsewhere.) Analysis of morphometric data may de- fine precisely the degree of difference ex- isting among taxa, but does not necessarily provide a basis for establishing taxonomic rank. As well, insular species may present unusual difficulties if conclusions are de- rived solely from macromorphological data, without consideration of potentiality for ge- netic drift and of the rate at which morpho- logical divergence may have taken place. The small mammals on St. Lawrence Is- VOLUME 108, NUMBER 4 land, with the possible exception of the varying lemming there, Dicrostonyx exsul Allen, have been isolated for only about 10,000 years or less; that island was the last of the Beringian highlands to become sep- arated from the continents by rising sea lev- el at the end of the Pleistocene period (Hop- kins 1959, 1976). In addition to the shrew and the varying lemming, the indigenous terrestrial mammals of the island (excluding the arctic fox, Alopex lagopus (L.), which immigrates and emigrates freely on the sea- ice) consist of the northern red-backed vole, Clethrionomys rutilus (Pallas); the northern vole, Microtus oeconomus (Pallas); and the arctic ground squirrel, Citellus parryi (Richardson). The mammalian faunas of the other Beringian islands (not including the upper Aleutian Islands) are more depauper- ate: a single species, Microtus abbreviatus Miller, on the St. Matthew Islands; the brown lemming, Lemmus sibiricus. (Kerr), on St. George Island and Sorex pribilofensis on St. Paul Island (Pribilof Islands), and L. sibiricus and Dicrostonyx vinogradovi Og- nev, on Vrangel’ Island. With the exception of the varying lemmings, S. pribilofensis, and S. jacksoni (about which questions have remained) all of the small mammals on the Beringian Islands are recognized as being only subspecifically distinct from their pre- cursors on. the continents. All but the brown. lemmings are characterized in part by larger size, most strongly defined in the voles. Mi- crotus abbreviatus on the St. Matthew Is- lands is not only much larger than any of the continental subspecies (Alaska and NW Canada), but it exhibits the greatest degree of phenotypic divergence among the insular taxa mentioned (Rausch & Rausch 1968). Microtus oeconomus on St. Lawrence Is- land also is very large (TL up to at least 210 mm); only M. o. koreni Allen, inhab- iting the valleys of the Kolyma and _ Indi- girka Rivers, in Chukotka, approaches it in size. The red-backed vole on St. Lawrence Island (TL up to at least 155 mm) is larger than subspecies of Clethrionomys rutilus in either Chukotka or Alaska, approaching the 725 northeastern Siberian red-grey vole, C. ru- focanus (Sundevall) in size. Comparisons have shown that the karyotypes of S. ci- nereus, L. sibiricus, M. abbreviatus, and M. oeconomus are indistinguishable from those of conspecifics on the continents (L. sibir- icus and M. oeconomus are holarctic). Clethrionomys rutilus is also indistinguish- able karyotypically, except that chromo- somal polymorphism, involving a Robert- sonian rearrangement, was observed among animals collected at one locality (Sevuokuk Mountain) on St. Lawrence Island, possibly indicating that chromosomal evolution is taking place (Rausch & Rausch 1975b). Like C. rutilus in Alaska and Chukotka, C. rutilus on St. Lawrence Island has a meta- centric Y-chromosome (Vorontsov et al. 1978). When the arvicolids were crossed with respective conspecifics from the Alas- kan mainland, the offspring were found to be interfertile for numerous generations (Rausch & Rausch 1968, 1975a, 1975b, and unpubl.). Intergrades were bred within each successive generation. Such breeding was carried to about the 30th generation in the case of C. rutilus albiventer X C. r. daw- soni, and to at least several generations with the other species. With successive genera- tions, the offspring of M. abbreviatus fish- ert Merriam X M. a. muriei Nelson, from the Alaskan mainland, increasingly resem- bled the latter in size and cranial character- istics (R. L. Rausch, unpubl.). Such inves- tigations involving S. c. jacksoni have not been attempted, since shrews are compara- tively difficult to breed in captivity. Rela- tive to the biological species-concept, we conclude that none of the aforementioned species, with exceptions as noted, is repro- ductively isolated. The varying lemmings on the Beringian Islands, D. exsul on St. Lawrence Island, and D. vinogradovi on Vrangel’ Island, are karyotypically distinct (Rausch 1977, Ko- zlovski & Khvorostianskaia 1978). Both are restricted to habitat at higher elevations. On St. Lawrence Island, the wet tundra that covers 60% of the surface-area is occupied 726 by the northern vole and the shrew, and the lowlands of Vrangel’ Island are inhabited by brown lemmings. The limitation of vary- ing lemmings to areas of higher elevation on the two islands may be attributable to the competitive superiority of the northern vole and the brown lemming, respectively (Rausch & Rausch 1975a). Among the spe- cies on the Beringian islands, Dicrostonyx spp. may have been the most ancient col- onizers of the highlands that are now is- lands. Lemmings of the genus Praedicros- tonyx evidently spread into eastern Beringia more than a million years ago, where they were replaced by Dicrostonyx (subgenus Misothermus) in pre-Mindel time (Zazhigin 1976). Dicrostonyx (subgenus Dicrostonyx) first appeared in deposits of Riss age. The degree of karyotypic diversity among Re- cent varying lemmings and the vast extent of their geographic distribution in the Ne- arctic (including apparently all of the 1is- lands of the Canadian Arctic Archipelago as well as Greenland) are indicative of long isolation of the respective populations. Of the small mammals on the Beringian is- lands, probably only the varying lemmings had been long established in restricted hab- itat by the time the mammals of the other species were separated from continental populations at the end of the last glacial pe- riod. In insular populations, phenotypic changes sufficient to permit characteriza- tion of subspecies may take place compar- atively rapidly [cf. Degerbgl (1939) for Apodemus sylvaticus islandicus Thiene- mann; Huxley (1943) for Mus musculus L.; and Cameron (1958) for mammals on New- foundland]. All of the small mammals on islands within the area of Beringia were de- scribed originally as independent species. The period of time elapsing since their sep- aration has been sufficient to mark most of those taxa as different, but only infraspe- cifically so, from mainland populations. With the exception of the varying lem- mings, and possibly Sorex pribilofensis, none is known to have become reproduc- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON tively isolated during the relatively brief in- terval since the end of the Pleistocene ep- och. Acknowledgments Many of the shrews used in this study were collected by the late Dr. Francis H. Fay, in connection with our investigation of the epidemiology of alveolar hydatid dis- ease on St. Lawrence Island. There, the hospitality and assistance of the Siberian Yupik residents greatly facilitated our work. At the Arctic Health Research Center, col- lections were made by Dr. Francis S. L. Williamson, and specimens were provided also by Mr. Donald Hartbauer, the late Dr. Christine Heller, Mr. Leonard Peyton, and Mr. Davis Rhodes, and at the Alaska Native Medical Center, by Dr. Joseph Wilson. Mr. L. Alan LeMaster and Mrs. Shirley Le- Master kindly provided space for laboratory work at Gakona. Mr. Warren Matumeak, Dr. Thomas EF Albert, and other members of the Department of Wildlife Management, North Slope Borough, Barrow, Alaska, gave sup- port in connection with field-work. At the University of Washington, Ms. Gladys Crespo generously advised concerning karyological procedures. We express our sincere thanks for these contributions. Literature Cited Bee, J. W., & E. R. Hall. 1956. Mammals of northern Alaska on the arctic slope.—University of Kan- sas, Museum of Natural History, Miscellaneous Publications 8:1—309. Cameron, A. W. 1958. Mammals of the islands in the Gulf of St. Lawrence. National Museum of Canada, Bulletin 154, Ottawa, 165 pp. Dannelid, E. 1989. Medial tines on the upper incisors and other dental features used as identification characters in European shrews of the genus So- rex (Mammalia, Soricidae).—Zeitschrift ftir Saugetierkunde 54:205—214. . 1994. Comparison of pigment and other den- tal characters of eastern palearctic Sorex (Mam- malia: Soricidae). Pp. 217-231 in J. FE Merritt, G. L. Kirkland, Jr., and R. K. Rose, eds., Ad- vances in the biology of shrews. Carnegie Mu- seum of Natural History, Special Publication No. 18, Pittsburgh, 458 pp. VOLUME 108, NUMBER 4 Degerbgl, M. 1939. The field mouse of Iceland, its systematic position (Apodemus sylvaticus gran- diculus subsp. nov.) and biology.—Zoology of Iceland 4:39-51. Hall, E. R. 1981. The mammals of North America. 2nd Edition. Vol. 1. John Wiley & Sons, New York, 600 + 90 pp. Hoffmann, R. S., & R. S. Peterson. 1967. Systematics and zoogeography of Sorex in the Bering Strait area.—Systematic Zoology 16:127—136. Hopkins, D. M. 1959. Cenozoic history of the Bering land bridge.—Science 129:1519-1528. (Khopkins). 1976. Istoriia urovnia moria v Beringii za poslednie 250,000 let. Pp. 9-27 in V. L. Kontrimavichus, ed., Beringiia v Kaino- zoe. Akademiia Nauk SSSR, Vladivostok, 594 Pp. Hutterer, R. 1993. Order Insectivora. Pp. 69-130 in D. E. Wilson and D-A. M. Reeder, eds., Mam- mal species of the world. A taxonomic and geo- graphic reference. 2nd Edition. Smithsonian In- stitution Press, Washington, D.C., 1206 pp. Huxley, J. 1943. Evolution, the modern synthesis. Harper and Bros., New York, 645 pp. Iudin, B. S. 1969. Novye dannye po systematike nek- otorykh vidov zemleroek (Soricidae) Palearktiki i Nearktiki.—Acta Theriologica 14:21—34. 1971. Nasekomoiadnye mlekopitaiushchie Sibiri (opredelitel’). Akademiia Nauk SSSR, Novosibirsk, 171 pp. Ivanitskaia, E. Iu., & A. I. Kozlovskii. 1985. Karioti- py palearkticheskikh zemleroek-burozubok podroda Otisorex c kommentariami po siste- matike i filogenii gruppi “‘cinereus.”’—Zoologi- cheskii Zhurnal 64:950-953. Junge, J. A., & R. S. Hoffmann. 1981. An annotated key to the long-tailed shrews (genus Sorex) of the United States and Canada, with notes on middle American Sorex.—Occasional Papers of the Museum of Natural History, The University of Kansas, Lawrence, Kansas, No. 94, 48 pp. Kozlovskii, A. I., & L. P. Khvorostianskaia. 1978. Khromosomnye nabory kopytnykh lemmingov Ostrova vrangelia. Pp. 100-105 in V. L. Kon- trimavichus, ed., Fauni i zoogeografiia mleko- pitaiushchikh severo-vostoka Sibiri. Akademiia Nauk SSSR, Vladivostok, 168 pp. Kurtén, B., & E. Anderson. 1980. Pleistocene mam- mals of North America. Columbia University Press, New York, 442 pp. Levan, A., K. Fredga, & A. A. Sandberg. 1964. No- menclature for centromeric position on chro- mosomes.—Hereditas 52:200—220. Martell, A. M., & A. M. Pearson. 1978. The small mammals of the Mackenzie Delta region, Northwest Territories, Canada.—Arctic 31:475— 488. Matthey, R. 1945. Lévolution de la formule chro- 27. mosomiale chez les vertébrés.—Experientia 1: 50-60, 78-86. Meylan, A. 1967. Chromosomes of four species of shrews (Soricidae, Insectivora).—Mammalian Chromosomes Newsletter 8:187—190. 1968. Formules chromosomiques de quel- ques petits mammiferes nord-américains.—Re- vue Suisse de Zoologie 75:691—-696. Okhotina, M. V. 1984. Otriad Insectivora Bowdich, 1821—nasekomoiadnye. Pp. 31-72 in V. G. Krivosheev, ed., Nazemnye mlekopitaiushchie dal’nego vostoka SSSR. Opredelitel’. Nauka, Moskva, 358 pp. Pavlinovy, I. Ia., & O. L. Rossolimo. 1987. Sistematika mlekopitaiushchikh SSSR. Izdatel’stvo Mos- kovskogo Universiteta, Moskva, 284 pp. Rausch, R. L. (Raush). 1977. O zoogeografii neko- torykh beringiiskikh mlekopitaiushchikh. Pp. 162-175 in V. E. Sokolov, ed., Uspekhi sovre- mennoi teriologii. Nauka, Moskva, 296 pp. , & V. R. Rausch. 1968. On the biology and systematic position of Microtus abbreviatus Miller, a vole endemic to the St. Matthew Is- lands, Bering Sea.—Zeitschrift fiir Saugetier- kunde 33:65—99. , & . 1975a. Taxonomy and zoogeog- raphy of Lemmus spp. (Rodentia: Arvicolinae), with notes on laboratory-reared lemmings.— Zeitschrift fiir Sdugetierkunde 40:8-34. , & . 1975b. Relationships of the red- backed vole, Clethrionomys rutilus (Pallas), in North America: karyotypes of the subspecies dawsoni and albiventer.—Systematic Zoology 24:163—170. Seabright, M. 1972. The use of proteolytic enzymes for the mapping of structural rearrangements in the chromosomes of man.—Chromosoma 36: 204-210. Stroganov, S. V. 1956. Novyi dlia fauny Sibiri vid zemleroiki.—Trudy Biologicheskogo Instituta, Vypusk 1, Zoologicheskii, pp. 11-14. Sumner, A. T. 1972. A simple technique for demon- strating centromeric heterochromatin.—Experi- mental Cell Research 75:304—306. Van Zyll de Jong, C. G. 1976. A comparison between woodland and tundra forms of the common shrew (Sorex cinereus).—Canadian Journal of Zoology 54:963-973. 1982. Relationships of amphiberingian shrews of the Sorex cinereus group.—Canadian Journal of Zoology 60:1580—1587. . 1991. Speciation in the Sorex cinereus group. Pp. 65-73 in J. S. Findley and T. L. Yates, eds., The biology of the Soricidae. Special publica- tion, The Museum of Southwestern Biology, No. 1, University of New Mexico, Albuquer- que, 91 pp. Volobouev, V. T., & C. G. van Zyll de Jong. 1994. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Chromosome banding analysis of two shrews of the cinereus group: Sorex haydeni and Sorex ci- nereus (Insectivora, Soricidae).—Canadian Journal of Zoology 72:958—964. Vorontsov, N. N., E. A. Liapunova, E. Iu. Ivanitskaia, Ch. E Nadler, B. Kral, A. I. Kozlovskii, & R. S. Khoffman. 1978. Izmenchivost’ polovykh khromosom mlekopitaiushchikh. Soobshchenie I. Geograficheskaia izmenchivost’ stroenniia Y- khromosomy u_ polevok roda_ Clethrionomys (Rodentia, Microtinae).—Genetika 14:1432— 1446. Youngman, P. M. 1975. Mammals of the Yukon Ter- ritory. National Museums of Canada, Publica- tions in Zoology, No. 10. Ottawa, 192 pp. Zaitsev, M. V. 1988. O nomenklature zemleroek-bu- rozubok roda Sorex fauny SSSR.—Zoologi- cheskii Zhurnal 67:1878—1888. Zazhigin, V. S. 1976. Rannie etapy evoliutsii kopyt- nykh lemmingov (Dicrostonychini, Microtinae, Rodentia)—kharakternykh predstavitelei su- barkticheskoi fauny Beringii. Pp. 280-288 in V. L. Kontrimavichus, ed., Beringiia v Kainozoe. Akademiia Nauk SSSR, Vladivostok, 594 pp. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(4):729-747. 1995 Taxonomy of the genus Lycalopex (Carnivora: Canidae) in Argentina Gabriel E. Zunino, Olga B. Vaccaro, Marcelo Canevari, and Alfred L. Gardner (GEZ, OBV, MC) Museo Argentino de Ciencias Naturales, Division de Mastozoologia, Buenos Aires, Argentina; (ALG) Biological Survey, National Biological Service, National Museum of Natural History, Washington, D.C. 20560, U.S.A. Abstract.—Previously treated as species of Pseudalopex, Argentine members of the genus Lycalopex (L. griseus, L. gymnocercus, and L. culpaeus) are ex- amined to clarify the taxonomic status of each named form. Principal com- ponents analyses of 26 cranial measurements of 151 adult specimens and 11 pelage characters of 111 specimens, clearly distinguish L. culpaeus from the other two taxa. Lycalopex griseus and L. gymnocercus show clinal variation in cranial measurements and pelage characters. Qualitative cranial characters, tra- ditionally used as diagnostic for L. griseus and L. gymnocercus, revealed great nongeographic variation. We conclude that L. griseus and L. gymnocercus are conspecific, and should be known as L. gymnocercus. Therefore, we recognize only two species of the genus Lycalopex (L. culpaeus and L. gymnocercus) in Argentina. We also use this opportunity to review synonymies of the recognized species in Lycalopex. Although the taxonomic status of South American canids has been reviewed at length, classification at generic and subge- neric levels has not been completely re- solved (Berta 1988, Kraglievich 1930, Langguth 1975). Three of the living Argen- tine foxes (Lycalopex culpaeus, L. gymno- cercus, and L. griseus) have been alterna- tively included in the genera: Pseudalopex Burmeister, 1856 (Cabrera 1931, 1932, 1940; Thomas 1914a; Berta 1987, 1988; Nowak 1991; Wozencraft 1993), Dusicyon Hamilton-Smith, 1839 (Cabrera 1958, Cor- bet and Hill 1991, Clutton-Brock et al. 1976, Langguth 1969, Osgood 1934, Simp- son 1945, Wozencraft 1989), or Canis Lin- naeus, 1758 (Kraglievich 1930, Langguth 1975, Van Gelder 1978). Their inclusion in each of these genera was based on the de- gree of comparability with D. australis (Kerr, 1792), type species of genus Dusi- cyon, aS determined from studies of exter- nal, cranial, and dental characters. Cabrera (1931, 1932) and Berta (1987, 1988) sug- gested that culpaeus, gymnocercus, and gri- seus were sufficiently different from D. australis to place them in the genus Pseu- dalopex, an arrangement followed by Woz- encraft (1993). Cabrera (1931, 1958) con- sidered Lycalopex Burmeister, 1854 (type species Canis vetulus Lund) separate from Pseudalopex Burmeister, 1856 (type species Canis magellanicus Gray). However, Berta (1987, 1988) and Wozencraft (1993) treated them as congeneric, but used Pseudalopex for the genus. Corbet and Hill (1991) also treated them as congeneric, but used Dusi- cyon for the genus. We follow Berta and Wozencraft in treating Lycalopex vetulus as congeneric with species previously as- signed to Pseudalopex, but use Lycalopex as the valid generic name because it has two-years priority. The taxonomic status of Argentine foxes has rarely been evaluated since their origi- nal descriptions. Three living species of ge- nus Lycalopex currently are recognized in Argentina: the culpeo fox, L. culpaeus, 730 which inhabits mainly grasslands (pampas) and deciduous forests from southern Pata- gonia to Ecuador; the patagonian gray fox, L. griseus, which occurs from Atacama, Chile, and Santiago del Estero, Argentina, south to Tierra del Fuego, and is sympatric with L. culpaeus in part of its range; and the pampas gray fox, L. gymnocercus, which inhabits the humid grasslands (pam- pas) in southern Brazil, northern Argentina, Uruguay, Paraguay, and eastern Bolivia (Cabrera & Yepes 1940, Cabrera 1958, No- wak 1991, Wozencraft 1993). However, in- formation on distributional limits of L. gri- seus and L. gymnocercus in central Argen- tina, especially in the provinces of La Pam- pa, Cordoba, and San Luis, is scanty (Kraglievich 1930, Cabrera 1932). Clutton-Brock et al. (1976) concluded that L. culpaeus and L. gymnocercus were phenetically close, and suggested that they should be considered conspecific because of cranial and pelage similarities. The skull of L. griseus has little to distinguish it from that of L. culpaeus except for smaller size and lack of an interparietal crest. Our objectives in this paper are to ana- lyze geographic variations of dental, crani- al, and pelage characters and to evaluate the taxonomic relationships of L. culpaeus, L. gymnocercus, and L. griseus. We also use this opportunity to correct several errors, omissions, and misallocations of names dis- covered during our research, and to provide a synonymy that includes first usage of unique name combinations applied to the recognized species in Lycalopex. Materials and Methods We examined 151 skulls and 111 skins (see Appendix 1) deposited in the following collections: Museo Argentino de Ciencias Naturales “‘Bernardino Rivadavia’? (MACN) and Instituto Miguel Lillo de Tucuman (IML). We used only specimens of adults with intact skulls and complete data in the sta- tistical analyses. We considered specimens PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON to be adult if the basisphenoid-basioccipital suture was closed and the permanent den- tition was complete. We measured 30 cra- nial and dental characters with a dial caliper to the nearest 0.1 mm on each skull, with bilateral characters always measured on the left side. Measurements are defined as follows: 1) condylobasal length: distance from anterior edge of premaxillae to posteriormost sur- face of occipital condyles; 2) palatal length: distance from anterior margin of premaxil- lae to anteriormost point on posterior edge of palate; 3) minimum rostral length: dis- tance from anterior margin of premaxillae to anteriormost orbital margin; 4) facial length: distance from anterior edge of pre- maxillae to posterior end of nasal bones; 5) bullar length: from sharp anterior end of tympanic bulla diagonally to posterior mar- gin adjacent to paroccipital process; 6) maxillary toothrow length: alveolar length of upper toothrow including canine and sec- ond upper molar; 7) P4—M2 length: alveolar length of upper toothrow including fourth upper premolar and second upper molar; 8) minimum rostral height: rostral height be- hind canines; 9) height of braincase: height measured at right angle to basisphenoid, ex- cluding sagittal crest; 10) interorbital con- striction: least distance between orbits; 11) postorbital constriction: least distance across constriction just posterior to postor- bital processes; 12) rostral width at canines: width across the rostrum at canines, mea- sured from outer margins of alveoli; 13) rostral width at first upper molars: width across the rostrum at first upper molars, measured from outer margins of alveoli; 14) zygomatic breadth: greatest distance across outer margins of zygomatic arches; 15) mastoidal width: greatest distance across mastoid processes; 16) braincase width; greatest width across parietals; 17) width across occipital condyles: greatest distance between outer margins of occipital condyles; 18) width between postglenoid processes: least distance between inner margins of postglenoid processes; 19) P4 VOLUME 108, NUMBER 4 length: greatest alveolar length of fourth upper premolar; 20) P4 width: greatest al- veolar width of fourth upper premolar; 21) ml length: greatest alveolar length of first lower molar; 22) ml width: greatest alveolar width of first lower molar; 23) rostral height at P2-P3: rostral height measured between second and third upper premolars; 24) man- dibular length: length of mandible from an- teriormost part of mandibular symphysis to posteriormost part of articular process; 25) mandibular toothrow length: alveolar length of lower toothrow including canine and third lower molar; 26) ml—m3 length: al- veolar length of lower molars; 27) alveolar length of upper canine: greatest alveolar length of upper canine; 28) nares length: distance from anterior margin of premaxil- lae to anteriormost margin of nasal bones; 29) width across postorbital processes: greatest distance across outer margins of postorbital processes; 30) length between postorbital constriction and postorbital pro- cesses: least distance measured along sag- ittal suture. The last four measurements were not included in statistical analyses due to their high individual variability. We grouped collecting sites of gray foxes into six OTUs, each representing a geo- graphically discrete region (Fig. 1, Appen- dix 1). One specimen from La Paz, Men- doza, coded as unknown, was assigned to the central-western (CWT) OTU using a stepwise discriminant-function analysis with a probability of 61%. All specimens of L. culpaeus were included in a single OTU. Sexual dimorphism was assessed only in the gray fox CCT OTU (19 doa, 14 28) by multivariate analysis of variance (MAN- OVA) and discriminant-function analysis. Samples for the other taxa were too small to evaluate for dimorphism. The mean and standard deviation of each measurement were calculated for each OTU. Principal component analysis was done using NT-SYS programs (Rohlf 1992) to assess phenetic overlap among individ- uals. Each character was standardized to a 731 Fig. 1. for specimens listed in Appendix 1. Squares indicate culpeo foxes; circles or dots, gray foxes; solid symbols correspond to specimens used in cranial analyses; open symbols indicate specimens used in pelage analysis. Abbreviations refer to OTUs, which are identified in Appendix 1. Map of Argentina showing collecting sites mean of zero and a standard deviation of one. Environmental factors vary greatly over the ranges of latitude (approximately 24°) and elevation (sea level to over 3000 me- ters) represented by the samples analyzed. Therefore, to determine if variation in size was associated with environmental factors, a second principal component analysis was performed using the means of measure- ments for each OTU. Pearson’s correlation coefficients were calculated between the first principal component scores and lati- 732 tude, longitude, elevation, mean annual temperature, and average annual rainfall. Since each OTU comprised more than one collecting site, weighted means were cal- culated for each environmental factor. Kraglievich (1930) used two cranial characters (crests on the skull and distance between postorbital constriction and post- orbital processes) to distinguish gymnocer- cus from patagonicus (= griseus of au- thors). He believed that the presence of a sagittal crest and a long interorbital region distinguished gymnocercus from patagoni- cus, which he identified as having a lyri- form sagittal area and a short distance be- tween postorbital constriction and postor- bital processes. We examined geographic and nongeographic variation of both char- acters in a sample of 81 specimens repre- senting areas from which only the presence of gymnocercus had been cited. We scored three possible conditions for each character as follows: Temporal ridges, A) fused to form a well-developed sagittal crest, B) not fused and enclosing a narrow lyriform sag- ittal area, or C) not fused and enclosing a wide lyriform sagittal area; distance be- tween postorbital constriction and postor- bital processes: A) long (10.0—15.0 mm), B) medium (7.0—9.0 mm) or C) short (4.0—6.0 mm). We analyzed geographic variation in 11 pelage characters (as described by Cabrera 1932) with color determinations made un- der natural light. Characters were scored as follows: A) chin spot: absent = 0, poorly marked = 1, very dark = 2; B) thigh spot: absent = O, poorly marked = 1, very dark = 2; C) dorsal longitudinal band: absent = OQ, poorly marked = 1, very dark = 2; D) underparts: brownish = O, cream = 1, gray = 2; E) sides: reddish = O, brownish = 1, gray = 2; F) upperparts: reddish = O, gray = 1, brownish = 2; G) dorsal guard hairs banded (agouti): fine = O, intermediate = 1, thick = 2; H) length of dorsal guard = hairs: short = 0, medium = 1, long = 1; D ears: reddish = O, pale brown = 1, dark brown = 2; J) head: reddish = O, gray = PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 1, brownish = 2; K) thigh: reddish = 0, brownish = 1, gray = 2. Specimens were grouped into six OTUs, each representing a geographically discrete region (Fig. 1, Ap- pendix 1). We used principal component analysis to evaluate phenetic overlap. We also examined the microscopic struc- ture of dorsal guard hairs taken from 48 specimens including both gray and culpeo foxes. Samples were first washed in an al- cohol-ether solution (1:1) and then bleached for 24 hours in hydrogen peroxide. Then hairs were washed with fresh water, fixed to a glass slide with a thin layer of vinilic glue, and examined at 320 diameters. Type and disposition of cuticular scales and the type of medulla follow the nomenclature adopted by Chehébar & Martin (1989). Results Cranial characters.—Miultivariate anal- ysis of variance (MANOVA) of the gray fox CCT OTU revealed no significant sex- ual dimorphism (P = 0.62; n = 33) in 20 of the 26 measurements analyzed. Discrim- inant analysis showed an overlap between sexes with females being more variable. Skulls of L. culpaeus (CCU OTUV) clearly are larger than those of the other groups, which are all gray foxes (see Table 1 for means and standard deviations). Braincase width (16) was one of the more important measurements for separating gray and cul- peo foxes. The skulls of the gray fox OTUs are not only smaller, but show little varia- tion except for size increase toward the east and northeast, suggesting that variation is clinal. All variables in the principal components analysis had positive correlations on com- ponent I, which indicates a general size fac- tor (Table 2). The second component was bipolar, with both positive and negative loadings. The first component (Fig. 2) clearly separated L. culpaeus (CCU) from the gray foxes (griseus and gymnocercus). The OTUs of gray foxes overlapped with a general increase in size from west to east. VOLUME 108, NUMBER 4 133 Table 1.—Cranial measurements for Argentine culpeo and gray foxes. Mean and one standard deviation (in parentheses) for each sample (OTU; sample acronym over sample size) measured. Measurements (left column) defined in text; OTUs identified in Appendix | under specimens examined for cranial analysis. CCU CST CWT CNW CCT CET CNT 58 7 14 8 33 15 16 1 164.66 122.69 122.83 122.98 138.79 141.53 125.91 (7.13) (4.15) (4.12) (8.90) (5.44) (7.56) (2.86) 2 85.87 64.80 65.11 64.49 TESS V2M2 64.90 G92) (3.36) (2.29) (5.70) (ALY) (3.56) @22) 3 73.59 54.09 53.74 239 60.49 61.24 53.81 (4.50) (3.04) (2.45) (4.93) (3.19) Gy) @32) 4 81.36 60.00 SIM)il 57.03 68.21 68.81 59.01 (4.65) (3.72) (3.43) (6.52) (2.72) (4.77) (2.90) 5) 20.66 18.44 17.90 18.39 18.92 19.76 19.91 (1.02) (1.07) (1.19) (0.73) (1.25) (0.93) (0.97) 6 20.99 15.19 14.83 15.64 17.60 17.97 16.24 (1.53) (0.92) (1.05) (1.44) (1.25) (1.18) (1.13) 7 43.66 Somg Neal) 37.88 40.15 40.10 38.03 (1.20) (0.76) (1.21) (2.75) (1.15) (1.57) (1.80) 8 24.99 20.00 18.84 19.16 AMS) 22.66 20.44 (1.67) (1.03) (1.37) (1.86) (1.35) (1.66) (0.75) 9 73.66 58.66 No 57.06 63.01 63.50 S293 (2.81) (1.66) G72) (4.09) (2.42) (2.91) (1.91) 10 30.42 PES oI/N| 25.34 25.76 27.41 DAESS 26.40 (1.29) (0.86) (1.22) (1.69) (1.12) (1.22) (1.05) JU 16.52 IAQ 12.84 WAT) 13.79 14.19 13.19 (0.87) (0.42) (0.71) (0.91) (0.77) (0.57) (0.66) 12 7.86 6.06 5.98 6.46 6.92 7.00 6.66 (0.49) (0.28) (0.51) (0.49) (0.44) (0.35) (0.37) 13 16.77 373) NBO 13.40 14.88 15.11 14.00 (0.83) (0.40) (0.89) (1.31) (0.71) (0.98) (0.73) 14 7.01 5.67 5.70 5.78 6.20 6.25 5.84 (0.37) (0.20) (0.36) (0.42) (0.32) (0.42) (0.31) 15 128.61 95.40 94.35 93.61 106.57 108.78 96.79 (6.10) (4.27) (3.51) (8.38) (4.40) (5.85) (2.59) 16 S59) 65.14 64.94 63.81 70.55 71.01 65.44 (3.30) (2.05) (2.29) (5.02) (2.81) G33) G93) 17 29.58 24.71 24.58 25.03 26.53 27.06 25-50) Ca) (0.98) (1.14) (1.66) (1.28) (1.10) (0.91) 18 28.15 23.04 21.94 21.48 Deol 26.47 23.19 (1.94) (1.41) (1.45) (2.50) (1.55) (2.30) (1.61) 19 26.55 PP SpQy3) 24.19 23.65 25.18 Madey) 24.19 (1.86) (1.28) (1.35) (1.25) (1.68) (1.67) (1.65) 20 Poesy 19.21 18.67 MWS 22.50 23.20 20.59 (1.84) (CIID) (0.91) (2.05) (1.26) (ta7/s)) (0.97) 72) 46.70 36.74 35.07 SUID 40.39 42.00 38.72 (1.80) Ges) (1.76) (2.45) GES») (2.09) (0.97) 22 88.79 64.27 63.27 65.06 (B59 UT) 67.60 ©2) @52) (2.09) (@./))) (3.54) (4.46) (2.09) Dp 54.75 42.49 41.95 42.00 46.62 48.13 44.53 (2.50) (1.43) (1.10) (2.89) (1.88) (2.99) (23) 24 51.88 44.66 44.30 44.05 46.76 46.93 44.58 (1.48) (0.94) (0.86) (1.65) (1.42) (1.97) (0.91) 25 DOWD 23.34 Ma Joop2) 23205 26.02 D023} 24.35 (1.63) (1.18) (1.41) (2.08) (1.48) (1.39) (0.92) 26 38.48 31.47 SED 31.85 34.32 34.53 35513 GE39) (0.96) (0.80) (1.47) (1.59) (1.78) EZ) 734 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 2.—Character loadings of cranial variables of Argentine culpeo and gray foxes (n = 151) on the first two principal components. Measurements defined in text. Component Component Vanable me OP Le Mee enn ee ee ae Variable I 0 1 0.211 —0.037 14 0.193 —0.103 2 0.209 —0.042 15 0.210 —0.026 3 0.205 —0.041 16 0.208 —0.059 4 0.206 0.021 17 0.197 —0.143 5 0.149 0.022 18 0.184 0.286 6 0.197 0.012 19 0.108 0.885 Tf 0.191 0.111 20 0.206 0.029 8 0.192 —0.228 ZAM 0.205 —0.012 9 0.208 —0.060 Mp) 0.206 0.064 10 0.198 —0.132 23 0.206 0.003 11 0.199 —0.107 24 0.203 0.078 12 0.184 —0.250 25 0.197 —0.096 13 0.194 —0.109 26 0.200 0.024 % variance 84.68 3.26 C2 = 118 =(8 al -4 ae @) z 4 6 8 10 C1 Fig. 2. Distribution of 151 specimens of Argentine foxes based on cranial data on the first two principal components. Symbols represent OTUs, identified in Appendix 1 under cranial analysis, as follows: BH—CCU; [J—CST; @—CwT: O—ENW: A—CCE A—CET; V—CNE VOLUME 108, NUMBER 4 Table 3.—Argentine gray foxes from areas where only gymnocercus was believed to occur, segregated on the basis of three developmental conditions of tem- poral ridges and three comparative lengths of interor- bital region of the skull. Development of temporal ridges: A, fused into sagittal crest; B, not fused and enclosing narrow lyriform sagittal area; C, not fused and enclosing wide lyriform sagittal area. Distance in midline from plane of interorbital constriction to plane of postorbital processes: A, long (10.0—-15.0 mm); B, medium (7.0—9.0 mm); C, short (4.0—6.0 mm). Temporal Interorbital Condition ridges distance A 15 (19%) 33 (41%) B 23 (28%) 20 (25%) C 43 (53%) 28 (34%) Total 81 81 The La Pampa OTU (CCT) was the most variable (Fig. 2), spanning the gap between the western OTUs and the Buenos Aires sample (CET). In the analysis based on means of gray fox OTUs, the first component explained 90.9% of the variance. As all variables had a positive loading, component I represents a general size factor. First component scores showed a significant positive correlation with rainfall (r = 0.90; P < 0.05) which increases toward the northeast. Cranial characters of Argentine gray fox- es are highly variable. Our analysis (Table 3) of the development of temporal ridges and length of interorbital region, once con- sidered diagnostic for separating griseus from gymnocercus, shows that these cranial features have little or no value for distin- guishing among gray foxes. Pelage characters.—Principal compo- nents analysis of pelage characters showed high variation among the groups (Table 4, Fig. 3) with the first two components ex- plaining only 57.8% of the variance. How- ever, these results are in agreement with the principal component analysis based on cra- nial characters; in both analyses the culpeo foxes (culpaeus) are clearly separated from the gray foxes (griseus and gymnocercus). The L. culpaeus group differs from the UD Table 4.—Character loadings of coded pelage vari- ables of Argentine foxes (n = 111) on the first two principal components. Variable PC I PC Il A 0.392 —0.010 B 0.355 —0.201 C 0.118 0.548 D =(0)117/33 0.108 E 0.318 —0.313 F 0.234 —0.281 G 0.195 0.326 H 0.217 0.551 I 0.354 0.068 J 0.372 0.159 K 0.414 —0.173 % of variance 43.00 14.81 other groups in the completely white chin and the reddish thigh lacking a posterior dark spot. Patterns of gray fox pelage char- acteristics can be correlated with geograph- ic areas represented by the OTUs (Table 5). Microscopic examination of guard hairs revealed no differences in either the type of medulla or the type and arrangement of cu- ticular scales. The medulla is segmented and scales are lanceolate and imbricated in all skins examined. Discussion and Conclusions We found no significant sexual dimor- phism in gray foxes; small samples sizes precluded testing for sexual dimorphism in culpeo foxes. The absence of sexual dimor- phism has been reported in other canids (Waithman & Roest 1977, Gingerich & Winkler 1979). Multivariate analyses of both cranial and pelage characters confirm that L. culpaeus differs significantly from gray foxes in Ar- gentina. These differences support species status for L. culpaeus and argue against the Clutton-Brock et al. (1976) conclusion that culpeo and gray foxes are phenetically close and might be conspecific. Gray foxes, however, show clinal varia- tion in skull size correlated with increasing annual precipitation. Gray foxes also are 736 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON C1 Fig. 3. Distribution of 111 specimens of Argentine foxes based on pelage characters on the first two principal components. Symbols represent OTUs, identified in Appendix | under pelage, as follows: H-—PCU; A—PET; O—PNW; V—PNT; @—PWT; A—PCT. Table 5.—Qualitative pelage characteristics of Argentine gray foxes. Pelage characters described in text; OTUs identified in Appendix 1 under specimens examined for pelage analysis; OTU acronym followed by sample size in parentheses. Character PNW (6) PWT (9) PNT (25) PET (14) PCT (30) Chin spot Very dark Variable-dark Very dark Very dark Very dark Thigh spot Very dark Variable-dark Very dark Very dark Very dark Dorsal band Absent—poorly Absent—poorly Absent—poorly Very dark Variable marked marked marked Underparts Brownish Cream Brownish-cream Cream Brownish-cream Sides Brownish Brownish-gray Brownish-gray Brownish Brownish Upperparts Brownish Brownish Brownish Brownish Brownish Agouti Fine—medium Fine—very fine Fine—medium Thick—medium __ Variable Guard hair Variable Variable Variable Variable Variable length Ears Reddish Dark brown Reddish-pale Pale brown Variable-brown brown Head Reddish Brownish Reddish-brownish Brownish Brownish Thigh Reddish Reddish Brownish Brownish Brownish VOLUME 108, NUMBER 4 smaller when sympatric with L. culpaeus. Data on geographic variation and clinal trends in size of canids are scarce. Fuentes & Jaksic (1979) found clinal variation in size from north to south in L. culpaeus and L. griseus in Chile, and suggested that pro- gressive increments of habitat overlap to- wards the south were compensated by pro- gressively greater differences in body size and in mean prey size. We found the cranial characters used by Kraglievich (1930) to distinguish L. gym- nocercus from L. griseus to be highly vari- able even within the same population. Kraglievich (1930:51) claimed that gym- nocercus frequently has an extensive sag- ittal crest and a greater distance between the interorbital region and postorbital processes than has patagonicus (= griseus of au- thors). We scored three possible conditions for each of these two cranial characters in 81 gray foxes previously identified as gym- nocercus from areas where only gymnocer- cus had been assumed to occur. Our results (Table 3) showed high individual and geo- graphic variation indicating that the char- acters have little or no taxonomic value for discriminating among gray foxes. Kragliev- ich (1930) also considered postorbital con- striction (measurement 11) to be narrower in gymnocercus than in patagonicus (= gri- seus Of authors), but our analysis revealed relatively little variation when all groups of gray foxes are compared (Table 1). Most of the morphological differences ascribed to the named forms of gray foxes had been based on type specimens, which for griseus and gymnocercus come from lo- calities approximately 3500 km apart. Some of these characters, once considered diag- nostic, have not proven useful for taxonom- ic purposes. For example, in accord with Clutton Brock et al. (1976), our results sug- gest that the absence of a sagittal crest in canids may be associated with small size. Chromosomal analyses of these species (Gallardo & Formas 1975, Vitullo & Zuleta 1992) show a 2N = 74, NE = 76 karyotype with all-acrocentric autosomes. This karyo- 137 type, considered primitive for South Amer- ican canids, does not distinguish among these species. Because our samples of gray foxes Showed clinal variation in size and color pattern as well as considerable nongeo- graphic variation in qualitative characters considered diagnostic by previous workers, we conclude that gray foxes of Argentina are conspecific. Lycalopex gymnocercus 1s the oldest available name for the gray foxes of Argentina. Taxonomy Lycalopex Burmeister, 1854 Synonyms.— Canis: Molina, 1782:293 (not Canis Lin- naeus, 1758). Vulpes: Martin, 1837:11 (not Vulpes Frisch, 1775). Procyon: Fischer, 1814:178 (not Procyon Storr, 1780). Dusicyon Hamilton-Smith, 1839:248 (part; described as a section of Chaon Hamil- ton-Smith, 1839). Cerdocyon Hamilton-Smith, 1839:259 (part; described as a section of Chaon Hamilton-Smith, 1839). Lycalopex Burmeister, 1854:95 (described as a section of Canis Linnaeus); type spe- cies Canis vetulus Lund, 1842a, by sub- sequent designation (Thomas 1914a: 352). Pseudalopex Burmeister, 1856:24 (de- scribed as a section of Canis Linnaeus); type species Canis magellanicus Gray, 1837a (= Vulpes magellanicus Gray, 1837b), by subsequent designation (Tho- mas 1914a:352). Thous: Gray, 1869:514 (part; not Thous Hamilton-Smith, 1839). Lupulus: Trouessart, 1897:304 (part; not Lupulus Blainville, 1830). Nothocyon: Wortman and Matthew, 1899: 124 (part; not Nothocyon Matthew, 1899). Pseudolopex Philippi, 1903:157 (incorrect 738 subsequent spelling of Pseudalopex Bur- meister, 1856). Pseudolycos Philippi, 1903:157; proposed as a subgenus of “unsere gr6sseren Fiichse”’ (our larger foxes [Chilean]); no species mentioned. Eunothocyon J. A. Allen, 1905:152; type species Canis sladeni Thomas, 1904, by original designation. Angusticeps Hilzheimer, 1906:114 (pro- posed as a subgenus of Canis Linnaeus, 1758); type species Canis (Angusticeps) reissi Hilzheimer, 1906, by monotypy. Viverriceps Hilzheimer, 1906:116 (lapsus for Angusticeps; not Viverriceps Gray, 1867). Microcyon Trouessart, 1906:1186 (pro- posed as a subgenus of Speothos Lund, 1839a); type species Speothos riveti Trouessart, 1906, by original designation. Lycalopex culpaeus (Molina, 1782) Synonyms.— Canis culpaeus Molina, 1782:293; type lo- cality “‘Chili,” restricted to “Santiago province’’ by Cabrera (1931:62). Clanis]. Vulpes chilensis Kerr, 1792:144; type locality “‘Chili.”” Based exclusively on Canis culpaeus Molina. C[anis]. Magellanicus Gray, 1837a:88 (no- men nudum). Vulpes magellanica Gray, 1837b:578; type locality ‘“‘Magellan’s Straits” [= Port Famine; Gray 1843:61], Magellanes, Chile. Cerdocyon Magellanicus: Hamilton-Smith, 1839:266 (name combination). Canis (Pseudalopex) lycoides Philippi, 1896:542 (p. 2 in reprint); type locality ““insulis Tierra del Fuego”’ [= Bahia Fe- lipe; Wolffsohn 1921:514], Magellanes, Chile. [Canis (Lupulus)| magellanicus: Trouessart, 1897:306 (name combination). Canis montanus Prichard, 1902:260; type locality “‘South-eastern Patagonia.”’ Pre- occupied by Canis montanus Marsh, 1871. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Canis amblyodon Philippi, 1903:158; type locality “provincia Valparaiso,’’ Chile. Canis albigula Philippi, 1903:159; type lo- cality “‘provinciis centralibus,’’ Chile. [Canis (Cerdocyon)| magellanicus: Troues- sart, 1904:233 (name combination). [Canis (Cerdocyon)| lycoides: Trouessart, 1904:234 (name combination). [Canis (Cerdocyon)| prichardi Trouessart, 1904:234 (new name for Canis montanus Prichard, 1902). Canis (Angusticeps) reissii Hilzheimer, 1906:116; type locality “‘Quito,”’ Pichin- cha, Ecuador. Speothos Riveti Trouessart, 1906:1185; type locality “‘Alchipichi, province de Pichin- cha (Equateur), altitude de 2101™.” Canis (Cerdocyon) magellanicus Riveti: Trouessart, 1910:12 (name combination). C[anis]. Riveti: Cabrera, 1912:63 (name combination). Ps|eudalopex|. c{ulpaeus|. magellanicus: Thomas, 1914a:357 (name combination). Ps({eudalopex|. c{ulpaeus|. culpaeus: Tho- mas, 1914a:357 (name combination). Ps{eudalopex]|. lycoides: Thomas, 191A4a: 357 (name combination). Ps|eudalopex]. c{ulpaeus| reissii: Thomas, 1914a:357 (name combination). Pseudalopex culpaeus andina Thomas, 1914a:357; type locality “‘Esperanza, near Mt. Sajama, Province of Oruro [La Paz], Bolivia. Alt. 4000 m.”’ Pseudalopex culpaeolus Thomas, 1914a: 359; type locality ““Santa Elena,’’ Sori- ano, Uruguay. Considered a composite (skin of P. culpaeus; skull of P. gymno- cercus) by Langguth (1967) who selected the skin as lectotype, thus treating the taxon as an objective synonym of P. inca Thomas, 1914a, and a subjective syn- onym of P. culpaeus andina Thomas, 1914a. Pseudalopex inca Thomas, 1914a:361; type locality ““Sumbay, Arequipa, Peru. Alt. 4000 m.”’ Considered a composite (skin of P. gymnocercus; skull of P. culpaeus) by Langguth (1967) who selected the skull as lectotype, thus treating the taxon VOLUME 108, NUMBER 4 as an objective synonym of P. culpaeolus Thomas, 1914a, and a subjective syn- onym of P. culpaeus andina Thomas, 1914a. Canis culpaeus reissi: Osgood, 1914:172 (name combination). Pseudalopex smithersi Thomas, 1914b:573; type locality “Sierra de Cordoba,”’ Cor- doba, Argentina; restricted by Cabrera (1958:232) to Pampa de Achala, 2200 m. Canis culpaeus andinus: Osgood, 1916:211 (name combination). Pseudalopex magellanicus: 1919:1 (name combination). Pseudalopex culpaea Thomas, 1921:385 (unjustified emendation of gender termi- nation for culpaeus). Pseudalopex reissii: LOnnberg, 1921:23 (name combination). Canis ferrugineus Huber, 1925:9; type lo- cality “la Cordillera [de los Andes], entre low rios Mendoza, Atuel, Neuquén y Collun-Cura,”’ Argentina. [Canis (| Pseudalopex{)| gymnocercus cul- paeola: Kraglievich, 1930:52; (name combination). [Canis (| Dusicyon[)] lycoides: Kraglievich, 1930:58; (name combination). [Canis (| Dusicyon[)] culpaeus magellani- cus: Kraglievich, 1930:58; (name com- bination). [Canis (| Dusicyon[)] culpaeus andinus: Kraglievich, 1930:59; (name combina- tion). [Canis (| Dusicyon[)] culpaeus reissii: Kraglievich, 1930:59; (name combina- tion). [Canis (| Dusicyon[)] culpaeus Riveti: Kraglievich, 1930:60; (name combina- tion). [Canis (| D[usicyon)]. smithersi: Kragliev- ich, 1930:60; (name combination). [Canis (| Dusicyon[)] inca: Kraglievich, 1930:61; (name combination). Pseudalopex culpaeus andina: Cabrera, 1931:63 (name combination). Pseudalopex culpaeus reissii: Cabrera, 1931:63 (name combination). Lonnberg, 739 Pseudalopex culpaeus smithersi: Cabrera, 1931:63 (name combination). Pseudalopex culpaeus magellanica: Ca- brera, 1931:63 (name combination). Pseudalopex culpaeus lycoides: Cabrera, 1931:63 (name combination). Dusicyon (Dusicyon) culpaeus: Osgood, 1934:49 (name combination). Canis magellanicus priscus Spillmann, 1938:387 (nomen nudum). Dusicyon culpaeus andinus: Osgood, 1943: 64 (name combination). Dusicyon culpaeus lycoides: Osgood, 1943: 66 (name combination). Dusicyon culpaeus magellanicus: Osgood, 1943:65 (name combination). Dusicyon {(Dusicyon)] culpaeolus: Cabrera, 1958:229 (name combination). Dusicyon {(Dusicyon)| culpaeus reissii: Ca- brera, 1958:232 (name combination). Dusicyon [|(Dusicyon)| culpaeus smithersi: Cabrera, 1958:232 (name combination). Dusicyon [(Dusicyon)]| inca: Cabrera, 1958: 235 (name combination). Comments.—Lycalopex culpaeus occurs from the Andes of Colombia to Tierra del Fuego, is widely distributed in Patagonia, but is not known from Paraguay, Uruguay, or Brazil. The species has not been revised since Cabrera’s (1958) catalog in which he recognized five subspecies in addition to the nominate form. Langguth (1967) pre- sented convincing arguments that Pseudal- opex culpaeolus and P. inca, both described by Thomas (1914a), were based on com- posites. Langguth reduced both names to the synonymy of Pseudalopex (= Lycalo- pex) culpaeus. Lycalopex gymnocercus (Fischer, 1814) Synonyms.— Canis lagopus Molina, 1782:272; type lo- cality ““Arcipelago di Chiloe,’’ Chiloé, Chile. Preoccupied by Canis lagopus Linnaeus, 1758. Procyon gymnocercus Fischer, 1814:178; based solely on ““‘L- Agourachay” of Aza- 740 ra (1801:317); therefore, type locality is Paraguay; restricted by Cabrera (1958: 235) to vicinity of Asuncion. Can|is]. brasiliensis Schinz, 1821:220; type locality ““Brasilien und Paraguay.’’ Based solely on ““L-Agourachay of Azara (1801: 317); therefore, restricted type locality is vicinity of Asuncion, Paraguay (Cabrera 1958:235). Canis Azarae s. Brasiliensis: Rengger, 1830:143 (name combination). C[anis]. griseus Gray, 1837a:88 (nomen nu- dum). Vulpes fulvipes Martin, 1837:11; type lo- cality “island of Chiloe,”’ restricted by Darwin (in Waterhouse 1839) to “‘sea- beach at the southern point of the is- land,’’ near San Pedro channel, Chiloé, Chile. Vulpes griseus Gray, 1837b:578; type lo- cality ““Magellan,’’ Magellanes, Chile. Cerdocyon fulvipes: Hamilton-Smith, 1839: 257 (name combination). Clanis]. protalopex Lund, 1839a:223 (no- men nudum). Canis protalopex Lund, 1839b:32 (nomen nudum). Canis protalopex Lund, 1840:54, text to plate 28 (fig. 9); type locality ““Rio das Velhas,’’ Lagoa Santa, Minas Gerais, Brazil. Canis |(Pseudalopex)| gracilis Burmeister, 1861:406; type locality “die buschige Pampa in den Umgebungen Mendozas,”’ Mendoza, Argentina. Canis patagonicus Philippi, 1866:116; type locality ““Magellans Strasse,” Magella- nes, Chile. Pseudalopex griseus: Gray, 1869:512 (name combination). Thous fulvipes: Gray, 1869:514 (name com- bination). Canis azarae fossilis Gervais and Amegh- ino, 1880:36 (nomen nudum). Canis Azarae m. fossilis Ameghino, 1889: 298; type locality ““Rio Lujan y Cafiada de Rocha en los partidos de Mercedes y Lujan, provincia de Buenos Aires,”’ Ar- PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON gentina. Preoccupied by Canis familiaris fossilis Pictet, 1853. Canis Azarae, m. antiguus Ameghino, 1889:298; type locality ““Rio Lujan en los partidos de Mercedes y Lujan, provincia de Buenos Aires,”’ Argentina. Canis azarae (var. fulvipes): Mivart, 1890a: fig. 25 (name combination). [Canis (Thous)] griseus: Trouessart, 1897: 307 (name combination). [Canis (Thous)] gracilis: Trouessart, 1897: 308 (name combination). Canis domeykoanus Philippi, 1901:168 (p. 4 in reprint); type locality ““Provincia de Copiapo,”’ Chile. Canis rufipes Philippi, 1901:168 (p. 4 in re- print); type locality not given, assumed to be Chile. Most likely a lapsus for Ca- nis fulvipes (Martin, 1837). Canis maullinicus Philippi, 1903:158; type locality “‘provincia Llanquihue ad occi- dentem lacus Llanquihue, loco ‘Nueva Braunau,’’ Chile. Canis trichodactylus Philippi, 1903:158; type locality “‘provincia Valdivia,’’ Chile. Canis torquatus Philippi, 1903:159; type locality “‘Puerto Montt,’’ Llanquihue, Chile. [Canis (Cerdocyon) azarae| protalopex: Trouessart, 1904:233 (name combina- tion). [Canis (Cerdocyon) azarae] antiguus: Trouessart, 1904:233 (name combina- tion). [Canis (Cerdocyon) azarae| fulvipes: Trouessart, 1904:233 (name combina- tion). [Canis (Cerdocyon)| griseus: Trouessart, 1904:234 (name combination). [Canis (Cerdocyon) griseus] gracilis: Trouessart, 1904:234 (name combina- tion). [Canis (Cerdocyon)| domeykoanus: Troues- sart, 1904:234 (name combination). Cerdocyon griseus: J. A. Allen, 1905:155 (name combination). Pseudalopex azarica Thomas, 1914a:360; type locality ““Mar del Plata, S. E. Bue- nos Ayres,”’ Argentina. VOLUME 108, NUMBER 4 Canis (Pseudalopex) gymnocercus: Os- good, 1915:143 (name combination). Pseudalopex domeykoanus: Cabrera, 1917: 27 (name combination). Canis Domeycoanus: Wolffsohn, 1918:61 (incorrect subsequent spelling of domey- koanus Philippi). Pseudalopex zorrula Thomas, 1921:383; type locality ““Chumbicha, Catamarca[, Argentina]. Alt. 500 m.”’ P{seudalopex|. domeycoanus: Wolffsohn, 1921:514 (incorrect subsequent spelling of domeykoanus Philippi). Cl[anis]. cinereo argenteus: Larrafiaga, 1923:344 (not Canis cinereo argenteus Schreber, 1775). C[erdocyon]. tlhous]. brasiliensis: G. M. Allen, 1923:56 (name combination). [Canis (| Pseudalopex{)| patagonicus grac- ilis: Kraglievich, 1930:50 (name combi- nation; however, gracilis has_ priority Over patagonicus). [Canis (| Pseudalopex|)] patagonicus zor- rula: Kraglievich, 1930:51 (name com- bination). [Canis (| Pseudalopex|)] gymnocercus at- tenuatus Kraglievich, 1930:54; type lo- cality “los Estados brasilefios de Rio Grande del Sur, Parana y tal vez Matto Grosso.”’ Pseudalopex gymnocercus gymnocercus: Cabrera, 1931:64 (name combination). Pseudalopex gymnocercus antiguus: Ca- brera, 1931:64 (name combination). Pseudalopex gracilis zorrula: Cabrera, 1931:65 (name combination). Pseudalopex gracilis domeykoanus: Ca- brera, 1931:65 (name combination). Pseudalopex gracilis patagonicus: Cabrera, 1931:65 (name combination). Pseudalopex fulvipes: Cabrera, 1931:66 (name combination). Dusicyon (Dusicyon) gymnocercus: Os- good, 1934:49 (name combination). Dusicyon (Dusicyon) griseus: Osgood, 1934:49 (name combination). Dusicyon griseus domeykoanus: Osgood, 1943:69 (name combination). 741 Dusicyon griseus maullinicus: Osgood, 1943:70 (name combination). Dusicyon fulvipes: Osgood, 1943:71 (name combination). Ducicyon griseus domeicoanus Mann, 1950:5 Gncorrect subsequent spellings of Dusicyon Hamilton-Smith, 1839 and do- meykoanus Philippi, 1901). Dusicyon [(Dusicyon)] griseus gracilis: Ca- brera, 1958:233 (name combination). Dusicyon [|(Dusicyon)]| gymnocercus antig- uus: Cabrera, 1958:234 (name combina- tion). Comments.—Lycalopex gymnocercus Oc- curs from southern Bolivia and Brazil to Ti- erra del Fuego. Previously considered to represent two species, gymnocercus was the name used for the northern and northeastern populations; griseus, for the western and southern populations. Kraglievich (1930: 49) used the name Canis (Pseudalopex) pa- tagonicus Philippi, 1866, for the taxon cur- rently known in the literature as Pseudalo- pex griseus (Gray) because he, along with Cabrera (1931), believed Canis griseus Gray (1837a) to be preoccupied by Canis griseus Boddaert, 1784. However, Gray’s name is a nomen nudum; the correct origi- nal name combination is Vulpes griseus Gray (1837b) a junior synonym of Lycalo- pex gymnocercus. Instead of patagonicus, Cabrera (1931:65) used Pseudalopex grac- ilis Burmeister, 1861, as the earliest avail- able name for the taxon he later (1958:233) called Dusicyon griseus (= Lycalopex gym- nocercus). Because we have not attempted a complete revision of L. gymnocercus, we are reluctant to identify any populations as representing subspecies. Nevertheless, the name griseus Gray is available for the southernmost Argentine and Chilean pop- ulation and fulvipes Martin is the appropri- ate name for the Chiloé Island and adjacent mainland population, which until recently had been considered a separate species. Canis azarae fossilis has been attributed to Gervais and Ameghino (1880:36); how- ever, the taxon was not described in that 742 publication, hence it is a nomen nudum be- cause the references to Bravard are not val- id indications, since Bravard’s work was never actually published (Simpson 1940). The name dates from Ameghino (1889) and is preoccupied. Wozencraft (1993:284) in- cluded [Canis] entrerianus Burmeister, 1861, under Pseudalopex gymnocercus, but the type is a male Cerdocyon thous (see Thomas 1914a:359, footnote; Cabrera 1931:61; Berta 1982). Lycalopex sechurae (Thomas, 1900) Synonyms.— Canis sechurae Thomas, 1900:148; type lo- cality ““Sullana,”’ Piura, Peru. [Canis (Cerdocyon)| sechurae: Trouessart, 1904:234 (name combination). Pseudalopex sechurae: J. A. Allen, 1916: 122 (name combination). [Canis (| Pseudalopex{)| sechurae: Krag- lievich, 1930:51 (name combination). Dusicyon (Dusicyon) sechurae: Osgood, 1934:49 (name combination). Comments.—Lycalopex sechurae 1s monotypic and restricted in distribution to the coastal desert and arid inter-Andean val- leys of northwestern Pert and southern Ec- uador. Its southern limits in western Pert are not known. Osgood (1914) recorded it in Depto. La Libertad, and it may occur far- ther south along the Pacific coast. Lycalopex vetulus (Lund, 1842) Synonyms.— Canis azarae: Lund, 1837:324 (not Canis azarae Wied, 1824). Canis vetulus Lund, 1842a:5; type locality “Rio das Velhas’s Floddal,’’ Lagoa Santa, Minas Gerais, Brazil. Canis fulvicaudus Lund, 1843:20; type lo- cality ““Rio das Velhas’s Floddal,’’ Lagoa Santa, Minas Gerais, Brazil. Vulpes vetulus: Gerrard, 1862:88 (name combination). [Lycalopex fulvicaudus| var. 1. chilensis PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Gray, 1869:511; type locality “‘Chiloe,”’ Chile. According to Thomas (1904:236), the type locality is wrong. Canis parvidens Mivart, 1890a:76; type lo- cality “Brazil.” Canis urostictus Mivart, 1890a:81; type lo- cality “Brazil.” [Canis (Thous)| parvidens: Trouessart, 1897:308 (name combination). [Canis (Thous)]| urostictus: Trouessart, 1897:308 (name combination). Nothocyon urostictus: Wortman and Mat- thew, 1899:125 (name combination). Nothocyon parvidens: Wortman and Mat- thew, 1899:126 (name combination). Canis sladeni Thomas, 1904:235; type lo- cality ““Santa Anna de Chapada,”’ Matto Grosso, Brazil. [Canis (Nothocyon)| parvidens: Trouessart, 1904:235 (name combination). [Canis (Nothocyon)]| urostictus: Trouessart, 1904:235 (name combination). E|unothocyon]. sladeni: J. A. Allen, 1905: 152, footnote (name combination). E[unothocyon]. urostictus: J. A. Allen, 1905:152, footnote (name combination). E[{unothocyon]. parvidens: J. A. Allen, 1905:152, footnote (name combination). Canis (Eunothocyon) vetulus: Ihering, 1911:206 (name combination). Canis Vitulus Huber, 1925:1 (incorrect sub- sequent spelling of Canis vetulus Lund, 1842a). Lycalopex vetulus: Kraglievich, (name combination). [Lycalopex| vetulus fulvicaudus: Kragliev- ich, 1930:43 (name combination). Dusicyon (Lycalopex) vetulus: Osgood, 1934:49 (name combination). P{seudalopex]. vetulus: Berta, 1987:458 (name combination). 1930:43 Comments.—Lycalopex vetulus is mono- typic and found only in Brazil, although Berta (1987) mentioned a fossil from Ar- gentina. Mivart’s names Canis parvidens and C. urostictus have been cited from his report in the Proceedings of the Zoological Society of London (1890b); however, that VOLUME 108, NUMBER 4 paper was published in August 1890 (Dun- can 1937) after his monograph on the Can- idae, which was listed in volume 12(8) of Nature Novitates for May 1890. Literature Cited Allen, G. M. 1923. The pampa fox of the Bogota savanna.—Proceedings of the Biological Soci- ety of Washington 36:55-58. Allen, J. A. 1905. Mammalia of southern Patagonia. In W. B. Scott, ed., Reports of the Princeton University Expeditions to Patagonia, 1896— 1899, 3(part 1):1—210, 29 pls. 1916. List of mammals collected for the American Museum in Ecuador by William B. Richardson, 1912—1913.—Bulletin of the American Museum of Natural History 35:113-— IWS Ameghino, E 1889. Contribuci6n al conocimiento de los mamiferos fosiles de la Republica Argenti- na.—Acta Academia Nacional de Ciencias de la Republica Argentina en Cérdoba, Buenos Ai- res 6:xxxii + 1—1027 pp.; atlas, 98 pls. Azara, EF 1801. Essais sur l’histoire naturelle des quadrupedes de la province du Paraguay. Trad- uits sur le manuscrit inédit de l’auteur, Pra. M. L. E. Moreau-Saint-Méry. Charles Pougens, Paris, 1:1xxx + 366 pp. Berta, A. 1982. Cerdocyon thous.—Mammalian Spe- cies 186:1-4. 1987. Origin, diversification, and zoogeog- raphy of the South American Canidae. Pp. 455— 471 in B. D. Patterson and R. M. Timm, eds., Studies in neotropical mammalogy. Essays in honor of Philip Hershkovitz.—Fieldiana, Zool- ogy 39:vii + 1-506. . 1988. Quaternary evolution and biogeogra- phy of the large South American Canidae (Mammalia: Carnivora).—University of Cali- fornia Publications, Geological Sciences 132:x + 1-149, 10 pls. Burmeister, H. 1854. Systematische Uebersicht der Thiere Brasiliens, welche wahrend einer Reise durch die Provinzen von Rio de Janeiro und Mi- nas Geraés gesammelt oder beobachtet wurden von Dr. Hermann Burmeister. Sdugethiere (Mammalia). Georg Reimer, Berlin, 1:x + 342 pp. 1856. Erlauterungen zur Fauna Brasiliens, enthaltend Abbildungen und ausfiihrliche Be- schreibungen neuer oder ungeniigend bekannter Thier-Arten. Georg Reimer, Berlin, ix + 115 pp., 32 pls. . 1861. Reise durch die La Plata-Staaten, mit besonderer Riicksicht auf die Beschaffenheit und den Culturzustand der Argentinischen Re- publik Ausgfiihrt in den Jahren 1857, 1858, 743 1859 und 1860. H. W. Schmidt, Halle, 2:vi + 538 pp., 1 map. Cabrera, A. 1912. Catalogo Metddico de las Colec- ciones de Mamiferos del Museo de Ciencias Naturales de Madrid.—Trabajos del Museo de Ciencias Naturales, Numero 11:1—147, 4 pls. . 1917. Mamiferos del viaje al Pacifico veri- ficado de 1862-1865 por una comision de na- turalistas enviada por el gobierno Espafol.— Trabajos del Museo Nacional de Ciencias Na- turales, Serie Zoologica, Numero 31:1—62. . 1931. On some South American canine gen- era.—Journal of Mammalogy 12:54—-67. . 1932. Sinopsis de los canidos argentinos.— Revista del Centro de Ingenieros Agr6nomos y Centro de Estudios de Agronomia 145:489— 501. 1940. Notas sobre carnivoros sud- americanos.—Instituto del Museo de la Univ- ersidad Nacional de La Plata 5(29):1—22. . 1958. Catalogo de los mamiferos de América del Sur. Revista del Museo Argentino de Cien- cias Naturales ““Bernardino Rivadavia,”’ Zool- ogia 4(1):1—308. [Dated 1957; published 27 March 1958—see notice on p. 308.] , & J. Yepes. 1940. Mamiferos Sud-america- nos. Compania Argentina de Editores, Buenos Aires, 370 pp. Chehébar, C., & S. Martin. 1989. Guia para el recon- ocimiento de los pelos de los mamiferos de la Patagonia.—Donana, Acta Vertebrata 16(2): 247-291. Clutton-Brock, J., G. B. Corbet, & M. Hills. 1976. A review of the family Canidae, with a classifi- cation by numerical methods.—Bulletin of the British Museum (Natural History), Zoology 29(3):119-199. Corbet, C. B., & J. E. Hill. 1991. A World list of mammalian species. Third edition. Natural His- tory Museum Publications, London, and Oxford University Press, New York, viii + 243 pp. Duncan, E M. 1937. On the dates of publication of the Society’s ‘Proceeding,’ 1859-1926. With an Appendix containing the dates of Publication of ‘Proceedings,’ 1839-1858, compiled by the late FE H. Waterhouse, and of the ‘Transactions,’ 1833-1869, by the late Henry Peavot, originally published in PZ.S. 1893, 1913.—Proceedings of the Zoological Society of London 107(A): 71-84. Fischer, G. 1814. Zoognosia tabulis synopticis illus- trata. Volumen tertium. Quadrupedum reliquo- rum, cetorum et montrymatum descriptionem continens. Nicolai Sergeidis Vsevolozsky, Mos- quae, 3:xxiv + 1-732. Fuentes, E. R., & EF M. Jaksic. 1979. Latitudinal size variation of chilean foxes: tests of alternative hypotheses.—Ecology 60(1):43-47. 744 Gallardo, M., & J. R. Formas. 1975. The karyotype of Dusicyon griseus (Carnivora, Canidae).—Ex- perientia 31(6):639-640. Gerrard, E. 1862. Catalogue of the bones of Mam- malia in the collection of the British Museum. British Museum (Natural History), London, iv + 296 pp. Gervais, H., & E Ameghino. 1880. Les mammiféres fossiles de 1 Amérique du Sud. FE Savy, Paris, xii + 225 pp. Gingerich, P. D., & D. A. Winkler. 1979. Patterns of variation and correlation in the dentition of the red fox, Vulpes vulpes.—Journal of Mammalo- gy 60(4):69 1-704. Gray, J. E. 1837a. [Characters of some new species of Mammalia in the Society’s Collection, with remarks upon the dentition of the Carnivora, and upon the value of the characters used by M. Cuvier to separate the plantigrade from the digitigrade Carnivora.|—Proceedings of the Zoological Society of London 1836:87-88. 1837b. Description of some new or little known Mammalia, principally in the British Museum Collection.—The Magazine of Natural History, and Journal of Zoology, Botany, Min- eralogy, Geology, and Meteorology 1:577—587. . 1843. List of the specimens of Mammalia in the collection of the British Museum. British Museum (Natural History), London, xxviii + 216 pp. . 1869. Notes on the skulls of the species of dogs, wolves, and foxes (Canidae) in the col- lection of the British Museum.—Proceedings of the Zoological Society of London 1868:492— S25) Hamilton-Smith, C. 1839. The natural history of dogs. Vol. I. The Naturalists Library, W. Jardine, ed. W. H. Lizars, Edinburgh, 9:frontispiece, i—xiv, 15-267, 33 pls. Hilzheimer, M. 1906. Papio mundamensis, Felis de- liensis, Canis reissii und andre neue Sduge- thiere.—Zoologischer Anzeiger 30:109-118. Huber, A. 1925. El zorro y su piel in la Republica Argentina.—Republica Argentina, Ministerio de Agricultura, Secci6n Propaganda e Informes, Numero 360, 11 pp. Ihering, H. von. 1911. Os mammiferos do Brazil Mer- iodinal.—Revista do Museo Paulista, Sao Paulo 8:147-275. Kerr, R. 1792. The animal kingdom or zoological sys- tem, of the celebrated Sir Charles Linnaeus. Class I. Mammalia: containing a complete sys- tematic description, arrangement, and nomen- clature, of all the known species and varieties of the mammalia, or animals which give suck to their young; being a translation of that part of the systema naturae, as lately published, with great improvements, by Professor Gmelin of PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Goettingen. Together with numerous additions from more recent zoological writers, and illus- trated with copperplates. A. Strahan, T. Cadell, and W. Creech, Edinburgh, xii, 1-32 + 32 (un- numbered) + 33—400, 7 pls. Kraglievich, L. 1930. Craneometria y clasificacion de los canidos sudamericanos, especialmente los argentinos, actuales y fdsiles——Physis 10:35— 13.5 7 DS. Langguth, A. 1967. Sobre la identidad de Dusicyon culpaeolus (Thomas) y de Dusicyon inca (Tho- mas).—Neotropica 13:21—28. . 1969. Die siidamerikanischen Canidae unter besonderer beriicksichtgung des mahnenwolfes Chrysocyon brachyurus MWliger.—Zeitschrift fiir Wissenschaftliche Zoologie 179(1—2):1—188. 1975. Ecology and evolution in the South American canids. Pp. 192—206 in M. W. Fox, ed., The wild canids, their systematics, behav- ioral ecology and evolution. Van Nostrand Reinhold Co, New York, xviii + 508 pp. Larranaga, D. A. 1923. Escritos. Instituto Histérico y Geografico del Uruguay, Montevideo, 2:1—512 + 2 tables. Linnaeus, C. 1758. Systema naturae per regna tria na- turae, secundum classes, ordines, genera, spe- cies, cum characteribus, differentiis, synonymis, locis. Editio decima, reformata. Laurentii Salvii, Holmiae, 1:1—824. Lonnberg, E. 1919. Remarks on some South Ameri- can Canidae.—Arkiv for Zoologi, Stockholm 12(13):1-18. 1921. A second contribution to the mam- malogy of Ecuador with some remarks on Caenolestes.—Arkiv f6r Zoologi, Stockholm 14(4):1-104. Lund, P W. 1837. Om Huler i Kalksteen 1 det indre af Brasilien der tildeels indeholde fossile Knok- ler. Anden Afhandling. Lappa da Cerca Gran- de.—Kongelige Danske, Videnskabernes Sel- skabs Naturvidenskabelige og Mathematiske Afhandlinger 6:307—332, pls. 1-3. . 1839a. Coup-d’oeil sur les espéces éteintes de Mammiferes du Brésil, extrait de quelques mémoires présentés 4 ]’Académie royale des Sciences de Copenhague.—Annales des Sci- ences Naturelles, Seconde Série 11:214—234. . 1839b. Pattedyrene.—Kongelige Danske, Vi- denskabernes Selskabs Naturvidenskabelige og Mathematiske Afhandlinger, 82 pp., 13 pls. [Re- paged preprint of Lund 1841a.] . 1840. Fortsaettelse af Pattedyrene.—Konge- lige Danske, Videnskabernes Selskabs Natur- videnskabelige og Mathematiske Afhandlinger, 56 pp., 11 pls. [Re-paged preprint of Lund 1841b.] 1841a. Blik paa Brasiliens Dyreverden for sidste Jordomvaeltning. Anden Afhandling: Pat- VOLUME 108, NUMBER 4 tedyrene.—Kongelige Danske, Videnskabernes Selskabs Naturvidenskabelige og Mathematiske Afhandlinger 8:63—144, pls. 1-13. 1841b. Blik paa Brasiliens Dyreverden for sidste Jordomvaeltning. Tredie Afhandling: Fortsaettelse af Pattedyrene.—Kongelige Dan- ske, Videnskabernes Selskabs Naturvidenska- belige og Mathematiske Afhandlinger 8:219- DY Xe . 1842a. Fortsatte Bemaerkninger over Brasi- liens uddéde Dyrskagning.—Kongelige Danske Videnskabernes Selskabs Naturvidenskabelige og Mathematiske Afhandlinger 3:1—16. [Re- paged preprint of Lund 1842b.] . 1842b. Fortsatte Bemaerkninger over Brasi- liens uddéde Dyrskagning.—Kongelige Danske Videnskabernes Selskabs Naturvidenskabelige og Mathematiske Afhandlinger 9:121—136. 1843. Fortsaettelse af Pattedyrene. Om de nulevende og uddéde Arter af Rovdyrenes Fam- ilie paa det tropiske Brasiliens indre hoisletter. Foérste Afdeling: Hundegruppen.—Kongelige Danske Videnskabernes Selskabs Naturviden- skabelige og Mathematiske Afhandlinger 5:1-— 82, 7 pls. [Re-paged preprint of Lund 1845.] . 1845. Blik paa Brasiliens Dyreverden for sid- ste Jordomvaeltning. Femte Afhandling. Fort- saettelse af Pattedyrene. Om de nulevende og uddéde Arter af Rovdyrenes Familie.—Konge- lige Danske Videnskabernes Selskabs Naturvi- denskabelige og Mathematiske Afhandlinger 11:1—82, pls. 40-46. Mann F, G. 1950. Nuevos mamiferos de Tarapaca.— Investigaciones Zooldgicas Chilenas 2:4—6. Martin, W. 1837. [Observations upon a new fox from Mr. Darwin’s Collection (Vulpes fulvipes).|— Proceedings of the Zoological Society of Lon- don 1837:11-12. Mivart, St. G. 1890a. Dogs, jackals, wolves, and fox- es: a monograph of the Canidae. R. H. Porter, London, xxxvi + 216 pp., 45 pls. [Published by April, 1890; see Nature Novitates, Berlin 12(8): 171.] . 1890b. Notes on the South-American Cani- dae.—Proceedings of the Zoological Society of London 1890:98—113. [Published August, 1890; see Duncan 1937.]| Molina, G. I. 1782. Saggio sulla storia naturale del Chili. Stamperia di S. Tommaso d’ Aquino, Bo- logna, 367 pp., 1 map. Nowak, R. M. 1991. Walker’s mammals of the World. Fifth edition. The John Hopkins University Press, London, 2:x + 643-1629. Osgood, W. H. 1914. Mammals of an expedition across northern Peru.—Field Museum of Natu- ral History, Zodlogical Series 10:143—185. . 1915. The name of Azara’s agouarachay.— 745 Proceedings of the Biological Society of Wash- ington 28:142-143. 1916. Mammals of the Collins-Day South American Expedition.—Field Museum of Nat- ural History, ZoGdlogical Series 10:199-216, plates 6—7. 1934. The genera and subgenera of South American canids.—Journal of Mammalogy 15: 45-50. 1943. The mammals of Chile.—Field Mu- seum of Natural History, Zoological Series 30: 1-268. Philippi, R. A. 1866. Ueber ein paar neue Chilenische Saugethiere.—Archiv ftir Naturgeschichte 32(1):113-117. . 1896. Dos animales nuevos de la fauna Chi- lena.—Anales de la Universidad de Santiago de Chile 94:541—546. [Reprint pages 1-6.] 1901. Nueva especie Chilena de zorras.— Anales de la Universidad de Santiago de Chile 108:167—170, 1 pl. 1903. Einige neue Chilenische Canis-Ar- ten.—Archiv fiir Naturgeschichte 69(1):155-— 160. Prichard, H. H. 1902. Through the heart of Patagonia. D. Appleton and Company, New York, xvi + 346 pp., 40 pls. Rengger, J. R. 1830. Naturgeschichte der Saeugethiere von Paraguay. Schweghauserschen, Basel, xvi + 394 pp. Rohlf, EF J. 1992. NTSYS-PC: numerical taxonomy and multivariate analysis system. Version 1.70. Ser. 1459CV. Exeter Software. Schinz, H. R. 1821. Das Thierreich eingetheilt nach dem Bau der Thiere als Grundlage ihrer Natur- geschichte und der vergleichenden Anatomie von dem Herrn Ritter von Cuvier. Erster band. Saéugethiere und Vogel. J. G. Cotta’schen Buch- handlung, Stuttgart und Tiibingen, 1:xxxviii + 894 pp. Simpson, G. G. 1940. The names Mesotherium and Typotherium.—American Journal of Science 238:518—521. . 1945. The principles of classification and a classification of mammals.—Bulletin of the American Museum of Natural History 85:xvi + 1—350. Spillmann, E 1938. Die fossilen Pferde Ekuadors der Gattung Neohippus.—Palaeobiologica 6:372— 383), Thomas, O. 1900. New South-American mammals.— Annals and Magazine of Natural History, Series 7, 5:148-153. . 1904. On the mammals collected by Mr. A. Robert at Chapada, Matto Grosso (Percy Sladen Expedition to Central Brazil).—Proceedings of the Zoological Society of London 1903, 2(2): 232-244, plate 27. 746 1914a. On various South-American mam- mals.—Annals and Magazine of Natural His- tory, Series 8, 13:345-—363. . 1914b. Three new S.-American mammals.— Annals and Magazine of Natural History, Series roy S25) 7/38) 7/5)a . 1921. New foxes of the genera Cerdocyon and Pseudalopex from northern Argentina.— Annals and Magazine of Natural History, Series 9, 7:381-385. Trouessart, E.-L. 1897. Catalogus mammalium tam viventium quam fossilium. Fasciculus II. Car- nivora, Pinnipedia, Rodentia I. R. Friedlander and Sohn, Berolini, 1:219-452. . 1904. Catalogus mammalium tam viventium quam fossilium. Quinquennale supplementium, anno 1904, Fascicule 1, pp. 1-288. R. Friedlan- der and Sohn, Berolini, vii + 929 pp. . 1906. Sur une espéce nouvelle du genre Ic- ticyon (Speothos), provenant de |’ Equateur.— Comptes Rendus des Séances de |’ Académie des Sciences 143:1184—1186. 1910. Mammiféres de la Mission de l’Eq- uateur d’apres les collections formées par le Dr. Rivet. Pp. 1-32, 8 plates in Mission du Service Géographique de l’Armée pour la mesure d’un arc de méridien équatoriale en Amérique du Sud sus le Controle Scientifique de 1’ Académie de Sciences, 1899-1906. Tome 9, Zoologie. Fas- cicule 1.—Mammiferes, Oisseux, Trochilidae. Ministére del’Instruction Publique, Paris. [Dated 1911; published November 24, 1910.] Van Gelder, R. G. 1978. A review of canid classifi- cation.—American Museum Novitates 2646:1-— 10. Vitullo, A. D., & G. A. Zuleta. 1992. Cytogenetics and fossil record: confluent evidence for speci- ation without chromosomal change in South American canids.—Zeitschrift ftir Saugetier- kunde 57:248—250. Waithman, J., & A. Roest. 1977. A taxonomic study of the kit fox, Vulpes macrotis.—Journal of Mammalogy 58:157—164. Waterhouse, G. R. 1839. Mammalia. Pp. 49-97 in C. Darwin, ed., The zoology of the voyage of the H.M.S. Beagle under the command of Captain Fitzroy, R. N., during the years 1832-1836. Smith, Elder and Company, London, 2:xii + 97 pp., 35 pls., 1838-1839. Wolffsohn, J. A. 1918. Resefia de los trabajos publi- cados desde 1895 por autores nacionales y ex- tranjeros sobre la mamalogia chilena.—Actes de la Société Scientifique du Chili 23:57—79. . 1921. Catalogo de craneos de mamiferos de Chile colectados entre los afios 1896—1918.— Revista Chilena de Historia Natural 25:511— S28). Wortman, J. L., & W. D. Matthew. 1899. The ancestry PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON of certain members of the Canidae, the Viver- ridae, and Procyonidae.—Bulletin of the Amer- ican Museum of Natural History 12:109-138, 1 table, 1 pl. Wozencraft, W. C. 1989. Classification of the Recent Carnivora. Pp. 569-593 in J. L. Gittleman, ed., Carnivore behavior, ecology and evolution. Cornell University Press, Ithaca, New York, xvi + 620 pp. . 1993. Order Carnivora. Pp. 279-348 in D. E. Wilson and D. M. Reeder, eds., Mammal spe- cies of the World: a taxonomic and geographic reference, Second Edition. Smithsonian Institu- tion Press, Washington, D.C., xvii + 1206 pp. Appendix 1 List of specimens included in the statistical analyses of craniometry and pelage. OTU acronyms for cranial analyses start with the letter C; those for pelage char- acter analysis start with a “P.”’ Museum acronyms: Museo Argentino de Ciencias Naturales, MACN; In- stituto Miguel Lillo, Tucuman, IML. Craniometric Analysis Lycalopex culpaeus (culpeo fox).—CCU (n = 58). (MACN) JUJUY, Abra Blanca (2): 38.039, 4.041; Li- zoite (1): 41.055. NEUQUEN, Catan-Lil (53): 15138— 1S 142, 151475 WS tS 115545 1S los wAlSGo stale 15173-15177, 15181, 15183-15184, 15188, 15190— 15192, 15194-15196, 15204, 15207-15208, 15211, 15220) 15221, 15223=15227. ls23i 5235-15236" 15238-15240, 15243-15244, 15247-15248, 15250— 15253. SANTA CRUZ, San Julian (2): 19221-19222. Lycalopex gymnocercus (gray fox).—CST (7). (MACN) CHUBUT, Lago Blanco (1): 224; Rio Mayo (2): 20207—20208. (MACN) SANTA CRUZ, Rio Ro- bles (2): 15692, 16321; (IML) SANTA CRUZ (2): 421—422. CWT (14). (MACN) MENDOZA, La Paz (1): 1737. (MACN) NEUQUEN, Zapala (1): 14902. (MACN) RIO NEGRO, Pilcaniyeu (4): 20205-20206, 20276, 20278: General Roca (8): 24.050, 24.052—24.054, 24.066, 24.079—24.081. CNW (8). (MACN) CATAMARCA, Andalgala (3): 50.419-50.420, 50.432; Salar de Pipanaco (1): 51.170; Singuil (1): 53.002; IML) CATAMARCA, Belén (1): 895. (MACN) SAN JUAN, Ischigualasto (1): 13781. (IML) SAN JUAN, Valle Fértil (1): 1178. CCT (33). (MACN) LA PAMPA, Caleu-Caleu (15): 13327, 1333159133372 15748=15752. 9157544 49189; 49.148-49.149, 49.159-49.160, 49.167; Carro Que- mado (11): 16120, 16127, 16130, 16136, 16140- 16141, 16145, 16147-16150; General Acha (7): 50.489, 50.492-50.493, 50.495-50.496, 50.499, 50.502. CET (15). (MACN) BUENOS AIRES, (1): 285; Puerto Quequén (1): 14409; Bolivar (2): 15363- 15364; Coronel Sudrez (4): 15387-15390; Balcarce VOLUME 108, NUMBER 4 (4): 24.133, 24.143, 24.148, 24.156; Azul (1): 26.028; Punta Médano (1): 26.163; Juarez (1): 54.133. CNT (16). (MACN) CORDOBA, La Paz (1): 29.035; Pampa de Olaén (1): 39.191; Soto (1): 39.194. (MACN) JUJUY, (1): 32.252. (MACN) SALTA, Me- tan (1): 14323; Dragones (1): 36.480. (MACN) SAN- TIAGO DEL ESTERO, Cerro Quemado (2): 30.210— 30.211; IML) SANTIAGO DEL ESTERO, La Banda (1): 908; Giménez (1): 959. (MACN) TUCUMAN, Tapia (1); 26.129; Monteros (1): 28.182; (IML) TU- CUMAN, Trancas (2): 192, 495; Tafi del Valle (2): 545, 836. Specimens Analyzed for Pelage Characters Lycalopex culpaeus (culpeo fox).—PCU (27). (MACN) RIO NEGRO, Valcheta (1): 14546; El Manso (1): 16419. (MACN) NEUQUEN, Aluminé (2): 13473, 13475; Catan-Lil (8): 15021-15024, 15030, 16169, 16171, 19223. (MACN) SANTA CRUZ, Puerto San Julian (3): 19221-19222, 50.449. (MACN) CHUBUT, Valle Lago Blanco (1): 207. (MACN) LA RIOJA, Ve- lazco (1): 34.316. (MACN) JUJUY, Cochinoca (1): 39.493; Lizoite (2): 41.055—41.056; Santa Catalina (2): 41.164-41.165. (MACN) TIERRA DEL FUEGO, Rio Grande (3): 50.480—50.482. (MACN) SALTA, San Antonio de los Cobres (2): 26.189—26.190. Lycalopex gymnocercus (gray fox).—PET (14). (MACN) BUENOS AIRES, Puerto Quequén (1): 747 14409; Bolivar (1): 15363; Sierra de La Ventana (9): 36.046—36.050, 36.052—36.055; Rocha (2): 39.713, 89 TS yuarez, (1): 54.133: PNW (6). (MACN) SAN JUAN, Ischigualasto (1): 13781. (MACN) LA RIOJA, Villa Union (1): 34.562. (MACN) CATAMARCA, Andalgala (3): 50.419- 50.420, 50.432; Singuil (1): 53.002. PNT (25). (MACN) CORDOBA, (1): 31.195; Bialet Massex())2 23921902 Soton (i) 393194. sitay Baz): 29.035; Valle de Los Reartes (1): 25.166; Sobremonte (1): 13299; Sierra de Achala (1): 25.110. (MACN) CHACO, (1): 30.206. (MACN) SALTA, Aguaray (1): 36.228; Metan (2): 14319, 14323; Dragones (2): 36.180, 36.479; Rio Carapari (1): 36.477. (MACN) SANTIAGO DEL ESTERO, Colonia Dora (1): 42.011. (MACN) FORMOSA, Ingeniero Juarez (5): 47.133-47.136, 47.138. (MACN) SAN LUIS, Chaca- buco (1): 14707; Chosmes (1): 49.224. (MACN) TU- CUMAN, Burruyact (1): 30.150; Tapia (1): 26.129; Monteros (1): 28.182. PWT (9). (MACN) MENDOZA, (2): 38.002-— 38.003; Capital (1): 17827. (MACN) NEUQUEN, Ca- tan-Lil (1): 17828; Junin de Los Andes (1): 38.223. (MACN) CHUBUT, Valle Lago Blanco (4): 209-212. PCT (30). (MACN) LA PAMPA, Lihuel-Calel (1): 15601; Limay-Mahuida (1): 16357; Caleu-Caleu (14): 49.134-49.136, 49.148, 49.159, 49.167, 49.174— 49.181; General Acha (14): 50.483—50.488, 50.492, 50.494, 50.498, 50.501, 50.503—50.505, 50.507. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(4):748-750. 1995 A new genus of the Compositae: Paramiflos (Espeletiinae) from Colombia Jose Cuatrecasas Department of Botany MRC-166, Smithsonian Institution, Washington, D.C. 20560, U.S.A. Abstract.—Paramiflos is a new genus from the high Andes of Colombia that belongs to the subtribe Espeletiinae of the Compositae. It differs significantly from the related genus Espeletiopsis and all the other genera of Espeletiinae in the structure of the involucre, which has the outer phyllaries sharply different from the inner phyllaries. During a study of all the taxa of the Es- peletiinae, while trying to place the old spe- cies Espeletia glandulosa within the genus Espeletiopsis, special attention was paid to the uncommon structure of its involucre. Almost all the features characteristic of this species agree with other species of Espele- tiopsis, except for the structure of the phyl- laries, which are of two types. The sterile (outer) ones are fewer (only 5-7), broad, herbaceous, green, and nearly glabrous. They are distinct from the fertile (inner) phyllaries subtending the ray florets, which differ in their smaller size and in their shape. The latter are scarious, loosely clasp- ing, and partially surround the floret. The contrast in these phyllaries distinguishes E. glandulosa not only from Espeletiopsis, but from all other genera of Espeletiinae. The involucral arrangement resembles rather that seen in Smallanthus Mackenzie (a seg- regate of Polymnia in the subtribe Melam- podiinae. Because of these differences, here is established a new genus, Paramiflos. The name means “flower of the paéramo,”’ the environment that is characteristic of the high tropical Andes. Paramiflos Cuatrecasas, gen. nov. Caulirosula trunco erecto brevi, foliis li- nearibus copiosissimis conferte rosulatis. Folia viridia coriacea angustissima margi- nibus revolutis. Plantae pachycaules univ- ersaliter glanduliferae. Inflorescentiae plu- res simultaneae axillare prodientes, quam folia duplo longiores vel ultra; axis erectus rigidus superne ramosus, ramis ramulisque alternis bracteatis in corymbo dispositis; pars inferior indivisa nuda ebracteata vel bractea singula distali. Capitula radiata cir- culo ligularum amplo, 20—35 mm diametro. Involucrum 13—16 mm diametro, cum 5—7 phyllariis exterioribus herbaceo-membran- aceis ovato-ellipticis vel obovatis vel oblon- gis attenuatis acutis, valde inaequalibus sae- pe asymmetricis 10-18 mm longis 7-11 mm latis, cum phyllariis interioribus satis distinctis bene discretis; aliquando 1-3 phyllariis similaribus sed minoribus inter- mediis cum phyllariis exterioribus. Phyllar- ia interiora herbacea uniformiter subverti- cillata, plerumque biseriata subaequalia, el- liptica acuta, dorso rigido convexo margi- nibus scariosis, 7-8 mm longa, 3-4 mm lata, flores femineos vel fructus amplecten- tia. Flosculi radii feminei corolla ligulata lutea, tubulo piloso et glandulifero. Fructus nigrescens trigonus faciebus laevibus. Flos- culi disci pseudohermaphroditi corolla tub- ulosa lutea pentamera parce pilosa et glan- dulifera; antherae 1.8—-2 mm longae, basi breviter sagittatae; stylo apice conico bilo- bato papilloso, ovario sterile. Grana pollinia sphaeroidalia, sphaeroidale-subprolata, sphaerodale prolata, (23.5)27—31 wm pol. diametro, (23)24-27 wm aeq. diametro, spinis 4.5 wm longis, numero 12. VOLUME 108, NUMBER 4 749 Fig. 1. Paramiflos glandulosus (Cuatrecasas) Cuatrecasas. A. Terminal part of inflorescence branchlet sup- porting three flower buds. B. Extended involucrum, abaxial view. C. Capitulum. D. Row of inner phyllaries with the outer phyllaries removed. E. Disk flower. Type: Espeletia glandulosa Cuatrecasas 23, pl. Hl. 1940. Type: Colombia: Boy- [=Paramiflos glandulosus (Cuatrecasas) aca: Alto de Canutos, Paramo de Guan- Cuatrecasas]. tiva, south side, 3000 m elev., 3 Aug 1940, Cuatrecasas 10360 (holotype COL; isotypes EK US). Espeletiopsis glandulosa (Cuatrecasas) Cuatrecasas, Phytologia 35(1):55. 1976. Paramiflos glandulosus (Cuatrecasas) Cuatrecasas, comb. nov. Figs. 1, 2 Espeletia glandulosa Cuatrecasas, Revista Acad. Colomb. Cienc. 3:434, f. 17, 18, Distribution.—Subpdéramo and paramo 750 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON WLE Fig. 2. Photographs of Paramiflos glandulosus at El Bosque, in a paramo close to Alto de Canutos, 3250 m elevation, in Boyaca, Cordillera Oriental, Colombia. Cuatrecasas photos C-1683 and I-3863. habitats of the Eastern Cordillera of the An- des in the Department of Boyaca, Colom- bia. Discussion.—Paramiflos glandulosus as first published was distinguished by the caulirosulous habit, with dense, linear, very narrow, stiff leaves; by monochasial, cor- ymbose, deciduous, axillary inflorescences with ebracteate scapes; by bright yellow heads; and by the general glandulosity of the plant. Later, because of some of these charac- teristics, the species was transferred to Es- peletiopsis (Cuatrecasas, 1976). More re- cently, I realized that the structure of the involucre in this species differs substantial- ly from the one common in that genus and in all other Espeletiinae. In E. glandulosa as described, the invo- lucre has 5-7 discretely separate, herba- ceous outer phyllaries that are different in shape and size from the pseudoverticillate inner phyllaries that are more like the re- ceptacular pales. This arrangement is oth- erwise unknown in Espeletiopsis and the other genera of Espeletiinae, which have a gradual and progressive change from the outermost phyllaries to the inner phyllaries and the receptacular pales. The distinctive structure observed in P. glandulosus is similar to one already known in some other genera of the Compositae, e.g., Smallanthus (Melampodiinae). How- ever, the characteristic alternate leaves and bracts (not frondose), the triangular non- striate, non-furrowed blackish achenes, the pachycaulous and caulirosular life-form of the plant, and the decidous, monochasial, axiallary infloresences clearly define Par- amiflos as anew genus of Espeletiinae close to Espeletiopsis. Literature Cited Cuatrecasas, J. 1976. A new subtribe in the Helian- theae (Compositae): Espeletiinae—Phytologia 35(1):43-61. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 108(4):75 1-752. 1995 INTERNATIONAL COMMISSION ON ZOOLOGICAL NOMENCLATURE Applications published in the Bulletin of Zoological Nomenclature The following Applications were published on 30 June 1995 in Vol. 52, Part 2 of the Bulletin of Zoological Nomenclature. Comment or advise on any of the appli- cations is invited for publication in the Bulletin and should be sent to the Executive Secretary (I.C.Z.N.), % The Natural History Museum, Cromwell Road, London SW7 5BD, U.K. Case No. 2900 Porites Link, 1807, Galaxea Oken, 1815, Mussa Oken, 1815 and Dendro- phyllia Blainville, 1830 (Anthozoa, Scleractinia): proposed conser- vation. 2903 Tropidoptera Ancey, 1889 (Mollusca, Gastropoda): proposed designation of Endodonta wesleyi Sykes, 1896 as the type species. 2946 PLUTONIINAE Bollman, 1893 (Arthropoda, Chilopoda) and PLUTONI- INAE Cockerell, 1893 (Mollusca, Gastropoda): proposed removal of homonymy. Cubaris murina Brandt, 1833 (Crustacea, Isopoda): proposed conservation of both the generic and specific names. Xerammobates Popov, 1951 (Insecta, Hymenoptera): proposed designation of Ammobates (Xerammobates) oxianus Popov, 1951 as the type spe- cies. Melissodes desponsa Smith, 1854 and M. agilis Cresson, 1878 (Insecta, Hy- menoptera): proposed conservation of the specific names. Rhabdomeson Young & Young, 1874 (Bryozoa): proposed designation of Rhabdomeson progracile Wyse Jackson & Bancroft, 1995 as the type species. Nectria Gray, 1840 (Echinodermata, Asteroidea): proposed designation of Nectria ocelleta Perrier, 1875 as the type species. Phyllophis carinata Giinther, 1864 (currently Elaphe carinata; Reptilia, Ser- pentes): proposed conservation of the specific name. Aptornis Owen, [1848] (Aves): proposed conservation as the correct original spelling. 752 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Opinions published in the Bulletin of Zoological Nomenclature The following Opinions were published on 30 June 1995 in Vol. 52, Part 2 of the Bulletin of Zoological Nomenclature. Copies of these Opinions can be obtained free of charge from the Executive Secretary, I.C.Z.N., % The Natural History Museum, Cromwell Road, London SW7 5BD, U.K. Opinion No. 1803 Robulina nodosa Reuss, 1863, (currently Lenticulina nodosa; Foraminiferi- — da): neotype confirmed as the name-bearing type. 1804 Cristellaria humilis Reuss, 1863 (currently Astacolus humilis; Foraminiferi- da): neotype replaced by rediscovered lectotype, and Rotalia schloenbachi (currently Notoplanulina? schloenbachi; Foraminiferi- da): placed on the Official List. Doris grandiflora Rapp, 1827 (currently Dendrodoris grandiflora) and Dor- idopsis guttata Odhner, 1917 (currently Dendrodoris guttata) (Mol- lusca, Gastropoda): specific names conserved. Ammonites nodosus (currently Ceratites nodosus; Cephalopoda, Ammonoi- dea): specific name attributed to Schlotheim, 1813, and a lectotype designated. Johnstonia Quatrefages, 1866 (Annelida, Polychaeta): conserved. Mastotermes darwiniensis Froggatt, 1897 and Termes meridionalis Froggatt, 1898 (currently Amitermes meridionalis) (Insecta, Isoptera): neoty- pes retained following rediscovery of syntypes. Bruchus Linnaeus, 1767, Ptinus Linnaeus, 1767 and Mylabris Fabricius, 1775 (Insecta, Coleoptera): conserved. Cryptophagus Herbst, 1792, Dorcatoma Herbst, 1792, Rhizophagus Herbst, 1793 and Colon Herbst, 1797 (Insecta, Coleoptera): conserved as the correct original spellings, and Lyctus bipustulatus Fabricius, 1792 ruled to be the type species of Rhizophagus. COLYDMDAE Erichson, 1842 (Insecta, Coleoptera): given precedence over CER- YLONIDAE Billberg, 1820 and ORTHOCERINI Blanchard, 1845 (1820); and Cerylon Latreille, 1802: Lyctus histeroides Fabricius, 1792 des- ignated as the type species. ELMIDAE Curtis, 1830 (Insecta, Coleoptera): conserved as the correct original spelling, and the gender of Elmis Latreille, 1802 ruled to be femi- nine. Alestes Miiller & Troschel, 1844 (Osteichthyes, Characiformes): conserved. Catharacta antarctica lonnbergi Mathews, 1912 (currently Catharacta skua lonnbergi) and Catharacta skua hamiltoni Hagen, 1952 (Aves, Charadriiformes): subspecific names conserved. INFORMATION FOR CONTRIBUTORS Content.—The Proceedings of the Biological Society of Washington contains papers bearing on systematics in the biological sciences (botany, zoology, and paleontology), and notices of business transacted at meetings of the Society. Except at the direction of the Council, only manuscripts by Society members will be accepted. Papers are published in English (except for Latin diagnoses/descriptions of plant taxa), with an Abstract in an alternate language when appropriate. Submission of manuscripts.—Submit three copies of each manuscript in the style of the Proceedings to the Editor, complete with tables, figure captions, and figures (plus originals of the illustrations). Mail directly to: Editor, Proceedings of the Biological Society of Wash- ington, National Museum of Natural History NHB-108, Smithsonian Institution, Washington, D.C. 20560. (Do not submit manuscripts to an associate editor). Presentation.—Clarity of presentation, and requirements of taxonomic and nomenclatural procedures necessitate reasonable consistency in the organization of papers. Telegraphic style is recommended for descriptions and diagnoses. The style for the Proceedings is described in ““GUIDELINES FOR MANUSCRIPTS for Publications of the BIOLOGICAL SOCIETY OF WASHINGTON” a supplement to Volume 103, number 1, March 1990. Authors are encour- aged to consult this article before manuscript preparation. Copies of the article are available from the editor or any associate editor. The establishment of new taxa must conform with the requirements of appropriate interna- tional codes of nomenclature. Decisions of the editor about style also are guided by the General Recommendations (Appendix E) of the International Code of Zoological Nomenclature. When appropriate, accounts of new taxa must cite a type specimen deposited in an institutional col- lection. Review.—One of the Society’s aims is to give its members an opportunity for prompt pub- lication of their shorter contributions. Manuscripts are reviewed by a board of Associate Editors and appropriate referees. Proofs.—Authors will receive first proofs and original manuscript for correction and ap- proval. Both must be returned within 48 hours to the Editor. Reprint orders are taken with returned proofs. Publication charges.—Authors are required to pay full costs of figures, tables, changes in proofs ($3.00 per change or revision), and reprints. Authors are also asked to assume costs of page-charges. The Society, on request, will subsidize a limited number of contributions per volume. If subsidized manuscripts result in more than 12 printed pages, the additional pages must be financed by the author(s). Multiple authorship will not alter the 12 page limit (each author will be viewed as having used his/her 12 subsidized pages). Payment of full costs will facilitate speedy publication. Costs.—Printed pages @ $60.00, figures @ $10.00, tabular material @ $3.00 per printed inch per column. One ms. page = approximately 0.4 printed page. CONTENTS Paraturbanella solitaria, a new psammic species (Gastrotricha: Macrodasyida: Turbanellidae), from the coast of California M. Antonio Todaro Rissoella ornata, a new species of Rissoellidae (Mollusca: Gastropoda: Rissoelloidea) from the southeastern coast of Brazil Luiz Ricardo L. Simone Description of a new viviparous species of Dentatisyllis (Polychaeta: Syllidae) from Belize with an assessment of growth and variation, and emendation of the genus David E. Russell New genera for two polychaetes of Lepidonotinae Marian H. Pettibone Two new species of Opisthotrochopodus (Polychaeta: Polynoidae: Branchinotogluminae) from the Lau and the North Fiji Bac-arc Basins, southwestern Pacific Ocean Tomoyuki Miura and Daniel Desbruyéres Two species of Oxyurostylis (Crustacea: Cumacea: Diastylidae), O. smithi Calman, 1912 and O. lecroyae, a new species from the Gulf of Mexico Daniel Roccatagliata and Richard W. Heard Leucothoe laurensi, a new species of leucothoid amphipod from Cuban waters (Crustacea: Amphipoda: Leucothoidae) James Darwin Thomas and Manuel Ortiz A new species of the genus Discorsopagurus (Crustacea: Decapoda: Paguridae) from Japan, previously known as D. schmitti (Stevens) Tomoyuki Komai Redescription of Sudanonautes faradjensis (Rathbun, 1921), a fresh-water crab from Central Africa (Brachyura: Potamoidea: Potamonautidae) Neil Cumberlidge Lithoscaptus pardalotus, a new species of coral-dwelling gall crab (Crustacea: Brachyura: Cryptochiridae) from Belau Roy K. Kropp Charybdis hellerii (Milne Edwards, 1867), a nonindigenous portunid crab (Crustacea: Decap- oda: Brachyura) discovered in the Indian River lagoon system of Florida Rafael Lemaitre Two new species of freshwater crabs of the genus Hypolobocera from Colombia (Crustacea: Decapoda: Pseudothelphusidae) Martha R. Campos and Gilberto Rodriguez Redescription of a unique feather star (Echinodermata: Crinoidea: Comatulida: Comasteridae) with the diagnosis of a new genus Charles G. Messing Gymnotus maculosus, a new species of electric fish (Chlordata: Teleostei: Gymnotoidei) from Middle America, with a key to species of Gymnotus James Sperling Albert and Robert Rush Miller A new species of the genus Bryconamericus Eigenmann, 1907 from southern Brazil (Ostar- iophysi: Characidae) Luiz R. Malabarba and Andreas Kindel Gobiodon acicularis, a new species of gobioid fish (Teleostei: Gobiidae) from Belau, Micro- nesia Antony S. Harold and Richard Winterbottom South American Toney habitat Leptodactylus (Amphibia: Anura: Leptodactylidae) with descrip- tion of two new species W. Ronald Heyer The taxonomic status of the shrew of St. Lawrence Island, Bering Sea (Mammalia: Soricidae) R. L. Rausch and V. R. Rausch Taxonomy of the genus Lycalopex (Carnivora: Canidae) in Argentina Gabriel E. Zunino, Olga B. Vaccaro, Marcelo Canevari, and Alfred L. Gardner A new genus of the Compositae: Paramiflos (Espeletiinae) from Colombia Jose Cuatrecasas International Commission on Zoological Nomenclature Table of Contents, Volume 108 Index to New Taxa, Volume 108 SDS 560 568 Si 583 596 613 617 629 637 643 649 656 662 679 687 695 Calg, 729 748 TS) iV vil a i . , LIBRARIES UTIO AIIM nn 3 9088 01205 2338 | PGi ates PORES OTS oaaymthy nip) ae bs wea ane. Uo ve eae Fin 36 PR Hew in Ainge gelle, ae Vera ce tog tty i : ABA Pn 5 We aes er Te Vora he Sect Lore sie ne FNS jek ISAs : : ake ‘ : : i at PPD HT a een reenter - 2 2 ? Se ery ayo ape , 2 3 2 i . sip teee AP be Simp ts LUG ENE Sp Ae ge Sega eo es SB SIR PE Se ace Fein UR SiN Le SpE oy FE HME ete FES HG Peat ree Santi 3 te Seas < 5 FRA ee sata seeps TO Rp ee ee YN = Sales dey ot Ae Soe ta a “Ge é “ ¥ aes 3 : : u : ae Z " tee Hes : Nhe asf 23: “ae : c Meee ate Pt Anemnwa tangy 4 z 5% = ee Sones epaiate: Bia tRet fi ; : ; eons S praret cate i A f SIRENS Hat 9a I GI RE es 2 meee 4 c 7 z : z ‘ s : 7 : : j rent { pan aera Seance Page a OM dish a ye ela A ad : weed OSIM 2 F : Uy : ° z % ; SCR INA! GAR ar rng geen reste Reaeap ey Nesebar eer ohun roa Rl aaenn Neuss o5 sat ise gases § e rie atk 4 5 , mae ¥ AU GEA FF 3 Gre NT CAR s A : Past ; PO Vieds Ph a ; i ne : . : eee: Mt : ‘ rs ne Pay ea ys! Ts / : ; > FRARON ve PAn Meteo j atp! fd waar Bes Ba A IS ep SNS SNES ARS, spa ott ae 5 a i ‘i 4 ‘ * 7 acteal E Dogme ane Sie eee Zest . . Z Berry oes 3 : ; ‘ F e epi eateel pale og ea 4 ea nie femur Magee a PA ae Ree < 5 2 ais g 5 % : é y PAG se ee pedal sa ig yause genes ray sents ran pS erik BRN bee y , ¥ 2 2 : s