ASAE rans ate ane anno mee Tanne Adee wt sar ee SNe Ts ctctbesaemi ae! Gens versie peihete Red apres TE aes ea sine Etat Dataszen oRS reece wn we een a Foceyer eayrenSe EGO: See rea BOS Reserve tre ra etait Bet ce eae STA aes RE mee eee nat» Soothe Sea E MN Cafe THIN sp ope are eon pee iene Soe IOS se Pont eatian i’ sedate tek Bae Aadietners tery sone Ow ciara mE Ree eae de ee a Pom ett ane st spe et ita! pene ee ST ee REE ae easton ed Eom awa ES Ce ae an tn tek mR NES cited APM gekinastet NT Ee eatin armen tn en ot aed BAREIS: ana Rete EE Neen Seto poate Fr nant ninneinteate Sern cet een nition CNT Si Tent ase ES Sesto Liens Tene pay eeeeier trea Patra sogtawind Moe iter tte raee Getto ne DAF seein pbeang eth Ten mension Or Garvee Scat er i memet AT ete SEE pen Rae ie Fale Sie M ne ETT Sas Smtne yale ROE 5 vie roa a AD Peganamnnnitn puterd par Sr Pye oom ee hoo ees ee Sa Guezinat mv Plena Lf ipee ate SAN Er oe se arinmny peeTIE on test aast te ad eA Set a Tok Naren poet stint cpeeasrs eamcret oe eres eae Me eh eS Sebyseansteretnde ele sprstitn © on eat nsw ban on jenn Fara 9 ST atonal oe ve Foperiten kth Yam bate Pot shee nPafIaM eat awe ta oma feet SoMa” seaport rata june star” eee Hat Soiy ptene™ a ent Te : AAU pilin sat RSH Menace rns Fy slate Set Seg Vea Hea na OSU AT - Nara hesshenian bin’ Mating Montes Pv Rastschs Tess Fem ean Doak RAED spp ct Me Tea tOP ONE ASN UN AER, ¥ SE OM aR RAR years AY Re SEF = ip itetety MPSS APS APE NAY ot 5 5 choy hasnt a mvc Rea ttl sta AAAI NBONS ae sikh Gist Pawar en cea eto OS OTE Ge aN s/h ASS MNO MOET A PRO e rrr Pea Jogheciehes apart Masnlcethy Py etuaeR TEAM Rs Mel shen gates Rt, im ORM MATES wenden Vee Mabe BCT 8 w= Uno atc afeaPe AN tei eer rs preyenite eran any ih a Serer! pre en treet peapeeverer tier tui tana ne seman RE Hane aye ma panna an to en DANS lisesi poten fe MANA Aa Thre Ay mtv sale MunsvareasteeTan iret WA, EDotayaAa harbor Pea yee aba Re EON ageism A oestT a, onthe fee pearly eon Nn rere vn Mah ease orat qoeetey Stbee adams AAs paerrer Tee SEU i Oe Le aang Kotte eA MSDE Sue Be x BicMhatiaBD ARSE Fo ANA Rapti a hoe TALE IS WS eg dg Rp esta dha es a A es aor Jeane dey. ts Sm Beatem wales Sails SIS Ne) SF es maaan wad ave aseeconS/Nisyasuimlamitra aiboe ct Jicprawate nite fe desis tthe Ani feasho eer Tena be ofiby mn tanhiteantze Ns @nD Nn arth Sos aan cag HRSA FT ey Heracles Meanie title Sa beitente eae os Laas PME TL et aU on ee yen ged eerste Saree swith = Bpatnctetts 90 sath ns aB phen r anpn hte PMT CENA let svagmanthalve Sete arene ferent thes Perera) net gee Sneed MN Teter nye ae re re SRL RENE SY sehpenesk net eniy afb son ae Data Tha 9 87M testa aN bores iene : ee ah piety neanatinste np FUSES Yen aOh 5 srerresas WMS SREP sage 3 . : Ly tata AMIE ea Ven ofan ria tenata mn ostston é f ea ASTIN Ney Ra ns nekeus i fog Bee Mimange ses Menten htt Aa NO Fema a AoE. Ihe fy ams Fra bgt tn Bes RSMEANS NE sersesliste tere ths ae aha tater ee gt RATE ete ey ik Stina rely AR BO ape cantante aaa Denes Ae RAN TD AA 4 pret : eon batremeay acess sameneeatel “ nr en enone onneencannert tap hes toesr ame nt ie : Resin ashe fe ne eh eT ‘ eee rca batinaceaimnencaiiomatst rare tk ale gas ma Rent efirmacncranence i Ra rena tcecsunianeuenes agrestis Mannan coctan a mpmme nnn Qtndy ter bade aneaten sme na tomsee since Eeere ener 94 haa pee TAR I em Seer enT) ca puneta taNentnding gare ye cay ean ae MU ey, Baa een es eT ai aie Pe as pat HATH nee earee ee rey pane tent tee SS 7 Oo : § NAG ‘3 5 : O Se > ISS O Je Oo < Z E Wo Zz: E Z iS > ST NT ee = > = m2) ove ” Z 75) z ” _NVINOSHLINS SAl¥V¥d!1 LIBRARIES INSTITUTION a > 7p) > ” ” o lu = uw es. a a oa 3 = (aed Bo (se = al x At < e Se 1S 4 = Cs 4 = =. fat rae ss rsa) (e) oO =s Oo — z Fa aly = =| LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVYUdIT LIBRA S3a1iYVYSIT LIBRARIES INSTITUTION INSTITUTION INSTITUTION Saiuvudil NOILALILSNI NVINOSHLINS S31¥vual 2 = 3 Z Z Ze = =: S ar S 5 = z S 2 2 B = 2 E - 2 i a BOT Wit ad os a 2 LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI w > 2) = ” = wl +2 Z a 2 & a = oc = ze x < 2 < at < pa oc = ox S foe S (aa) = m _— mM = = Ye: 2) — oO _ 2) al za - a 7 ae Zz NOILALILSNI _NVINOSHLIWS S3IYVYaIT LIBRARIES SMITHSONIAN INSTITUTION _ NOILAL ~ z= Be z C S : is z Oo > — ow — ive) = a \ = Bs) = ee 5 > Yo WN = > ra = = rN NY Zz = ‘tee = 2p) = “i ars a ce 4 Z : Z LIBRARIES SMITHSONIAN INSTITUTION NOILNJILSNI_ NVINOSHLINS Saluvudi7 LIBRA = = 4 LL; = z \ a YM zZ is 2 2 a > = Se >" = oe 2 ” z = a x, NOILNLILSNI_ NVINOSHLINS SA3l1YVYdIT LIBRARIES INSTITUTION NOILAL LIBRARI ES SMITHSONIAN NOILNLILSNI LIBRARIES NOILNLILSNI LIBRARIES NOILNLILSNI LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNI NVINOSHLINS S3Z1YVYUEIT LIBRA Saluvuadin INSTITUTION S3aluvugil INSTITUTION INSTITUTION NOILALILSNI NVINOSHLINS S3I14YVYEIT LIBRARIES SMITHSONIAN INSTITUTION NOILNI NVINOSHLINS S31iu¥vual WM E 2 = Ss =) ees iy VAgrs A pet fy = = : = ‘ Wiens f (@) mA, Y Or: NY Soa 4? 8ZY IRS a NEE? a: se: 5 CN E@ hohe ies era - : . ” at Zz i LIBRARIES SMITHSONIAN INSTITUTION NOILMIILSNI NVINOSHLINS SAJIYVYSEIT LIBRA = > ee ” = [EVI = nf fy? 2 (EMR = Gam 2 SSS = Gam) 2 Gav a bets ——, < = S = Ss = = = Zz = z 4 iD) as 2 1@) =f (@) a f 8 = g 2 g Ss S é = Ba FE Zz S . = = >" xs x > op) Fa ” Pa 2p) ae LE SHLINS S3IYVYNEIT LIBRARIES INSTITUTION NOILALILSNI_ NVINOSHLINS : ff 2 2 : sis ” s n sae a NS =a Pe Se oc S LQ \ ne = < ie < a NS. Xs = = a = ns = RNAS oR a = = S = > FS ISONIAN INSTITUTION NOILNLILSNI. NVINOSHLINS S3I1uvadll LIBRARIES SMITHSONIAD (a z Cc z — a Wy ce iB a 2 ae By a e oD ne i Z ISHLINS SAZIYVYUEIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI (ep) ae oes Ww Zz a w Pe = Ve Cie = x ea 1s nes + 4 mn SS \ —) > ~. = za = Be O We. W Sle oO Se SS ae (@) iy eo XK S mp) \ D> 7 2 = z E Z F > i = ee = > = 2 Ww a * Zz wo Loe Fa 4 Ww ISONIAN INSTITUTION NOILMIILSNI NVINOSHLINS S3IYVYSIT LIBRARIES SMITHSONIAD 4 = Ww ae us 2 ee) 4 i : . = o = as a ee = :, ke bia Sa WW = 2 = (ees = a NS 2 ri g : z m > >’ = : > Ww) Fe ” 2 Y”) “ Zs ie LIBRARIES INSTITUTION Pola Snia WwiNest ss lal a ue us = ul Z uw a oc a = tea ow = = fi 3 S jaa) ed mo. a feo) = mS = oO are oO = fe) a =z aly Zz ial z= THSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVUdIT LIBRARIES SMITHSONI/ is Zz 5 2 ia z par Oo = Seu ro) ia Oo isa) — s = 10.9) _ ee = 2 WY ‘3 = ec = Lon y - : EN = : 2 OE = = . = ~ Yy NOSHLIWS S23 luvud bog) BRARI ES SMITHSONIAN INSTITUTION |, NOILMLILSNI_ NVINOSHLI) Ee ce \S = iN ee Gig) ez 2 pal fe BANS = . = = <= B fp Oo QS = 5 al = > Ss : > ; = ae (op) * Zz ” my Fa ; (Tp) THSONIAN _ INSTITUTION NOILOLILSNI NVINOSHLIWS SAIMVYGIT_LIBRARIES SMITHSONI/ z S ee ” 2 ep ipn ats Mis ” Srey ay a Kei et . NS = sn a Ee Sere’ Gi, PROCEEDINGS of the Biological Society of Washington VOLUME 98 1985 Vol. 98(1) published 20 March 1985 Vol. 98(2) published 30 August 1985 Vol. 98(3) published 16 May 1985 Vol. 98(4) published 4 December 1985 WASHINGTON PRINTED FOR THE SOCIETY EDITOR BRIAN KENSLEY ASSOCIATE EDITORS Classical Languages Invertebrates GEORGE C. STEYSKAL THOMAS E. BOWMAN Plants Vertebrates DAvIpD B. LELLINGER ; RICHARD V. VARI Insects ROBERT D. GORDON 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 1984-1985 OFFICERS President DONALD R. DAVIS Vice President AUSTIN B. WILLIAMS Secretary GORDON L. HENDLER Treasurer LESLIE W. KNAPP COUNCIL Elected Members J. LAURENS BARNARD MAUREEN E. DOWNEY FREDERICK M. BAYER LOUIS S. KORNICKER ISABEL C. CANET STORRS L. OLSON TABLE OF CONTENTS Volume 98 Baez R., Pedro. Eupleurodon peruvianus (Rathbun, 1923); a species of crab newly re- corded from Chile (Crustacea: Decapoda: Brachyura) eect Barnard, J. L., and Janice Clark. A new sea-cave amphipod from Bermuda (CMD ea ee FS) a aa ac eee el nee bee rc A cee ee Bauer, Raymond T. Penaeoid shrimp fauna from tropical seagrass meadows: species composition, diurnal, and seasonal variation in abundance. eee Becker, Jonathan J. Pandion lovensis, a new species of osprey from the Late Miocene (3 28) i CO) aa ie tare eee eae ee cahoots as OR See oe OE Oe I Bernard, Ernest C. Two new species of Protura (Insecta) from North America........... Bowman, Thomas E. The correct identity of the pelagic amphipod Primno macropa, with a diagnosis of Primno abyssalis (Hyperiidea: Phrosimidae) . eee Bowman, Thomas E. Heteromysoides dennisi, a new mysidacean crustacean from Cem- etery ‘Cave: (Grand! Balvarmal [slaty cee accede cece caer ae Bowman, Thomas E. Thermosphaeroma cavicauda and T. macrura, new sphaeromatid iSOpodsHromyMexicanvhOtts prin gs eee Bowman, Thomas E., and Boris Sket. Remasellus, a new genus for the troglobitic swim- ming Florida asellid isopod, Asellius parvius SteeVes .........cecceeeeeeceeeeeeeeeeeeeeeeevve cence ene Bowman, Thomas E., Jill Yager, and Thomas M. Iliffe. Speonebalia cannoni, n. gen., n. sp., from the Caicos Islands, the first hypogean leptostracan (Nebaliacea: ING Dalla ae) rec ont loser ta Me ete ane ee ee lee te es i ee Brakoniecki, Thomas F., and Clyde F. E. Roper. Lolliguncula argus, a new species of loliginid squid (Cephalopoda: Myopsidae) from the tropical eastern Pacific... Brinkhurst, Ralph O. The generic and subfamilial classification of the Naididae (An- ME Ga Oli soc hac tea) ie sees eR A sd ete oleae ee ct Brinkhurst, Ralph O. Varichaetadrilus minutus (Brinkhurst, 1965) new combination for Psammoryctides (?) minutus (Oligochaeta: Tubificidae) ee eect Brinkhurst, Ralph O., and Michael T. Barbour. A new species of Aulodrilus Bretscher (Oligochaeta: Tubificidae) from North Ammerica accent Brinkhurst, Ralph O., and Kathryn A. Coates. The genus Paranais (Oligochaeta: Na- 1Gidae)einsINorthiyA wm eri Caer se ar ee reve sce ae ae MR edo see Brinkhurst, Ralph O., and Robert J. Diaz. Isochaetides columbiensis new species (Oli- gochaeta: Tubificidae) from the Columbia River, Oregon Buden, Donald W. New subspecies of thick-billed vireo (Aves: Vireonidae) from the Caicos Islands, with remarks on taxonomic status of other populations... Buden, Donald W. A new subspecies of Common Ground-dove from Ile de la Tortue, Haiti, with taxonomic reappraisal of Bahaman populations (Aves: Columbidae).......... Cairns, Stephen D. Three new species of Stylasteridae (Coelenterata: Hydrozoa)................ Campos, Martha R., and Gilberto Rodriguez. A new species of Neostrengeria (Crustacea: Decapoda: Pseudothelphusidae) with notes on geographical distribution of the Cannatella, David C. The systematic status of Syrrhopus juninensis Shreve (Anura: Leptodactylidac) te 220 ohm eenean Re ee ek nS ON ee en Oe ee BR Carleton, Michael D., and C. Brian Robbins. On the status and affinities of Hybomys planifrons (Miller, 1900) (Rodentia: Muricae). eee ceneeeeneeeetnnnnnene Clark, Guy T., and Ronald Fricke. A new species of dragonet, Synchiropus randalli, from Easter Island (Teleostei: Calliomyrmidae) cece ence eeeneeneeneeeeeenetnetene Clark, Janice, and J. L. Barnard. Lucayarina catacumba, new genus, new species, a Bahamian sea-cave amphipod (Crustacea: Amphipoda: Lysianassidae).....- Collette, Bruce B. Zenarchopterus ornithocephala, a new species of freshwater halfbeak (Pisces: Hemirhamphidae) from the Vogelkop Peninsula of New Guinea... Cuatrecasas, Jose. Studies in neotropical Senecioneae IV. New taxa in Senecio and Ga breniel aie sn esate yt eS Nell i, 1 PRES ER PR Sot oe eee Te aaa Ee 561-563 1048-1053 177-190 314-320 72-80 121-126 945-948 1042-1047 554-560 439-446 47-53 470-475 655-656 931-934 303-313 949-953 591-597 790-798 728-739 718-727 774-777 956-1003 539-543 243-254 107-111 623-626 Cutler, Edward B., and Norma J. Cutler. A revision of the genera Phascolion Théel, and Onchnesoma Koren and Danielssen (Sipuncula) - i ecceeeceeeceeccsccescesessessesnseeeseseeseceeeeeeeeeeeeeeeeeeeeee Dauer, Daniel M. A new species of Scolelepis (Polychaeta: Spionidae) from Lizard Island, PANUIS Uiteal lita eeseeomnne ween teen oN, eee Aes Bee aN oN en aerial EE Davis, Dale. The Oligochaeta of Georges Bank (NW Atlantic), with descriptions of four Flint, Oliver S. Jr., and Elisa B. Angrisano. Studies of neotropical caddisflies XXXV: The immature stages of Banyallarga argentinica Flint (Trichoptera: Gal ATVOCE Fak AAC) ee aN ae he ace wl ee AD tei ESE hd hte Ee. Formas, J. R. The voices and relationships of the Chilean frogs Eupsophus migueli and E. calcaratus (Amphibia: Anura: Leptodactylidae) 0... cccccccccccceeeeeeeeeeeeceeeceeeeeeeeeeceeeeeenennens Green, Timothy M. Pinnotheres jamesi synonymized with P. reticulatus (Decapoda: VBS RENG) YAU eer er a Pe Se Oe nee Hafner, John C. New kangaroo mice, genus Microdipodops (Rodentia: Heteromyidae), ATE TINOLG al @ paty GIN evel Cl aera aes hed err ccs eons ea Se Te ee Haman, Drew. Review of the foraminiferal genus Orbignynella Saidova, 1971 ................ Hart, C. W., Jr., R. B. Manning, and T. M. Iliffe. The fauna of Atlantic marine caves: evidence of dispersal by sea floor spreading while maintaining ties to deep waters...... Hawkes, Clayton R., Theodore R. Meyers, and Thomas C. Shirley. Larval biology of Briarosaccus callosus Boschma (Cirripedia: Rhizocephala) .. eee cee eee eeneeeee Heyer, W. Ronald. New species of frogs from Boracéia, Sao Paulo, Brazil... Hobbs, Horton H., Jr., and Paul H. Carlson. A new member of the genus Distocambarus (Decapoda: Cambaridae) from the Saluda Basin, South Carolina. Hobbs, Horton H., Jr., and Henry W. Robison. A new burrowing crayfish (Decapoda: Cambaridae) from southwestern Arkansas eee cece eeeeeeeeeeecenenenneeeneeeeeeeeeeceneeneeeeeeeeeneenennnnnnnnee Hoffmann, Robert S. The correct name for the Palearctic brown, or flat-skulled, shrew SES OK CAA OU OT ALUS 22 Us sake t BN eR tnt Ne oe lel varie cet eS fd Ne Pa al ere ch ent Hooker, Allan. New species of Isopoda from the Florida Middlegrounds (Crustacea: RG Tal Geta Cl a) eet htt ese Set coy uP Wg epied ell ans RR ee IG A heed a Se Huddleston, Richard W., and Drew Haman. Mississippiellidae, a new eulobosinid (@ihhecamocbinids) ifamilya(Rrotoz0a) = Kathman, R. D. Synonymy of Pristinella jenkinae (Oligochaeta: Naididae)................... Keirans, James E., and William G. Degenhardt. Aponomma elaphense Price, 1959 (Acari: Ixodidae): Diagnosis of the adults and nymph with first description of the Kensley, Brian, and Richard Heard. A new species of the genus Spinianirella Menzies (Crustacea: Isopoda: Janiridae) from the western Atlantic. cece Kensley, Brian, and William Tobias. Redescription of Heterocarpus laevis A. Milne Edwards (Crustacea: Decapoda: Pamdalicace) occ eee eeeeeeeeeeeeeeneneeeeeeeceennnnnnn Kornicker, Louis S. Azygocypridina lowryi, a new species of myodocopid ostracode from bathyal depths in the Tasman Sea off New South Wales, Australia. Kornicker, Louis 8S. Thaumatoconcha porosa, a new species of abyssal ostracode from the Indian Ocean (Halocyprida: Thaumatocyprididae) eee eeeceeeeeeeeeeeeeeeeee eens Kornicker, Louis S., and Thomas M. Iliffe. Deeveyinae, a new subfamily of Ostracoda (Halocyprididae) from a marine cave on the Turks and Caicos Islands... Kritsky, Delane C., Walter A. Boeger, and Vernon E. Thatcher. Neotropical Monogenea. 7. Parasites of the Pirarucu, Arapaima gigas (Cuvier), with descriptions of two new species and redescription of Dawestreama cycloancistrium Price and Nowlin, 1967 (Dactylogyridae: Ancyrocephalimae) cece cccceceesestnsessecseeseccenceneeceneenceneensennnnnnnnussusensneneceeeeeeeeees Kropp, Roy K., and Raymond B. Manning. Cryptochiridae, the correct name for the family containing the gall crabs (Crustacea: Decapoda: Brachyura). ee Kudenoy, Jerry D. Four new species of Scalibregmatidae (Polychaeta) from the Gulf of Mexico, with comments on the familial placement of Mucibregma Fauchald and FFT COC Kees Si ea dO a ns Pe a hen gaa ete Lazo-Wasem, Eric A. Idunella smithi, a new species of marine amphipod (Gammaridea: Liljeborgiidae) from the east coast of the United States ie eeeeeeeeeeeeeeeeeeeeeeeeeee 809-850 678-68 1 158-176 295-302 687-697 411-415 611-614 1-9 622 288-292 935-944 657-671 81-89 1035-1041 17-28 255-280 10-12 1022-1027 711-717 682-687 237-242 698-704 1012-1021 476-493 321-331 954-955 332-340 705-710 Lellinger, David B. Nomenclatural and taxonomic notes on the pteridophytes of Costa Ricas"Panama;andColombia, ll... ee ee ee eee Lewis, Julian J., and John R. Holsinger. Caecidotea phreatica, a new phreatobitic isopod crustacean (Asellidae) from southeastern Virginia ccc cccccccccccecccccececeseenceesceceeneeutnnnceeeneeeeeee Menez, Ernani G., and Hilcondida P. Calumpong. Halophila decipiens, an unreported SCASTASS fn rata HMMA py pie San a eaten Morgan, Gary S. Taxonomic status of the Swan Island hutia, Geocapromys thoracatus (Mammalia: Rodentia: Capromyidae), and the zoogeography of the Swan Islands Werte borate fen a cs cca ate eee he el tI pee OR, Ps Upice tinh hi 08) eee ee eee Nelson, Joseph S., and John E. Randall. Crystallodytes pauciradiatus (Perciformes) a new creediid fish species from Easter [Slarnd cee ccccccccceteescecceeceeesntnnececcenseesentenseeeeeenee Norden, Arnold W., and Beth B. Norden. A new entocytherid ostracod of the genus Dactylocytherei ues We oat CNTR R EE sn 5, sete seen eee de Aah OEE) TREY ANN SEES SRE A ee ee Ohwada, Takashi. Redescription of the nephtyid polychaete Aglaophamus minusculus hartimiams, 19 GS ii ss Nese ia eles ed Side Be Se leno eee Eke le ee Mee ed Olson, Storrs L. A new species of Siphonorhis from Quaternary cave deposits in Cuba (Avies:Caprimnulgidac) is 5 eae Be Dea oe Ns ee en Od Baad Olson, Storrs L. A new genus of tropicbird (Pelecaniformes: Phaethontidae) from the Middle Miocene Calvert Formation of Maryland ceeeceeeeeceeceeeeeceeceeecceceeeeenennnnnnenne Page, Lawrence M. A new crayfish of the genus Orconectes from the Little Wabash River system of Illinois (Decapoda: Cammbaricace) cece een ceceeeeeeeeeeeeeeenneeneeeeeteetttttntnccee Pawson, David L. Psychropotes hyalinus, new species, a swimming elasipod sea cu- cumber (Echinodermata: Holothuroidea) from the north central Pacific Ocean.............. Pereira, Guido. Freshwater shrimps from Venezuela III: Macrobrachium quelchi (De Man) and Euryrhynchus pemoni, n. sp.; (Crustacea: Decapoda: Palaemonidae) from MDE eas Gothen |S ea RN Batak cheer i aed na A i RO, PR SD 2h 2 Pérez Farfante, Isabel, and Brian Kensley. Cryptopenaeus crosnieri, a new species of shrimp, and a new record of C. sinensis (Penaeoidea: Solenoceridae) from Australian SW ALOT S oe are lee re EN ee Aad rae ot oe nee et OT ANY. ON eke rae 8 Perkins, Philip D., and Paul J. Spangler. Quadryops, new genus, and three new species of arboreal Dryopidae (Insecta: Coleoptera) from Panama and Ecuador .............0..............- Perkins, Thomas H. Chrysopetalum, Bhawania and two new genera of Chrysopetalidae (Polychaeta), principally from FLOrida cece eeeeeeeeecnennnnnnennnneeneeenneeeeeneneeeeenenneeneees Pettibone, Marian H. Polychaete worms from a cave in the Bahamas and from exper- imental wood panels in deep water of the North Atlantic (Polynoidae: Macelli- Cephialinae se arriy© thoi ae) ee ea Su RS Pettibone, Marian H. An additional new scale worm (Polychaeta: Polynoidae) from the hydrothermal rift off westerm Mexico at 21°N occ ccceeennnnnnnnneneeeeeeeseeeeeneeneenee Pettibone, Marian H. Additional branchiate scale-worms (Polychaeta: Polynoidae) from Galapagos hydrothermal vent and rift-area off western Mexico at 21°N_ ww... Pettibone, Marian H. New genera and species of deep-sea Macellicephalinae and Har- mothoinae (Polychaeta: Polynoidae) from the hydrothermal rift areas off the Gala- pagos and western Mexico at 21°N and from Santa Catalina Channel... Reid, Janet W. Calanoid copepods (Diaptomidae) from coastal lakes, state of Rio de Janeiro, Brazil tees te Se ae eee ee ee ere 2 AM EES Wife anand beh Ayeeee mt aw Rice, Mary E. Description of a wood-dwelling sipunculan, Phascolosoma turnerae, new SPECISS ences pe Ned ee eet es eed ge eee Petal ate A ee Ps Robbins, C. Brian, and Henry W. Setzer. Morphometrics and distinctness of the hedge- hog genera (Insectivora: Erimaceidae) ce eeeceeeceneceeccecseeseeceeeeeeesttstntnnnnnnensnececcenenenenneeeeenecstettennneet Rowe, F. W. E., and David L. Pawson. Loisettea amphictena, new genus, new species, from the sublittoral of northwestern Australia (Echinodermata: Holothuroidea)............ Sassaman, Clay. Cabirops montereyensis, a new species of hyperparasitic isopod from Monterey Bay, California (Epicaridea: Cabiropsidae) cc eeeeeeeeeceeeeeeeeeeeceeeeeteetetetcennneennee Schmidt, Robert E., and Carl J. Ferraris, Jr. A new species of Parotocinclus (Pisces: oricaridae) fron: Guyana se ee ee A Siegel, Jeffrey A., and Thomas A. Adamson. First record of the genus Zebrias (Pisces: Pleuronectiformes: Soleidae) from the Philippine Islands, with the description of a TNC WW SPC CLC Sesser uso feo a aft TA ye ROEM DG eee Mead Dee oe eee nein Ae ee es 366-390 1004-1011 232-236 29-46 403-410 627-629 604-610 526-532 851-855 564-570 523-525 615-621 281-287 494-510 856-915 127-149 150-157 447-469 740-757 574-590 54-60 112-120 672-677 778-789 341-346 13-16 Smith, Alan R., and David B. Lellinger. New tropical American species of Thelypteris (CLE RieeUs aya) RN a eee ee ee ee eee ee Spangler, Paul J. A new species of the aquatic beetle genus Dryopomorphus from Borneo (Crlleororisiens TBllriiQabe yee 05 tia Ye) yer eee ee eee Springer, Victor G. Oman ypsilon, a new genus and species of blenniid fish from the Tiny GlPbzaray OSIM casa ekki i oe NUMER Steadman, David W., and Gary S. Morgan. A new species of bullfinch (Aves: Embe- rizinae) from a Late Quaternary cave deposit on Cayman Brac, West Indies............ Thomas, James Darwin, and J. L. Barnard. Perioculodes cerasinus, n. sp., the first record of the genus from the Caribbean Sea (Amphipoda: Oedicerotidae) 000.0000. Thomas, James Darwin, and J. Laurens Barnard. Two new species of two new gammarid genera (Crustacea: Amphipoda) from the Florida keys... ccs ccccccccsccseesssscesceeeeetetenneneneeee Thomas, James Darwin, and J. L. Barnard. A new marine genus of the Maera group (Crustacea: Amphipoda) from Belize 2... eeeesessnseccccoccccccecceceeceeneeneeneeeeseeetetttttttntennnnenenneneseeeseee Thomas, Richard, Roy W. McDiarmid, and Fred G. Thompson. Three new species of thread snakes (Serpentes: Leptotyphlopidae) from Hispaniola eee Thompson, Fred G. Columbinia vasquezi, a new clausiliid land snail from Bolivia......... Timm, Robert M., and Roger D. Price. A review of Cummingsia Ferris (Mallophaga: Trimenoponidae), with a description Of twO MEW SPECIES... ec eeeeeceeeecceeceeeeeeeeeeeeeeeeeeeeeennenee Toll, Ronald B. The reinstatement of Bathypolypus faeroensis (Russell, 1909) (Octopoda: VB Een sy OO) Nf CY Oa Wz YS) Oe eee 2 Pe ee pret ae ee pee Tumer, Richard L. Microphiopholis, replacement name for Micropholis Thomas, 1966 (Ophiuroidea: Amphiuridae), non Huxley, 1859 (Amphibia: Dissorophidae)................... Vari, Richard P. A new species of Bivibranchia (Pisces: Characiformes) from Surinam, SIUM OMT TIES OTM CY] OTN UAS coca a cease sect ote a eee pete Vari, Richard P., and Jacques Géry. A new curimatid fish (Characiformes: Curimatidae) fre Tree PAT AZ OTN ERAS TN ae ea a se er le SE ae a re an Vari, Richard P., and Michael Goulding. A new species of Bivibranchia (Pisces: Cha- raciformes) from the Amazon River Basin eecccceceeceeeeeeceeeeenneeneeeeeeeeeeeeeeceeeeeneeeeeeeeeenenennnnnnnee Vila, Irma, and Nibaldo Bahamonde N. Two new species of Stratiodrilus, S. aeglaphilus and S. pugnaxi (Annelida: Histriobdellidae) fromm Chaitle cece ceecceeccceceeeeeeeeteeteeteee Watson, Ronald E., and Ernest A. Lachner. A new species of Psilogobius from the Indo- Pacific with a redescription of Psilogobius mainlandi (Pisces: Gobiidae)............................ Weems, Robert E. Miocene and Pliocene Molidae (Ranzania, Mola) from Maryland, Virginia, and North Carolina (Pisces: Tetraodomtiformes) i eeeeeeceeeeeeeeeeeecce eee ee eee Weitzman, Marilyn. Hyphessobrycon elachys, a new miniature characid from eastern paracuaya (niScesa@haracitonrics) ma Wicksten, Mary K., and Michel E. Hendrickx: New records of caridean shrimps in the Grullifge@ fa @ allio remn a A CX CO sea i a re Wilkins, Kenneth T. Pocket gophers of the genus Thomomys (Rodentia: Geomyidae) fromuthepPleistocener@teR lori a area eee eerste eee Williams, Jeffrey T. Cirripectes imitator, a new species of western Pacific blenniid fish Wolfe, G. W., and P. J. Spangler. A synopsis of the Laccornis difformis groups with a revised key to North American Species of Laccornis Des Gozis (Coleoptera: TDD y('0 Su cH che) estas a a Laan Eade aa Seer ceca Woodwick, Keith H., and Terry Sensenbaugh. Saxipendium coronatum, new genus, new species, (Hemichordata: Enteropneusta): the unusual spaghetti worms of the Galapagos Ruftehiy dro thermal Gvemts ies ena en a es tan ler a Zug, George R. A new skink (Reptilia: Sauria: Leiolopisma) from Fiji Zullo, Victor A., and Norman F. Sohl. Scalpelloid barnacles from the Upper Cretaceous Ofsoutheastenn North Carolina cc eset eee 916-930 416-421 90-97 544-553 98-106 191-203 630-635 204-220 768-773 391-402 598-603 1028-1029 511-522 1030-1034 1054-1061 347-350 644-654 422-438 799-808 571-573 761-767 533-538 61-71 351-365 221-231 636-643 INDEX TO NEW TAXA VOLUME 98 (New taxa are indicated in italics; new combinations designated n.c.) PROTOZOA Mississippiellidae 22/8 vss Wire 5 gan Sone See eee ee Eee ien Le eed eee te eee 11 COELENTERATA Hydrozoa @rypthe li ay 7721C1 OD O11 cere sere a Se ee 736 Lepidopora cryptocymas 729 OLY SULCH OP OV Gy ts cote le, Ollate nS Ritg Rie lee eats, Aa a lh 733 PLATYHELMINTHES Trematoda Dawestremascycloancistnigides 225. eee ee 325 DUTIES ae ree ae Oe Ey aoe 327 SIPUNCULA @nehmesormaisteens tru pois! Hee sa ecco econ ee sacectescsccinseem ee 844 Phascolion (Jsomya) it se AAD) CYT 0) 0) 2) ae 821 SENT US Creo mul ee! ss Co aes ae ee ee ae 835 Pe aASCOlOS OMA E CAC oP aE a es cee rs cece detcnecel cae 54 ANNELIDA Polychaeta Asclerocheilus Wiexicares 0s We tes foe reed A EN oe ree ees ee tae re a ee a 332 BOLI VD GIVIGS xe ates el ee tae le Ns RE cea ee ee eee Peer vee ee 137 charleneae 138 Bathykermadeca turnerae 133 Bhawania heterosetarnic ests ss re ee tees shee a ee he eg Ul 899 IBRANGHINOLO SIU se oe ee ee a es ot Se 449 OT ASSO Ue Bn Wines RO ANS TTR, ed Pell 1 a Sin Ra UAE REE Soo Lett 9 Se Ek be ae 457 hessleri...... 450 sandersi.... 453 Branchinotogluminae 447 Branchiplicatindesek—. «sa T ee: Se ey ee ee eee SEs ie et Ei 150 IBrANGCH pli CAUUS Sin Cee ase RA rE ee ANT Ae aretha don A 150 CUDTCUS 2h ee ie 151 Chrysopetalum eurypalea 890 5 (lola tc(21/1)1 | 886 ROR NGNCONLEZ GE noo nen tlh ane eI NN pale RE ce alien A 877 HELV OD ACA e ree > kee FARA Bes tor emiee pede? ek Ma eee err aR et Ey 884 TO TVUAVICD soso oe a a ee nL ems ee DO 874 Harmothoe macnabi 749 vagabunda 146 FAV OL OP Gl Cee re Nes ee eR eT ete er es ee eR ee ee 908 DUSDDETUOS Ge te ee te OON IE ER a eb ee Be aR a Tn 908 Hyboscolex quadricincta 334 evensteiniell ai ee ee eh Le EO ee eee A 740 PERI COLLD Le 125 oR DVR WS AN iS eo ee nn Rela aD rte Re 741 IMiacelliGephalayea la pacers 1 Sie ae seers et sac elcsecescetesscscs tut WEL EUR WN 746 Natopolynoe 747 SEES HTLLE /1 [pear eae ew ash ed er tea neh NEL ed a te Lt a 747 Neolipobranchius blakei 336 OUST ORV OGI OD OAUS tee see ere aah Rae een asehcactb lassen aN 459 LLY E111 S RA te NS I eh Neath Na da relnrree nse ene i RL 459 130 131 337 SG@le le pUSW/1U/ 6/1172 GS CC eee a a ee weet a eal AEM 678 BS AiMreA LT CO GINGLINT SHCZCSU CLD) DULLES po ast a nt che Ae eet iceman tanevec eee, BULA INES 348 FDU T CSG Ue a a er nee etn Mee eet nn ee A PEE ee 349 PINT CID LO 0G Cre APRIL Ee a Os edad ou dene NOLL La A LETS 904 GLEE OL ID YUU ce Ra se A ac Pe lw li at art ese cigs NN ABR 905 Oligochaeta Aalslockalns (QaGHOAB ee ee ee ee 159 PN Ora SHy7 tL CL CI LOT Ch aa a pe i a ee ol edad ortestcioal tee Ne nd De 931 TIGSYG CLANS EIS OY DT BEA ZI aw aL a ee 949 HEBeATATV AT SEO TeeATY CLUS RIN eet cee te re eager inch ta ples eee 310 Phallodrilus GILES 1 CY; Lie ae ae Nah oe Bhd NN a etal Wal aeeautonte aclu NOR ET 164 (LO V;S OSD CUT YIAUIT CC Ch tN rh ele sD il cn eat ted 162 PE TESUE ELC | Ca RAMON i Sse ae em eo ci al etl rteaecae bi Ra da RRA ee Ret 472 {Ura pO OA TA SVS ELH 18711 ees ae ene a ee take pte ee ne La 167 AN feared G Fn eae Gea Gh ra D ua Spor AN UC US Ns Cart ae el ee er 655 ARTHROPODA Crustacea LU OH UDA eae ee ee ee ere ee [OES OTIS cece ee SS SI OEE NE Le Rea Net gp CR oe PNAS OGY JOT LIEN AGL OM Tay | sores aaa nee Ue EY eae ce eee Cabirops montereyensis Caecidotea phreatica ............ Cryptopenaeus crosnieri Dactylocythere SCOCOS 2... Me os pid aie ei ee eee ee DD COV CY ee er om TO a Pate ENA aN ee Nee NO MEE A onel WLP UN a NU RN al ES TD UTC UES meet ee Re or Pure aoe RN RO 2 Sat PA na BID) COV CV ULC Cire rea id te Nae ot NN NA PN Pa a a em “Diaptomus” azureus DL LLET TALLIES ES Reese ee A ea es Nee eS SII ates RNS Sones re eee rE Distocambarus (FitzCambarus) YOUNIS T n-ne cceeeeee cece cece eee enennnnnnvneeeeeeeeeeeeeeenneeeseeeeeeeeneenennnnncenunneesseeseeseeeeceeeee DUTT OS US peer etd oe Pato ta ie Ren OER nF Aa NSN Dita Ate PAOD Se NGI BO lo Eb Re ae te ar en CUM OU at ae a ea Ae I a ct ce oo EO a Euryrhynchus Deron... Fallicambarus (Fallicambarus) harpi Gnathostenetrioides PUgio ........ eee. Heteromysoides dennisi IT Ghana NA W5 7727 6/2 Raerrmmee erent ee em Nel ye ae Ne | ren YE ct Nn eee ne a neem hy A A i Janira biunguicula DUCAYATING oes catacumba..................... Mesanthura hopkinsi WG EGY CO) acs Sh AI TO ae Ft a BB AB FOC Sa A as PT PEE RG ROOT Sc TE I ee Pe ECCS ceil eh ae HON kao oh se eg AEN eae est a Sc np a Hare a TANG IEBTY WYER ON OW ENT a YE AY Ls ee NR ee er pe ae ER Neostrengeria charalensis @RCOMECLES SLQTTAGH En a Bs eco eS Oe 564 eric tl OGES) COR GST US cs ee 99 Pleurocopé flonidensis sc. 2) eet ie a ee oe ee 257 Podobothrus DOT IRI CIS 1S. oscctntes aon 2) Oe a a eee 1049 ROWSE 11S oe is Pee tg Irn SR a eR ote 554 PSY O70 P18) 010 a ns ec Meee EP es eS ete ese eae ee 191 looensis 192 Speonebailia.................. 439 cannoni 440 Spinianirella serrata 682 MhaumatocOncha POVrOS@ 28s see sae ee ec ae 1013 TWhermvOsphaerOmmia CQVICQUGG cc cess cesses eee 1042 VG CTUTO oe Virgiscalpellum inornatum um mi PSia G11 0 os oa a ae OX AG Sa ee a ee a nti oe Tey O PONTO US| SO Ol sconce eee a eevee eee eae ee Eosentomon adakense NGACCOINISHCLT CN é he EP tae el 2 ve 3 vi ite Mad 5000 ft., Owyhee Co., Idaho. This view, facing north, illustrates the general topography of the area. The holotype was collected near a sandy wash (center of photograph), whereas other specimens were collected in seemingly less-favorable habitat on the mesas to the west. goids, 0.76; arching of cranial dome, 6.28; mandibular length, 10.38; and angular bifurcation, 0.58. Diagnosis and description. — Kangaroo mice of this subspecies are small in over- all cranial measurements, including: greatest length of skull, bullar length, max- illary breadth, and greatest length of incisive foramina (Table 1). Further, in restrictus the ratio between the least and greatest expanse of the lateral face of the zygoma is large (see Table 1), which reflects a lesser relative development of a “zygomatic plate” along the zygoma. Also, the pterygoids are narrow and the nasals are short. With respect to colorimetric characters, this subspecies is char- acterized by having a small value for excitation purity of the dorsal pelage (Table 2). Chromosomally, restrictus has the 42-a karyotype (Hafner 1981), which has 42 chromosomes and a fundamental number of 70 (five pairs of acrocentric autosomes). Comparisons.—Cranial and colorimetric variables of restrictus and the only other two subspecies, pallidus and ruficollaris (Hafner 1981), are presented in Tables 1 and 2. From pallidus, (the main western subspecies that largely surrounds restrictus in distribution), restrictus differs in having shorter nasals, larger mea- surements for the least expanse of lateral face of zygoma and smaller values for excitation purity of the dorsal pelage. Kangaroo mice of the subspecies restrictus differ from ruficollaris (the eastern M. pallidus subspecies) in having the 42-a 8 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON mo ie 1 wh di i = ae + si See : ss b, u 4 . = ane. * : eee Sg A code AEE gaan speed BS eat . “ aes ae at ae ee Bet? een Fig. 2. Type-locality of Microdipodops pallidus restrictus: 8.9 mi. S, 1.2 mi. E Mina, 4400 ft., Mineral Co., Nevada. Kangaroo mice of this subspecies are found on and immediately about these semi-stabilized sand dunes (view is facing north). karyotype instead of the 42-6 (totally biarmed autosomes; see Hafner 1981) karyo- type. Comments.—The form restrictus, although distributed in the western portion of the species’ range, is morphologically unlike the other western subspecies, pallidus, and remarkably similar to ruficollaris of the east. Interestingly though, restrictus differs markedly in karyotype from the eastern subspecies and possesses the same karyotype found throughout the majority of the range of pallidus. Hence, restrictus combines characters found in both of these other forms. In addition to the morphological and chromosomal differentiation of restrictus, this new sub- species is biochemically quite distinct. Indeed, the degree of biochemical differ- entiation between restrictus and samples of pallidus (S = 0.60) and ruficollaris (S) = 0.62) is commensurate with that observed between pallidus and ruficollaris (S = 0.64; see Hafner 1981). The geographic distribution of restrictus is very limited in comparison to pal- lidus and ruficollaris. This subspecies is known from one isolated locality in southeastern Mineral County and more collecting is needed to document the extent of its geographic range. The habitat at the type-locality is characterized by semi- stabilized sand dunes several meters in height (Fig. 2). The vegetation is sparse and the flora is dominated by Sarcobatus and Atriplex. This restricted sand dune system is separated from other known localities of M. pallidus by approximately VOLUME 98, NUMBER 1 9 30 km of alkali-caked and/or hardpan, gravelly soils, which are not suitable for M. pallidus habitation. Specimens examined.—A total of 14 specimens was examined (MVZ 159969- 159982) all from the type-locality. Etymology.—The subspecific epithet, restrictus, is derived from the Latin root restrict, and was chosen to reflect both the small geographic range of the new subspecies and its isolation from other populations of the species. Acknowledgments Much of the information presented here stems from a dissertation presented to the Graduate School, University of California at Berkeley. I thank J. L. Patton for advice and support during my research on kangaroo mice at the Museum of Vertebrate Zoology. For assistance in the field, I gratefully acknowledge M. S. Hafner and P. M. Hafner. E. R. Hall and D. T. Wright drew my attention to Microdipodops in Idaho. I thank J. A. White for kindly providing detailed locality information about Idaho kangaroo mice and for loaning specimens. Appreciation is extended to M. D. Carleton, D. J. Hafner, M. S. Hafner, and J. L. Patton for critical evaluation of the manuscript. I also acknowledge most gratefully the as- sistance and encouragement shown me by P. M. Hafner in all phases of the study. Financial support for this study was provided, in part, by a grant (Doctoral Dissertation Research in Systematics) from the National Science Foundation (DEB- 7822038), and further benefited by support from the Louise M. Kellogg Fund and the Museum of Vertebrate Zoology. Lastly, I am grateful to the Smithsonian Institution for support during the final examination of specimens. Literature Cited Davis, W. B. 1939. The Recent mammals of Idaho.—The Caxton Printers, Ltd., Caldwell, Idaho, 400 pp. Hafner, D. J., J. C. Hafner, and M. S. Hafner. 1979. Systematic status of kangaroo mice, genus Microdipodops: morphometric, chromosomal, and protein analyses.—Journal of Mammalogy 60:1-10. ——__, , and . 1984. Skin-plus-skeleton preparation as the standard mammalian mu- seum specimen.— Curator 27:141-145. Hafner, J.C. 1976. Specific and evolutionary relationships of kangaroo mice, genus Microdipodops. — M. S. thesis, Texas Tech University, Lubbock, vii + 139 pp. . 1981. Evolution, systematics, and historical biogeography of kangaroo mice, genus Micro- dipodops.—Ph.D. dissertation, University of California, Berkeley, viii + 269 pp. Hall, E.R. 1941. Revision of the rodent genus Microdipodops.— Field Museum of Natural History, Zoological Series 27:233-277. . 1946. Mammals of Nevada.— University of California Press, Berkeley, x11 + 710 pp. Patton, J. L. 1967. Chromosome studies of certain pocket mice, genus Perognathus (Rodentia: Heteromyidae).— Journal of Mammalogy 48:27-37. Schitoskey, F., Jr. 1968. Notes on the morphological variation in the dark kangaroo mouse.— Southwestern Naturalist 13:243-248. Moore Laboratory of Zoology and Department of Biology, Occidental College, Los Angeles, California 90041. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 10-12 MISSISSIPPIELLIDAE, A NEW EULOBOSINID (“THECAMOEBINID”) FAMILY (PROTOZOA) Richard W. Huddleston and Drew Haman Abstract.—The genus Mississippiella Haman, 1982, based on the uniqueness of test composition and construction, along with distinctive apertural morphology, is transferred from the family Paraquadrulidae Deflandre, to the new family Mississippiellidae. The Paraquadrulidae was, prior to the discovery of the genus Mississippiella by Haman (1982a, b) the only thecamoebian family known to possess an external secreted calcite test (Loeblich and Tappan 1964). This led Haman (1982a, b) to assign Mississippiella to this family despite the fact that Paraquadrula Deflandre was known only from western European mosses. Loeblich and Tappan (1964) along with Ogden and Hedley (1980), accept the Paraquadrulidae as a distinct family based on test composition. Page (1982), however, places the genus Paraquadrula in the family Nebelidae (which he equates with the Hyalosphenidae), but does acknowledge that ““The one calcareous genus (Paraquadrula) is sometimes separated into a family.”’ The uncertainty of system- atic placement of this calcareous entity is typical of the overall arcellacean clas- sification, a state of affairs lamented upon by Medioli and Scott (1983). We subscribe to the viewpoint that until stability is attained in arcellacean classification, newly discovered forms should be retained as distinct entities if they differ from previously described taxa. This viewpoint is followed even when the new forms occur with less than optimal abundance. If additional information on the new organism becomes available, which clearly demonstrates a biological affinity to another taxon, it is at that time that any systematic grouping should be undertaken. Presumptive grouping may lead to erroneous observations and conclusions. The retention of the new entity may occur at any level within the hierarchal system and simply depends upon the existing classification scheme. Consequently, based on the significant morphological differences between Para- quadrula and Mississippiella, plus the fact that the former genus is placed in an accepted family, we remove Mississippiella from the Paraquadrulidae and estab- lish the family Mississippiellidae to accommodate the form. A revision of the family Paraquadrulidae to accommodate Mississippiella is not believed prudent at this time and would only add to the confusion, detailed above, concerning this family. We are fully cognizant that other workers on the Arcellacea may differ with our philosophy but we believe it to be expedient at present as the ““micro- classification of the group remains. . . in a chaotic state” (Medioli and Scott 1983: 6). Page (1982) claimed that the test of Paraquadrula is formed by a calcium replacement of the original silica; however, other workers believe that the qua- drangular plates characteristic of this genus are formed of secreted calcite (Loeblich and Tappan 1964; Ogden and Hedley 1980). We favor the second opinion. X-ray energy dispersant analysis of Mississippiella utilizing a Kevex 10-channel recorder VOLUME 98, NUMBER 1 11 showed the test composition to be almost pure calcite with no trace of silica (Haman 1982a, Fig. 27). Paraquadrula has been described as having a test com- posed of “quadrangular calcareous plates which may have thickened rims” (Loe- blich and Tappan 1964:C34). The presence of plates or “‘scales” is a consistent feature found in nearly all nonproteinaceous thecamoebinids possessing a secreted test of either calcareous or siliceous composition. Mississippiella clearly diverges from this pattern by displaying autogenous intergrowths of euhedral calcite rhombs comprising the test. Ogden and Hedley (1980:3) diagnosed the shells of testate amoebae as being “‘proteinaceous, agglutinate, siliceous, or calcareous in composition” and that their morphology would normally “‘consist of one single aperture.” This last criterion cannot be regarded as a critical thecamoebinid feature as demonstrated by the estuarine-marine genus J7richosphaerium Doflein which possess multiple circular apertures. Mississippiella has an “‘aperture composed ofa series of predominantly ovate openings, variable in number, 10-18 arranged in a crescentic series’ (Haman 1982a:366). While the majority of thecamoebinids have a single aperture (Ogden and Hedley 1980), forms with multiple apertures exist (Page 1982). In addition, Ogden and Hedley (1980) illustrate numerous species which display apertural modifications in the form of circum-apertural pores. Grospietsch (1965) reported that one form of Hyalosphenia papilio Leidy has a simple concave aperture, while another has several apertural pores. The openings present on tests of Mississip- piella are believed to represent the true aperture and not pores as illustrated by Ogden and Hedley (1980). Paraquadrula possesses a terminal aperture that is described as oval (Ogden and Hedley 1980) or as oval to slitlike (Loeblich and Tappan 1964) and differs markedly from Mississippiella. Mississippiella differs from all thecamoebinid taxa described to date based on apertural characters. The possession by Mississippiella of a secreted, calcitic, non-porcellaneous, imperforate, unichambered test is sufficient to warrant placement within the Ar- cellinida as defined by Sheehan and Banner (1973), despite the fact that infor- mation on the pseudopodia is not currently available and the fact that Mississip- piella displays features not previously known to exist in thecamoebinids. Page (1982) divides the Arcellinida into two suborders, the Eulobosina and the Trichosida. The second suborder is characterized by the single family Trichos- phaeriidae and the single genus Trichosphaerium. Sheehan and Banner (1973) published a detailed study on this genus and determined that Trichosphaerium possesses an external, flexible spicular test with quasi-permanent apertures. Such characteristics clearly exclude Mississippiella from this suborder; therefore, this genus is placed in the Eulobosina. The affinities and differences of Mississippiella to and from other eulobosinid taxa have been detailed above. From this discussion it is evident that Mississippiella does not demonstrate equitability with any es- tablished thecamoebinid genus or family, and warrants the establishment of a new family, the Mississippiellidae. As the family is monotypic, the characters of the family are those of the genus. Mississippiellidae, new family Diagnosis.—Test free, unilocular, circular to subcircular in outline, discoidal to hemispherical in side view, ventral surface flat to concave; dorsal surface convex; aperture composed of series of predominantly ovate openings, variable 12 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON in number, 10-18, arrranged in crescentic pattern towards one side of ventral surface; wall composed of autogenous intergrowths of euhedral calcite rhombs. Type-genus. — Mississippiella Haman, 1982. Type-species. — Mississippiella multiapertura Haman, 1982. Type-illustrations. —Haman (1982:holotype, pl. 2, figs. 11-15; paratypes, pl. 2, figs. 16-22). Etymology.—The family name is derived from the type-locality, the Mississippi Delta. Acknowledgments We thank Chevron Oil Field Research Company and Chevron U.S.A., Inc., for publication permission. We are also grateful to F. S. Medioli for his constructive comments. Literature Cited Grospietsch, T. 1965. Monographische Studie der Gattung Hyalosphenia Stein (Rhizopoda, Tes- tacea).— Hydrobiologia 26:211-—241. Haman, D. 1982a. Modern thecamoebinids (Arcellinida) from the Balize Delta, Louisiana. —Trans- actions, Gulf Coast Association of Geological Societies 32:353-376. —. 1982b. Modern thecamoebinids (Arcellinida) from Balize Delta, Louisiana. Abstracts: Gulf Coast Association of Geological Societies and Gulf Coast Section Society of Economic Pale- ontologists and Mineralogists Meeting, October 27-29, 1982, Houston, Texas.— American As- sociation of Petroleum Geologists Bulletin 66:1430-1431, Sep 1982, (Issued Nov 1982). Medioli, F. S., and D. B. Scott. 1983. Holocene Arcellacea (Thecamoebians) from eastern Canada. — Cushman Foundation for Foraminiferal Research Special Publication No. 21:1-63. Loeblich, A. R. Jr.,and H. Tappan, 1964. Sarcodina, Chiefly ““Thecamoebians” and Foraminiferida. In R. C. Moore, ed., Treatise on Invertebrate Paleontology.— University of Kansas Press, pt. C, Protista 2, 1:1-900. Ogden, C. G., and R. H. Hedley. 1980. An atlas of freshwater testate amoebae.— British Museum (Natural History); Oxford University Press, 222 pp. Page, F. C. 1982. Lobosa. Jn S. P. Parker, ed., Synopsis and Classification of Living Organisms. — McGraw-Hill, 1:510-517. Sheehan, R., and F. T. Banner. 1973. Trichosphaerium—an extraordinary testate rhizopod from coastal waters.— Estuarine and Coastal Marine Science 1:245-260. (RWH) Scientific Research Systems, 11044 McGirk, El Monte, California 91731; (DH) Chevron Oil Field Research Company, P.O. Box 466, La Habra, California 90631. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 13-16 FIRST RECORD OF THE GENUS ZEBRIAS (PISCES: PLEURONECTIFORMES: SOLEIDAE) FROM THE PHILIPPINE ISLANDS, WITH THE DESCRIPTION OF A NEW SPECIES Jeffrey A. Seigel and Thomas A. Adamson Abstract. — Zebrias lucapensis is described from two specimens trawled in Lin- gayen Gulf, Luzon, Philippines, and represents the first record of the soleid genus Zebrias from the Philippine Islands. Of the 16 previously described nominal species of Zebrias, the new species most closely resembles Z. crossolepis from coastal China. Zebrias crossolepis differs from the new species in number of bands on the body, size of the ocular-side pectoral fin, and relative length of the spinulated portion of the ocular-side body scales. The genus Zebrias Jordan and Snyder includes 17 nominal species of small, banded soles found throughout the Indo-West Pacific from east Africa, the Red Sea, Persian Gulf, India, Australia and Tasmania, to coastal China, Taiwan, and Japan. Chabanaud (1934) listed eight species of Zebrias from the coasts of Aus- tralia, India, the ““Malay Archipelago,” “Indo-China’”’ and Japan. Subsequent authors (Herre and Myers 1937; Ochiai 1963, 1966; Punpoka 1964; Cheng and Chang 1965; Smith 1965; Rama Rao 1967; Talwar and Chakrapani 1967; Dor 1970; Kailola 1974; Scott 1975; Joglekar 1976; Hussain and Khan 1981; Shen and Lee 1981) have recognized from one to 11 species throughout the Indo-West Pacific, excluding the Philippine Islands. Herre (1953) and more recent authors (Murdy 1979; Rau and Rau 1980; Schroeder 1980; Fourmanoir 1981; Murdy et al. 1981) have not recorded Zebrias from the Philippine Islands. Among collections made to assess the fish fauna of the Hundred Islands region of western Lingayen Gulf, Luzon, Philippines, were two specimens of an unde- scribed species of Zebrias, which constitute the first record of the genus from the Philippine Islands and form the basis for the following description. Methods and Materials Counts and measurements follow Hubbs and Lagler (1947) and Ochiai (1963). Standard length (SL) is used throughout and rounded to the nearest 0.5 mm. Scale terminology follows Chabanaud (1934), Ochiai (1963) and Joglekar (1976). Ma- terial is deposited in the Natural History Museum of Los Angeles County (LACM). One specimen of Brachirus selheimi was borrowed from the Australian Museum, Sydney (AMS). Zebrias lucapensis, new species Fig. 1 Holotype.—LACM 37436-6, 83.5 mm; Philippines, Lingayen Gulf, vicinity of Hundred Islands, local fisherman, trawl, 9 Mar 1978. 14 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Paratype.—LACM 37436-8, 83.0 mm; collected with holotype. Diagnosis.—A species of Zebrias that differs from Z. zebra, Z. altipinnis, Z. fasciatus, Z. japonicus, Z. synapturoides and Z. maculosus by having eyes con- tiguous (without interorbital space); from Z. guagga by lack of orbital tentacles; from Z. craticula and Z. cancellata by dorsal and anal fins not fully contiguous with caudal fin; from Z. callizona by number of dorsal-fin rays (82-87 in Z. callizona, 71-72 in Z. lucapensis); from Z. annandalei, Z. cochinensis and Z. regani by number of pored lateral line scales (102-105 in Z. annandalei; 90 in Z. cochinensis; 82—90 in Z. regani; 63-68 in Z. lucapensis); and from Z. keralensis and Z. crossolepis by total number and status (single or paired) of bands on head and body (13 single bands in Z. keralensis; nine paired (18) bands in Z. crossolepis; 16-18 mostly unpaired bands in Z. /ucapensis). Description. —(Meristics and morphometrics for the paratype are indicated in parentheses when differing from holotype.) Dorsal-fin rays 72 (71); anal-fin rays 56; pectoral-fin rays i1, 11; pelvic-fin rays 5, 4 (5, 5); caudal-fin rays 18; lateral line scales 63 (68); vertebrae 9 + 34 = 43. Body depth 2.70 (2.72) in SL; head length (HL) 4.64 (4.88) in SL; caudal length 3.63 (3.60) in SL; upper eye diameter 4.25 (4.59) in HL; pectoral length 6.9, 6.0 (6.25, 5.14) in HL. Eyes contiguous, upper eye slightly in advance of the lower; no interorbital scales; orbital tentacles absent. Anterior nostril on ocular side with long tube reaching to anterior margin of lower eye when depressed; posterior nostril on ocular side with a short tube, not reaching lower eye; nostrils on blind side inconspicuous. Mouth subterminal, gape extending to vertical through center of lower eye; teeth small, in bands, present only at rear of upper and lower jaws on blind side. Blind side of head with small papillae anteriorly. Opercular margins fringed with minute tentacles. Scales strongly ctenoid on both sides, those of ocular side with 8 to 15 marginal spinules and 5 to 11 basal grooves, those of blind side with 12 to 16 marginal spinules and 9 to 13 basal grooves; spinulated portion of scale 33—40% of its total length. Tubed lateral-line scales cycloid; lateral-line system well developed on both sides of body. Dorsal fin originating slightly in front of anterior margin of upper eye and extending to base of caudal fin, with which it is only partially contiguous. Anal fin originating just posterior to pelvic fins, similar to dorsal fin in attachment to caudal fin. Caudal fin relatively long, extending beyond ends of posteriormost dorsal- and anal-fin rays. Dorsal- and anal-fin rays bifid distally. Pectoral fins broad-based, subequal, connected to branchiostegal membranes dorsally. Pelvic fins of equal size, free from anal fin; separate (not united) but connected proximally by skin. Color.— After fixation in formalin and storage in ethanol, ocular side light greyish-brown, with 16-18 narrow, dark brown bands on body extending onto dorsal and anal fins; bands about as wide as interspaces between them, a few occurring in pairs (paratype); edges of bands darker than midsections. Pectoral fin of blind side dark brown, that of ocular side dusky. Proximal and distal areas of caudal fin dark brown. Pelvic fin of blind side darker than that of ocular side. Blind side of body generally white with scattered dots of pigmentation becoming dusky toward body margins. Dorsal and anal fins on blind side dark brown. Distribution.— Known only from Lingayen Gulf, Luzon, Philippines. VOLUME 98, NUMBER 1 15 Fig. 1. Zebrias lucapensis, holotype, LACM 37436-6, 83.5 mm SL, vicinity of Hundred Islands, Lingayen Gulf, Luzon, Philippines. Etymology.—The name /ucapensis is an adjective designated in honor of the people of the coastal village of Lucap, Pangasinan Province, Philippines. Remarks.—Of the 16 previously described, nominal species of Zebrias, Z. lucapensis most closely resembles Z. crossolepis Cheng and Chang from coastal China. Zebrias lucapensis may be distinguished from Z. crossolepis by number of bands on the body (16-18 mostly unpaired bands in /ucapensis; 9 paired bands in crossolepis), size of the ocular-side pectoral fin (14-18% HL in /ucapensis; 15.7- 20.8 [X = 18.5%] HL in crossolepis), and by relative length of the spinulated portion of ocular-side body scales (33-40% of scale spinulated in /ucapensis, 45— 57% in crossolepis). Acknowledgments We thank J. R. Paxton (AMS) for the loan of Brachirus selheimi, S. C. Shen (National Taiwan University) for translating Cheng and Chang’s Chinese key to Zebrias, Cheng Pao-Shan (Academia Sinica, Peking) for gifts of specimens of Z. crossolepis, D. M. Cohen and R. J. Lavenberg (LACM) for review of the manu- script, A. G. K. Menon (Zoological Survey of India) and C. C. Swift (LACM) for helpful comments and suggestions. Special thanks to Mr. and Mrs. M. DeGuzman (Magsaysay, Pangasinan, Philippines) for their help during field work in Lingayen Gulf and to Mr. and Mrs. A. Adamson (Manila, Philippines) for their support and assistance. Drawing by M. Butler (LACM), manuscript typed by T. Togiai (LACM) and photography by R. Meier (LACM). Partial support of travel and field expenses provided by the LACM Foundation. Literature Cited Chabanaud, P. 1934. Les soleides du groupe Zebrias. Definition d’un sous-genre nouveau et de- scription d’une sous-espece nouvelle.— Bulletin de Société Zoologique de France 59:420-436. Cheng, P. S., and Y. W. Chang. 1965. Studies on the Chinese soleid fishes of the genus Zebrias, with description of a new species from the South China Sea.— Acta Zootaxonomica Sinica 2(4): 267-278. Dor, M. 1970. Contributions to the Knowledge of the Red Sea No. 44. Nouveaux poissons pour la fauna de la Mer Rouge.—Bulletin of the Sea Fisheries Research Station, Haifa 54(44):7-28. 16 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fourmanoir, P. 1981. Poissons (premiere liste). (IN) Resultats des Campagnes MUSORSTOM. I. Philippines (18-28 Mars 1976).—Collection Memoirs ORSTOM 91:85-102. Herre, A. W. C. T. 1953. Checklist of Philippine fishes.— United States Department of the Interior, Fish and Wildlife Service Research Report 20:1-977. , and G. S. Myers. 1937. A contribution-to the ichthyology of the Malay Peninsula. — Bulletin of the Raffles Museum, Singapore, Straits Settlements 13:5-75. Hubbs, C. L., and K. L. Lagler. 1947. Fishes of the Great Lakes Region.—Cranbrook Institute of Science, Bulletin 26:1-186. Hussain, S.M.,andJ.A.Khan. 1981. Species of family Soleidae (Pleuronectiformes) from Pakistan. — Biologia 27(1):19-32. Joglekar, A. 1976. Ona new species of the genus Zebrias Jordan and Snyder, 1900 (Pisces: Soleidae) from the Kerala Coast (India).— Zoologischer Anzeiger Jena 197(1/2):67-70. Kailola, P. 1974. Additions to the fish fauna of Papua New Guinea. III.— Department of Agriculture, Stock and Fisheries, Port Moresby, Research Bulletin 12:59-89. Murdy, E.O. 1979. Fishery ecology of the Bolinao artificial reef.—Kalikasan, Phillipine Journal of Biology 8(2):121-154. —, C. J. Ferraris, Jr., D. F. Hoese, and R. C. Steene. 1981. Preliminary list of fishes from Sombrero Island, Philippines, with fifteen new records.— Proceedings of the Biological Society of Washington 94(9):1163-1173. Ochiai, A. 1963. Fauna Japonica Soleina (Pisces).— Biogeographic Society of Japan, Natural Science Museum: 1-114. —. 1966. Studies on the comparative morphology and ecology of the Japanese soles.— Misaki Marine Biological Institute, Kyoto University, Special Report 3:1-97. Punpoka, S. 1964. A review of the flatfishes (Pleuronectiformes = Heterosomata) of the Gulf of Thailand and its tributaries in Thailand. —Kasetsart University Fishery Research Bulletin 1:1— 86. Rama Rao, K. V. 1967. A new sole Zebrias cochinensis from India.—Journal of the Zoological Society of India 19(1 & 2):99-100. Rau, N., and A. Rau. 1980. Commercial marine fishes of the central Philippines (Bony Fishes).— Deutsche Gesellschaft fiir Technische Zusammenarbeit (GTZ) 92:1-623. Schroeder, R. E. 1980. Philippine shore fishes of the western Sulu Sea.— National Media Production Center, Manila, Philippines: 1-266. Scott, E.O.G. 1975. Observations on some Tasmanian fishes: Part XXI.— Papers and Proceedings of the Royal Society of Tasmania 109:127-173. Shen, S. C., and C. H. Lee. 1981. Study on sole fishes (Family Soleidae) from Taiwan.—Bulletin of the Institute of Zoology, Academia Sinica 20(2):29-39. Smith, J. L. B. 1965. The sea fishes of southern Africa.—Central News Agency, Ltd., South Africa. Fifth Edition, 580 pp. Talwar, P. K., and S. Chakrapany. 1967. A new flatfish of the genus Zebrias Jordan and Snyder (Soleidae) from the Orissa Coast, India.— Proceedings of the Zoological Society of Calcutta 20: 119-121. (JAS) Section of Ichthyology, Natural History Museum of Los Angeles County, 900 Exposition Boulevard, Los Angeles, California 90007; (TAA) Adamson Phil- ippines, Inc., 51 E. Colorado Boulevard, Pasadena, California 91105. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 17-28 THE CORRECT NAME FOR THE PALEARCTIC BROWN, OR FLAT-SKULLED, SHREW IS SOREX ROBORATUS Robert S. Hoffmann Abstract.—Examination of the holotypes of Sorex roboratus Hollister, 1913, and Sorex vir Allen, 1914, and comparisons with series of other Palearctic species of Sorex reveals that S. roboratus is the oldest valid name for this taxon, and that S. vir must be regarded as a junior synonym for the species. Geographic variation in Sorex roboratus is reviewed, and a new taxonomic arrangement proposed. During the period of indiscriminate taxonomic “lumping,” the number of species of Sorex recognized in the fauna of the Soviet Union was reduced to six (Bobrinskii et al. 1944) or seven (Ellermann and Morrison-Scott 1951). Several taxa consid- ered subspecies of Sorex araneus by these workers were revealed by subsequent investigations to be independent species—S. asper, S. caucasicus, S. daphaeno- don, S. isodon, S. raddei, S. arcticus (=S. tundrensis, Junge et al. 1983), S. un- guiculatus, and a species usually called S. vir (Stroganov 1957, Dolgov 1967, Corbet 1978). However, Yudin (1971, and references cited therein) maintained that the correct name for this latter species was S. roboratus. Neither he nor Dolgov were able to examine the holotypes of these two taxa, which are housed in the Museum of Comparative Zoology of Harvard University and the National Museum of Natural History, Washington, D.C., respectively, and which are crucial to the question. I have examined both holotypes, compared them with relevant material of other Palearctic Sorex (20 nominal taxa, 274 individuals including 14 other ho- lotypes; see “Specimens Examined’’) and conclude that Yudin’s view, that S. roboratus is the oldest valid name for this species, is correct, for the reasons given below. Three external measurements were taken from collector’s labels. Seven cranial measurements were made to the nearest 0.1 mm with dial calipers. These include: 1) condyloincisive length (CIL): occipital condyles to anterior tips of first incisors; 2) cranial breadth (CB): greatest breadth of braincase, approximately across the mastoids; 3) maxillary breadth (MB): greatest distance between tips of maxillary processes; 4) M2 to M2 breadth (M2—M2): greatest distance between anterior labial margins of second upper molars; 5) interorbital breadth (IOB): least lateral diameter of skull at anterior end of orbits, just posterior to maxillary process; 6) palatoincisive length (PIL); posterior margin of palate in midline to anterior tips of first incisors; 7) upper tooth row length (UTRL): greatest length of tooth row from posterior margin of alveolus of M3 to anterior tip of I1. These measurements were selected for the speed and accuracy with which they could be made, and some differ from those of earlier investigators. For example, condy- lobasal length is a traditional measure of skull length, but it is usually difficult and time-consuming to measure the anterior margin of the premaxillary bone. 18 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Employing the anterior tips of the first incisors for the anterior terminator results in a more accurate and repeatable measurement among specimens of comparable incisor wear and rotation. Individual specimens classed as overwintered adults were excluded from comparisons, because of the amount of incisor wear and rotation. Sorex roboratus Hollister, 1913 Hollister (1913a) based this name on a specimen from 5 mi S Dapucha (=Ta- pucha), Altai Mountains, Gorno-Altaisk Autonomous Oblast, R.S.F.S.R., U.S.S.R. This locality was not located on maps, but Hollister (1913b) described it as about 125 miles southeast of Biisk, along the ““post road”’ (Chuiskii Trakt) that follows the valley of the Katun and Chuya rivers. This would place Tapucha (Fig. 1) somewhere between the present towns of Shebalino and Tuekta, probably near Seminskii Pass. The holotype is a skin and badly broken skull, a male with unworn teeth, trapped by Hollister (orig. no. 4451), 7 August 1912, in a dense Pinus cembra forest, and represents one of ten Sorex captured by Hollister from two locations in the Altai Mountains. Hollister (1913a) believed that the nine speci- mens other than the holotype of S. roboratus were assignable to S. araneus bo- realis. Stroganov (1957) has shown, however, that the name S. borealis should be restricted to those populations of S. “‘arcticus”’ (i.e., S. tundrensis) inhabiting the tundra and northern taiga zone from the Yenesei River to Chukotka (see map, Fig. 2, in Junge et a/. 1983), thus throwing into question the identity of Hollister’s shrews; moreover, Stroganov (1957) placed S. roboratus as a subspecies of S. araneus. Dolgov followed Stroganov in regarding S. roboratus as a synonym of S. ara- neus. However, Dolgov (1964, 1966, 1967) and Yudin and others (1967, 1979) recognized that four medium to large species of Sorex were geographically sym- patric in the Altai Mountains—S. araneus, S. arcticus (=tundrensis), S. isodon, and the largest species, which they named S. roboratus or S. vir, respectively. I have re-examined nine of the ten Sorex referred to by Hollister, and find that they represent three species rather than two as he thought. Six from Tchegan- Burgazi (=Chagan-Burgazy) Pass are all Sorex tundrensis, probably S. t. schnit- nikovi (Junge et al. 1983). From Tapucha, in addition to the holotype of S. roboratus, there are two specimens of S. araneus. Sorex vir G. Allen, 1914 This name is based on a specimen captured by J. Koren at Nizhni Kolymsk, Yakutsk A.S.S.R., R.S.F.S.R., U.S.S.R. (Fig. 1). It was part of a large series of skins and skulls of this species taken during the fall—-winter of 1911, and now for the most part preserved at the Museum of Comparative Zoology. Koren also obtained several other species of Sorex at that time, including S. tundrensis borealis (incl. S. t. ultimus), S. daphaenodon sanguinidens, and S. caecutiens koreni. In contrast to S. roboratus, S. vir has been regarded as polytypic. Its distribution is eastern Palearctic—S. v. vir occurring from the lower Ob River to Chukotka; S. v. thomasi in Transbaikalia, and S. v. platycranius in a restricted range in the Ussuri and Amur river regions of the Soviet Far East (Fig. 1). Like S. roboratus in the Altai, S. vir is the largest of the sympatric Sorex that co-occur VOLUME 98, NUMBER 1 19 N70 W10 20 40 100 160 W180E 170 N70 N6o vr We ae oe 4 Ride Sf (4) pq, yi 180E IBNizhni olymsk Chukotka \__ Gichiga a Sees a " L866, SS jBdvovosivirsk Ob! 4S fo Baikal ) 1 Khabarovsk & Log, ‘ : A 5 — K } C } ral NN Turochak > UW te eS eee Barnaul & ‘ 2 Irkutsk Obt.| Wa ae on aie Me. AV \ SF os \ & Kebezen 7. _ ac ae [5 nee) 3S LON Sy —— --A@Tapucha. ———— ee Ce ie Z / se > tee Cc Sas — \ " AEX Race (= "Bchagah Ru aTCS pace yess 7 a > Primorsk A iN S (SSS = =, iS ya ee 6 oe ord So Li <> 2K = / QAltai Mts [Sayan mts_ mee W70 80 90 100 110 120 130 140W Fig. 1. Map of Siberia and adjacent Mongolia, showing type localities of Sorex roboratus, and collecting localities for some samples used in this study. Open symbols: type localities; solid symbols, other specimens examined. Triangles, S. roboratus; circles, S. araneus; squares, S. tundrensis. Dotted line, eastern and southern range boundary of S. araneus; dashed line, western and southern range boundary of S. tundrensis, dash-and-dotted line, western and southern range boundary of S. roboratus. in eastern Siberia (except in the Ussuri region, where the even larger S. mirabilis is found). Comparisons The basic question to be answered is: to which species of shrew should the holotype of S. roboratus be assigned? Cranial and external measurements are presented for selected specimens in Table 1. In most cranial dimensions, both S. roboratus and S. vir are larger than shrews of the S. araneus and S. tundrensis populations with which they occur. Where S. araneus and S. tundrensis occur together in the Altai, the former species is larger in most dimensions. However, S. araneus exhibits some size variation, being nearly as large as S. roboratus at Kebezen’ but only a little larger than S. tundrensis at Tapucha (Fig. 1). Specimens from Barnaul and Toguchin are intermediate in size. Fortunately, other characters help to separate the four nominal species. Pelage colors differ, both in summer and winter, in that in S. roboratus and S. vir the back and sides are the same color, whereas in S. araneus and S. tundrensis the sides are lighter than the back. This contrasting “‘side stripe’ is most prominent in adult S. tundrensis in winter pelage. The side stripe in adult S. araneus in winter pelage is more buff and contrasts less, while in neither species does it contrast as strongly with dorsal color in summer pelage. In juveniles of both S. araneus and S. tundrensis the side stripe is usually faint. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 20 87 87 82 87 87 L7 L7 v p v u 56-98 GOS OPE 5 ISS ISS AUIS SIU RSIS 6b-7r S'€8-L9 aduey 6£0° €v0" £70" £70" €£0° 8€0° 190° Sst c9'l €8'€ wast 16 $6 8'€ V's Ss 66 6°02 671 cbb 6LL x ,UdZIQ9J “ [GO ‘INY “YSlelpV-OUIOD ‘snauniv snauviv Xa10y 16 v6 = = ss 16 C1 cI Ov 08 16 v6 = = 8°¢ 76 We SI cE cL Terry “YSAoleqeuy ‘sniupsddjojd snjDIOGOA XAOS L p 4 c c L L 61 61 61 u 6'8-S'8 OCR8 OSE OWS i7S4es WHE} WOKE vI-Il Ip—-O€ €L-8S osuey £90" 7S0° ar. = ae 801° byl 961° 18° ell Ww AS+ L'8 6°8 9€ 64 v's 76 9°07 67I 9°SE 1°S9 x BITOSUOJ] “GO “ASMA ‘IsSWUO0Y] SNJDAOGOA XALOS 81 0@ 81 81 81 07 07 8r Lv 8b u 06-78 SOL OSE ESa57 —OC=S's OG SIEVE SGiKe €p-l€ L8-8S oduey 970" 0) 1€0" LZ0" EO" 0" SLO" eI" SE S83" Ww Ast 9°8 06 ge cs LS $6 6°02 Lvl y9€ OIL x 9°8 0°6 6€ CS 8°¢ v6 6°07 9r1 Le STL adMofoH YSUIA[OY IUYZIN “exoyNYO “7A snjoiogod xa10y’ S€°6 8°6 Is ms L's 66 617 €l ep IL TO YSIIQISOAON ‘SHJDAOGOA SNIDAOGOA XAOS’ O€ O€ O€ O€ O€ O€ O€ € € € u 953 VORS6G Sere SSS O85 EWR SMT Ie Ll cb-S ‘OF €8-9L osuey S€0° 6£0° 610° 020" S70" 101° LSO° BE" ey a wast £6 8°6 Or€ aks 8°¢ L'6 812 SG) TIP 0°6L x (udZBQoY “TGC ‘INV “YSIei[V-OUWIOH ‘snjoLOgGOA SNJDAOGOA XALOG' £6 = a S°S = 8°6 0'%Z SI Iv IL ad 00H eyondey “[qO ‘NY “ASlel[y-OUIOD ‘snjo10gos SNIDAOGOA XAIOS) TaLn ‘Id dO! ZN=zN aw t: (@) TID JOO} pul [ey Apoq-praH ‘XAOS ULLIDGIS JO SUOISUOUIIp [eIUeID pue [eUIOIXq—"] IQR L 21 VOLUME 98, NUMBER 1 € 6 € € € C € 9 9 9 u 8 L-OL €8-IL SISSIES Srv TL VS-8P 0'6-C 8 ¢8I-OLT CCI-II Le—-OE c9-SS osuey CEC Sel OCI col OLT ea Bor 8c L60° SLI adas+ E/E SL ce CPV OES 98 OLI Lit Ove 86S w BI[OSUOJ “GO “YSINAI] “YSIIQIsoOAON “dss sisuaipun] Xasogy v v v v v v v 9 9 9 u SL-CL 6L-GL CECE Oe a 0S-9'P € 6-6 8 ¢8I-9'LT CcI-Il Ov-Te L9-19 osuey ole LIT LvO 1€0° ¢80° c60° CoG LV sel cll wast CL LL ve i I 06 81 (a CSE € v9 x sseq Izeding-ueseyD “[qO ‘INV “YSIVI[V-OUIODH ‘MO0YIUJIUYIS SIsuaApUN] XAa10S' € € € € € ¢ ¢ “¢ iS ¢ u 16-98 Cc 6-16 Ov-8'€ c $-0'S 9°S-G'S 101-96 L0c-S'61 €I-S Cl ¢ 8r-CL CS8-PL osuey L80° L10 Adv} Adv) 6c0 680° 661° Il 60° 06 1 was+ 88 C6 6€ rs 9°¢ 86 C0C LCI LSv 66L x 1dO YSIIQISOAON ‘[neuleg “OIA ‘Tery “Ysieipy ‘SnauDiD SnauDdiD Xa10y' 6L 8 ve 9 os v6 681 cl 6€ co 18 C8 ee CPV. OES £6 L 8 cl Le v9 eyondel “[qO INV “YSIel[V-OUIOD ‘snauviv SnauviD Xa10y TaLn Tld dol 7N-N aw ao 19 100} PUIH TeL Apoq-peoH ‘panunuoy—"] 21921 22 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Skull of Sorex roboratus roboratus, juvenile, from Mirnyi, Toguchinsk. rai., Novosibirsk. Obl., U.S.S.R. (UMZ 13005). Dorsal, lateral, and ventral views; enlarged occlusal view of upper tooth Tow. Qualitative dental characters also distinguish the taxa. In both S. roboratus and S. vir in lateral view the anterior pair of maxillary unicuspid teeth (1, 2) are large and subequal, and the next two (3, 4) are markedly smaller and again subequal although the posterior unicuspid of each pair (2, 4) is usually smaller than the anterior (1, 3) (Fig. 2). In S. araneus and S. tundrensis the first two unicuspids are also large and subequal, but the unicuspid series tends to decrease in size more gradually from unicuspids 2 through 4 (Figs. 3, 4). However, the toothrow in S. tundrensis appears more crowded than that of S. araneus, and the unicuspids are quadrate in shape, to wider than long (Fig. 4), whereas in S. araneus the toothrow is less crowded, and the unicuspids are usually longer than wide (Fig. 3) (Junge and Hoffmann 1982; Junge et al. 1983). The holotypes of S. roboratus and S. vir are also characterized by the posterior margins of the maxillary molariform teeth, particularly P4 and M2, being deeply excavated (Fig. 2) (Corbet 1978). In contrast, the posterior margins of the upper VOLUME 98, NUMBER 1 23) Fig. 3. Skull of Sorex araneus araneus, juvenile, from Tapucha, Gorno-Altaisk. Aut. Obl., U.S.S.R. (USNM 175438). Dorsal, lateral, and ventral views; enlarged occlusal view of upper tooth row. molariform teeth of S. araneus and S. tundrensis are only moderately excavated (Figs. 3, 4). In S. roboratus and S. vir P4 is also relatively longer and narrower than in the other two species. Thus, the holotype of S. roboratus shares cranial and external characters with S. vir, including the holotype, that distinguish it from both S. araneus and S. tundrensis with which it is sympatric in the Altai Moun- tains. Sorex roboratus is not assignable to any other species of Sorex in the Altai (S. isodon is smaller, with gradually decreasing unicuspid size and narrow rostrum; S. caecutiens, S. minutus and S. minutissimus are much smaller). Thus, S. ro- boratus and S. vir should be regarded as conspecific, and the former name has priority. Sorex roboratus varies geographically in size and relative proportions. Sorex r. roboratus from the Altai Mountains and the surrounding region is the largest, with a broader, more massive skull. The holotype of S. araneus tomensis Ognev, 1921, formerly assigned to S. araneus, belongs to this large subspecies, and must be considered a junior synonym of S. r. roboratus. Sorex roboratus vir is smaller, 24 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 5mm DKB ‘33 Imm Fig. 4. Skull of Sorex tundrensis schnitnikovi, juvenile, from Chagan-Burgazy Pass, Gorno-Altaisk. Aut. Obl., U.S.S.R. (USNM 175430). Dorsal, lateral, and ventral views; enlarged occlusal view of upper tooth row. with a generally narrower, less massive skull but with wider interorbital region; the tail is also much shorter. Sorex vir jacutensis Dukelski, 1928, (originally described as a subspecies of S. araneus), which was recognized as a subspecies by Stroganov (1957), is similar in size and proportions to S. r. vir, and I regard it, as well as S. vir turuchanensis Naumov, 1931, and S. dukelskiae Ognev, 1933, to be junior synonyms of S. r. vir. This subspecies inhabits central and eastern Siberia from the Sayan Mountains to Lake Baikal on the south, and from the Ob River to Chukotka on the north. The last two subspecies I recognize have more limited ranges. Sorex roboratus thomasi inhabits Transbaikaliya, and is the small- est geographic race, being only slightly larger than sympatric populations of S. tundrensis, though easily distinguished by cranial morphology (see above). Sorex macropygmaeus araneoides Ognev, 1921, has been considered a junior synonym VOLUME 98, NUMBER 1 25 of S. araneus (Bobrinskii et al. 1944) or S. c. caecutiens (Stroganov 1957; Corbet 1978), but examination of the holotype reveals that it is virtually identical with the holotype of S. r. thomasi. Both were taken near Barguzin, share the diagnostic cranial features of S. roboratus, and are larger than sympatric S. tundrensis and S. caecutiens. The range of S. r. platycranius adjoins that of S. r. thomasi along the Amur River and extends to the Primorskii Krai. This subspecies is larger than S. r. thomasi, being about the size of S. r. vir, but it has a relatively longer tail and rostrum; it is known from only a few specimens. Taxonomic Conclusions The correct names and synonyms for the subspecies and other nominal taxa of Sorex roboratus are summarized below: Sorex roboratus roboratus Hollister, 1913. Type locality, Tapucha, Gorno-Altaisk Autonomous Oblast, R.S.F.S.R., U.S.S.R. S. araneus tomensis Ognev, 1921. Type locality, vic. Barnaul, Altaisk. Krai, R.S.F.S.R., U.S.S.R. Sorex roboratus vir Allen, 1914. Type locality, Nizhni Kolymsk, Yakutsk. A.S.S.R., R.S.F.S.R., U.S.S.R. S. araneus jacutensis Dukelski (=Dukel’skaya) 1928. Type locality, Suntar, Yakutsk. A.S.S.R., R.S.F.S.R., U.S.S.R. S. vir turuchanensis Naumov, 1931. Type locality, Artyugin River, tributary of Yenesei River, Turukhansk. rai., Krasnoyarsk. Krai, R.S.F.S.R., U.S.S.R. S. dukelskiae Ognev, 1933. Type locality, Yanov Stan, on Turukhana River, Krasnoyarsk. Krai, R.S.F.S.R., U.S.S.R. Sorex roboratus thomasi Ognev, 1921. Type locality, Budarman River, tributary of Sosnovka River, vic. Barguzin, Buryat-Mongolsk. A.S.S.R., R.S.F.S.R., U.S.S.R. S. aranoides Ognev, 1921. Type locality, Sosnovka River valley, near Barguzin, Buryat-Mongolsk. A.S.S.R., R.S.F.S.F., U.S.S.R. Sorex roboratus platycranius Ognev,. 1921. Type locality, vic. Voroshilov, Pri- morsk. Krai, R.S.F.S.R., U.S.S.R. Specimens examined Abbreviations: AMNH, American Museurn of Natural History, New York; BMNH, British Museum (Natural History), London; FMNH, Field Museum of Natural History, Chicago; KU, Museum of Natural History, University of Kansas, Lawrence; MCZ, Museum of Comparative Zoology, Harvard University, Cam- bridge; MGU, Moscow State University, Moscow; MVZ, Museum of Vertebrate Zoology, University of California, Berkeley; UMZ, University of Montana Zoo- logical Museum, Missoula; USNM, National Museum of Natural History, Washington; ZIN, Zoological Institute, Academy of Sciences, Leningrad. Sorex roboratus roboratus. (51) U.S.S.R: R.S.F.S.R., Gorno-Altaisk Aut. Obl., Tapucha (USNM 175436, holotype); Kebezen’sk. rai. (MGU 73624, -27, -29, 74325, -28, -30, -40, -46, -48, -49, -52, -54 through -56, -62, -64, -65 -66, -70, -71, -73, -74, -76, -77, -79, -82, -85, -87, 77742, 82182) Altaisk. Krai, vic. Barnaul (ZIN 9173, holotype S. araneus tomensis). Novosibrisk. Obl., Togu- 26 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON chinsk. rai., vic. Mirnyi (UMZ 13005); Buryat-Mongolsk. Aut. Obl., 100 mi W Lake Baikal (BMNH 12.4.1.1, -2); Irkutsk. Ob., Alzamai (BMNH 14.11.1.8); Listvyank (BMNH 15.3.9.1 through 9; 14.11.1. 2 through 7). Sorex roboratus vir. (52) U.S.S.R.: R.S.F.S.R., Yakutsk. A.S.S.R., Nizhni Ko- lymsk (MCZ 15068, holotype; 15018, -23 through -31, -33 through -38, -40, -42 through -44, -46 through -51, -53 through -59, -64 through -66, -70 through -74, -82; AMNH 38880, -81; FMNH 34107 through -09; MVZ 81167); Suntar (MGU 4819, holotype S. araneus jacutensis); Krasnoyarsk. Krai, Turukhansk. rai., Artyugin R., trib. Yenesei (MGU 13455, holotype, S. dukelskiae); Yanov Stan, Turukhana R. (ZIN 10561, holotype, S. vir turuchanensis). Sorex roboratus thomasi. (10) U.S.S.R.: R.S.F.S.R., Buryat-Mongolsk. A.S.S_R., Budarman R., trib. Sosnovka R., vic. Barguzin (ZIN 11904, holotype); Sosnovka R. valley, vic. Barguzin (ZIN 11905, holotype, S. aranoides). MONGOLIAN PEOPLES REPUBLIC: Tov Aymak, 45 mi N Urga (=Ulan Bator) (AMNH 45573, -80, -81, -90; FMNH 39317, -18; MCZ 45588, -89). Sorex roboratus platycranius. (3) U.S.S.R.: R.S.F.S.R., Primorsk. Krai, vic. Vo- roshilov (ZIN 2392, holotype); Khabarovsk. Krai, 60 mi N Khabarovsk, Nelta River (AMNH 85468, -69). Sorex araneus araneus. (40) U.S.S.R.: R.S.F.S.R., Zapadnaya (=Kalininsk.) Obl., Ostashkovsk. rai., Petropavlovsk (MCZ 32778 through -80); Leningradsk. Obl., Valdansk. rai., Yakonova (MCZ 32781); Novosibirsk Obl., Toguchinsk. rai., vic. Mirnyi (UMZ 13003, -04); Al/taisk. Krai, near Barnaul (BMNH 28.10.24.1, -.2; MCZ 23952, -53); Gorno-Altaisk Aut. Obl., Tapucha (USNM 175437, -38); Kebezen’sk. rai. (MGU 73632, -35, -36, -39, 734185, -87, -88, -91 through -95, -97, 74204 through -13, -15, -18, -20, -24); Turochaksk. rai. (MGU 73671). Sorex tundrensis borealis. (51) U.S.S.R.: R.S.F.S.R., Koryaksk. Nats. Okr., Gi- chiga (AMNH) 18626, -27, -33, -37, -42, -49); Yakutsk. A.S.S.R., Nizhni Ko- lymsk (AMNH 38884, -85; MCZ 14992, -95, -97 through -99, 15001, -02, -69, -76, -78 through -80, -84); Chukotsk. Nats. Okr., Palyavaam River (MGU 88237, -44, -48, -50, -53, -58, -59, -63 through -65, -68, -69, -71 through -73, -77 through -79, -82, -85, -88, -89, -92, -96, -300, -01, -04, -05, -10, -11). Sorex tundrensis baikalensis. (6) U.S.S.R.: R.S.F.S.R., Chitinsk. Ob., Gornyi Zer- entui (MGU 2710, holotype); Primorsk. Krai, Nadezhdinsk. Obl., Razdol’noe (KU 121366). MONGOLIAN PEOPLES REPUBLIC: Tov Aymak, 15 mi N Urga (=Ulan Bator) (AMNH 45595; MCZ 20744); 45 mi N Urga (FMNH 39312, -14). Sorex tundrensis petschorae. (3) U.S.S.R.: R.S.F.S.R., Arkhangel’sk. Obl., Pvim- va (ZIN 8459, holotype); Tyumensk. Obl., Purovsk. rai., Samburg (KU 121364- 65). Sorex tundrensis schnitnikovi. (7) U.S.S.R.: R.S.F.S.R., Gorno-Altaisk. Aut. OBl., Tchegan-Burgazi (=Chagan-Burgazy) Pass (USNM 175429 through -31; MCZ 14373 through -75); Kazakh S.S.R., Vostochno-Kazakhstansk. Obl., Kopal (ZIN 8641, holotype). Sorex tundrensis sibiriensis. (12) U.S.S.R.: R.S.F.S.R., Kemerovsk. Ob1., Kol’chugino Sta., 4 km SE Leninsk- Kuznetskii (ZIN 6527, holotype); Novo- sibirsk. Obl., vic. Novosibirsk (USNM 253025); Krasnoyarsk. Krai, Sev. Vos- tochnoe, 40 km SE Minussinsk (MGU 13452, holotype S. jenissejensis); ““Yene- sei region” (probably Irkutsk. Obl., Alzamai) (BMNH 14.11.1.9 through -11); VOLUME 98, NUMBER 1 Di “Irkutsk region” (probably Irkutsk. Obl., Listvyanka) (BMNH 14.11.1.12, -13); Buryat-Mongolsk. A.S.S.R., Sayan Mts., 100 mi W of Lake Baikal (BMNH 12.4.1.3. through -.5, -.6, holotype S. centralis). Sorex isodon isodon. (28) U.S.S.R.,: R.S.F.S.R., Gorno-Altaisk. Aut. Obl., Turo- chaksk. rai. (MGU 73642, -47); Kebezen’sk. rai(MGU 73643 through -46, -51, -52, -56, -58, -61 through -65, -67, -72 through -79, -81, -82, -88); Buryat- Mongolsk. A.S.S.R., Sosnovka River, [N of] Barguzin (MGU 13486, holotype). Sorex caecutiens caecutiens. (12) U.S.S.R.: R.S.F.S.R., Gorno-Altaisk. Aut. ODbI., Buisk. rai. Ongudai (ZIN 6409, holotype, S. c. altaicus) Turochaksk. rai. (MGU 110062); Kebezen’sk. rai. (MGU 73999, 74000, -02, -05, -06, -08, -15, -55, -58); Tomsk. Obl., Listshchi River (UMZ 13009). Acknowledgments Support of National Science Foundation Grant No. DEB 80-04148 has been essential in this work, which is part of a joint research program on Holarctic mammals sponsored by the Academies of Science of the U.S. and the U.S.S.R., and the bilateral Environmental Protection agreement (Project 0.2.05-7104). Drs. B. S. Yudin, Institute of Zoology, Siberian Branch, Academy of Sciences of the U.S.S.R., Novosibirsk, and V. A. Dolgov, Department of Vertebrate Zoology, Moscow State University, have provided much advice on Palearctic Sorex for many years, although they do not share all of my opinions. To them, and to the curators and institutions who made specimens available to me, I express my thanks. I also thank the University of Kansas for the sabbatical leave that provided me the time to complete this project, Rosetta Herman and Kathy McManness, who typed the manuscript, and Deb Bennett, who drafted the figures. G. B. Corbet, British Museum (Natural History), suggested improvements in an early draft, as did several anonymous reviewers. Literature Cited Allen, G. M. 1914. Notes on the birds and mammals of the arctic coast of east Siberia. Mammals. — Proceedings of the New England Zoological Club 5:49-66. Bobrinskii, N. A., B. A. Kuznetsov, and A. P. Kuzyakin. 1944. Opredelitel’ mlekopitayushchikh SSSR. [Guide to the mammals of the USSR.]—“‘Sovetskaya Nauka,”’ Moscow, 440 pp. Corbet, C. B. 1978. The mammals of the Palearctic region. A taxonomic review. — British Museum (Natural History) and Cornell University Press, London and Ithaca, N.Y. [viii] + 314 pp. Dolgov, V.A. 1964. Ravnozubaya burozubka—Sorex centralis Thomas, 1911 (Mammalia, Soricidae) v faune Sovetskovo Soyuza.— Zoologicheskii Zhurnal 43:898-903. Dolgov, V. A. 1966. Vidy Palearkitcheskikh zemleroek rod Sorex i ikh individual’naya 1 populyat- sionnaya izmenchivost’.— Dissertation summary, Moscow State University, 20 pp. Dolgov, V. A. 1967. Rasprostranenie i chislennost’ Palearkticheskikh burozubok (Insectivora, So- ricidae).— Zoologicheskii Zhurnal 46:1701-1712. Dukel’skaya (=Dukelski), N. M. 1928. Opyt obzora fauny mlekopitayushchikh gosudarstvennovo Il’menskovo zapovednik. — Moscow [original not seen; from Stroganov, 1957; pagination not given.] Ellerman, J. R., and T. C. S. Morison-Scott. 1951. Checklist of Palaearctic and Indian mammals 1758-1946.—British Museum (Natural History), London, 810 pp. Hollister, N. 1913a. Two new mammals from the Siberian Altai.—Smithsonian Miscellaneous Col- lections 60(24):1—3. Hollister, N. 1913b. Mammals collected by the Smithsonian—Harvard expedition to the Altai Mountains, 1912.—Proceedings of the United States National Museum 45:507-532. 28 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Junge, J. A., and 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 94:1—48. Junge, J. A., R. S. Hoffmann, and R. W. Debry. 1983. Relationships within the Holarctic Sorex arcticus-Sorex tundrensis species complex.— Acta Theriologica 28:339-350. Naumov, S. P. 1931. Mlekopitayuschie i ptitsy Gydanskovo poluostrova.— Trudy Polyarnaya Kom- missiya, Akademiya Nauk SSSR, Leningrad, vol. 4. [original not seen; from Stroganov, 1957; pagination not given.] Ognev, S. I. 1921. Materialy dlya sistematiki nasekomoyadnykh mlekopitayushchikh Rossii. —Ez- hegodnik Zoologicheskovo Musei Akademiya Nauk 13:311-350. Ognev, S. I. 1933. Materialy po sistematike Palearkticheskikh zemleroek.—Byulletin Nauchno- issledovanie, Institut Zoologii, Moscow State Univ. 1:57-63. Stroganov, S. Yu. 1957. Zveri Sibiri. Nasekomoyadnie. [Mammals of Siberia. Insectivores.]—Aca- demiya Nauk SSSR, Moscow, 267 pp. Yudin, B. S. 1962. Ekologiya burozubok (rod Sorex) Zapadnoi Sibiri.—Trudy Biologicheskovo Instituta, Sibirskoe Otdelenie, Akademiya Nauk USSR, No. 8:33-134. Yudin, B. S. 1971. Nasekomoyadnye mlekopitayushchie Sibiri. [Insectivorous mammals of Sibe- ria]— Nauka, Novosibirsk, 169 pp. Yudin, B.S.,and L.I. Barsova. 1967. Zemleroiki kedrovykh lesov priteletskovo ochaga kleshchevovo entsefalita. In Prioroda ochagov kleshchev entsefalita na Altae.— Nauka, Novosibirsk, pp. 51- 59. Yudin, B. S., L. I. Galkina, and A. F. Potapkina. 1979. Mlekopitayuishchie Altae-Sayanskoi gornoi strany. [Mammals of the Altai-Sayan montane region].— Nauka, Novosibirsk, 196 pp. Museum of Natural History and Department of Systematics and Ecology, Uni- versity of Kansas, Lawrence, Kansas 66045. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 29-46 TAXONOMIC STATUS AND RELATIONSHIPS OF THE SWAN ISLAND HUTIA, GEOCAPROMYS THORACATUS (MAMMALIA: RODENTIA: CAPROMYIDAE), AND THE ZOOGEOGRAPHY OF THE SWAN ISLANDS VERTEBRATE FAUNA Gary S. Morgan Abstract.—Comparisons of external and cranial characters in the three Recent species of Geocapromys demonstrate that the Swan Island Hutia, G. thoracatus, is a distinct species, rather than a subspecies of the Jamaican Hutia, G. brownii, as it has been regarded by most recent authors. Based on derived characters of the zygomatic arch region, G. thoracatus and G. brownii are closely related and constitute a species-group within Geocapromys. The third extant species of the genus, G. ingrahami from the Bahamas, and at least four extinct species referable to Geocapromys form a second species-group. All available information pertaining to the recent extinction of the Swan Island Hutia is reviewed. Zoogeographic analysis of the Swan Islands vertebrate fauna reveals that the majority of species have been derived from the West Indies, substantiating placement of the Swan Islands in the West Indian Subregion of the Neotropical Region. Overwater dis- persal is advocated to explain the origin of the fauna, as the geological history of the Swan Islands precludes vicariance as a tenable biogeographic hypothesis. ““There are some islands which are so small and isolated that the mere presence, or the mere absence, of certain birds and mammals on them, lends them just that touch of interest which they would not otherwise possess.” Percy Lowe (1911: 38). Geocapromys thoracatus, a member of the endemic West Indian hystricognath rodent family Capromyidae, is known only from Little Swan Island, a tiny (2 km?), remote, limestone island located in the northwestern Caribbean Sea. The taxo- nomic status of G. thoracatus has been unclear from the beginning. When True (1888) described G. thoracatus as a subspecies of the Jamaican Hutia, Capromys brachyurus Hill, 1851 (=C. brownii Fischer, 1830), he compared only external characteristics of the two forms. Jamaican specimens were not available to True, so his comparisons were based on Hill’s original description of C. brachyurus in Gosse (1851). True (1888:470) distinguished thoracatus from brachyurus on the basis of “‘. .. the white band of fur across the breast, the gray throat, and brown and ochreous (not blackish) hind feet .. . .”» He noted that the Little Swan speci- mens, of which he had two, closely resembled C. brachyurus in size and propor- tions. This last observation is puzzling, because all later workers have mentioned the considerably larger size of the Jamaican animal. In his original description of Capromys ingrahami from East Plana Cay, a small island in the Crooked-Acklins group in the south-central Bahamas, J. A. Allen (1891) compared the species with C. thoracatus. He noted that these two species 30 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON were similar in overall coloration, but that C. ingrahami was a smaller animal with a relatively longer tail, and also differed in cranial features. Except for the narrower jugal and less pronounced lateral jugal fossa of C. ingrahami, most of Allen’s characters are of doubtful value as judged from larger samples. Chapman (1901), in his revision of Capromys, erected the subgenus Geocapro- mys to include the short-tailed members of the genus: G. brownii, G. thoracatus, and G. ingrahami. Compared with the other two species, Chapman found that the skull of G. brownii is larger, lacks supraorbital processes and is relatively broader at the interorbital constriction. In a direct comparison of G. brownii and G. thoracatus, Chapman (1901:321) made the following observations: ““The ac- quisition of the above-mentioned specimen of C. brownii (=brachyurus auct.) permits, for the first time, actual comparison of thoracatus with the form to which it has generally been supposed to be subspecifically related. The result shows the two animals to differ widely from each other in color, dimensions, particularly of the ears, and in cranial characters. In fact, thoracatus proves to be much more closely related to ingrahami, from which indeed it is to be distinguished externally only by size, the two known specimens of thoracatus agreeing exactly in color with the prevailing type of C. ingrahami as it is shown by a series of twelve specimens in the America Museum.” Based on the differences between G. brownii and G. thoracatus, Chapman regarded the latter as a full species. G. M. Allen elevated Geocapromys to generic rank and in so doing, he noted (1917:8-9), ““Three living species are included in this genus. Of these, Geocapro- mys brownii, of Jamaica, is the largest. The two others, G. thoracatus of Little Swan Island, and G. ingrahami of Plana Keys, Bahamas, are smaller, and much more resemble each other in their gray type of coloring than they do the large dark brown animal of Jamaica. As Chapman pointed out, these may indicate two species groups.”’ Allen also mentioned that the incisors are very pale yellow in G. brownii and G. ingrahami and ivory white in G. thoracatus; however, every specimen of G. brownii I have examined has dark, yellowish-orange incisors. In a report on the generic characters of Geocapromys, Miller (1929) noted that the ascending process of the maxilla is essentially vertical in G. ingrahami and posteriorly sloping in G. brownii and G. thoracatus. Lawrence (1934) presented a key summarizing the cranial characters of all the then known species of Geo- capromys, both living and extinct. She distinguished G. brownii based on its larger size, anterior inflation of the frontals, constriction of the frontals posterior to the supraorbital processes, and short, bluntly triangular supraorbital processes. Geo- capromys thoracatus was differentiated from G. brownii by its smaller size and from G. ingrahami by the pronounced jugal spine and more vertical posterior margin of the jugals. Geocapromys thoracatus was recognized as a distinct species from the time Chapman (1901) first accorded it full specific rank until Mohr (1939) relegated it to its current status as a subspecies of G. brownii. Almost all authors have followed Mohr in calling the Swan Island Hutia, G. brownii thoracatus (Clough 1972, 1976; Hall 1981; Oliver 1976, 1977; Varona 1974). Furthermore, Mohr (1939) reduced Geocapromys to its former status as a subgenus of Capromys, a usage that has gained favor with some recent authors (Hall 1981; Varona 1974). However, based on nine external and cranial characters, I recognize Geocapromys as a distinct genus. External features which differentiate Geocapromys from Capromys are the VOLUME 98, NUMBER 1 31 short tail, reduced first digit on the front foot, and the shorter, finer fur. Cranial characters separating the two genera are discussed in detail by Morgan (1977) and Woods and Howland (1979). The most significant diagnostic features of Geocap- romys noted by these authors are: the less arched or more procumbent incisors, the origin of the upper incisor root capsule high on the maxilla above the P’*, the broad vertically or posteriorly oriented superior zygomatic root of the maxilla, the tendency toward anterior convergence of the upper tooth rows, the labial inclination of the occlusal surface of the cheek teeth, and the presence of an additional anterolingual re-entrant (anteroflexid) on the P,. In a study of the living and extinct species of Geocapromys undertaken to determine the affinities of an undescribed extinct species of the genus from the Cayman Islands (Morgan 1977), it became clear to me that the external and cranial differences between G. thoracatus and G. brownii represented distinctions between species, not subspecies. Many of these differences have been pointed out by pre- vious authors, but they have not been adequately summarized. For the sake of completeness, I have included G. ingrahami in my descriptions and analyses. Methods and specimens. — External measurements are those of the original col- lectors and were taken from skin labels. Cranial measurements were taken with dial calipers and rounded to the nearest 0.1 mm. In addition to the standard mammalian cranial measurements, as defined in DeBlase and Martin (1974), other measurements were taken. Those that require further explanation include: the internal width of palate taken between the anterolingual edges of alveoli of P* and M! and at the posterior palatal margin; the occlusal lengths of the upper and lower tooth rows are measurements of the teeth—not alveoli; length of dentary taken from posterior tip of angular process to anterior extension of incisor alveolus; and length of mandibular symphysis taken from posterior edge of alveolar sheath to anterior extension of incisor alveolus. Only adults were measured, as deter- mined by the eruption and wear on M3 and by fusion of the basioccipital and basisphenoid. For two reasons the sexes were combined in the statistical analyses. First, the sex was not recorded on 25 of the 65 specimens examined. Second, the presence of an enlarged clitoris in female Geocapromys increases the likelihood that individuals were incorrectly sexed in the field. For example, of the 25 sexed specimens of G. thoracatus and G. ingrahami examined, only three were recorded as females. Descriptive morphological terms are standard except certain terms used to describe hystricognathous rodents (Woods and Howland 1979). Dental terminology follows Wood and Patterson (1959:287). I examined and measured the majority of existing specimens of Recent Geo- capromys, including all specimens from the following museums: American Mu- seum of Natural History (AMNH), British Museum (Natural History) (BMNH), Florida State Museum (UF), Museum of Comparative Zoology (MCZ), and Na- tional Museum of Natural History, Smithsonian Institution (USNM). The fol- lowing specimens were examined: Geocapromys brownii (23; 9 64, 6 °°, 8?) JAMAICA: Portland Parish, John Crow Mountains, AMNH (7), MCZ (2), UF (1); Stony Hill, UF (1); St. Thomas Parish, Cuna Cuna, AMNH (1), MCZ (1), USNM (2); St. Catherine Parish, Worthy Park, UF (2); no specific locality, AMNH (2), BMNH (2), MCZ (3). Geocapromys ingrahami (21; 9 66, 1 2, 11?) BAHAMAS, East Plana Cay, AMNH (8, including type), BMNH (2), MCZ (8), USNM (3). 32 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Geocapromys thoracatus (21; 13 66, 2 22, 6?) SWAN ISLANDS, Little Swan Island, AMNH (2), BMNH (7), MCZ (10), USNM (2, including type). Morphological Comparisons External characters.—The most obvious difference among the three species of Geocapromys is size (Table 1): G. brownii is largest in body size, G. thoracatus is intermediate, and G. ingrahami is smallest. As noted by many previous workers, G. thoracatus and G. ingrahami are similar in overall coloration. Both species have grayish-brown upperparts and a light brown to tan venter. In these two species, the majority of hairs on the back are light brown at the base with tan tips, but interspersed with these, especially in the middle of the back, are longer unbanded dark brown hairs. These darker hairs, together with the paler banded hairs, produce the grayish-brown color of the dorsum. Geocapromys thoracatus has a cream-colored collar 1-3 cm in width that extends transversely across the chest between the front limbs, hence the specific epithet. Geocapromys ingrahami lacks this bar, but some individuals do have a cream-colored spot between the front limbs that may extend posteriorly to the genital region. Compared to the two smaller species, G. brownii is a much darker animal, being dark reddish- brown to blackish-brown on the dorsum and medium brown on the belly. The dorsal guard hairs are alternately banded reddish-brown and dark brown or black, with longer black hairs interspersed. Unlike the other two species, G. brownii has no lighter colored spots or bars on the underside. The tail is short in all Geocapromys, but it varies among the three species (Table 1). The tail is shorter than the hind foot and sparsely furred in G. brownii, ap- proximately equal in length to the hind foot and sparsely furred in G. thoracatus, and longer than the hind foot and densely furred with short, reddish-brown hairs in G. ingrahami. The difference in tail length between the three species appears to be correlated with the number of caudal vertebrae. The longest-tailed species, G. ingrahami, has an average of 19 caudal vertebrae, G. thoracatus has 17, and G. brownii 14. The ears also differ in size (Table I) and morphology among the living members of the genus. Geocapromys thoracatus has comparatively large ears that appear to be almost naked, although both the internal and external surfaces have a sparse covering of short, fine hairs. Geocapromys brownii has small ears that are covered by a dense mat of short, fine hairs and in addition has two tufts of longer hair on the inner surface of the pinna, one above and behind the meatus and the other directly posterior to the meatus on the ventrolateral margin of the ear. Geocap- romys ingrahami has intermediate-sized ears that are clothed with long, poste- riorly directed hairs. As in G. brownii, there are two tufts of hair along the inner dorsal margin of the ear, although the tufts are much more prominent in G. ingrahami. The presence of long ear tufts in G. ingrahami appears to be unique within the Capromyidae. Most species of Capromys have nearly naked ears, resembling those of G. thoracatus. Cranium.— The most obvious difference between the cranium of Geocapromys brownii on the one hand and. G. thoracatus and G. ingrahami on the other, is the larger size of the former (Table 1, Figs. 1, 2). Geocapromys brownii averages 15-— VOLUME 98, NUMBER 1 33 Table 1.—External, skull, and dental measurements (in mm) of Geocapromys brownii, G. thoracatus, and G. ingrahami. The mean, standard deviation, sample size (in parentheses), and observed range, respectively, are given for each measurement. Measurement Geocapromys brownii Geocapromys thoracatus Geocapromys ingrahami Length of head and body 410 + 24 (12) 338 +4 (5) 308 + 21 (4) 372-448 334-343 280-326 Length of tail 48 +8 (13) 65+5 (5) Vi se 5 (7) 40-64 57-70 70-85 Length of hindfoot 70+6 (14) 66+2 (5) 58 +5 (7) 60-78 64-70 53-65 Length of ear (from notch) 20+1 (5) 26+1 (8) 19+4 (7) 19-21 24-28 15-24 Number of caudal vertebrae 144+1 (7) 17+1 (8) 19+1 (5) 12-15 15-18 18-19 Greatest length of skull 81.1 + 3.7 (19) 68.6 + 1.8 (19) 63.2 + 1.2 (16) 75.1-87.0 65.9-73.0 61.5-65.6 Condylobasal length 75.3 + 4.0 (14) 63.6 + 2.0 (18) 59.3 + 1.4 (12) 68.9-8 1.6 60.4-67.9 57.5-61.6 Zygomatic breadth 43.7 + 2.5 (18) 34.8 + 1.3 (18) 32.8 + 1.3 (18) 39.1-48.4 32.7-36.9 30.3-34.9 Breadth at auditory meatus 29.5 + 1.1 (17) 25.0 + 0.9 (17) 24.4 + 1.0 (15) 27.2-30.9 23.4—27.1 23.2-26.5 Breadth of frontals anterior to 23.9 + 1.2 (18) 17.8 + 0.8 (21) 17.1 + 0.8 (19) supraorbital processes 20.5-26.1 16.3-19.3 15.3-18.5 Breadth of frontals posterior to 19.7 + 1.2 (19) 19.9 + 0.8 (21) 17.9 + 0.7 (19) supraorbital processes 18.0—22.6 18.8—21.3 17.1-19.3 Breadth of superior zygomatic 5.5 + 0.9 (19) 4.5 + 0.3 (20) 3.2 + 0.6 (18) root of maxilla 4.0-7.6 4.0-5.2 1.9-4.3 Breadth of palate anterior to P* 3.4 + 0.5 (18) 2.6 + 0.3 (21) 2.4 + 0.3 (19) 2.6—4.2 2.3-3.1 1.9-2.8 Breadth of palate between P* 3.4 + 0.6 (16) 2.9 + 0.3 (19) 2.2 + 0.3 (19) and M! 2.4—4.6 2.3-3.7 1.6—2.8 Breadth of palate at posterior 6.4 + 0.4 (18) 5.6 + 0.3 (19) 4.8 + 0.4 (19) palatal margin 5.5-7.0 5.3-6.5 3.8-5.6 Length of upper diastema 19.6 + 1.1 (19) 17.0 + 0.6 (20) 15.4 + 0.5 (18) 17.9-21.6 16.0-18.5 14.7-16.6 Alveolar length of upper tooth 19.3 + 0.7 (19) 15.2 + 0.5 (21) 15.6 + 0.5 (19) TOW 18.0-20.3 14.4-15.9 14.8-16.4 Occlusal length of upper cheek 18.1 + 0.7 (11) 14.1 + 0.6 (21) 14.6 + 0.6 (18) teeth 17.0-19.4 12.8-15.4 13.3-15.6 Greatest length of mandible 55.8 + 2.7 (17) 45.2 + 1.6 (18) 41.3 + 1.7(15) 50.7-59.8 41.4-48.3 38.0—43.2 Length of lower diastema 13.7 + 1.0 (19) 12.8 + 0.7 (20) 10.1 + 0.6 (18) 12.6-16.0 11.5-14.3 9.2-11.3 Alveolar length of lower tooth 19.1 + 1.0(19) 14.6 + 0.5 (20) 15.2 + 0.6 (17) TOW 17.5—21.0 13.3-15.4 13.9-16.2 Occlusal length of lower cheek 18.7 + 1.0 (12) 14.4 + 0.6 (20) 14.6 + 0.6 (17) teeth 17.3-20.8 13.5-15.5 13.7-15.6 34 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 20% larger than G. thoracatus, depending on the measurement, with almost no overlap between the smallest G. brownii and the largest G. thoracatus. G. ingra- hami is slightly smaller (6-8%) than G. thoracatus, but there is broad overlap between them in some measurements. One of the most characteristic features of Geocapromys brownii is the prominent inflation of the frontals medial to the anterior edge of the orbits. The degree of frontal inflation varies individually, but all show some evidence of it. When viewed laterally, the frontals are seen to form a noticeable bulge in the dorsal profile of the skull (Fig. 1D). The inflation of the frontal sinuses begins immediately posterior to the nasofrontal suture, extends posteriorly to the level of the supraorbital processes, and is also present on the orbital wall dorsal to the lacrimals. The frontals are not inflated in G. thoracatus or G. ingrahami. Although the functional significance of frontal inflation is not known, it is present in several other groups of hystricognath rodents, particularly in the porcupines Coendou and Hystrix. Anterior to the frontoparietal suture, the interorbital region is strongly constricted in G. brownii, but not in the other two species. The anteriorly inflated and pos- teriorly constricted frontals of G. brownii are thus considerably broader anterior to the supraorbital processes than posterior to them. This contrasts with G. thorac- atus, in which the frontals are always broader posterior to the supraorbital pro- cesses, and with G. ingrahami, in which the two measurements are nearly equal. Most specimens of Geocapromys brownii have a moderate to strongly developed sagittal crest formed by the convergence of weak temporal crests. The temporal crests are stronger in G. thoracatus and G. ingrahami, but never meet to form a sagittal crest in the latter, and only rarely meet to form a weak crest in the former. Compared to the other two species, G. ingrahami has an inflated and foreshortened braincase and a constricted pterygoid region. The auditory bullae of G. ingrahami are also inflated, whereas those of G. brownii and G. thoracatus are not. In posterior view, the bullae of G.ingrahami project ventral to the occipital condyles, whereas in the other two species the ventral surface of the bullae is always dorsal to the condyles. Additionally, the bullae of G. ingrahami are comparatively shorter, broader anteriorly, and have a larger external auditory meatus. The combination of the inflated bullae and a narrower basioccipital results in the anterior portions of the bullae being in closer approximation in G. ingrahami than in its congeners. In all Geocapromys, the lateral jugal fossa is present, being particularly large in G. brownii, slightly smaller in G. thoracatus, and reduced in G. ingrahami. The enlarged jugal fossa of G. brownii is partially the result of a prominent jugal spine on the posteroventral edge of the jugal. The jugal spine is present, but smaller in G. thoracatus and absent in G. ingrahami. In lateral view, the posterior portion of the zygomatic arch appears to be rotated ventrally in G. brownii and G. thorac- atus in comparison to that of G. ingrahami or Capromys. The downturning or flexion of the zygomatic arch region is best observed in the relationship between the ventral border of the jugal and the alveolar margin of the upper cheek teeth. The ventral border of the zygomatic arch is inclined relative to the alveolar margin — Fig. 1. Dorsal (A-C) and left lateral (D-F) views of cranium of Geocapromys species. A, D, G. brownii, MCZ 11040, Jamaica; B, E, G. thoracatus, AMNH 34547, Little Swan Island; C, F, G. ingrahami, MCZ 29427, East Plana Cay, Bahamas. All photographs are natural size. VOLUME 98, NUMBER 1 35 36 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON VOLUME 98, NUMBER 1 37 and projects ventral to it in G. brownii and G. thoracatus. In most specimens of these two species the jugal spine, the ventralmost portion of the zygomatic arch, is ventral to the occlusal surface of the cheek teeth. In G. ingrahami, the ventral border of the zygomatic arch is essentially parallel to the alveolar margin and does not project ventral to it. Apparently, as a result of the ventral rotation of the zygomatic arch, the superior zygomatic root of the maxilla in G. brownii and G. thoracatus is inclined posteriorly relative to a perpendicular line drawn through the alveolar margin. In addition, the superior zygomatic root in these two species is relatively broader than in G. ingrahami or Capromys. The superior zygomatic root of G. ingrahamiis perpendicular or inclined slightly anteriorly. The zygomatic arches in G. brownii are broadest anteriorly, whereas in G. thoracatus and G. ingrahami the lateral margins of the zygomatics are parallel. The upper tooth rows converge anteriorly in all Geocapromys more so than in Capromys. Geocapromys ingrahami shows the greatest tendency toward this con- vergence, G. brownii the least. The upper tooth rows of G. ingrahami do not diverge as strongly as in G. brownii and G. thoracatus and as a result, the internal nares are constricted laterally. The anterior convergence of the upper tooth rows and constriction of the internal nares are carried to an extreme in several extinct species of Geocapromys in which the tooth rows nearly meet anteriorly. In contrast to the condition in G. ingrahami, the opening of the internal nares is constricted vertically in G. thoracatus due to the dorsal inclination of the palate from anterior to posterior. The median ridge of the palate has been used to distinguish species of Geocapromys, but I found this character to be highly variable in all species except G. thoracatus. Geocapromys thoracatus has a small spinous process that projects 1-2 mm beyond the posterior palatal margin along the midline; this process was not observed in any other species of the genus. Both the alveolar and occlusal lengths of the upper tooth rows are shorter in G. thoracatus than in G. ingrahami, even though the former is larger in most other cranial measurements. This derives from the comparatively small cheek teeth of G. thoracatus, a feature discussed in greater detail in the Dentition section. Mandible.—The coronoid process is triangular and vertical in Geocapromys thoracatus and G. ingrahami, whereas in G. brownii the tip of the coronoid is curved posteriorly. The masseteric crest is broad and rounded laterally in G. thoracatus, but is narrower in G. brownii and G. ingrahami. The pterygoid shelf of the angular process is also broader in G. thoracatus. The articular surface of the condyloid process is anteroposteriorly elongate in G. brownii and G. ingra- hami, but is nearly circular in G. thoracatus. A postcondyloid process is well developed in G. brownii, somewhat smaller in G. ingrahami, and reduced in G. thoracatus. As in the upper cheek teeth, the lower teeth in G. thoracatus are shorter and narrower than in the other species and consequently, the alveolar and occlusal lengths of the tooth rows are shorter. The mandibular tooth rows are closer in G. ingrahami than in the other two species, but are not as convergent Se Fig. 2. Ventral views of cranium (A-C) and occlusal outlines of maxillary cheek teeth (D-F) of Geocapromys species. A, D, G. brownii, MCZ 11040, Jamaica; B, E, G. thoracatus, AMNH 34547, Little Swan Island; C, F, G. ingrahami, MCZ 29427, East Plana Cay, Bahamas. Photographs of crania (A-C) are natural size, occlusal outlines of cheek teeth (D-F) are 4 natural size. 38 mm PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON VOLUME 98, NUMBER 1 39 as are their upper counterparts. Geocapromys ingrahami has a relatively shorter diastema than do the other two species. Dentition.—In the cheek teeth of Geocapromys, cement is present on that por- tion of the crown not enclosed by the continuous enamel band, except on the anterior margin of P* and the posterior margin of M,;. However, the cement is reduced on all cheek teeth of G. thoracatus in comparison to its congeners. In particular, the cement in G. thoracatus is thin on the anterior edge of P, and absent from the anterolingual edges of M,-M; and the posterolabial edges of P*- M2. The edges of the labial lophs and lingual lophids on the upper and lower molars respectively, are more prominent in G. thoracatus because the flexi and flexids are not filled entirely to their margins with cement as they are in G. brownii and G. ingrahami. The thinner layer of cement on the anterior and posterior margins of the cheek teeth in G. thoracatus almost certainly accounts for the shorter tooth row lengths observed in this species. The presence of a small anterolingual re-entrant (anteroflexid) on P, is char- acteristic of all Geocapromys. This may be a primitive character, as an anteroflexid or anterofossetid on P, is found in echimyids, the presumed sister group of the capromyids. However, the presence of an incipient anteroflexid in some specimens of Capromys pilorides indicates that the anteroflexid of Geocapromys may be secondarily derived from the condition found in Capromys and hence, not strictly homologous with the flexid located in the same position in echimyids. Whether the anteroflexid on P, is a primitive character or a neomorph, its morphology differs among the species of Geocapromys and is useful in distinguishing them. In G. brownii, the anteroflexid is well developed, extending to the midline of P, in most specimens and separating the anterolophid into two parts. The medial lingual re-entrant (mesoflexid) nearly contacts the hypoflexid. The posterolingual re-entrant (metaflexid) contacts or nearly contacts the posterior enamel band of the hypoflexid. The anteroflexid is well developed in G. thoracatus, but it does not extend to the midline of P,, nor does it appear to separate the anterolophid into two distinct lophids. The mesoflexid and metaflexid are as in G. brownii, except that the metaflexid never distorts the posterior margin of the hypoflexid as it does in some specimens of G. brownii. The anteroflexid is invariably present in G. ingrahami but it is very small compared to that in the other two species. The mesoflexid of G. ingrahami is broad and shallow. The metaflexid extends farther lingually than the mesoflexid, but does not contact the hypoflexid. The differences in the upper dentition between the species of Geocapromys are not as pronounced as in the lowers. The paraflexus and hypoflexus of all upper cheek teeth in G. brownii and G. thoracatus are in contact along the longitudinal midline, effectively isolating the protoloph. This is also true of the upper molars in G. ingrahami, but the paraflexus and hypoflexus are not in contact on P%. Color of the incisors has been used in the past to distinguish species of Geo- capromys. This character must be used with care because a change in incisor pam Fig. 3. Lateral (A—C) and occlusal (D—F) views of left mandible and occlusal outlines of mandibular cheek teeth (G-I) of Geocapromys species. A, D, G, G. brownii, MCZ 11040, Jamaica; B, E, H, G. thoracatus, AMNH 34547, Little Swan Island; C, F, I, G. ingrahami, MCZ 29431, East Plana Cay, Bahamas. Photographs of mandibles (A—F) are natural size, occlusal outlines of cheek teeth (G—I) are 4x natural size. 40 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON coloration can result from certain preparation techniques or from long submersion in preservative fluids. Still, there is a consistent difference in incisor color between G. brownii, in which the incisors are almost always bright yellowish-orange and G. thoracatus and G. ingrahami in which the incisors are pale yellow or cream- colored. Relationships of the Recent species of Geocapromys.—The three Recent species of Geocapromys can be readily distinguished from one another on the basis of external, cranial, and dental features. In addition, there are four or five extinct species in the genus, several of which are as distinctive as the Recent species. Three extinct species of Geocapromys have been described from Cuba, an un- described species is known from cave deposits in the Cayman Islands, and fossils of an undetermined number of forms have been recovered from nine of the Bahama Islands. Because of the large number of Geocapromys fossils known, I am postponing a detailed discussion of the intrageneric relationships to a future paper reviewing the extinct species. Although the evolutionary history of Geo- capromys is incomplete without inclusion of the fossils, it seems appropriate to discuss several characters that separate the Recent species into distinct lineages or species-groups, as these will prove useful when the fossil taxa are considered. Before discussing the relationships within Geocapromys, it is necessary to review briefly the broader affinities of the genus in order to establish my criteria for the determination of primitive and derived character states. The Echimyidae and Capromyidae are closely related based on several shared derived characters: the presence of a lateral process of the supraoccipital, the presence of a lateral jugal fossa, and the retention of dP{. These and other cranial and dental features suggest strongly that echimyids are the mainland group from which capromyids were derived. Compared to echimyids, capromyids are derived in the possession of high-crowned, rootless, and evergrowing cheek teeth; the presence of cement on the tooth crowns; the absence of a metaloph/metalophid on all cheek teeth; and prominent paraoccipital processes that stand apart from the bullae. Capromys resembles echimyids and differs from Geocapromys in the possession of a long, well-furred tail; thin, anteriorly oriented superior zygomatic root of the maxilla; narrow jugal; relatively large orbit; and more highly arched, nonprocumbent in- cisors. Based on the widespread occurrence of these characters in the Echimyidae and other capromyids, they are considered primitive in Capromys. In the following analysis, the presence of a particular character in echimyids and Capromys would be the basis for regarding that structure as primitive in Geocapromys. Similarly, a character present in one or more species of Geocapromys, but not in Capromys or echimyids, would be considered derived. Geocapromys browniiis derived relative to its congeners, Capromys, and echim- yids in several cranial features, including the inflated frontal sinuses, the posterior constriction of the frontals, the well developed sagittal crest, and the anteriorly broadened zygomatic arches. Derived features of G. thoracatus are the dorsally sloping palate, the vertical constriction of the internal nares, the posterior palatal spine, and the relatively small teeth. Geocapromys brownii and G. thoracatus share a unique complex of derived characters in the zygomatic arch region. These include the broad, posteriorly oriented superior zygomatic root, deeper jugal, smaller orbit, and ventral rotation of the entire zygomatic arch. Based on these derived features, G. brownii and G. thoracatus appear to represent a lineage distinct from other Geocapromys species and are here designated the brownii VOLUME 98, NUMBER 1 41 species-group. The close relationship of these two forms is not surprising consid- ering that most recent workers have regarded them as conspecific. Geocapromys ingrahami is distinguished from G. brownii and G. thoracatus by its shortened, inflated braincase, constricted pterygoid region, incipiently con- vergent upper tooth rows, lateral constriction of the internal nares, and inflated auditory bullae. Within the Capromyidae these characters are all derived. Cap- romys nanus also has a shortened, swollen braincase and inflated bullae, but is clearly not closely related to G. ingrahami and appears to have developed these features independently. The extinct taxa of Geocapromys from Cuba, the Baha- mas, and the Cayman Islands possess most of the derived characters present in G. ingrahami, although the expression of these characters varies significantly between species. My preliminary analysis of these extinct forms indicates that they are most closely related to G. ingrahami and are here grouped with the latter in the ingrahami species-group of Geocapromys species. Extinction of the Swan Island Hutia.— From the time of its discovery in 1887 by Charles Townsend, the naturalist aboard the U.S. Fish Commission Steamer Albatross, until its extinction less than a century later, Geocapromys thoracatus was known to occur only on Little Swan Island. Even though Great Swan Island is less than 0.5 km west of Little Swan, is larger, and supports a more luxuriant vegetation, there is no evidence that the hutia ever occurred there. This is anal- ogous to the situation in the Bahamas where G. ingrahami lives in large numbers on East Plana Cay, but has never been found on West Plana Cay, an island of similar size only 3 km to the west (J. A. Allen 1891; Clough 1972). Clough and Fulk (1971) do note that there is very little exposed limestone on West Plana Cay compared to East Plana Cay. The coral rock on West Plana Cay is covered by soil and supports a thicker, higher vegetation than does the eastern island. Most of the limestone in the central portion of Great Swan Island is also soil-covered, whereas Little Swan is almost totally devoid of soil. No species of Geocapromys is known to dig its own burrows; instead they live in natural cavities, especially small caves and solution holes in limestone karst areas. It seems probable that Great Swan Island and West Plana Cay are unsuitable for hutias because these islands lack large areas of exposed limestone and the myriad of caves and solution holes which are the inevitable result of the chemical weathering of a limestone terrain in the West Indies. This hypothesis has some corroboration in the distri- bution of G. brownii in Jamaica, where it occurs primarily in areas where there are extensive limestone outcrops. Naturalists in the early part of this century found hutias to be extremely abun- dant on Little Swan Island. Lowe (1911:114) “‘. . . saw at least a dozen others [in addition to the two he had already captured] running about and bolting into the big crevasses with which the island is seamed.”’ George Nelson collected 15 Swan Island Hutias for the Museum of Comparative Zoology in March and April of 1912. Even Lord Moyne, who was the last person to collect Geocapromys thorac- atus, found hutias so abundant on Little Swan in 1937 that (Moyne 1938:82) “... four men from the western island with neither nets nor traps caught twelve alive for us in about two hours.” Based on this evidence G. thoracatus must have become extinct extremely rapidly, as it was last seen alive sometime in the early 1950’s. The Swan Islands suffered a devastating hurricane in 1955, and in the late 1950’s or early 1960’s a box of unwanted cats was released on Little Swan (Clough 1976). The combination of these two events undoubtedly resulted in the 42 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON CARIBBEAN SEA of t= Cayman Islands Soiree Swan Islands ; = Jamaica Middle America Little 5.85 mm). Male anterior protarsal claw a little contorted and foliate. Metafemur of male without fringe of long setae on posterior margin ................ deltoides (Fall) Elytra more coarsely punctate; moderately attenuated posteriorly, more strongly so in male. Size smaller (<5.85 mm). Male protarsal claw distinctly contorted and foliate (Figs. S—7). Metafemur with or without fringe of setae. Color light to dark reddish brown ................. 4 Pronotum piceous, darker than head or elytral base. Male metafemur without fringe of long setae along posterior margin. Aedeagal apex renlexed- broad lyatniangulamatilp sen eeeeee eae aeee latens (Fall) Pronotum reddish brown at least discally, little or not darker than head or base of elytra. Metafemur of male with fringe of long setae on posterior marin iscerwas iseeeteS. we) oleae ween ee 5 Male anterior protarsal claw expanded; apex acute (Fig. 5) ......... ERS A Ses ey Rg ops oh Ti te Sg a nei Seber bel vn difformis (LeConte) Male anterior protarsal claw truncate; with either slender anterolateral process (Fig. 6) or long robust anterolateral process (Fig. 7) ........ 6 Male anterior protarsal claw with extremely elongate robust lateral process (Fig. 7). Tip of aedeagus distinctly bent; bent portion long in lateral vacws( Pigs ean Sar yee schusteri, new species Anterior protarsal claw broadly truncate; with short lateral process (Fig. 6). Tip of aedeagus feebly bent; bent portion short in lateral view (Figs Ti, oes. Fos: SEL Ce ee etnieri, new species Acknowledgments Curators from the following institutions have generously lent specimens for this study: National Museum of Natural History, Smithsonian Institution (NMNH); VOLUME 98, NUMBER 1 71 California Academy of Sciences (CAS); Cornell University (CU); American Mu- seum of Natural History (AMNH); Old Dominion University (ODU); Museum of Comparative Zoology at Harvard (MCZ); New York State Museum (NYSM); and the University of Minnesota (UM). Our thanks are extended to the curators of those collections. We acknowledge and thank Dr. Guenter A. Schuster, Dr. J. F. Matta, and Ms. Belinda S. Wunderlin for their assistance with collecting some of the specimens included in this study. We also thank Mrs. Susann Braden, Smithsonian Institution scanning electron microscopist, for the micrographs and Michael Druckenbrod, former Smithsonian Institution biological illustrator, for preparing the line drawings of the aedeagi (Figs. 8-16). Literature Cited Des Gozis, M. 1914. Tableaux de détermination des Dytiscides, Notérides, Hyphydrides, Hygro- biides, et Haliplides de la fauna Franco-Rhénane.— Miscellanea Entomologica 21:81-128. Fall, H. C. 1923. A revision of the North American species of Hydroporus and Agaporus.—John D. Sherman, Mt. Vernon, New York, 129 pp. . 1937. A new Agaporus (Dytiscidae-Coleoptera).— Entomological News 48(1):10-12. Leech, H. B. 1940. Description of a new species of Laccornis, with a key to the Nearctic species (Coleoptera, Dytiscidae).— Canadian Entomologist 72(6):122-128. LeConte, J. L. 1855. Analytical table of the species of Hydroporus found in the United States with descriptions of new species.— Proceedings Academy of Natural Sciences, Philadelphia 7:290- 299. Malcolm, S. E. 1971. The water beetles of Maine: including the families Gyrinidae, Haliplidae, Dytiscidae, Noteridae, and Hydrophilidae.— University of Maine Technical Bulletin 48:1—49. Spangler, P. J..and R. D. Gordon. 1973. Descriptions of the larvae of some predaceous water beetles (Coleoptera: Dytiscidae).— Proceedings of the Biological Society of Washington 86(22):261- 278. (GWW) Department of Entomology and Economic Zoology, Cook College, P.O. Box 231, New Jersey Agricultural Experiment Station, Rutgers University, New Brunswick, New Jersey 08903; (PJS) Department of Entomology, National Mu- seum of Natural History, Smithsonian Institution, Washington, D.C. 20560. Addendum Since submission of this paper for publication, specimens of Laccornis oblongus Stephens were discovered from Canada with the following locality information: Tununuk, N.W.T., 10 Aug 1930, O. Bryant/Mackenzie River 1930 Trip, Lot 114 O. Bryant. This is the first discovery of a Palearctic species of Laccornis in the Nearctic region; as far as we know, L. oblongus is the only species of Laccornis with a circumboreal distribution. Male specimens of L. oblongus can be included in the above key by making couplet 1 trichotomus by inserting the following addition at the beginning of couplet 1; the rest of couplet 1 remains unchanged. 1. Male anterior protarsal claw unmodified (i.e. neither toothed nor broadened and contorted); an- tennal segments 3-7 not broadened ......................e cece ceeeeeees oblongus Stephens The aedeagus of L. oblongus is illustrated in Leech (1940). PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 72-80 TWO NEW SPECIES OF PROTURA (INSECTA) FROM NORTH AMERICA Ernest C. Bernard Abstract.—Eosentomon adakense, n. sp. (Eosentomidae) from the Aleutian Is- lands and Neocondeellum americanum, n. sp. (Protentomidae) from Tennessee are described and figured. Eosentomon adakense differs from other Eosentomon species by the possession of long middle and hind empodia, sensillum b'1, labral setae, and six setae on sternites IX and X. A key to Eosentomon species with long hind empodia is given. Neocondeellum americanum differs from all other Neo- condeellum species by the possession of two minute sensilla on the foretarsus in addition to the normal five. This paper contains the descriptions of two species of Protura, one from Adak Island in the Aleutian chain and the other from eastern Tennessee. The new species of Neocondeellum Tuxen and Yin described in this paper represents the first record of the genus from North America. Methods Specimens were collected by a sugar flotation-centrifugation method (Jenkins 1964) and preserved in 95% ethanol, then mounted in polyvinyl alcohol-lacto- phenol and dried in a 60°C oven to expand and clear the bodies. Terminology used in this paper is largely that of Tuxen (1964). Holotypes and paratypes num- bered in the text are deposited in the U.S. National Museum of Natural History. Eosentomon adakense, new species Figs. 1, 3-15; Table 1 Color and dimensions. — Body fragile, sclerites yellowish only on the last several abdominal segments. Length of holotype female 836 um, that of allotype male 759 wm. Length of foretarsus without claw 82 wm in female, 80 um in male. Length of head without labrum 109 um in female, 116 um in male. LR = 12, PR = 9.6 (range = 9.1-10.3); TR = 5.7 (4.9-6.4). Morphology.—Pseudoculus broadly oval with a weak longitudinal line and a pair of weak, oval depressions in the center (Fig. 11). Clypeal apodeme small but stout, the sides connected only anteriorly. Labrum (Fig. 5) short and inconspic- uous, the anterior edge shallowly V-shaped; labral setae present, not extending past anterior edge of labrum. Rostral setae not inflated. Mandible (Fig. 9) with three teeth, two of them terminal and inner tooth subterminal. Galea of maxilla (Fig. 7) broad, with two exterior spines, more anterior one long, rounded median digit bent outward, and two inner digits bent slightly inward. Outer lobe of lacinia short, slightly curved, inner edge smooth; inner lobe slender, strongly curved (Fig. 8). Six outer setae of labium in three groups: one anterior, three in middle, two posteriorly (Fig. 6); anterior tip of labium with few irregular, minute teeth and three weak thickenings. VOLUME 98, NUMBER 1 Table 1.—Chaetotaxy of Eosentomon adakense n. sp. Formula Composition (Dorsal) Thorax I 4 II 8 al, 2,3,4 14 jo Il, 1", A, Ao Bs 3, @! III 10 al, 2, 3, 4, 4’ 14 jo Jl, I DB, 3 a! Abdomen I 4 al,2 10 jo ll, 1, 2 3B, 3 II-IV 10 al, 2,3,4,5 16 p 1, 1’, 2, 2’, 3, 4, 4’, 5 V-VI 8 al,2,4,5 16 p li; 1,252", 3;4,4%,,5 VII 6 a2,4,5 16 p 1, 1’, 2, 2’, 3, 4, 4’, 5 VIII 4 al,3 9 OG, tl, 1”, 4s IX-XxI 8 p 1, 2, 3,4 XII 9 (Ventral) Thorax I-II 8 6 Ill 10 8 Abdomen I 4 a2,3 4 p1,2 I-III mo) a2,3,4 4 p1,2 IV-VII 6 a2, 3,4 10 jp) Il, A, BW 2, 8 VIII 2 a2 7 jG i, WY, 2 IX-X 6 jo Hl, "5 2 XI 8 p 1, 2, 3,4 XII 10 Empodium of foretarsus nearly as long as claw, EUI = 0.98 (0.9-1.1). Empodia of middle and hind legs well-developed (Figs. 12-13), EUII = 0.59 (0.54—- 0.63), EUIII = 0.58 (0.58-9.59). Central lobe of praecosta not indented. Squama genitalis of female (Fig. 15) weakly sclerotized, caput processus nearly straight distally but curving proximally into corpus processus. Median sclerotizations not seen. Filum processus longer than stylus; stylus tip broadly rounded. Male squama genitalis identical to that of other Eosentomon species. Chaetotaxy.—Paired sensilla behind the pseudoculus minute, pyriform (Fig. 10). 74 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Figs. 1-2. Eosentomon adakense: 1, Dorsal view of holotype female. Neocondeellum americanum: 2, Dorsal view of holotype female. Scales in um. In foretarsus, sensillum 5'/ present; sensillum a short and slender, not reaching seta y2; b reaching the base of G6, c reaching y3; d broad and long; //, e, and g spatulate, f2 short. Sensillum a’ broad and long; b’2 about as long as d and twice the length of b'/; c' present, reaching to the base of seta 66. Sensillum ¢/ closer to a3 than to a3’; BS = 0.86; t2 slender, t3 clavate (Figs. 3—4). Foretarsal pits located level with seta a2 and slightly posterior to seta y. Thoracic and abdominal setal compositions as given in Table 1. Accessory setae pl’ and p2’ of tergites II-VI nearly twice the length of pl and p2 (Fig. 14). Collection data.— Holotype female (Type #101410) and allotype male extracted from soil collected by M. P. Williams at Kuluk Bay, Adak Island, Aleutian Islands, Alaska, 13 Aug 1978. Diagnosis.—Eosentomon adakense can be separated from most other Eosen- tomon species by the presence of long hind empodia. From the fifteen previously described species with long hind empodia, FE. adakense can be differentiated by the following characteristics: sensillum b’1 present, empodium of middle leg long, labral setae present, sternites IX and X with six setae. The following key will serve to differentiate those species of Eosentomon which have long hind empodia. The key is based primarily on original and revised descriptions from several sources. VOLUME 98, NUMBER 1 75 Figs. 3-11. Eosentomon adakense: 3, Foretarsus, dorsal view; 4, Foretarsus, ventral view; 5, Labrum, dorsal view; 6, Right prelabium, ventral view (palpus not drawn); 7, Left galea of maxilla; 8, Left lacinia of maxilla; 9, Left mandible; 10, Pseudoculus; 11, Right side of head, dorsal view, and enlargement of cephalic sensillum. (50 um scale applies to Fig. 10, 20 um scale to all others.) 76 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 2.—Chaetotaxy of Neocondeellum americanum n. sp. Formula Composition (Dorsal) Thorax I 4 I-III mG al, 2,3 14 p 1, 1’, 2, 2’, 3, 4, 4’ Abdomen I 4 al,2 14 p 1, 1’, 2, 2’, 3, 4, 4’ II-VI 4 al,2 14} p 1, 2, 2’, 3, 4, 4’, 5 VII 4 al,5 18 pls 1’, 2,.2', 3: 3% 45455 VIII _@ al, 3,5 i) p 1, 1’, 2, 3, 4,5 IX 12 DI, WA 5 35! x 10 p 1, 2, 3, 4,5 XI 8 p 1, 2, 3,4 XII 9 (Ventral) Thorax I 10 II 12 Ill 14 Abdomen I 4 al,2 4 pl, 1’ II-III a alee 5 DiGwle 2 IV-V 4 fl il, 7 8 jm il, 1, 3 VI-VII 4 al,2 9 me i, I, 453 VIII 6 jo) Hl, 1 2 IX-X 4 jo) Wl, 2 XI 6 XII 6 1 Seta p2 missing on terg. 5, right side, of holotype. ? Holotype with an a c seta on stern. III. Is waSensillumebalipresent? soe. -5. el ees Sees ae 2 6 - Sensillumeb ivabsentuccts eM Os ee ee oe eee 2 DD, Sensillumptiscloserm toysetara3) thanitolaS see eee eee 3 ~ Sensillumitivery closeitosctajas see eee ae ee eee 5 3.) sternite Vill with tworanteriomsctace sae ee een ee eee 4 - Stemnite VIII without anterior'setae >....).............: cocqueti Conde 4.) sStermites PDX—X with smmsetaey- 54. 6 sane eens brevicorpusculum Yin - Stemitess Xo Xewathi founmseta ca eee ee sociale Bernard 5. | Abdominal tergites V and VI with ten anterior setae; sensillum c’ well- developed: seo. 2. Wl ee a ee notiale Tuxen & Imadate I Abdominal tergites V and VI with eight anterior setae; sensillum c’ very shorty Mam miiforn sear ee eae Sn ee eae erwini Copeland VOLUME 98, NUMBER 1 Wy CHE Sterinite Vv lllewithie7, P=SClLAG: 6. ts bb hee ns hn wie tention ee 7 - Stermite VIIE with 9 p-setae .........20-.08s.8-.- eu: novemchaetum Yin REE SCHIST VS ORESSING ea os cc daoys etic oe sie oa bee aemme er dechiia tegen aetna Nees 9 - SensillmamcraOSen tae ey ean iee. vc-cee. called yee ie ak Me cttw ol had dae woe 8 8. Filum processus more than two-thirds length of stylus ...... meihwa Yin - Filum processus less than half length of stylus ............ babai Imadate 9. Empodium on middle leg more than half length of claw ............ 10 - Empodium on middle leg less than half length of claw ............. 11 10. Labral setae present; sternites IX and X with six setae .... adakense, n. sp. - Labral setae absent; sternites [IX and X with four setae ..saharense Conde InleuelbabraleSctde presente. 5 siho sr. oe Se) sk ele ole AS 6 pbs ou 12 BIE TAIESELAC ADSCINE Oe fie puts st tek | heme oe le 2 15 IP eee Sterner VillinwithsewOranterion setae 9s. sn.. sss sa. s4s hss sea. 13 - Sternite VIII without anterior setae ................... pallidum Ewing 13. Tergite VII with six anterior setae; tergite XI with six setae ......... 5: 0:60 0 Bibs! 0 SagheNe Siete Gaeta ea oe meee ere Pn ee ea ae brassicae Bernard - Tergite VII with eight anterior setae; tergite XI with eight setae ..... 14 14. Corpus processus strongly sclerotized, slender ............ pruni Bernard - Corpus processus weakly sclerotized, broad .... pinusbanksianae Bernard 15. Sensillum fl twice as long as f2; sternites IX—X with six setae ....... 5.0 Sey See ERIC ATR NES en BIC. ate ar ie nae ete ty Mem Ae udagawai Imadate - Sensilla f1 and f2 of equal length; sternites IX—X with four setae .... 2 0-6 020 BRA ARERR Reinet Oy OR Seige taint eh Al Mea NGA Deure G) ce Aol simile Conde Imadate (1974) examined several Chinese Eosentomon species described by Yin (1965) and determined that some possessed long hind empodia. These are included in the key above, with the exception of E. orientalis Yin. Imadate could find no differences between E. orientalis and E. udagawai Imadate, 1961, and was inclined to consider them synonyms. Thus E. orientalis is not included in the key. Bernard (1975) described E. pinusbanksianum from Michigan. Dr. George C. Steyskal (in litt., 29 Oct 1976) pointed out that the specific name was incorrectly formed and needed alteration. It is therefore emended to E. pinusbanksianae. Neocondeellum americanum, new species Figs. 2, 16-27; Table 2 Color and dimensions. — Body slender, sclerotization prominent only on the last five abdominal segments. Length of holotype female 787 um, that of two paratype females 784 wm (range = 771-798 um). Length of foretarsus without claw 42 um for all females; length of head without labrum 83 um for females. LR = 19; PR = 10.0, TR = 3.0. Morphology.—Pseudoculus oval with posterior prolongation about one-third its length, median line very weak (Fig. 17). Labrum very short, rounded (Figs. 16-17), with one pair of setae. Mandible with 3 minute, rounded teeth, 2 terminal and 1 subterminal, and prominent median groove (Fig. 19); lobes of lacinia straight and stout anteriorly, bluntly pointed; galea broad, thickened only along outer edge, with two diverging, pointed lobes terminally and third, minute, pointed lobe on exterior edge. Maxillary palpus with only 1 sensillum-like seta. Labial 78 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON O 34 24 Figs. 12-24. Eosentomon adakense: 12, Middle tarsus; 13, Hind tarsus; 14, Tergite VI, left side, 15, Female squama genitalis. Neocondeellum americanum: 16, Head, dorsal view; 17, Pseudoculus; 18, Labrum; 19, Mouthparts of left side (left to right: maxillary palpus, galea, lacinia, mandible); 20, Canal of maxillary gland, proximal portion; 21, Labial palpus; 22, Foretarsus, exterior view; 23, Foretarsus, interior view; 24, Arrangements of minute foretarsal sensilla on a paratype. (50 um scale applies to Fig. 16, 20 um scale to all others.) VOLUME 98, NUMBER 1 719 Figs. 25-27. Neocondeellum americanum: 25, Appendages of first and third abdominal segments; 26, Tergites VIII-IX, left side; 27, Female squama genitalis. palpus with terminal tuft of seta-like lobes and 5 other setae (Fig. 21). Canal of maxillary gland similar to that of other members of genus, posterior portion swollen for more than half its length then narrowed to slightly enlarged terminus; calyx round (Fig. 20). Empodium of foretarsus short, about one-sixth length of claw; middle and hind legs similar to those of other Neocondeellum species. Appendages of first abdom- inal segment with 4 setae, those of second and third segments with 3 setae each (Fig. 25). Upper edge of gland cover on tergite VIII with 5 teeth; lower edge smooth except for 1 tooth on interior corner (Fig. 26). Female squama genitalis with enlarged arms on basal apodeme; acrostylus conoid, with narrowly rounded tip and | or 2 indentations on interior surface (Fig. 27). Males not seen. Chaetotaxy. —Head with 4 setae between pseudoculi (Fig. 16). Foretarsus characterized by presence of 2 minute sensilla ventrally, one ap- parently reduced, modified 63, the other a doubled 64 and thus labeled 64’ (Figs. 23-24); these 2 sensilla appearing bent in profile on holotype (Fig. 23), but ap- pearing peg-like when seen less obliquely on paratype (Fig. 24). Sensilla a, b, and t3 of equal length and longer than a’ and f (Figs. 23-24). Thoracic and abdominal compositions as given in Table 2. Tergite VIII with 3 anterior and 6 posterior setae on each side; tergite [IX with 6 setae on each side (Fig. 26). Collection data.—Holotype female (Type #101411) and one paratype female extracted from soil collected under turf and yellowwood trees (Cladrastis lutea (Michx.) K. Koch) on the west side of the Plant Sciences Building, University of Tennessee, Knoxville, 24 Feb 1984; and one paratype female collected from the same locality 15 Feb 1978. 80 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Diagnosis.— Neocondeellum americanum differs from the other four species of the genus (Tuxen and Yin 1982) by the sensillar nature of seta 63 and the presence of an extra sensillum here designated 64’. Additionally, the maxillary sensilla of N. americanum consist of one typical sensillum and a setiform sensillum, a di- chotomy shared only with N. matobai (Imadate, 1973; see Tuxen and Yin 1982). Literature Cited Bernard, E. C. 1975. A new genus, six new species, and records of Protura from Michigan.—The Great Lakes Entomologist 8:157-181. Imadate, G. 1973. Contributions towards a revision of Japanese Protura.— Revue d’Ecologie et de Biologie du Sol 10:603-628. —. 1974. Fauna Japonica. Protura (Insecta).—Tokyo: Keigaku Publishing Company, Ltd. 351 pp. Jenkins, W. R. 1964. A rapid centrifugal-flotation technique for separating nematodes from soil.— Plant Disease Reporter 48:692. Tuxen, S. L. 1964. The Protura.—Hermann, Paris, 360 pp. , and Yin Wen-Ying. 1982. A revised subfamily classification of the genera of Protentomidae (Insecta: Protura) with description of a new genus and a new species. — Steenstrupia 8:229—249. Yin Wen-Ying. 1965. Studies on Chinese Protura. I. Ten species of the genus Eosentomon from Nanking-Shanghai regions.—Acta Entomologica Sinica 14:71—92. [In Chinese with English summary.] Department of Entomology and Plant Pathology, University of Tennessee, Knoxville, Tennessee 37901-1071. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 81-89 A NEW MEMBER OF THE GENUS DISTOCAMBARUS (DECAPODA: CAMBARIDAE) FROM THE SALUDA BASIN, SOUTH CAROLINA Horton H. Hobbs, Jr. and Paul H. Carlson Abstract.— The burrowing crayfish Distocambarus (Fitzcambarus) youngineri is described from two localities in the southwestern part of Newberry County, South Carolina. In most respects it resembles the sympatric D. (F.) carlsoni, but the first pleopod of the male differs little from that of its more primitive congeners assigned to the nominate subgenus. Within the Saluda River basin of South Carolina, a number of isolated colonies of crayfishes belonging to the genus Distocambarus have been found, and inas- much as the ranges of the three species represented in them are sympatric, we are puzzled that we have not found them to be at least occasionally syntopic. The species described here appears to have established fewer populations than have Distocambarus (D.) crockeri Hobbs and Carlson (1983:421) and D. (Fitzcambarus) carlsoni Hobbs (1983:430), both of which also frequent the Savannah watershed. Despite our being unable to locate additional localities for it other than the two that have been discovered in the southwestern part of Newberry County, we believe that other colonies must exist elsewhere in the Saluda Basin. Because of the combination of characteristics exhibited by this crayfish, to assign it to either of the subgenera that have been recognized requires an emendation of the existing diagnoses (Hobbs 1983:429, 430). The following are offered in support of our concept of the affinities of this crayfish that are discussed under “‘Relationships.”’ Subgenus Distocambarus Hobbs, 1981 Diagnosis.—Chela of male with mesial margin of palm distinctly longer than width of palm; carpus of cheliped weakly expanded distally; second pereiopod lacking conspicuous fringe of setae on merus; areola in animals with carapace length greater than 20 mm usually less than 15 times as long as wide; color although variable (brownish, greenish, or lavender) never with reddish or pinkish suffusion; female with annulus ventralis moving through arc of at least 50 degrees, and postannular sclerite conspicuously large, almost as long as annulus ventralis; bur- rows usually with only one or two openings to surface and consisting basically of single subvertical passageway and secondary short branch leading to surface. Comprising two species: Distocambarus (D.) crockeri Hobbs and Carlson (1983: 421) and D. (D.) devexus (Hobbs, 1981:302). Subgenus Fitzcambarus Hobbs, 1983 Diagnosis.—Chela of male with mesial margin of palm distinctly shorter than width of palm; carpus of cheliped expanded distally; second pereiopod with con- 82 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON spicuous fringe of setae extending distally from at least midlength of merus distally onto dactyl; areola in animals with carapace length greater than 19 mm seldom less than 15 times longer than wide; color variable (brownish, greenish, blue, or pinkish lavender) but usually pinkish lavender; female with annulus ventralis moving through arc of no more than 50 degrees, and postannular sclerite not conspicuously large, its length not greater than half that of annulus ventralis; except during droughts, burrows almost always with multiple openings to surface and usually with complex system of galleries. Comprising two species: Distocambarus (F.) carlsoni Hobbs (1983:430) and D. (F.) youngineri (described herein). Distocambarus (Fitzcambarus) youngineri, new species Fig. 1 Diagnosis.— Body and eyes pigmented, latter small but well developed. Rostrum without marginal spines, tubercles, and median carina. Carapace with | to several small cervical tubercles. Areola 13 to 24 (average, 17.6) times as long as broad, and constituting 37.6 to 41.9 (average, 38.9) percent of entire length of carapace (41.9 to 46.6, average 44.6, percent of postorbital carapace length). Ventral surface of ischium of third maxilliped only partly obscured by plumose setae. First 3 pairs of pereiopods without conspicuous ventral brush of setae extending from basis to merus, although second pereiopod with distoventral half of merus and margins of more distal podomeres with fringe of long setae. First pair of pereiopods with ventral surface of merus densely tuberculate, corresponding surface of proximal part of both fingers lacking tubercles. First pleopods of first form male with small but distinct cephalic shoulder at base of conspicuous, corneous, subquadrangular, platelike caudodistally and somewhat mesially directed central projection; ce- phalic process vestigial at best, not clearly defined; mesial process broad basally, tapering distally in long pointed element directed caudodistally and slightly lat- erally. Mesial ramus of uropod with distomedian spine very small but almost or quite reaching margin of ramus. Female with anterior part of annulus ventralis membranous across which hingelike motion accomplished; postannular sclerite not half so long as annulus; first pleopods rudimentary, in form of small tuber- culiform prominences. Holotypic male, form I:—Cephalothorax (Fig. la, h) subovate, compressed laterally; maximum width of carapace greater than height at caudodorsal margin of cervical groove (12.1 and 10.4 mm). Abdomen distinctly narrower than thorax (9.0 and 12.1 mm). Areola 17.2 times as long as wide but with only 1 punctation in narrowest part. Cephalic section of carapace about 1.4 times as long as areola, latter comprising 41.9 percent of total length of carapace (45.8 percent of post- orbital carapace length). Surface of carapace mostly punctate, few small tubercles in anteroventral branchiostegal region. Rostrum broader than long with margins tapering gently from base and slightly more sharply from base of poorly defined acumen to tip; latter reaching base of ultimate podomere of antennular peduncle; rostral margins not conspicuously thickened; upper surface distinctly concave and bearing, in addition to usual marginal punctations, few large, comparatively deep punctations. Subrostral ridge moderately well developed and evident in dorsal aspect to end of basal third of rostrum. Suborbital angle obtuse but distinct. VOLUME 98, NUMBER 1 83 Fig. 1. Distocambarus (F.) youngineri (all from holotype except c, e from morphotype, and g, j from allotype): a, Lateral view of carapace; b, c, Mesial view of first pleopod; d, Epistome; e, f, Lateral view of first pleopod; g, Annulus ventralis and adjacent sternites; h, Dorsal view of carapace; 1, Antennal scale; j, m, Dorsal view of distal podomeres of cheliped; k, Caudal view of first pleopods; 1, Proximal podomeres of third, fourth, and fifth pereiopods. Postorbital ridge clearly defined but merging insensibly with carapace anteriorly. Branchiostegal and cervical spines absent, latter represented by very small tu- bercle. Abdomen shorter than carapace (21.5 and 24.6 mm). Pleura of second through 84 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON fifth abdominal segments rounded ventrally, lacking caudoventral angle. Cephalic section of scabrous telson with | strong, fixed spine in each caudolateral corner. Cephalic lobe of epistome (Fig. 1d) subtriangular with short anteromedian pro- jection, margins slightly elevated ventrally; main body of epistome with shallow median depression but lacking distinct fovea; epistomal zygoma broadly arched. Ventral surface of proximal podomere of antennular peduncle lacking spine. An- tennal peduncle with usual spine on ventral surface of basal podomere reduced to minute tubercle; flagellum reaching about midlength of areola. Antennal scale (Fig. 11) 2.4 times as long as broad, widest distinctly distal to midlength; greatest width of lamellar area little more than 1.5 times that of thickened lateral part. Third maxilliped overreaching antennal peduncle by length of ultimate podo- mere; mesial sector of ventral surface of ischium with clusters of stiff, long setae; lateral sector studded with clusters of shorter, plumose setae which also forming submarginal lateral row; merus with setae similarly disposed. Right chela (Fig. 1m) subelliptical in section, strongly depressed; palm 1.4 times as broad as length of mesial margin; length of latter almost one-third that of entire length of chela; most of palm studded with squamous to subsquamous tubercles. Mesial margin of palm with row of 6 (5 on left) tubercles flanked dorsolaterally by row of 6 (4 on left). Both fingers with well defined dorsomedian ridges flanked by setiferous punctations. Opposable margin of fixed finger with row of 3 tubercles, second from base largest, along proximal two-fifths of finger and single row of minute denticles extending between and distal to tubercles from base to corneous tip of finger; prominent tubercle also present below row slightly beyond midlength of finger; lateral margin of finger costate with row of 5 deep, setiferous punctations. Opposable margin of dactyl with 2 large tubercles on proximal half beyond which single row of minute denticles extending to corneous tip of finger, and shallow, rounded excavation present along basal part of finger; mesial margin with 1 or 2 tubercles proximally, followed distally by row of setiferous punctations. Ventro- lateral surface of chela with conspicuous longitudinal row of long setae which extending from bases of tubercles in proximal part of row and from ventrolateral punctations on finger. Carpus of cheliped distinctly longer than mesial margin of palm of chela (7.3 and 4.5 mm) and bearing slightly sinuous dorsal furrow; dorsal surface of podo- mere sparsely punctate, mesial surface with upper longitudinal row of 5 small squamous tubercles and cluster of 8 larger ones below row, distalmost of cluster largest; mesial surface with few additional small tubercles, and ventrodistal margin with single large tubercle articulating with proximal ventrolateral condyle on propodus. Merus with usual tubercles dorsally, distalmost larger than others; mesial face with few small scattered tubercles present on distal half; lateral surface sparsely punctate; ventral surface tuberculate: 11 tubercles in irregular lateral row and 12 in mesial row, other tubercles and plumose setae present between rows; tubercles in both rows increasing slightly in size distally, but none spiniform. Ischium with row of 4 tubercles ventromesially, otherwise punctate. Hook on ischium of third pereiopod (Fig. 11) simple, slightly flattened, but subacute, and barely overreaching basioischial articulation, not opposed by strong tubercle on basis. Ventral membrane of coxa of fifth pereiopod conspicuously VOLUME 98, NUMBER 1 85 setose. Plumose pubescence associated with sternum and coxae of all pereiopods very prominent (not shown in illustration). First pleopods (Fig. 1b, f, k) typical of genus in being symmetrical, not contig- uous at base, reaching coxae of third pereiopods, bearing strong caudoproximal lobe and broadly rounded proximomedian lobe, flexed caudally slightly distal to midlength, and lacking subapical setae. Terminal elements described in “‘Diag- nosis.” Uropods with only mesial lobe of proximal podomere bearing acute spine, that on lateral lobe rudimentary; mesial ramus with distomedian spine small and almost or quite reaching distal margin of ramus. Allotypic female. — Differing from holotype, other than in secondary sexual fea- tures, in following respects: acumen of rostrum with base more clearly defined and reaching midlength of ultimate podomere of antennular peduncle; subrostral ridge evident in dorsal aspect to base of acumen; cephalic section of telson with 2 spines in each caudolateral corner, more mesial pair smaller and movable; median fovea present on epistome; flagellum of antennule reaching caudal margin of carapace; ischium of third maxilliped lacking tufts of plumose setae between lateral marginal row and mesial sector, latter bearing clusters of long stiff setae: mesial margin of palm of right chela (Fig. 1j) with row of 5 tubercles (6 on left) flanked dorsolaterally by row of 4; lateral margin of fixed finger with row of 7 setiferous punctations; opposable margin of dactyl with row of 3 tubercles (left with 4), distalmost smallest; mesial surface of carpus with 5 small tubercles in- terspersed among 5 larger ones; small tubercle present on ventrodistal margin of carpus opposite ventral condyle on proximal margin of propodus. (See “‘Mea- surements.”’) Segment of sternum between fourth pereiopods with very narrow median fissure (expanding anteriorly) extending entire length. Annulus ventralis (Fig. 1g) hinged between caudal sclerotized part and semi- membranous anterior region (moving through arc of some 30 to 45 degrees), approximately 1.4 times as broad as long; anterior section broadly excavate; tongue descending into centrally located fossa from high sinistral wall; sinus originating in fossa near median line, extending sinistrally and making almost U- turn from which extending caudodextrally to median line where ending on caudal face of annulus. Postannular sclerite almost three-fourths as wide as annulus ventralis, but dis- tinctly less than half as long; central area elevated (ventrally) with anterior face convex and posterior surface flattened. Morphotypic male, form IIT.—Except in secondary sexual characters, differing in no conspicuous way from holotype, few differences noted, probably reflecting juvenile condition of specimen: subrostral ridges evident in dorsal aspect from orbit almost to base of acumen; cephalic lobe of epistome lacking anteromedian projection; flagellum of antenna almost reaching caudal margin of carapace; fixed finger of chela with only 2 tubercles in row on proximal opposable margin, and other tubercles on chela not arranged precisely as in holotype, but possessing same numbers on either right or left members; ventral surface of merus of cheliped with lateral row of 8 tubercles and mesial row of 10. Hook on ischium of third pereiopod tuberculiform, not projecting nearly so far 86 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON proximally as that in holotype, and plumose setae on sternum and coxae of pereiopods much shorter, not obscuring distal part of first pleopods; latter (Fig. lc, e) with oblique juvenile suture in proximal half of appendage, shoulder on cephalic surface situated slightly more proximally than in holotype; both terminal elements heavier and less attenuate, non-corneous central projection not nearly so lamelliform, and not even rudiment of cephalic process discernible. Color notes.—Holotypic male with carapace mostly pale bluish lavender fading ventrally to pinkish cream; antennal, mandibular, and most of hepatic and an- teroventral branchiostegal areas also pinkish cream. Caudodorsal ridge almost white. First through fifth abdominal terga brownish red suffused with lavender, fading caudally; tergum of sixth segment, telson, and uropods translucent, that of sixth segment with faint pink tint. Pleura translucent with pinkish cream blush. Dorsum of cheliped bluish lavender from midlength of merus distally, color intensified at distal end of merus and carpus and along mesial margin of palm and proximomesial part of dactyl; ventrolateral part of propodus and ventral and ventrolateral parts of all podomeres pinkish cream. Remaining pereiopods pinkish cream with lavender suffusion, latter intensified on distal half of merus and on proximal half of carpus. Eyestalks, antennules and antennae uniformly pale bluish lavender; third maxilliped and ventral part of body pinkish cream. Paratypic male, form I, with carapace distinctly reddish tan, fading to pink laterally, dorsolateral parts of rostrum and posterior flank of cervical groove darker than elsewhere. Abdomen as in holotype except tergum of sixth segment and marginal areas of pleura, uropods, and telson pale pinkish orange. Antennules and antennae reddish, only little paler than carapace, dorsodistal parts of podo- meres of peduncles darker than elsewhere. Chelipeds pinkish orange with brownish triangular mark on dorsodistal part of merus, similarly colored marks flanking groove on dorsal surface of carpus becoming very dark where merging distally. Chela with dorsal dark band across distal part of palm (ridge at base of dactyl almost black) and proximodorsal parts of both fingers dark, becoming lighter toward ends; dactyl darker than propodus; tubercles on mesial part of palm very dark. Ventral surface of chela deep pink. Other pereiopods pinkish orange with dark suffusion on distal part of merus and carpus, distal margin of merus especially dark. Third maxilliped and venter pinkish to purplish cream. Females almost concolorous: lavender to lavender pink with dark markings on dorsum of chelipeds from merus distally. Measurements (mm) Holotype Allotype Morphotype Carapace Height 10.4 12.2 8.5 Width 12.1 15.0 9.7 Entire length 24.6 30.6 20.1 Postorbital length DDS) 26.9 17.6 Areola Width 0.6 0.9 0.5 Length 10.3 ED 8.0 Rostrum Width 4.7 52 3.4 Length 4.3 5.0 3.0 VOLUME 98, NUMBER 1 87 Chela Length of mesial 4.5 5.5 2.8 margin of palm Width of palm 6.2 7.4 4.0 Length of lateral 13.2 16.5 8.7 margin Length of dactyl 8.5 10.1 4.5 Carpus of cheliped Width 4.3 5.4 3.1 Length 7.3 9.0 5.0 Abdomen Width j 9.0 11.1 UD Length 21.5 25.0 15.8 Types.— The holotypic male, form I, allotypic female, and morphotypic male, form II (numbers 208413, 208414, and 208415, respectively) are deposited in the National Museum of Natural History, Smithsonian Institution, as are the following paratypes: 4 éI, 2 II, 17 2, 9j 6, 14) 2, 3 2 with young. Type-locality.— Burrows in area adjacent to small woodland pool, about 50 m southwest of State Route 58 on Route 22 (11 airmiles due west of Newberry, Newberry County, South Carolina) (34°18'N, 81°48’W). There in a wooded area supporting a dense growth of Pinus, Quercus, and Nyssa, the chimneys constructed by members of Distocambarus (F.) youngineri mark the complex systems of galleries that have been excavated in the sandy clay soil. Range.—Even though intensive searches have been made to discover other colonies of this crayfish, until now it has been found in only one locality other than that cited above: roadside ditch on State Highway 121, 0.15 mile northeast of its junction with State Highway 34, just east of Silverstreet, and only nine airmiles from the type-locality. Specimens examined.— Thirty-one specimens have been collected from the type- locality as follows, 3 2, 1j 6, 2) 2, 17 Dec 1982, PHC, collector; 2 6I, 7 2, 2j 4, 3j 9, 2 2 with yng, 6 Mar 1983, PHC; 3 2, 1j 2, 22 Jun 1983, PHC; 2 @, 3j 6, 29 Oct 1983, PHC, GBH, HHH. From the other locality just east of Silverstreet, the following are available: 1 éII, 1J 2, 11:Nov 1983, PHC; 3 4I, 1 4II, 3 2, 3j 4, 7j 9, 1 2 with yng, 18 Feb 1984, PHC. Variations.— Most of the variations noted in this crayfish are indeed minor ones. The ratio of the length of the carapace to that of the areola in specimens from the type-locality ranges from 37.1 to 40.5 (average 38.9) whereas that in the other locality is 37.8 to 40.8 (average 39.1), the range of the corresponding ratios of the areola length to postorbital carapace length 42.2 to 46.1 (average 44.4), and 41.9 to 46.4 (average 45.1), and that of the corresponding ratios of the length to the width of the areola 13.0 to 24.0 (average 18.1) and 10.0 to 18.0 (average 16.4 mm), thus even though the areolae of specimens from the type-locality are slightly shorter and narrower than that found in the other locality, specimens from the two places cannot be separated with certainty. The rostrum exhibits perhaps the most conspicuous variations: it ranges from subtriangular to broadly subovate, and the acumen may or may not be rather clearly delimited basally; in all specimens it is somewhat deflected ventrally, and in one very strongly so; the apex of the acumen may not reach the base of the penultimate podomere of the antennular peduncle, or it may overreach it. The mesial surface of the chela bears 88 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON a row of 5 to 7 tubercles, and the row immediately dorsolateral to it consists of 4 to 6. The opposable margin of the fixed finger bears a row of 2 to 4 tubercles in addition to the more distoventrally situated tubercle, and the dactyl also exhibits a row of 2 to 4 of them. Size.—The largest specimen available is a female, which has a carapace length of 31.7 (postorbital carapace length 27.4) mm. The corresponding length of the smallest and largest first form males available are 12.1 (10.4) mm and 29.2 (25.6) mm. The smallest female carrying young (no ovigerous females have been found) has corresponding lengths of 27.1 (23.4) mm. Life history notes.—Two first form males, each found in a burrow with a female, were collected on 6 March 1983, and three were obtained on 18 February 1984 when some of the females taken were observed to bear sperm plugs in their annuli ventrales. A first form male and a female with carapace lengths of 26.5 and 29.4 mm, respectively, were found together in a burrow on 18 February 1984. Young with carapace lengths of 6.8 to 11.8 mm were obtained from the burrow of a female on 18 February 1984, and others with corresponding lengths of 7.9 to 11.0 with two females on 6 March 1983. Ecological notes.— Distocambarus (F.) youngineri is a primary burrower, and its abode is generally marked by three or four openings, one or more being sealed by a capped turret. Irregular horizontal galleries, 30-60 mm beneath the soil surface connect the openings, and one or two vertical to subvertical shafts penetrate the groundwater, reaching a maximum depth of about 60 to 70 cm below the soil surface. One of the burrows excavated in December, 1982, and all of those dissected in June and October, 1983, were much less complex than others investigated in December, 1982, and in March, 1983. These simple burrows consisted of only one subvertical shaft which led beneath the surface of the groundwater. During drought conditions the sandy-clay soil became so hard that excavating an entire burrow would have been exceedingly difficult, but there was evidence that some of the upper galleries had been filled with soil that presumably had been removed from the wet fundus of the main vertical shaft. As a result, the burrows seemed to be comparatively simple during these periods. On March 6, 1983, one pair was found in a horizontal passageway, only 30 to 40 mm beneath the soil surface. Most other specimens were encountered deep under the water in the somewhat enlarged chambers at the bases of the subvertical passages. Juveniles appear to have constructed simple burrows emanating from that occupied by the mother. The only features that appear to distinguish most burrows occupied by first form males from those of females are their proximity to the more permanent surface water and, except for those complex burrows in which a female is also present (presumably the male is the temporary occupant), they are comparatively simple. The small woodland pool at the type-locality was carefully sampled on two occasions, and on neither was a crayfish found in it. Moreover, no openings to flooded tunnels could be located. No other crayfishes were found in the immediate areas occupied by D. (F.) youngineri. Relationships.— This crayfish has its closest affinities with Distocambarus (F.) carlsoni, sharing with it, among other features a rostrum that is shorter than wide, a comparatively narrow areola, a reduced abdomen, a cheliped in which the mesial VOLUME 98, NUMBER 1 89 margin of the palm of the chela of the male is shorter than the maximum width of the podomere and shorter than the carpus, stocky first pleopods in which the central projection is more strongly recurved than in the other two members of the nominate subgenus, an annulus ventralis that is movable through an arc of no more than 50 degrees, and a wedgelike postannular sclerite that is not con- spicuous and not nearly so long as the annulus. It differs from D. (F.) carlsoni most conspicuously in the secondary sexual features of the male and in the more strongly tuberculate ventral surface of the merus of the cheliped. The shorter, broader, subquadrangular shape of the central projection of the first pleopod of the first form male is strikingly different from the elongate bladelike element in D. (F.) carlsoni, as is the mesial process which projects caudodistally rather than caudally. In the female, the cephalic region of the annulus ventralis is broadly and comparatively deeply excavate whereas in the typical form of D. (F.) carlsoni it is seldom more than narrowly, and usually quite shallowly excavate. Had this crayfish not been discovered there would be less evidence for our conclusion that D. (F.) carlsoni should be considered a congener of D. (D.) crockeri and D. (D.) devexus (Hobbs 1981:302). Possessing a first pleopod in the male that markedly resembles that of the last two mentioned species, but having a body and chelipeds that are clearly more like those of D. (F.) carlsoni, D. (F.) youngineri seems to tie the latter even more firmly to these two species. Moreover, the discovery that sperm plugs occur in the annuli ventrales of D. carlsoni and D. youngineri, a feature that to our knowledge has never been observed in a member of the genus Cambarus, lends greater assurance that D. carlsoni is not, as the first pleopods of the male suggest, a product of the cambaroid line of evolution. With little imagination one might visualize D. youngineri as an arrested state in the evolution of D. carlsoni from a more primitive D. crockeri-like ancestor. Etymology.—This crayfish is named in honor of our mutual friend, Edward M. Younginer of the Department of Health and Environmental Control, Columbia, South Carolina, who not only has offered us encouragement throughout the course of our study of the crayfishes of South Carolina, but also has assisted us in obtaining many of the collections that are now available. Acknowledgments We are grateful to Georgia B. Hobbs for her assistance in collecting some of the specimens cited herein and to J. F. Fitzpatrick, Jr., of the University of South Alabama, and C. W. Hart, Jr., of the Smithsonian Institution, for their criticisms of the manuscript. Literature Cited Hobbs, Horton H., Jr. 1981. The crayfishes of Georgia.—Smithsonian Contributions to Zoology 318:viii + 549 pages, 262 figures. . 1983. Distocambarus (Fitzcambarus) carlsoni, a new subgenus and species of crayfish (De- capoda:Cambaridae) from South Carolina. — Proceedings of the Biological Society of Washing- ton 96(3):429—439, 1 figure. , and Paul H. Carlson. 1983. Distocambarus (Decapoda: Cambaridae) elevated to generic rank, with an account of D. crockeri, new species, from South Carolina.— Proceedings of the Biological Society of Washington 96(3):420—428, 1 figure. (HHH) Department of Invertebrate Zoology, Smithsonian Institution, Wash- ington, D.C. 20560; (PHC) Department of Health and Environmental Control, 2600 Bull Street, Columbia, South Carolina 29201. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 90-97 OMAN YPSILON, A NEW GENUS AND SPECIES OF BLENNIID FISH FROM THE INDIAN OCEAN Victor G. Springer Abstract.—Oman ypsilon is described, based on a single specimen from Sur, Oman. The taxon differs from all other Omobranchini in having the following combination of characters: dorsal-fin spines X; pectoral-fin rays 13; infraorbital bones 4; nasal bones not touching or joined to each other; interorbital pores 2; median supratemporal sensory canal pore absent; ventral hypural plate fused to complex urostylar centrum. A key is given to the seven genera of Omobranchini. The classification of the omobranchin genera was last considered by Springer (1972). Subsequently (Springer and Gomon 1975; Springer 1981; Bath 1983), several new species, but no new genera, have been described in the tribe. The present study was initiated as a result of a query from Dr. Richard Win- terbottom, who had obtained a specimen of omobranchin that he was unable to assign to a genus or species. Dr. Winterbottom kindly allowed me to study the specimen and describe it. Although the specimen is assignable to the Omobran- chini, and superficially resembles species of Omobranchus, Omox, and Paren- chelyurus, its complex of characters does not clearly place it in any of the presently recognized omobranchin genera. Before I decided to describe a new genus for the specimen, I attempted a cladistic analysis of the Omobranchini to determine if the new species might be placed reasonably in one of the three genera it resembled. I spent considerable time searching for, and coding, characters for the analysis, which was performed using the well-known WAGNER 78 computer program developed by J. S. Farris. Ul- timately, insoluble problems with coding characters, lack of enough characters to resolve relationships, and peculiarities of the computer program convinced me to abandon the results of the analysis. I wish to emphasize that my decision was not based on disagreement with the results of the analysis but on dissatisfaction with the soundness and sufficiency of the data and program on which the analysis was based. Methods.— Measurements were made to the nearest 0.1 mm using needlepoint dial calipers. Standard length (SL) was measured from the mid-tip of the snout to the mid-base of the caudal fin. Vertical-fin elements were measured from the rear axil at the base of the element to the distal tip of the element, without allowance for curvature of the element. Gill-opening depth is the distance between the dorsalmost and ventralmost points of the opening. Other measurements are self-explanatory. The holotype was cleared using the trypsin method and stained with alizarin red s and alcian blue. Other methods and terminology follow those of Springer (1972). VOLUME 98, NUMBER 1 91 Key to the Genera of Omobranchini 1. Cirri present on rims of anterior and posterior nostrils; circumorbital pores 9 to 12 (usually 10); one or more supratemporal canal pores on occiput just anterior to median supratemporal pore ......... Laiphognathus Smith — No cirri present on rims of nostrils (anterior nostril may open at end of slender tube); circumorbital pores 6 to 10 (rarely more than 9, and modally less than 9, in any species); no supratemporal canal pores on occiput anterior to median supratemporal pore, which may be absent ......... 2 Pre Mandibular ipOrese. uid cx tees ieee ish. . ieeyesss. evsed cess liiQeie Gee ates 3 Slain Gils LATE IOOKESISH CEs hee Us Ewe eeieyers eteis hus oan Lert a hereers ete Gis Starnes « 4 3. Dorsal-fin spines modally 11 or 12; pectoral-fin rays modally 13 ...... 2b a IE 52 SR aN ce dg ee SES eee a ae Ee Parenchelyurus Springer — Dorsal-fin spines modally 7 to 10; pectoral-fin rays modally 14 to 16 mes 7 Harledsrorsdtaits gpbitepes ean yoe cheba el Bes yen dvs ase Enchelyurus Peters* 4. Dorsal-fin spines 10; ventral hypural plate fused to urostylar complex; median supratemporal pore absent; interorbital pores 2; gill opening ex- tending ventrally to level opposite base of dorsalmost pectoral-fin ray .. 00. 8 6-025 0 RE te IE se aE Sees oh ISR) YS a ane en Oman Springer — Dorsal-fin spines modally 11 to 13; ventral hypural plate autogenous; median supratemporal pore present or absent; interorbital pores modally 2 to 4; gill opening varying from restricted to area dorsal to level of pectoral-fin base to extending ventrally to level opposite ventralmost pec- CORT CMIMERAN ES Cee ray toi RRR eee mets Gee RIE ee ete a 5) 5. Median supratemporal pore usually present; interorbital pores modally 2 or 3; lateral-line tubes present or absent; gill opening restricted to area dorsal to level of 7th from dorsalmost pectoral-fin ray (usually restricted to level dorsal to 3rd from dorsalmost ray); infraorbital bones 4 or 5 (5 in most species); thin, fleshy crest on top of head present or absent .... 508 20 6 SAO RT aM IR MS OS | IR SIM te Omobranchus Ehrenberg — Median supratemporal pore usually absent; interorbital pores modally 4; lateral-line tubes absent; gill opening extending ventrally at least to level of 8th from dorsalmost pectoral-fin ray; infraorbital bones 3 or 4 (usually 4) en OnleshwackeSizOnutOp Olhcadmre were 4 ese nee 6 6. Total dorsal-fin elements 30 or 31; segmented anal-fin rays 20 or 21; nasal bonesnoimedsdornsallyary. 2 weg wncts ces oe ct ee Haptogenys Springer * Smith-Vaniz (1976:154) followed Springer (1972) in reporting that the postcleithra of Enchelyurus (Omobranchini) consist of a few fragments of bone. Actually, the ventral postcleithrum of Enchelyurus is always present and complete, and appears similar to that of Phenablennius (Phenablenniini; Springer and Smith-Vaniz 1972), which completely lacks the dorsal postcleithrum. The dorsal postcleithrum of Enchelyurus is either absent or reduced to a fragment of bone that appears near the dorsal end of the ventral postcleithrum. Other than Enchelyurus and Phenablennius, only Praealticus Schultz and Chapman (Salariini) among the Blenniidae has modified the structure of the postcleithra. In Praealticus, the dorsal postcleithrum is represented by a dorsally positioned fragment of bone; another fragment of bone occurs ventral to the dorsal fragment, but it is not clear whether this represents a second fragment of the dorsal post- cleithrum or a fragment of the ventral postcleithrum (Smith-Vaniz and Springer 1971, fig. 15). 92 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON — Total dorsal-fin elements 27 to 29; segmented anal-fin rays 17 to 19; nasal bones separated for their entire length ................... Omox Springert Oman, new genus Diagnosis.—A member of the blenniid tribe Omobranchini with: ventral hy- pural plate fused to complex urostylar centrum; basibranchials 2 and 3 reduced, present only as cartilage; frontals not fused to each other; basisphenoid well developed (but lacking belophragm); no kinethmoid (Springer 1968: fig. 13); dorsal and ventral postcleithra normal; nasal bones not touching each other; no cirri on head; median supratemporal sensory canal pore absent; interorbital pores 2. Etymology.— Oman is derived from the name of the country of origin of the type species. The gender is feminine; the stem is “oman.” Type-species.—Oman ypsilon Springer. Comparisons. — The genera of Omobranchini can be distinguished in the above key. A fuller comparison is given in Table 1. Oman ypsilon, new species Fig. 1 Description (based on holotype and only known specimen; characters in generic diagnosis not repeated here).— Dorsal fin X, 25; slightly notched between spines and rays; attached by membrane to caudal-fin origin. Anal fin II, 24; attached by membrane to caudal peduncle. Vertebrae 11 + 29. Epipleural ribs on vertebrae 1 to 12; pleural ribs on vertebrae 3 to 11. Pectoral fin 13 (both sides). Pelvic fins I, 2 (left pelvic fin with splint of bone, probably representing vestigial third ray, closely applied to base of innermost segmented ray); membrane between seg- mented rays incised one-half length of innermost ray. Caudal fin with 6 dorsal and 6 ventral procurrent rays, 7 dorsal and 6 ventral segmented rays, and one epural; posteriormost dorsal procurrent ray and posteriormost ventral procurrent ray each attached to its own roundish, plate-like procurrent cartilage; anterior 5 dorsal procurrent rays attached to elongate procurrent cartilage reaching anteriorly to point dorsal to tip of neural spine of third preural vertebra (PU3); anterior 5 ventral procurrent rays attached to elongate procurrent cartilage reaching ante- riorly to point ventral to tip of hemal spine of PU3; hypural 5 absent. Branchio- stegals 6. Upper jaw and lower jaw each with recurved canine tooth posteriorly on each side and 23 incisor teeth. Infraorbital bones 4, dorsalmost very small. Ventralmost point of gill opening opposite level of dorsalmost pectoral-fin ray. Sensory canal pores (left-right): Circumorbital (=supraorbital + infraorbital) 9-8; interorbital 2 (total); mandibular 3-3; preopercular 6-5; supratemporal 4-4 (no median predorsal pore); bi-pored lateral-line tubes 0-2, reaching posteriorly to vertical from interspace between dorsal-fin spines 2 and 3 (on right side of specimen). Measurements in mm (and as percent SL): Standard length 32.5; head length 7.0 (21.5); snout length 2.4 (13.5); orbital diameter 2.1 (6.5); interorbital width + Because of superficial external similarity, it is possible that Phenablennius (Phenablenniini) will key to Omox. Phenablennius differs most obviously from Omox in having 3 segmented pelvic-fin rays (versus 2) and 6 circumorbital sensory pores (versus 7 or 8, rarely 7). 93 ‘(quoMssuT[oH “y ‘qd Aq uMesp) ueWC “Img “Js WU ¢'Z7¢ g[BUI “ZOZOH Winaesny] OLIEJUGO [eAOY ‘uojisdd uvWGQ ‘| ‘BLq eens ——— 2 BBs a Sep VOLUME 98, NUMBER 1 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 94 natn < won ty Ay Vv d (QT 10 6 Aes) T1100 OI O€ (LI 10 €] Ajores) el LI 91 €1 67 Of 0} EC (ZI 10 6 Ajores) Il TI 01 6 VT v7 91 BI 6 Le 91 61 ({1 10 9 Ajosed) Ol TT 019 ({) unWO (¢) snandjayou Q1VOIPUT SOWeU SLIOUEs Joye sIoquinu jeonouusied °9 UvIPOU = SW -[e}q1o19}UT = OJ SuinsjUs9 Ie[AjsOIN x9] =<) ‘sjetyouviqiseq = qa ‘snouodoine = ny -JUasqe = V d S 4 peurol 8 (€ Ajyensn) E107 8¢ Ol 1Z 91 0¢ 61 0} LI cl (1) sduasojdvyy S (o1 Atfepow) ZI 016 d (Q Ajoret) ¢ 10 (p1 Ajored) v1 OCI 6C 91 SC Ol £7 01 61 17 9} 81 (ZI 10 OT Ajeze4) TI 91 OI (1) snyjpusoydivT jyeid jeindAy [e1jUsA = duio9 01 pasny = 04 -posny = 4 Sumniypeoisod jes1op SUOTIBIADIQQY “TUIyURIQOUIO JO pi9ues oY} BUOWe S1oJOvIvYO S (Z Ajyensn) 8 019 d IT 01 ¢ 100 (py Ajored) pI 10 €I SZ 01 VT (T1 Ajered) TT 91 OT 17 0} 81 Oc 01 LI (€] Ajores) €T OTT (Z) SnandjayquadDd Vv Ss plo€g Ss (L Ajeret) BIOL € (p Ajrensn) COE (Vv Ajjensn) V 10d €1 JO CI 97 91 HC Ol 61 0} LI LI 91 ST cl (Z) xowo Vv S G10 S (g 10 1 Ajjepout) OI 019 £ (p 10 | Ajores) p Ol | d OI 910 (€1 Atfepour) pl ZI €€ 01 97 ZI 91 O LT 01 81 97 01 OT (€{ 10 Z1 Atfepow) yt 01 OI (1Z) snysuDAqouo prouyyoury s[e]uoLy s[eIqioelyUuy souo0g |eSeN saiod OD soiod qV soiod Of o10d SW soqn} DUT]-[21918T uy [e10109g jepne) jepneseig deIQOLIO A. sAvi Uy-[eUuy she soutds uy-[esiod 1a}De1IeY) ‘sg1oads popnyout yo Joquinu dy, ‘jerssoa = A ‘gyeI1vdas = § ‘JUasoid = q :[ewWIOU = N — odq ‘jeiqioumnomls = OD ‘snoulse]iyieo jo uostedui0)—"T 219PL ‘jerodwioyeidns oS VOLUME 98, NUMBER 1 _ Oe a eee Vv Vv Vv V Vv Vv Vv d Vv Vv DIOA Vv V d I ny Vi1od N peoy uo Vi1od yso10 AYSsopy s[Lsou V uo THT) d odq TZ IO [ sjeindq ny dua Viod ¢ jeindAyY N ¢c pue | da —_—_—— OOOO Ooo (1) unwoO (¢) Snandjayou gq (1) sduasojdoyy (1) SnyjousoydivT (Z) Snandjayouasvg (Z) xouo (1Z) snysunaqouwg Jouovieyo _ CC OO rrr ‘ponunuoy—"] 21qeL 96 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 0.7 (2.2); predorsal length 7.6 (23.4); preanal length 4.5 (13.8); depth at anus 5.3 (16.3); caudal peduncle depth 2.3 (7.3); caudal peduncle length 2.4 (13.5); length of: 1st dorsal-fin spine 1.5 (4.6), 2nd dorsal-fin spine 1.7 (5.2), 5th dorsal-fin spine 2.6 (8.0), 10th dorsal-fin spine 2.7 (8.3), 5th dorsal-fin ray 3.4 (10.5), 10th dorsal- fin ray 3.8 (11.7), 15th dorsal-fin ray 4.2 (12.9), 20th dorsal-fin ray 3.9 (12.0), 25th dorsal-fin ray 2.2 (6.8), 1st anal-fin spine 0.5 (1.5), 2nd anal-fin spine 1.1 (3.4), 1st anal-fin ray 1.8 (5.5), 5th anal-fin ray 2.3 (7.1), 10th anal-fin ray 2.7 (8.3), 15th anal-fin ray 3.1 (9.5), 20th anal-fin ray 2.8 (8.6), 24th anal-fin ray 2.3 (7.1), longest segmented pelvic-fin ray 4.0 (12.3), shortest segmented pelvic-fin ray 2.8 (8.6), longest pectoral-fin ray 5.0 (15.4), longest caudal-fin ray 5.7 (17.5); gill opening depth 1.0 (3.1). Color pattern (preserved): Ground color of head and body pale. Prominent dark-dusky, U-shaped marking on head, with arm of U originating laterally on nape, extending anteroventrally along nape and over eye, down to and around snout tip, and up opposite side of head; sparsely distributed group of melanophores extending posteriorly from mid-postorbital margin; indications on body of about 9 narrow, irregularly-shaped bands of sparsely distributed melanophores separated by wider unmarked areas. Membranes of spinous dorsal fin covered with fine, evenly-distributed melanophores; segmented-ray portion with melanophores re- stricted mostly to distal margin of fin. Anal fin with hazy stripe of fine melano- phores along distal margin. Caudal fin with darkly dusky band of melanophores distally. Pectoral fins with sparsely distributed melanophores distally. Pelvic fins unmarked. Holotype.—Royal Ontario Museum 40208, male, 32.5 mm SL (now cleared and stained and partially dissected), Sur, Oman (22°35'39”N, 59°32’E), depth 4 m, sandy rock reef over rocky sand, 100 m from shore, | Jun 1981, B. N. G. Simm. Etymology.—The species name, here used as a noun in apposition, is based on the Greek name of the letter ““U,” in reference to the dark U-shaped marking on the anterodorsal surface of the head. Acknowledgments I am especially grateful to R. Winterbottom for allowing me to describe the new genus and species. Several colleagues discussed cladistic methodology with me and/or offered suggestions for improvement of early drafts of the manuscript, which included the cladistic analysis as well as the description of the new taxa: B. Chernoff, R. H. Gibbs, Jr., J. Russo, W. F. Smith-Vaniz, R. P. Vari, R. Voss, J. T. Williams, R. Winterbottom, and, in particular, G. D. Johnson. Literature Cited Bath, H. 1983. Omobranchus hikkaduwensis, n. sp. von Ceylon (Pisces: Blenniidae).— Senckenber- giana Biologica 64(1/3):25-30. Smith-Vaniz, W. F., and V. G. Springer. 1971. Synopsis of the Tribe Salariini, with Description of Five New Genera and Three New Species (Pisces: Blenniidae). —Smithsonian Contributions to Zoology 73:1—72. Springer, V.G. 1968. Osteology and Classification of the Fishes of the Family Blenniidae.— Bulletin of the United States National Museum 284:1-85. VOLUME 98, NUMBER 1 97 1972. Synopsis of the Tribe Omobranchini with Descriptions of Three New Genera and Two New Species (Pisces: Blenniidae).— Smithsonian Contributions to Zoology 130:1-31. 1981. Notes on Blenniid Fishes of the Tribe Omobranchini, with Descriptions of Two New Species. — Proceedings of the Biological Society of Washington 94(3):699-707. , and M. F.Gomon. 1975. Revision of the Blenniid Fish Genus Omobranchus with Descrip- tions of Three New Species and Notes on Other Species of the Tribe Omobranchini.—Smith- sonian Contributions to Zoology 177:1-135. , and W: F. Smith-Vaniz. 1972. A New Tribe (Phenablenniini) and Genus (Phenablennius) of Blenniid Fishes Based on Petroscirtes heyligeri Bleeker.—Copeia 1972(1):64—71. Division of Fishes, Department of Vertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 98-106 PERIOCULODES CERASINUS, N. SP., THE FIRST RECORD OF THE GENUS FROM THE CARIBBEAN SEA (AMPHIPODA: OEDICEROTIDAE) James Darwin Thomas and J. L. Barnard Abstract. —Perioculodes cerasinus, a probable cryptic fossorial amphipod with embedded white orbicular ommatidia in bright ruby eyes is described from To- bago, Belize, Florida Keys, and Biscayne Bay, Florida. The eyes are separated either into two lunes or combined side to side into one large irregular brow. This is the first western Atlantic record of a generic group heretofore confined to the warm eastern Atlantic and Indian Oceans. Close affinity appears to be with the type-species of the genus, P. Jongimanus, from the eastern Atlantic Ocean. The paucity of exploration for amphipods in the west tropical Atlantic is no better emphasized than by the present species which has come to light in collec- tions from as far spread as Tobago, Belize, and Biscayne Bay, Florida. These collections have been made in just the past two years. The species has been observed alive on several occasions. The ring-like eye in life is quite noticeable because of the jewel-like appearance: diamonds glinting from a bed of rubies. This new species has some affinities with the west African Perioculopsis lophopus Schellenberg, 1925, from the Gulf of Guinea but is also close to Perioculodes longimanus (Bate and Westwood) (see Sars 1895) distributed in the eastern At- lantic from Norway south to the Cape of Good Hope and into the warm Indian Ocean. Unfortunately, there are numerous taxonomic difficulties in this group as de- scribed below. Oedicerotidae Perioculodes Sars Perioculodes Sars 1895:312.—Lincoln 1979:338. Diagnosis.—Eyes when present forming anterodorsal ring from side to side, ring sometimes divided into two pieces. Peduncular articles of antennae 1 stout in male. Incisors projecting and toothed, each mandible with toothed lacinia mobilis, raker row, and weak, non-triturative molar bearing spines; mandibular palp feeble, in female articles 1 and 3 short, of equal length; in male article 3 elongate. Inner lobes of lower lip coalesced but fusion line occasionally marked as raphus. Inner plates of maxillae 1-2 poorly setose, outer plate of maxilla 1 with 7 spines, palp 2-articulate. Maxilla 2 feeble. Maxillipedal plates moderately to poorly armed, inner small, outer large, dactyl elongate and unguiform. Gnathopods 1-2 alike in both sexes, of similar size, with short article 5 (wrist) bearing long posterior lobe guarding article 6 completely, article 6 (hand) long VOLUME 98, NUMBER 1 99 and subrectangular, palm oblique, well defined. Coxa 6 not bevelled posteroven- trally. Epimera simple. Uropod 2 reaching near apex of uropod 3. Gills present on coxae 2-6, gill on coxa 4 largest and most adze-shaped, on coxa 6 smallest and most sausage-shaped. Normal oostegites present on coxae 3-5, thin, weakly setose, coxa 2 with vestigial oostegite bearing | seta. Variables. —Peduncle of antenna 2 short in male, with article 3 short, but oc- casionaly species with female antenna | bearing elongate articles, article 3 espe- cially elongate. Coxae variable, in type-species coxae 1—4 forming one group, and coxae 5—7 forming second group but in other species no distinct grouping; in other species with coxae 1-3 forming a short group, coxae 5-7 forming long group. Gnathopods elongate or not. Dactyls of pereopods 3—4 long, medium or short. Pereopods 3-7 variable in dimensions and armaments. Urosomites 2—3 separate or fused. Telson emarginate or rounded apically. Type-species.— Monoculodes longimanus Bate and Westwood (by monotypy). Remarks.—The variability of known characters in the species of this genus and the lack of knowledge of many characters precludes any division into genera at this time. Whether or not Perioculopsis Schellenberg (1925) is distinctive must also be reviewed when it is more adequately illustrated. Perioculodes cerasinus, new species Figs. 1-3 Description of holotype male “‘h” 1.57 mm.—Head extremely broad, as long as first 3 pereonites combined, rostrum thick and of medium forward extension; eyes in life ruby red with diamond white ommatidia sparkling from ruby matrix, divided into 2 arcs separated by small space, because of head-width eyes very large for body size. Antennae 1-2 short, extending equally, peduncle of antenna 1 short, article 1 scarcely extending beyond apex of rostrum, articles 2 and 3 slightly shorter and therefore scarcely longer than wide; primary flagellum about as long as peduncle, with 5 articles, one aesthetasc each on articles 3—4; accessory flagellum marked by vestigial hump and setule. Gland cone of antenna 2 (see special figure) small; article 4 of peduncle as long as article 1 of antenna 1, article 5 slightly shorter, flagellum shorter than peduncle, with 4 articles. Prebuccal complex bulbous anteriorly, ventral margin of upper lip weakly sin- uate and with weak ventral midprotrusion. Left mandibular incisor with 3 main teeth formed in phoxocephalid fashion, long margin between main teeth serrate, lacinia mobilis either absent or represented by first raker, raker spines thus 4 in number, simple, molar obsolescent, marked by triad of spines; article 1 of palp elongate, article 2 thick, with 3 setae, article 3 shorter than 1, stubby, setae = 2E. Inner lobes of lower lip fused, dome-shaped, with middle raphus, outer lobes widely spread, inflated, each with weak cone and weak mandibular lobe. Inner plate of maxilla 1 broad, recumbant (tilting towards outer plate), in illustrations shown tilted and straightened with one apical spine, inner plate broad and short, with 7 apical spines, palp weakly 2-articulate, symmetrical on both sides, with 5 apical and subapical spines. Maxilla 2 feeble, composed of two broad plates sparsely armed, inner plate with 2 medial setae towards apex. Inner plate of maxilliped ordinary, with 3 apical setae, innermost attached ventrally; outer plate 100 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Capital letters denote main parts in following list; lower case letters to left of capital letters ieee or in body of figure indicate modifications as per following list; lower case letters to right of capital letters indicate specimens described in captions: B, body; C, coxa; E, epimeron; F, accessory flagellum; G, gnathopod; H, head; J, prebuccal; L, labium; M, mandible; P, pereopod; R, uropod; S, maxilliped; T, telson; U, labrum; V, palp; W, pleon; X, maxilla; Y, oostegite; Z, gill; d, dorsal; 1, lateral; 0, opposite; r, right; v, ventral. Perioculodes cerasinus, unattributed figures, holotype, male “h”’ 1.5 mm; i = female “7? 1.81 mm. VOLUME 98, NUMBER 1 101 Fig. 2. Perioculodes cerasinus, unattributed figures, holotype, male ‘“‘h” 1.5 mm; i = female “i” 1.81 mm; n = male “n” 1.90 mm. 102 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON of ordinary size but sparsely armed, with only 5 medial spines, palp stout, dactyl elongate, unguiform, with short nail. Coxae 1-3 forming short group together, coxae 4—7 forming longer group to- gether, coxa 7 not distinguished from coxae 4-6, ventral setae sparse, coxa 1 bevelled anteroventrally, coxa 2 much narrower than coxae | and 3, coxa 5 scarcely bilobate. Hand of gnathopod 1 shorter than that of gnathopod 2, wrist lobe also slightly shorter, both lobes extending beyond proximal extent of palms. Pereopods 3-4 without major spines, all other armaments of setal form. Pereopod 5 especially small relative to body size, article 2 feeble, medial face of article 2 on pereopods 5—6 with vertical setal row; pereopod 7 not as elongate as in most oedicerotids, article 2 ovate, dactyl with 3 setae. Gills on coxae 2-6, that on 4 largest, mostly adze-shaped but becoming smaller and more sausage-shaped on coxae 5 and 6. Epimera 1-3 plain, rounded posteroventrally, each segment bearing single ven- trofacial spine. Peduncle of uropod 1 with 3 dorsolateral spines, long gap between spines 2 and 3. Urosomites 2—3 fused; peduncle of uropod 2 with one apicolateral spine; each peduncle with one apicomedial spine; each ramus of uropods 1-2 with one dorsal spine; outer ramus of uropod 2 shortened. Uropod 3 feeble, peduncle about half as long as peduncle of uropod 2, rami longer than peduncle, outer shorter than inner, both naked. Telson ovate, margins entire, apex with 2 separated setules and each apicolateral face with pair of setules. Female ‘i’ 1.81 mm.—Like male but with oostegites, no other conspicuous secondary sexual characters but this specimen larger and better developed than holotype: eyes larger and pair abutting medially (see illustration); epimeron 1 with 2 ventral spines; peduncle of uropod 1 with 5 lateral spines, gap still present, of uropod 2 with 2 dorsolateral spines; outer rami of uropods 1—2 with 2 spines each. Oostegites thin, that of coxa 2 rudimentary, with | apical seta, others long and thin, setal formulae of all as follows: anterior setae = 0-2-2-1, distal setae = 1-2- 2-2, posterior setae = O0-0-0-0, anterior setules = 0, posterior setules = O0-0-1-2. Male “‘g’’ 1.70 mm.— Better developed than holotype: epimeron | with 2 ventral spines; outer rami of uropods 1—2 with 2 dorsal spines; each ramus of uropod 3 with one dorsal spine. Male “n’’ 1.90 mm.—Left epimeron | with midfacial spine-seta, 1 ventral seta, right side with 3 facials and 2 ventrals (illustrated); spine on epimeron 3 very posteriad; peduncle of uropod 1 with 5 dorsolateral spines, no gap, uropod 2 with 2; outer rami of uropods 1—2 with 2 spines each; each ramus of uropod 3 with 1 spine. Juvenile ‘j’’ 1.30 mm.—Spines on epimera 1, 2, 3 = 1-1-0. Peduncle of uropod 3 with 3 dorsolateral spines, one gap, uropod 2 with 2 spines, rami of uropods 1-2 each with | spine, rami of uropod 3 naked. Male ‘‘p”’ 2.04 mm and male “‘q’’ 1.77 mm.—Eyes faded totally in preservative, thus apparent only as foam and bubbles inside head. Uropods like male “‘n.”’ Male “‘p”’ 2.04 mm and male “q’’ 1.77 mm.—Eyes faded totally in preservative, thus apparent only as foam and bubbles inside head. Uropods like male “‘n.”’ Notes.— Largest specimen from Bacha Shoal, Florida, specimen “i” 2.75 mm, with following spine counts: uropod | peduncle = 6, outer ramus 2-3, inner ramus 1; uropod 2 peduncle 2, outer ramus 2, inner ramus 1; uropod 3 outer ramus 1, VOLUME 98, NUMBER 1 103 Fig. 3. Perioculodes cerasinus, unattributed figures, holotype, male “‘h” 1.5 mm; g = male “g’”’ 1.7 mm; i = female “1”’ 1.81 mm; n = male “‘n” 1.9 mm. inner ramus |. The numerous peduncular spines of uropod 1 show no extravagant gaps. Article 2 of pereopod 7 with 8 posterior setae. Largest specimen from Belize, specimen “‘z’”? 2.85 mm, with following spine counts: uropod | peduncle 6 (no extravagant gaps), outer ramus 2, inner ramus 1; uropod 2 peduncle 2, outer ramus 2, inner ramus |. Illustrations.— Main view of upper lip shown obliquely from below. Etymology.— From “‘cerasinus,”’ cherry colored. Holotype.—USNM No. 195131, male “h” 1.57 mm, illustrated. Type-locality.—Tobago, Kilwyn Bay, 2 Oct 1983, 1 m, in algae on sand reef, coll. J. L. Barnard. Material.—The type-locality, female ‘i’? 1.81 mm (illustrated), juvenile ‘‘j” 1.30 mm and 4 other specimens, “‘k, 1, m, n.’”’ Florida Keys, Looe Key, Sta LKR 104 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 4H, 9 Oct 1983, 1 m, west end of rubble zone in backreef area, coll. J. D. Thomas, male “‘g”’ 1.70 mm (illustrated) and one other specimen.— Biscayne Bay, Florida, sta 35-4, 152 m east of Intracoastal Waterway Marker 65, 1.3 m, dense Thalassia and Halodule, 5 Mar 1983, coll. Biosystems Research Inc., Miami; specimen “‘p”’ 2.04 mm, specimen “‘q” 1.77 mm.—Bacha Shoal, Florida, 13 Oct 1982, nighttime suction dredge, coll. Dr. Iver Brook, 3 specimens, largest “1” = 2.75 mm.—Belize, region of Carrie Bow Cay, in South Water Cay Channel, 15 Jun 1980, 8.2 m, in patch reef and coral rubble with small attached coral heads, coll. J. D. Thomas (2 specimens). Ecology.—Cryptic, usually collected in formalin washes of coral, coral rubble, or other hard substrates. Probably occupying isolated sediment-filled areas or cavities. Color.—In life, pale white to ivory body, eyes deep cherry red. Red eye color persists in formaldehyde, fades in alcohol. Relationship.— The species of Perioculodes are very diverse but we cannot find any differences of generic value between the type-species, P. /ongimanus, and our species, as they are either integrated by attributes of other species or because the sexual dimorphism in antennae of some of the species has not yet been worked out. Perioculodes cerasinus differs from Perioculodes longimanus (Bate and West- wood) (eastern Atlantic), the type-species of Perioculodes Sars (1895), in the (1) short peduncle of female antenna | with article 3 being especially short; (2) pres- ence of a raphus on the inner plate of the lower lip; (3) shorter article 6 on gnathopods 1 and 2; (4) division of coxae 1-3 and coxae 4—7 into groups with coxa 7 not disjunctly shorter than others within its series; (5) short peduncle of uropod 3 which is about half as long as the peduncle of uropod 2 (but as long as the peduncle in P. /Jongimanus). In Perioculodes longimanus there is a sexual diversity in antennae which we have as yet not found in P. cerasinus. The first article of the mandibular palp is shorter and article 2 much more setose in Perioc- ulodes longimanus. Perioculodes cerasinus differs from the Indian Ocean Perioculodes megapleon Giles, 1888 (see also Pillai 1957; Rabindranath 1972; and Ledoyer 1973, 1979) in the short uropod 3, less setose maxilla 2 and antenna 2 in both sexes, the non- excavate telson, the much less armed dactyl of pereopod 7 and the much more sparsely setose pereopods 5-7 in general. Perioculodes megapleon appears to have the same coxal arrangement and the short article 3 on antenna | of the female as in Our species. Perioculodes serra Walker, 1904 (and see Ledoyer 1979) from the Indian Ocean has the short article 3 of antenna | and apparent short uropod 3 of our species, but the outer ramus of uropod 1 of P. serra is very short and the inner ramus grossly serrate; coxae 4—5 of P. serra have bent posteroventral lobes, mandibular palp articles 2-3 are strongly setose, the dactyls of pereopods 3 (and 4) are tiny, article 2 of pereopod 5 has a large posteroventral lobe and the general pereopod proportions are distinctive. Perioculodes acuticoxa Ledoyer, 1973, differs from P. cerasinus in the very elongate and thin gnathopods, the presence of an anteroventral cusp on coxa 3, the slightly more spinose uropods 1-2, the truncate telson, and the different proportions and setae patterns of pereopods 6 and 7. VOLUME 98, NUMBER 1 105 Perioculodes aequimanus Kossman, 1880, (see Stebbing 1906) is poorly de- scribed but has the same elongate gnathopods of P. acuticoxa and may be a synonym of that species. Perioculodes pallidus Griffiths, 1975, from South Africa in 39 meters, has nine possible generic characters of distinction from the type-species, P. Jongimanus, as follows: (1) lack of eyes; (2) long rostrum; (3) poor to undeveloped lacinia mobilis; (4) enlarged pereopod 6; (5) naked uropods; (6, 7, 8) short uropods 2-3 with unequal rami on uropod 3; (9) emarginate telson. Our species differs from Perioculopsis lophopus Schellenberg from Ghana in the (1) absence of a distoventral tooth on article 1 of antenna 1; (2) the smallness of the gland cone on antenna 2; (3) the well-toothed mandibular incisors; (4) the unbevelled posteroventral margin of coxa 6; (5) the equally extending rami of uropods 1-2 (in Perioculopsis the outer rami are shortened); and (6) the nonex- cavate posterior margin of the telson. The fusion of pleonites 5—6 in our species appears to be an apomorphic character quite remote from the same apomorphic expression in the Paracalliopiidae, an Indo-Pacific family with affinities to Oedicerotidae. Distribution.— Biscayne Bay, Florida to Florida Keys to Tobago and Belize, 1- 8 m. Acknowledgments We thank Mr. David Hardy of NOAA, USA, for his help with our work in Tobago. We thank Dr. Pat McLaughlin, whose specimens:were collected by Bio- systems Research Inc, as part of the Biscayne Bay Restoration and Enhancement Program under the direction of the Dade County Department of Environmental Resources Management and funded by the Florida Department of Environmental Regulation. Dr. Iver M. Brook also helped collect Biscayne Bay specimens. The first author was supported by grants DEB8121128 from the National Science Foundation, and the Scholarly Studies Program of the Smithsonian Institution. Looe Key material was collected by the first author under contract NA82AAA01 157 from the Sanctuary Programs Division, Office of Ocean and Coastal Resource Managemant, NOAA. For support in Belize we thank Klaus Riitzler, Director of the Smithsonian Western Atlantic Mangrove Program (SWAMP), and Mike Car- penter, also from the Smithsonian Institution, for his assistance in collecting. We thank Carolyn Cox Lyons of New York for inking our plates. Literature Cited Giles, G. M. 1888. No. 9. Further notes on the Amphipoda of Indian waters. Natural history notes from H.M.’s Indian marine survey steamer “Investigator,” Commander Alfred Carpenter, R.N., D.S.O., Commanding. —Journal of the Asiatic Society of Bengal 57:220-255, pls. 6-12. Griffiths, C. L. 1975. The Amphipoda of Southern Africa. Part 5.—Annals of the South African Museum 67:91-181, 21 fig. Kossman, R. 1880. Malacostraca.—Zoologische Ergenbisse einer Auftrage der K6niglichen Academie Wissenschaften zu Berlin. Reise im Kiistengebiete des Rothen Meeres 2(1):67—140, pls. 4-15. Ledoyer, M. 1973. Etude des amphipodes gammariens des biotopes de substrats sableux et sablo- vaseux de la region de Tuléar et de Nosy-Be (Madagascar.)— Tethys Supplement 5: 51-94, pls. 1-30. . 1979. Les gammariens de la pente externe du grand recif de Tuléar (Madagascar) (Crustacea 106 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Amphipoda).—Memorie del Museo Civico di Storia Naturale de Verona (II? Serie), Sezione Scienze della Vita 2:1—-150, figs. 1-91. Lincoln, R. J. 1979. British marine Amphipoda:Gammaridea.—British Museum (Natural History), London: v-vi, 1-658, figs. 1-280, pls. 1-3. Pillai, N. K. 1957. Pelagic Crustacea of Travancore. III. Amphipoda.—Bulletin of the Central Re- search Institute, University of Travancore 5:29-68, figs. 1-18. Rabindranath, P. 1972. Studies on gammaridean Amphipoda (Crustacea) from India.— Bulletin van de Zoologisch Museum, Universiteit van Amsterdam 2:155-172, figs. 1-7. Sars, G. O. 1895. Amphipoda.—An account of the Crustacea of Norway with short descriptions and figures of all the species.—Christiania and Copenhagen 1:1—-viii, 1-711, 240 pls., 8 suppl. pls. Schellenberg, A. 1925. Crustacea VIII: Amphipoda.—Jn W. Michaelsen, Beitrage zur Kenntnis der Meersfauna Westafrikas 3:111-204, 27 figs. Stebbing, T. R. R. 1906. Amphipoda I. Gammaridea.—Das Tierreich 21:1-806, 127 figs. Walker, A. O. 1904. Report on the Amphipoda collected by Professor Herdman, at Ceylon, in 1902.—Report to the Government of Ceylon on the Pearl Oyster Fisheries of the Gulf of Mamaar, Supplementary Report 17:229-300, 8 pls. (JDT) Newfound Harbor Marine Institute, Rt. 3, Box 170, Big Pine Key, Florida 33043; (JLB) NHB-163, Department of Invertebrate Zoology, National Museum of Natural History Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 107-111 ZENARCHOPTERUS ORNITHOCEPHALA, A NEW SPECIES OF FRESHWATER HALFBEAK (PISCES: HEMIRAMPHIDAE) FROM THE VOGELKOP PENINSULA OF NEW GUINEA Bruce B. Collette Abstract.— Zenarchopterus ornithocephala is the fifth freshwater species of the genus known from New Guinea. It resembles Z. alleni Collette and Z. robertsi Collette in having moderately high numbers of predorsal scales (more than 47), a connection of the supraorbital lateral-line canals across the top of the head, and a moderately large body size, but differs from those two species in having only the sixth anal fin ray enlarged in males instead of the fifth and sixth. It further resembles Z. alleni in having many vertebrae (49-50) but differs in having fewer predorsal scales (48-52 vs. 66), and a pectoral fin shorter than the head length. Dr. Gerald R. Allen recently spent four months collecting in the freshwaters of New Guinea (Allen 1984). According to Dr. Allen (pers. comm.), the highlight of his 1982 trip was a 10-day foray via single engine Cessna in Irian Jaya, the Indonesian western half of the island of New Guinea. Among the new fish species collected on the Vogelkop Peninsula (Fig. 1) were a new Melanotaenia, a new teraponid, and a new Zenarchopterus, the fifth known from the freshwaters of New Guinea. The purpose of this paper is to describe the new species of Zen- archopterus. Methodology follows my previous papers on halfbeaks (Collette 1974, 1982). Material examined is in the collections of the Lembaga Biologi Nasional (National Biological Institute), Bogor, Indonesia (LBN); National Museum of Natural History, Washington, D.C. (USNM); the Western Australia Museum, Perth (WAM); and the Zoédlogisch Museum, Universiteit van Amsterdam (ZMA). Zenarchopterus ornithocephala, new species Fig. 2 Diagnosis.—Sixth anal fin ray of adult male enlarged (Fig. 3A). Many total predorsal scales (48-52), position of predorsal scale that overlaps scales both anteriorly and posteriorly far anterior, 45 scales anterior to origin of dorsal fin. Upper jaw longer than wide (width divided by length 0.74—0.84); lower jaw about equal to head length (head length divided by lower jaw length 0.96-1.01). Pectoral fin much shorter than head length (head length divided by pectoral length 1.55- 1.72). No dorsal fin rays modified in males. Description. — Dorsal and anal fin rays 14; pectoral fin rays 10. Vertebrae (3 1- 32) + 18 = 49-50. Gill rakers on first arch (4—5) + (13-14) = 17-19, on second arch (0-2) + (14-15) = 14-16. Left and right supraorbital lateral line canals con- nected across top of head (Fig. 4A—B). Types.— Holotype: LBN 5419 (6, 124 mm SL); Irian Jaya; Vogelkop Peninsula; stream at Fruata, 2°59’S, 133°32'E; G. R. Allen and H. Bleher; 16 Nov 1982.— 108 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ar] a Ons EFAS y 5 a, , Mamberamo R. f N\ Vogelkop% Salta > ‘si re Peninsula N a Q : ce A \ Ss ‘) = yy y Set \) I; = Vas 2 NX = & , eye SOA ae ee NG —oepik R. =< $ aa | — OC OR mY °: ~ fo) ey eee 7 ISS Seay Lorentz R. ; N o Q * Z. ornithocephalus < = © Z. novaeguineae S Te ~ oy a Sees ay F < H O Z. kampeni A 5 & A x Z.alleni p \ (ky ANCA ) @ Z.robertsi asl S aw Se JAS SS Fly R. es 1 KumusiR. Sy Fig. 1. Location of New Guinea rivers known to contain freshwater species of Zenarchopterus: Z. ornithocephala on the Vogelkop Peninsula; Z. alleni in the Mamberamo; Z. kampeni in the Mam- beramo, Ramu, and Sepik; Z. robertsi in the Kumusi; and Z. novaeguineae in the Laloki, Fly, and Lorentz. Paratypes: WAM P-27868-004 (6, 93.9 mm SL) and USNM 266413 (6, 94.5 mm SL); Irian Jaya; Vogelkop Peninsula; stream at Senopi, 0°50’S, 132°56'E; G. R. Allen and W. Tins; 18 Nov 1982. Etymology.—A noun in apposition, from the Greek ornis, ornithos (bird) and kephale (head), after the Vogelkop (birdhead) Peninsula. Habitat.—Notes on the habitat were kindly provided by Dr. Allen. Both sites were rainforest streams with slight turbidity. The bottom consisted of sand and gravel. The stream at Fruata was in mainly flat country at an elevation of 90 m; the stream at Senopi was in hilly terrain with a moderately fast flow at an elevation of 460 m. The pH was 7.8 and water temperature 28.3°C at Fruata; 7.5 and 27.5°C at Senopi. Comparisons.— The five freshwater New Guinea species of Zenarchopterus share two specializations. They have more predorsal scales (33-66) than the other 13 species in the genus (26-34). They reach a larger maximum standard length (124— 173 mm SL) than the marine species (126 mm, Collette 1982). The next largest species are Z. buffonis (Valenciennes)— 126 mm; Z. caudovittatus (Weber)— 123 mm; Z. ectuntio (Hamilton-Buchanan)— 122 mm; and Z. dispar (Valenciennes) — 121 mm. Fig. 2. Zenarchopterus ornithocephala LBN 5419, holotype, 124 mm, male; Irian Jaya; Vogelkop Peninsula; Fruata. VOLUME 98, NUMBER 1 109 a Z) (\ Y if =x Fig. 3. Anal fin in males of three freshwater species of Zenarchopterus from New Guinea. A, Z. ornithocephala, holotype, LBN 5419; B, Z. robertsi, USNM 219229, 126mm SL; C, Z. alleni, holotype, ZMA 116.479, basal portion of rays 5 and 6 only. (B and C from Collette 1982:fig. 3A—B). Based on number of predorsal scales, maximum length, and male anal fin structure, the New Guinea freshwater species can be divided into two species groups: Zenarchopterus kampeni (Weber) and novaeguineae (Weber) have 33-47 predorsal scales; and the other three species have still higher counts, robertsi Collette and ornithocephala with 47-53 and alleni Collette with 66. Zenarchop- terus kampeni and novaeguineae reach the largest size, 161 and 173 mm SL respectively. The other three freshwater New Guinea species are smaller, maxi- mum sizes of 131 mm (robertsi), 130 mm (the unique holotype of alleni), and 124 mm (ornithocephala). Zenarchopterus kampeni and novaeguineae have the sixth anal fin ray enlarged and paddle-shaped (Collette 1982:fig. 3C—D); robertsi and alleni have the fifth and sixth rays greatly elongate (Fig. 3B—C), reaching beyond the caudal base; and ornithocephala has only the sixth ray enlarged: Both Z. ornithocephala and Z. alleni have the left and right supraorbital lateral- line canals connected with a single median pore (Fig. 4A—C). The left and right canals are nearly connected in Z. robertsi (Fig. 4D) and Z. novaeguineae, but are completely separated in Z. kampeni (Collette 1982:fig. 5B) and many marine species of Zenarchopterus. Comparative material.—In addition to collecting the three type-specimens of Z. ornithocephala, Dr. Allen and his associates collected additional material of three of the other four species of freshwater Zenarchopterus in Papua New Guinea from September to November, 1982. This material is listed here as an appendix to the material listed in Collette (1982). 110 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 4. Dorsal view of cephalic lateralis system in three freshwater species of Zenarchopterus from New Guinea. A, Z. ornithocephala, holotype, LBN 5419; B, Z. ornithocephala, paratype, USNM 266413; C, Z. alleni, holotype, ZMA 116.479. (Collette 1982:fig. 5A); D, Z. robertsi, paratype, USNM 219299. VOLUME 98, NUMBER 1 iti Z. kampeni. Sepik R. WAM P-27847 (13, 28.9-125) and USNM 266411 (4, 113-144); Kwatit R. at junction with Sepik R.; 4°05’S, 143°06’E. Z. novaeguineae. Oriomo R. WAM P-27815 (24, 41.1—118); Papua New Guinea; Oriomo R., 30 km upstream from mouth; 8°52’S, 143°11’E. Fly R. WAM P-27812 (6, 51.5-—148); trib. Nomad R., 1 km N of village; 6°18’S, 142°14’E.—USNM 266364 (1, 139); Nomad R., N of airstrip; 6°18’S, 142°14’E.— WAM P-27810 (1, 105); Hamami R. S of Nomad airstrip; 6°18'S, 142°14’E.— WAM P-27799 (4, 119-127); small creek 10 km S of Ningerum on Kiunga Rd.; 5°46’S, 141°08’E.— USNM 266363 (1, 113); Wai Somare R., 1 km S of Ningerum; 5°41’S, 141°09’E.—WAM P-27805 (1, 82.7); trib. of Ok Tedi R., 15 km N of Ningerum on Tabubil Rd.; 5°33’S, 141°16’E. Z. robertsi. Kumusi R. WAM P-27790 (9, 65.5-128) and USNM 266412 (4, 96.2—125); Kaili Cr. 12 km E of Kokoda; 8°55'S, 147°47'E. Acknowledgments My deep thanks go to Gerald R. Allen who collected the material of the new species, surmised that it was undescribed, and then graciously turned over his material and notes to me. Drawings were made by Keiko Hiratsuka Moore. Radiographs were taken by Ruth E. Gibbons. Localities were plotted by Ms. Gibbons on a drainage map supplied by Dr. Allen. Drafts of the manuscript were read by Gerald R. Allen and Austin B. Williams. Literature Cited Allen, Gerald R. 1984. Irian Jaya—The last frontier.— Tropical Fish Hobbyist 32(7):22-29. Collette, Bruce B. 1974. The garfishes (Hemiramphidae) of Australia and New Zealand.— Records of the Australian Museum 29:11-105. . 1982. Two new species of freshwater halfbeaks (Pisces: Hemiramphidae) of the genus Zen- archopterus from New Guinea.—Copeia 1982(2):265-276. National Marine Fisheries Service Systematics Laboratory, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 112-120 MORPHOMETRICS AND DISTINCTNESS OF THE HEDGEHOG GENERA (INSECTIVORA: ERINACEIDAE) C. Brian Robbins and Henry W. Setzer Abstract.—Five genera (Erinaceus, Atelerix, Hemiechinus, Paraechinus, Ae- thechinus) have been described in the hedgehog subfamily Erinaceinae. Using non-mensural characters, previous authors have recognized from one to five of these genera as valid. Population samples of the five named genera were compared using multivariate statistics on selected cranial measurements. Results of the statistical analyses, coupled with non-mensural characters, distribution, ecology, and fossil history, indicate that all five genera are distinct and all should be recognized. The subfamily Erinaceinae includes five nominal genera: Erinaceus Linnaeus, 1758, Atelerix Pomel, 1848, Hemiechinus Fitzinger, 1866, Paraechinus Troussart, 1879, and Aethechinus Thomas, 1918. The status of these names ranges from the recognition of a single genus (Erinaceus, Dobson 1882), three genera (Erinaceus, Hemiechinus, and Paraechinus—see Corbet 1974, 1978; Honacki, Kinman, and Koepp! 1982), four genera (Erinaceus, Hemiechinus, Paraechinus, and Atelerix— see Dorst and Dandelot 1969), to all five as valid genera (Thomas 1918; Cabrera 1925; Allen 1939; Simpson 1945). A recent classification (Nowak and Paradiso 1983), although recognizing three genera (Erinaceus, Hemiechinus, and Para- echinus), subdivides Erinaceus into subgenera as: genus Erinaceus, with subgenera Erinaceus (one species) and Atelerix (four species, including two attributable to Aethechinus. All categories above the species level include groups that are genetically and morphologically discontinuous between one another. Those taxa cannot be sat- isfactorily defined in absolute terms because of the possibility of the absence of a marked discontinuity between taxa of the same rank. However, a genus is generally regarded as containing one or more species phenetically separable from other genera by a decided gap between species clusters. For practical reasons, the more species in a species-group the smaller the gap needed to recognize it as a separate genus, and the smaller the number of species, the larger the gap needed to recognize it (Mayr 1969). The function of the genus is to group monophyletic (related) species and facilitate information retrieval. Mayr (1969) gave the following criteria for delimiting and ranking taxa: (1) distinctness (size of gap)— measured in terms of phenetic distance and the bio- logical significance of the difference; (2) evolutionary role (uniqueness of adaptive zone)— analyzed by its ecological significance and evolutionary history; (3) degree of difference—phenetically, the distance between means of two groups of species; (4) size of taxon—number of species; and (5) equivalence of ranking in related taxa. Various combinations of characters could be used to construct a key to differ- entiate each of the five genera or to recognize four, three, two, or only one genus. VOLUME 98, NUMBER 1 113 Because previous generic distinctions were based on non-mensural characters, we decided to compare samples of the named genera by using multivariate statistical analyses on selected cranial measurements to determine if another method would satisfactorily differentiate the genera. In the following analyses we determine the degree of difference between groups, the type of scatter of a cluster, and whether or not there is equivalence of ranking in related taxa. The following questions were then asked and answered using results from the statistical analyses: 1) Are species clusters evident? 2) Are there gaps between the species clusters? 3) Are the species clusters of uniform density? 4) Are the species clusters large and heterogeneous? 5) Do the individual clusters include only specimens considered to be a part of the same genus (sensu stricto)? It is also possible that a combination of mensural and non-mensural characters would better define the genera. Therefore, a sixth question was also asked: 6) Are species clusters (genera) better defined using results from statistical analyses as well as previously used non-mensural characters? Materials and methods.—To assess the degree of difference (distinctness) of these five genera, specimens from several localities within the range of each genus were selected to include as many of their taxa as were available. Nine cranial measurements were taken and subjected to the NIT-SYS multivariate statistical programs (Rohlf, Kishpaugh, and Kirk 1972) and the BMD computer programs (Dixon 1973). Character means of each sample were used when the sample size was greater than two. The standardized means or individual specimen measure- ments were used to compute matrices of average distance and correlation coef- ficients among the samples. The unweighted pair-group method using arithmetic averages (UPGMA) was used in the cluster analysis and presented as a phenogram. Non-metric multidimensional scaling (MDSCALE; Kruskal 1964a, b), were also provided by the NT-SYS programs. The BMD0O7M stepwise discriminant function analysis provided overall discrimination among the groups using the raw data from all individual specimens. Specimens used in the analyses.—Species names are those recognized by Hon- acki et al. 1982. Subspecies names are included for those samples which have been named, by geographic area, in various publications. These are followed by country localities and sample size. Sample numbers used in Figs. 1 and 2 are indicated in parentheses. Erinaceus europaeus europaeus—Germany, 10 (1); E. e. hispanicus—Spain, 3 (2); E. concolor— Turkey, 3 (3); Paraechinus aethiopicus— Morocco, 10 (4); Niger, 2 (6 and 7); Mauritania, 1 (5); P. a. dorsalis—Egypt, 4 (8); P. a. deserti— Egypt, 1 (11); P. micropus—W. Pakistan, 1 (9); P. hypomelas—Iran, 1 (10); Aethechinus algirus—Morocco, 11 (12); A. frontalis—South Africa, 17 and Namibia, 2 (13); Hemiechinus auritus libycus—Egypt, 12 (14); H. a. aegyptius—Egypt, 9 (15); H. a. auritus—Iran, 2 (16 and 17); Atelerix albiventris albiventris—Senegal, 22 (21); A. a. spiculus—Nigeria, 18 (19); A. albiventris—Ghana, 6 (20); Bourkina Fasso (Upper Volta), 10 (18). 114 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 2 Erinaceus 12 Aethechinus Paraechinus 5 16 14 Hemiechinus 15 17 Paraechinus Atelerix 2.0 1.6 eZ 0.8 0.4 0.0 Fig. 1. Phenogram based on distance matrix of NT-SYS analysis of five genera of hedgehogs. Numbers refer to taxa and localities listed in the text. The cophenetic correlation is 0.898. Results.—Some of the most common characters used by Anderson (1895), Miller (1912), Thomas (1918), Allen (1922), Cabrera (1925), and Corbet (1974, 1978) to compare or contrast the five genera are shown in Table 1. Morphological comparisons by Corbet (1974, 1978) led him to conclude that Erinaceus, Atelerix, and Aethechinus were congeneric (Erinaceus). Thus construed, Erinaceus incor- porates all of the characters listed in Table 1 for Erinaceus, Atelerix, and Ae- thechinus. The remaining characters in Table 1 characterize Hemiechinus and Paraechinus. Erinaceus, as defined by Corbet (1974, 1978), differs from both Hemiechinus and Paraechinus only by having a small postglenoid process. In addition, it differs VOLUME 98, NUMBER 1 115 Table 1.—Morphological characters used to differentiate the five genera of hedgehogs. Erinaceus Atelerix Aethechinus Hemiechinus Paraechinus 13 and C—no. of one two two two two roots ; P3—size normal reduced or normal normal reduced absent Postglenoid process smaller smaller smaller same larger vs. mastoid process Pterygoids and normal normal normal normal large bullae size Hallux normal generally normal normal reduced absent Posterior palatal narrow broad broad narrow narrow shelf Spine-part on crown present present present absent present Ears small small small large large from Hemiechinus by having a median spine-part on the crown of the head and from Paraechinus by having normal-sized pterygoids and bullae. Hemiechinus differs from the other two in lacking a median spine-part on the crown. It also differs from Paraechinus by having normal-sized pterygoids and bullae. Para- echinus is distinguishable from the other two genera by its inflated pterygoids and bullae. Results of the UPGMA clustering analysis of specimens are shown as a phe- nogram in Fig. 1, which reveals two large and distinct clusters. Erinaceus and Aethechinus are clustered together and separated from the other three genera. Population samples of the taxa in Erinaceus and Aethechinus, as a part of the same large cluster, form distinct clusters of their own that show quite distinctive morphometric (cranial measurement) or phenetic distance between them. In the second or bottom cluster, Ate/erix is distinct from both Paraechinus and Hemi- echinus. Figure 2, showing the MDSCALE projection, gives a better representation of the phenetic distance separating the five genera. The minimum spanning tree (Prim 1957) connects those samples closest in phenetic distance, yet shows the distances between the taxa compared. In this analysis, Erinaceus, Aethechinus, and Atelerix separate into distinct groups. The African samples of Paraechinus (numbers 4, 5, 6, 7, 8, 11) are also linked and clustered. The two samples of Paraechinus from Asia (9, 10) are linked but separated from the African samples by a specimen of Hemiechinus (17). The other three samples of Hemiechinus are linked. A discriminant function analysis (Fig. 3) provided additional information for evaluating the morphometric differences between the five genera. The discrimi- nant analysis shows that the degree of difference, as reflected by the distances between group means, is highly significant. The probability that all samples are allocated correctly to a particular group is P = 0.90. The probability for most is P = 0.95. The character vectors included in the figure (Power and Tamsitt 1973) show the cranial measurements and their relative contribution in the separation of the five clusters. 116 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Atelerix Paraechinus Aethechinus 12 Us (p-€) Paraechinus Erinaceus Fig. 2. Three-dimensional projection of MDSCALE analysis with minimum spanning tree for samples of hedgehog genera. For sample numbers refer to text. Stress is 0.033. The distance between groups (size of gap) should be inversely proportional to the number of species in each group (size of taxon) in Fig. 3. For example, Erinaceus is represented by the fewest species but is separated from the other genera by the largest gaps and distance between group means. Erinaceus also has a dense species cluster. The degree of difference among the other genera is nearly equal. The distinctness of those genera, as reflected by distance (gap) between groups, varies. Paraechinus and Hemiechinus have homogeneous clusters and are separated by a decided gap. Both are separated from the other groups by even larger gaps. Atelerix and Aethechinus are closest in gap distance but the mean distance between their clusters is significant and they are well separated from the other genera. Atel/erix has a dense and uniform cluster while the Aethechinus cluster is large, probably indicating greater heterogeneity. The NT-SYS principal components analysis based on a correlation matrix (not figured) showed that cranial-size measurements (first component) separated Er- inaceus and Aethechinus from the other three genera, of which Ate/erix has the smallest skulls. Erinaceus was separated from Aethechinus in the second com- ponent, which was influenced positively by breadth of braincase and negatively by length of palatal shelf. Although Erinaceus has slightly larger breadth of brain- case measurements than Aethechinus, its length of palatal shelf measurements are much smaller (see Table 1). Paraechinus has the largest breadth of braincase measurements relative to skull length, reflecting the inflated bullae in members of this genus. Except for Aethechinus, Atelerix has the longest length of palatal shelf measurements. VOLUME 98, NUMBER 1 117 CV I1=32.2 % Hemiechinus Atelerix CV 1=52.9 % \ LPOB MTR Aethechinus Erinaceus Fig. 3. Projection of the first two canonical variates in a discriminant analysis of samples of hedgehog genera. Arrows indicate vectors for the nine cranial measurements. Abbreviations are: CBAL—condylobasal length; LR—length of rostrum; MTR—alveolar length of upper molar tooth row; LPAL—length of palate; LPSH—length of palatal shelf; LPOB—least postorbital breadth; BR— breadth of rostrum; BBC—breadth of braincase; ZB—zygomatic breadth; N—Nigeria; G—Senegal; M=— Morocco; S—southern Africa; CV I and CV II—canonical variates one and two; percent of cranial measurement variation accounted for by each is indicated. Discussion and Conclusions.—The distinctiveness of each taxon based on the characters listed in Table 1 can be, as expected, subjective. The number of species recognized in each genus is also inconsistent among authors. Erinaceus is generally regarded as having one polytypic species, but Corbet and Hill (1980) recognized three species; Paraechinus contains three or four species; Aethechinus includes two or three species; Hemiechinus contains two or three species; and Atelerix contains several described taxa that need further study. Butler (1978) summarized the distribution and fossil history of the five genera. He reports that Erinaceus is known from the Miocene to Recent in Europe, and Pleistocene to Recent in Asia in deciduous and Mediterranean woodland vege- tation zones. Atelerix is known only from the Recent in Africa where it occurs throughout the savanna vegetation zones. Hemiechinus is known only from the Recent in southern Asia and northeastern Africa in semi-arid steppe or savanna regions. Paraechinus is known only from the Recent in southern Asia and northern Africa in desert regions. Aethechinus is recognized from the Recent in north and south African temperate to dry savanna habitats. These time and geographic ranges give an indication of evolutionary role, distribution, and ecological differences. 118 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON In Africa, four genera are geographically and ecologically separated. Hemi- echinus is in the northeastern part where its range slightly overlaps that of Ae- thechinus in Libya. Paraechinus occurs in all of the North African deserts and slightly overlaps Aethechinus in Morocco and Algeria. Atelerix is found south of the Sahara desert in the savanna zones and does not co-occur with any other genus in the north. Aethechinus occurs again in southern Africa and may slightly overlap Atelerix in the northern part of its (Aethechinus) range. Erinaceus is the only genus in Europe except for introduced Aethechinus which persists as localized populations in southern Spain and France. Three genera occur in Asia, with Erinaceus occupying the northernmost areas in the woodland steppe regions. It slightly overlaps Hemiechinus in some parts of its range but does not occupy the semi-arid steppe areas that Hemiechinus prefers. The desert-inhabiting genus Paraechinus slightly overlaps the range of Hemiechinus in some areas. The five genera all have unique adaptive zones. Those that share similar ecol- ogies are geographically separated. Different habitats and different distributions plus evidence from the fossil record (Butler 1978) indicate that these five also had different evolutionary histories. The differences in use of the environment are responsible for the width and distinctness of the gaps between the genera. The six questions posed in the introduction can all be answered. Species clusters are evident; there are gaps between the clusters; four clusters are dense and uni- form, the other is large and heterogeneous; each cluster includes only specimens considered to be a part of the same genus; and the species clusters are better defined using a combination of mensural and non-mensural characters. Paraechinus and Hemiechinus are not clearly separated by the mensural data in Figs. 1 and 2. However, using non-mensural characters (Table 1), these dis- tinctive genera are readily separable as was shown by Corbet (1974). This is supported by the data in Fig. 3 which show a decided gap and distance between group means. Using the results of these statistical analyses and the non-mensural characters from Table 1, five genera of hedgehogs are indicated. Morphological analyses also show that Erinaceus and Aethechinus are more closely related to each other and form a unit apart from the other three genera. If subgenera within Erinaceus (Nowak and Paradiso 1983) were justified, then they should be Eri- naceus and Aethechinus, not Erinaceus and Atelerix. Nowak and Paradiso (1983) included species of Aethechinus with Atelerix. Morphologically, Aethechinus and Atelerix are not closely related. Their fossil records (Butler 1978) and geographic distributions suggest to us that European and Asian Erinaceus could have given rise to Aethechinus, which is now represented by separate species in northern and southern Africa. Generic classifications using qualitative characters that recognize fewer than five genera need modification. Equivalence of ranking in related taxa, when ap- plying unweighted non-mensural characters, can only result in the recognition of one or five genera. The results from the morphometric analyses of cranial mea- surements support the recognition of five genera (Erinaceus, Aethechinus, Para- echinus, Hemiechinus, and Atelerix) as given by Thomas (1918), Cabrera (1925), Allen (1939), and Simpson (1945). Recognition of five genera of hedgehogs also groups closely related species and avoids unnecessarily complicated arrangements resulting from using any other taxonomic grouping. Such an interpretation is VOLUME 98, NUMBER 1 119 possible using non-mensural morphological characters coupled with results of the statistical analyses. Erinaceus: Relatively large animals with long and broad skulls; CBAL usually greater than 55 mm; [3 and C single rooted; P3 normal in size; postglenoid process smaller than mastoid process; pterygoids and bullae normal (not inflated); palatal shelf narrow; hallux well developed; a median spine-part present on crown of head; ears small, not projecting above head-spines. Hemiechinus: Medium to large animals with medium to large skulls; CBAL ranges from 45 to 55 mm; I3 and C double rooted; P3 normal; postglenoid process same size as mastoid process; pterygoids and bullae normal; palatal shelf narrow; hallux well developed; median spine-part on crown of head absent; ears large, projecting above head-spines. Aethechinus: Large animals with large skulls; CBAL ranges from 45 to 60 mm; I3 and C double rooted; P3 normal; postglenoid process smaller than mastoid process; pterygoids and bullae normal (not inflated); hallux well developed; palatal shelf broad; median spine-part on crown present; ears small. Paraechinus: Medium animals with medium length but broad skulls; CBAL ranges from 40 to over 50 mm; [3 and C double rooted; P3 reduced in size; postglenoid process larger than mastoid process; pterygoids and bullae markedly inflated; hallux reduced; palatal shelf narrow; median spine-part on crown present; ears large, projecting well above head-spines. Atelerix: Small animals with small skulls; CBAL usually less than 45 mm; [3 and C double rooted; P3 reduced or absent; postglenoid process smaller than mastoid process; pterygoids and bullae normal; palatal shelf broad; hallux usually absent, but if present then greatly reduced in size; median spine-part on crown present; ears small. Acknowledgments We thank curators and support staff from numerous U.S., European, and African museums who allowed both authors to examine specimens during their many visits. A majority of the specimens examined are housed at the National Museum of Natural History, Washington, D.C., and were collected during African and Asian country faunal surveys directed by HWS. We thank Dr. Eleanora I. Robbins for preparation of the illustrations. Drs. Alfred L. Gardner and Don E. Wilson, U.S. Fish and Wildlife Service, Smithsonian Institution, and Guy G. Musser, American Museum of Natural History, New York, provided helpful suggestions and criticism of the manuscript. Literature Cited Allen, G. M. 1939. A checklist of African mammals.—Bulletin of the Museum of Comparative Zoology 83:1-763. Allen, J. A. 1922. The American Museum Congo Expedition collection of Insectivora.— Bulletin of the American Museum of Natural History 47:1-38. Anderson, J. 1895. Ona new species of the genus Erinaceus from Somaliland.— Proceedings of the Zoological Society of London 27:414—421. Butler, P. M. 1978. Ch. 4. Insectivora and Chiroptera, pp. 56-68. Jn V. J. Maglio and H. B. S. 120 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Cooke, eds., Evolution of African mammals.—Harvard University Press, Cambridge, xiv + 641 pp. Cabrera, A. 1925. Genera Mammalium, Insectivora Galeopithecia.— Museo Nacional de Ciencias Naturales, Madrid, 232 pp. Corbet, G. B. 1974. Family Erinaceidae. In J. Meester and H. W. Setzer, eds., The mammals of Africa. An identification manual.— Smithsonian Institution Press, Washington, D.C. Part 1.4, p. 1-3. 1978. The mammals of the Palaearctic Region: a taxonomic review.—British Museum (Natural History), London, 314 pp. , and J. E. Hill. 1980. A world list of mammalian species. — British Museum (Natural History), London, viii + 226 pp. Dixon, W. J. (Ed.). 1973. BMD biomedical computer programs.— University of California Press, Berkeley, vii + 773 pp. Dobson, G. E. 1882. A monograph of the Insectivora, systematic and anatomical. Part 1, including the families Erinaceidae, Centetidae, and Solenodontidae.—John Van Voorst, London, iv + 96 pp. Dorst, J., and P. Dandelot. 1969. A field guide to the larger mammals of Africa.— Houghton Mifflin Company, Boston, 287 pp. Honacki, J. H., K. E. Kinman, and J. W. Koeppl (Eds.). 1982. Mammal species of the world: a taxonomic and geographic reference.— Allen Press, Inc., Lawrence, Kansas, ix + 690 pp. Kruskal, J. B. 1964a. Multidimensional scaling by optimizing goodness of fit for a nonmetric hy- pothesis. — Psychometrika 29:1—27. . 1964b. Nonmetric multidimensional scaling: a numerical method. — Psychometrika 29:1 15- 129. Mayr, E. 1969. Principles of systematic zoology.—McGraw-Hill Book Company, New York, xi + 428 pp. Miller, G.S. 1912. Catalogue of the mammals of Western Europe.—Trustees of the British Museum, London, xv + 1019 pp. Nowak, R. M., and J. L. Paradiso. 1983. Walker’s mammals of the world.—The Johns Hopkins University Press, Baltimore 1 and 2:1—-1362. Power, D. M., and J. R. Tamsitt. 1973. Variation in Phyllostomus discolor (Chiroptera: Phyllosto- matidae).— Canadian Journal of Zoology 51:461—468. Prim, R.C. 1957. Shortest connection networks and some generalizations. — Bell Systems Technique Journal 36:1389-1407. Rohlf, J. J., J. Kishpaugh, and D. Kirk. 1972. Numerical taxonomy system of multivariate statistical programs.—State University of New York, Stony Brook, 87 pp. Simpson, G. G. 1945. The principles of classification and a classification of mammals.— Bulletin of the American Museum of Natural History 85:xv + 350 pp. Thomas, O. 1918. The generic division of the hedgehogs.— Annals and Magazine of Natural History (9)1:193-196. Department of Vertebrate Zoology (Mammals), National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. (HWS) current ad- dress, 4615 NW 43rd Place, Gainesville, Florida 32611. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 121-126 THE CORRECT IDENTITY OF THE PELAGIC AMPHIPOD PRIMNO MACROPA, WITH A DIAGNOSIS OF PRIMNO ABYSSALIS (HYPERIIDEA: PHROSINIDAE) Thomas E. Bowman Abstract.—Two large species of Primno are recognized: P. macropa Guérin- Méneville, an inhabitant of the Subantarctic biotic province, and P. abyssalis (Bowman), a resident of the Subarctic biotic province. Primno macropa is rede- scribed, and records from USNS Eltanin cruises mapped. It is suggested that P. macropa expanded its range into the North Pacific during a period of cooling. The North Pacific population became isolated when warming occurred and evolved into a distinct species, P. abyssalis. In my revision of Primno (Bowman 1978) I identified as the type-species, P. macropa Guérin-Meéneville, a large form that appeared to be confined to Subarctic water in the North Pacific. I made this identification with some misgivings, since the type-locality was “les mers du Chile.” But I did so because the North Pacific form was the only one available to me that matched Guérin-Meéneville’s species in body length and in the form of pereopod 5. I had been unsuccessful in my attempts to obtain large specimens of Primno from the Southern Hemisphere. Recently, however, I received from Brain P. Boden specimens of a large form of Primno from the vicinity of Marion Island in the southern Indian Ocean. This form is very similar to the North Pacific form, but differs consistently in several details. I consider the two forms to be specifically distinct, and the Marion Island form to represent the true Primno macropa. Additional material of the true P. macropa has been identified in collections made during cruises of the USNS Eltanin for the U.S. Antarctic Research Program. Figure 2 shows the locations of these records. The North Pacific species identified as Primno macropa by Bowman (1978) is herein referred to Primno abyssalis (Bowman, in Fulton 1968). Primno abyssalis (Bowman, 1968) Fig. 1A-K Euprimno abyssalis Bowman, 1953:348-354, figs. 45-46, charts 114-116 [un- published].— Bowman, in Fulton, 1968:104, 109. Primno macropa Guérin-Méneville.—Thorsteinson, 1941:93—94, figs. 98—102.— Vinogradov, 1956:209.— Yoo, 1971a:59 [partim]; 1971b, passim; 1972a, pas- sim; 1972b:174.—Sanger, 1973:20, 1974:7.—Lorz and Pearcy, 1975:1445- 1446.—Bowman, 1978:3-8, figs. 1-2, 3a—c, 4.—Brusca 1981:43, fig. 15b.— Semenova, 1982:354—355 [partim], fig. 189. Euprimno macropus (Guérin-Méneville).— Wailes, 1929:161; 1931:41; 1933:9.— Behning, 1939:363. Types.— Holotype, adult 2, 14 mm in length, Scripps Institution of Oceanog- raphy Northern Holiday expedition sta 12, 10 Aug 1951, 40°34’N, 147°54.5’W, 122 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. A-K, Primno abyssalis 2: A, Head, lateral; B, Head, dorsal; C-I, Pereopods 1-7; J, Telson and uropods; K, Posteroventral corner of pleonite 3. L-N, Primno macropa: Posteroventral corners of pleonite 3 of three different females. VOLUME 98, NUMBER 1 123 Isaacs-Kidd midwater trawl, 1020-0 m, USNM 213613. Paratypes, 2 2, 14 mm in length, from same sample, deposited in collections of Scripps Institution of Oceanography. Etymology.—From the Latin “‘abyssus” (bottomless depth) + the adjectival suffix “‘*-alis,”’ referring to the bathypelagic distribution of the species. Diagnosis (emended from Bowman 1978).—Length up to 21 mm. Rostrum truncate, limited to space between antennae |. Middorsal spines more pronounced than in other species of Primno. Pleonite 3 with shallow concavity ventral to posteroventral spine, with or without low defining notch. Antenna | only slightly longer than head. Rudimentary 2 antenna 2 with few inconspicuous setae. Pere- opod 5 carpus with proximal 1-2 teeth short; long teeth slightly shorter than width of carpus. Pereopod 6 basis slightly more than '4 as wide as long, narrowed proximally, proximal part of anterior margin concave; ischium, merus and carpus narrower than in other species of Primno; merus about 7 length of carpus. Pere- opod 7 basis about as long as remaining segments combined. Uropod 3 with well developed medial shoulder. Distribution.—Subarctic biotic province, North Pacific. Primno macropa Guérin-Meéneville Fig. 1L—-N Primno macropa Guérin-Méneville, 1836:4, pl. 17, fig. la—f.—Bovallius, 1887: 28.—?Spandl, 1927:168—169.— Barnard, 1930:424—425 [partim]; 1932:287-288 [partim].— Mackintosh, 1934, passim.— Hardy and Gunther, 1935, passim.— Hurley, 1955:172-174, figs. 219-235; 1969:33, map 7.— Vinogradov, 1962: 22.—Semenova, 1982:354-355 [partim]. Primno menevillei Stebbing, 1888:1447-1448, pl. 179B. Primno antarctica Stebbing, 1888:1448-1451, pl. 209B. Euprimno macropus (Guérin-Méneville).—Bovallius, 1889:400—407 [partim].— Walker, 1907:9. Euprimno macropa Guérin-Meéneville var. menevillei Stebbing.— Monod, 1926: 50-51, fig. 49. Material. — Near Marion Island, Prince Edward Islands: Sta 6, no. 133, 46°52’S, 37°54’E, 100-0 m, 5 May 1983, 1 9; Sta 11d, 44°42’S, 22°30'W, 1000-500 m, 21 May 1983, 4 2; Sta 18, no. 41, 46°38’S, 38°05E, 100-0 m, 24 May 1983, 1 2. From collections of USNS £Eltanin in the Southern Ocean: localities shown in Fig. 2 (complete station data available upon request). Diagnosis.—Similar to Primno abyssalis but with the following differences: Body shorter, up to about 15 mm in length. Rostrum broad-truncate, conforming to contour of head, about 0.4 width of head. Middorsal spines slightly less well developed. Pleonite 3 with deeper concavity ventral to posteroventral spine, de- fined by a tooth. 2 antenna 1 about 4% longer than head. Rudimentary 2 antenna 2 with well developed cluster of delicate setae. Pereopod 6 not so slender; anterior margin of basis convex throughout length; merus about half length of carpus; posterior margin of propus with minute setae. Distribution. —Subantarctic biotic province. Relationships.—Primno macropa and P. abyssalis differ from the other species of Primno in their large size and in the pattern of teeth on the pereopod 5 carpus. 124 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON | acl Lf Fig. 2. Records of Primno macropa from cruises of USNS Eltanin (dots) and from Marion Island stations (stars). They also live at greater depths and in higher latitudes than the other species. I suggest that P. macropa is the older of the two large species and gave rise to P. abyssalis as follows: Originally P. macropa or its progenitor was limited to the Southern Hemisphere and there was no large Primno in the North Pacific. Cooling of the oceans by a few degrees during an ice age permitted P. macropa to expand its range into the North Pacific by way of the western coasts of South and North America. When the oceans warmed again the continuity was broken, leaving the North Pacific population isolated from that of the Southern Hemisphere. Since this isolation, the populations developed the small but constant differences that now require their recognition as distinct species. A number of species of zooplankton and nekton have a distribution pattern like that of P. macropa-abyssalis combined, occupying what McGowan (1974) calls the Subarctic and Subantarctic biotic province. VOLUME 98, NUMBER 1 125 Literature Cited Barnard, K. H. 1930. Amphipoda.—British Antarctic (“Terra Nova’’) Expedition, 1910, Natural History Report, Zoology 8(4):307—454. 1932. Amphipoda.— Discovery Reports 5:1—326, pl. 1. Behning, A. L. 1939. Die Amphipoda-Hyperiidea der den Fernen Osten der UdSSR umgrenzenden Meere. — Internationale Revue der Gesamten Hydrobiologie und Hydrographie 38:353—367. Bovallius, C. 1887. Systematical list of the Amphipoda Hyperiidea.—Bihang till Kungliga Vetens- kapsakademiens Handlingar 11(16):1—50. 1889. Contributions to a monograph of the Amphipoda Hyperiidea, part I:2, the families Cyllopodidae, Paraphronimidae, Thaumatopsidae, Mimonectidae, Hyperiidae, Phronimidae, and Anchylomeridae.— Kongliga Svenska Vetenskapsakademiens Handlingar 22(7):1—434, pls. 1-18. Bowman, T. E. 1953. The systematics and distribution of pelagic amphipods of the families Vibi- liidae, Paraphronimidae, Hyperiidae, Dairellidae, and Phrosinidae from the northeastern Pa- cific.— Unpublished Ph.D. thesis, University of California, Los Angeles, 430 pp. 1978. Revision of the pelagic amphipod genus Primno (Hyperiidea: Phrosinidae). —Smith- sonian Contributions to Zoology 275:1-23. Brusca, G. J. 1981. Annotated keys to the Hyperiidea (Crustacea: Amphipoda) of North American coastal waters.— Technical Reports of the Allan Hancock Foundation 5:1-76. Fulton, J. 1968. A laboratory manual for the identification of British Columbia marine zooplank- ton.—Fisheries Research Board of Canada Technical Report 55:1-141. Guérin-Méneville, F.-E. 1836. Description de quelques genres nouveaux de Crustacés appartenant a la famille des Hypérines.— Magazin de Zoologie 6(7):1-10, pls. 17-18. Hardy, A. C., and E. R. Gunther. 1935. The plankton of the South Georgia whaling grounds and adjacent waters, 1926—1927.—Discovery Reports 11:1—456. Hurley, D. E. 1955. Pelagic amphipods of the sub-order Hyperiidea in New Zealand waters, I: Systematics. — Transactions of the Royal Society of New Zealand 83(1):119-194. . 1969. Amphipoda Hyperiidea.— Antarctic map folio series, folio 11: Distribution of selected groups of marine invertebrates in waters south of 35°S latitude:32—34, sheets 1-2.— American Geographical Society. Lorz, H., and W. G. Pearcy. 1975. Distribution -of hyperiid amphipods off the Oregon coast.— Journal of the Fisheries Research Board of Canada 32(8):1442-1447. McGowan, J. A. 1974. The nature of oceanic ecosystems. Jn C. B. Miller, ed., The biology of the oceanic Pacific, 157 pp.,—Oregon State University Press, Corvallis, pp. 9-28. Mackintosh, N. A. 1934. Distribution of the macroplankton in the Atlantic sector of the Antarctic. — Discovery Reports 9:65—160. : Monod, T. 1926. Tanaidacés, isopodes et amphipodes.— Expédition Antarctique Belge, Résultats du Voyage de la Belgica en 1897-99, Rapports Scientifiques, Zoologie: 1-67. Sanger, G. A. 1973. Epipelagic amphipods (Crustacea) off Washington and British Columbia, Oc- tober—November 1971.—Northwest Fisheries Center MARMAP Survey I, Report 8:1-29. . 1974. Pelagic amphipod crustaceans from the southeastern Bering Sea, June 1971.—NOAA Technical Report National Marine Fisheries Service, Special Scientific Report, Fisheries 680: 1-8. Semenova, T. N. 1982. XIV. Fam. Phrosinidae Dana, 1853. In M. E. Vinogradov, A. F. Volkov, and T. N. Semenova, Amphipoda-Hyperiidea of the world ocean, 493 pp., ““Nauka,”’ Leningrad (Opredeliteli po Faune SSSR, Zoologicheski Instituta Akademii Nauk SSSR 132), pp. 348-357. [In Russian]. Spandl, H. 1927. Die Hyperiiden (exkl. Hyperiidea Gammaroidea und Phronimidae) der Deutsche Stidpolar-Expedition 1901—1903.— Deutsche Siidpolar-Expedition 1901-1903, 19 (Zoologie 11): 145-287, pl. 10. Stebbing, T. R.R. 1888. Report on the Amphipoda collected 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 29:xxiv + 1713 pp., 210 pls. [in 3 volumes]. Thorsteinson, E.D. 1941. New or noteworthy amphipods from the North Pacific coast. — University of Washington Publications in Oceanography 4(2):50-96. Vinogradov, M. E. 1956. Hyperiids (Amphipoda-Hyperiidea) of the western Bering Sea. —Zoolo- gicheskii Zhurnal 35(2):194—218. [In Russian]. 126 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON —. 1962. Hyperiids (Amphipoda, Hyperiidea) collected by the Soviet Antarctic Expedition on M/V Ob south of 40°S.— Biological Results of the Soviet Antarctic Expedition (1955-1958) 1: 5-35. [In Russian, English translation by Israel Program for Scientific Translations published in 1966.] Wailes, G. H. 1929. The marine zoo-plankton of British Columbia.—Museum and Art Notes, Vancouver 4:159-165. 1931. Amphipoda from British Columbia, part II.—Museum and Art Notes, Vancouver 6: 40-41. 1933. The plankton of the west coast of Vancouver Island, British Columbia.—Museum and Art Notes, Vancouver 7 (supplement 9):1—11. Walker, A.O. 1907. Crustacea, III: Amphipoda.— National Antarctic Expedition 1901-1904 (“Dis- covery’), Natural History 3:1—38, pls. 1-13. Yoo, K.I. 197la. Pelagic hyperiids (Amphipoda-Hyperiidea) of the western North Pacific Ocean. — Journal of the National Academy of Sciences, Republic of Korea, Natural Sciences Series 19: 39-89. —. 1971b. The biology of the pelagic amphipod, Primno macropa Guér., in the western North Pacific, 1: Systematics.— Korean Journal of Zoology 14(3):132-138. 1972a. The biology of the pelagic amphipod, Primno macropa Guér., in the western North Pacific, 2: Geographical distribution and vertical distributional pattern.—Korean Journal of Zoology 15(2):87-91. 1972b. Faunal studies on the amphipods in Korea.— Floral studies on some taxa of plants and faunal studies on some taxa of animals in Korea, R-72—82:163-181.— Ministry of Science and Technology. Department of Invertebrate Zoology, NHB 163, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 127-149 POLYCHAETE WORMS FROM A CAVE IN THE BAHAMAS AND FROM EXPERIMENTAL WOOD PANELS .IN DEEP WATER OF THE NORTH ATLANTIC (POLY NOIDAE: MACELLICEPHALINAE, HARMOTHOINAE) Marian H. Pettibone Abstract. — Pelagomacellicephala iliffei, n. gen., n. sp. a macellicephalin polynoid is described from a cave in Middle Caicos Island, Bathykermadeca turnerae, n. sp. and Bathybahamas charleneae, n. gen., n. sp. also in the Macellicephalinae, and Harmothoe ingolfiana and H. vagabunda, n. sp. in the Harmothoinae are described from four experimental “‘wood islands” in deep water of the North Atlantic. Of particular interest is a recent collection of some polynoid polychaetes from Conch Bar Cave, Turks and Caicos islands in the Bahamas, belonging to the Macellicephalinae, a subfamily of the Polynoidae known primarily from the deep sea, including bathyal and abyssal depths (Pettibone 1976). Four specimens were collected by Thomas M. Iliffe and Jill Yager in connection with their studies of cave faunas. They are referred, herein, to Pelagomacellicephala iliffei, n. gen., n. sp. The following comments on the habitat of the worms were kindly furnished by Dr. Iliffe. The Turks and Caicos Islands are geologically and geographically a southeast- ward extension of the Bahamas. The Bahama Platform, including the Caicos Bank, consists of a cap of shallow water carbonates thicker than the surrounding ocean is deep (Dietz et al. 1970). This suggests that a shallow water depositional en- vironment has been maintained in this region since at least the early Cretaceous. Slow subsidence, offset by deposition of coral and algal carbonates, has maintained the top of the platform nearly at sea level. The Caicos Bank is a shallow water platform bordered by a series of islands extending along the northern and eastern margins. The major islands consist of a broad expanse of low lying flat land facing the interior of the platform, while a range or series of ranges of eolian carbonate hills 20 to 40 m in elevation fronts the seaward margin, parallel to the long axis of the islands. The youngest hills closest to the seashore are unconsolidated dunes consisting of reef-derived car- bonate sand washed ashore and then blown inland by the prevailing trade winds. Those hills further inland are older, probably Pleistocene in age, and have been lithified into a hard eolian limestone. Caves are situated within these lithified dunes, while large circular collapsed shafts, called ““blue holes’’ (Dill 1977), are found in shallow waters of the interior platforms. Conch Bar Cave, reportedly in excess of 2.5 km long (Gregor 1981), is the largest and most significant cave in the Turks and Caicos group. The cave is developed in a coastal dune-derived hill (Conch Bar Hill) located 500 m inland from the open ocean. It consists of multiple levels, the lowest of which is per- manently flooded with tidal brackish waters 10 or more meters deep. The four 128 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON polynoids were collected from a broad shallow pool containing considerable amounts of organic detritus including leaves, twigs, and land snail shells in the sediments. Surface salinity in the pool was 23%o. Other animals found in the cave include the amphipod Spelaeonicippe provo, large numbers of the shrimps Bar- bouria cubensis and Typhlatya garciae, a new species of troglobitic mysid Sty- giomysis clarkei described by Bowman et al. (1984) and several copepods now under study. Another troglobitic polynoid was found inhabiting a sea water-flooded lava tube cave in the Canary Islands, the Jameos del Agua, collected by Wilkens and Par- zefall (1974). It was described as Macellicephala jameensis in the Macellicephal- inae by Hartmann-Schroder (1974), later revised and referred to a new genus, Gesiella, in the Harmothoinae by Pettibone (1976:64), and, still later, referred to a new subfamily, Gesiellinae, by Muir (1982:173). This extensive lava cave, despite being only 3000 to 5000 years old, contains fauna of both Tethyan and deep sea origins (Wilkens and Parzefall 1974; Iliffe et al. (1984). A number of studies have pointed out the importance of plant remains of shallow-water or epipelagic origin (wood, leaves, fruit, sea grass, Sargassum, etc.) to the macrofauna of the deep sea, and these are summarized by Torben Wolff (1979). For instance, wood is used as a substrate and food for opportunistic boring bivalve mollusks, such as Xylophaga. Such macrofaunal herbivores are available as prey for predators, as dead remains for scavengers, and their fecal pellets can be utilized by deposit feeders. Polychaetes were found to be the dominant group, seeking shelter under the bark and in cavities of the wood. Four species of deep-sea polynoids have been identified from wood panels that were submerged and later retrieved after one to three years at four permanent DSRV Alvin deep Atlantic bottom stations, in connection with the experimental studies being carried out by Ruth Turner on wood-boring mollusks (Pholadidae, Xylophagainae: Xylophaga and Xylorado). Turner (1973) found that after three and a half months some wood panels that had been pushed into the bottom sediment at a depth of 1830 meters were riddled by the burrows of the bivalve mollusks. These deep-sea wood-borers are opportunistic species and “‘.. . are the most important deep-sea organisms involved in converting woody plant material to available food sources (1) in the form of fecal pellets for detritus feeders, (2) as larvae or adults, exposed by distintegration of the wood, for predators, and (3) as dead remains for scavengers” (Turner 1973:1379). In connection with the ““wood-island”’ and panel experiments of Turner (1977, 1981), large ““wood islands” (12 blocks of spruce, 30 cm on a side) were placed by DSRV Alvin at two Deep Ocean Stations (DOS) south and southeast of Woods Hole, Massachusetts, in 1830 and 3506 meters, one in the Tongue of the Ocean (TOTO, TOWER 3), the deep “fjord” in the shallow Great Bahama Bank, in 2066 meters, and one off St. Croix in 3995 meters. Around these “‘wood islands,” a series of wood panels (60 by 15 by 3 cm) were pushed in the mud. When the areas were visited later, some panels were retrieved and replaced after variable lengths of time, from 9 months to three years. The panels were placed in retrieval boxes for fixation of the specimens at bottom temperature and pressure, assuring the best possible condition of the specimens. From the panels riddled by the boring xylophagids, a great variety of other organisms was collected, including detritus feeders and predators, and, among the latter, numerous polynoid worms. VOLUME 98, NUMBER 1 129 Table 1.—Polychaetes identified from wood panels in North Atlantic. DSRV Alvin “Wood Island” Stations Species DOS Ue TE DOS*) DN ELONOLowcr sin TNLSCzCroln nan = Polynoidae: Macellicephalinae: Bathykermadeca turnerae, n. 2 spec. 10 spec. 33 spec. 51 spec. sp. (some young) (many young) (many young) Bathybahamas charleneae, n. 2 spec. gen., n. sp. Harmothoinae: Harmothoe ingolfiana Ditlev- 372 spec. 184 spec. sen (many young) (many young) Harmothoe vagabunda, n. sp. 7 spec. 36 spec. (many young) Ampharetidae:* Decemunciger apalea Zottoli 69 spec. 26 spec. 1 spec. (64 juv.) (10 juv.) Guv.) Endecamera palea Zottoli 18 spec. * See Zottoli (1982). After preliminary sorting by Helene Ferranti and Charlene D. Long, assistants to Ruth Turner, the polynoid worms were sent to me for identification. From this collection of Polynoidae, four species were identified, two belonging to the usually deep water Macellicephalinae and two in the Harmothoinae. Zottoli (1982) re- ported on the polychaetes of the sedentary tube-dwelling family Ampharetidae, which were collected on some of the same panels. The collection sites and species of Polynoidae and Ampharetidae are sum- marized below and in Table 1. Locations of four experimental “wood islands”: 1. Deep Ocean Station-1 (DOS-1), 110 miles south of Woods Hole, Massa- chusetts, 39°46’N, 70°41’W, in 1830 meters. 2. Deep Ocean Station-2 (DOS-2), 190 miles southeast of Woods Hole, Mas- sachusetts, 38°18'24”N, 69°35'36”W, in 3506 meters. 3. Tongue of the Ocean, Bahama Islands (TOTO, TOWER 3), 24°53’12’N, 77°40'12”W, in 2066 meters. 4. Off St. Croix, Virgin Islands, 17°57'36”N, 64°48'36”W, in 3995 meters. The types and additional specimens are deposited in the National Museum of Natural History, Smithsonian Institution (USNM). Types of Harmothoe ingolf- iana, deposited in the Zoological Museum, Copenhagen (ZMC), were examined. Family Polynoidae Subfamily Macellicephalinae Hartmann-Schroder, 1971 Pettibone (1976:6—42) included seven genera in the subfamily Macellicephal- inae, the type-genus Macellicephala McIntosh, 1876, and six new genera: Bathy- catalina, Bathyeliasona, Bathyfauvelia, Bathykermadeca, Bathykurila, and Bathy- vitiazia. Pettibone (1979:384—388) emended Bruunilla Hartman, 1971, and added 130 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON it to the subfamily. Levenstein (1981:27—29) added Bathypolaria and Bathy- miranda. Levenstein (1982b:1291-—1296) added Bathytasmania and Bathynotalia and provided a key to the genera of the Macellicephalinae. However, her figure of the type-species of Bathynotalia, B. perplexa, shows the prostomium with lateral antennae (not mentioned in the text), in addition to the median antenna, palps and tentacular cirri (fig. e, in Levenstein 1982b). On this basis the genus is herein referred to Harmothoinae. Further, the prostomium of Bathymiranda mi- crocephala Levenstein (1981, fig. 2a) lacks both lateral and median antennae and, on this account, is referred herein to Polaruschakovinae Pettibone, 1976. Diagnosis of Macellicephalinae.— Body short, with relatively few segments (15— 24). Prostomium bilobed; median antenna with distinct ceratophore and distal style; paired palps; without lateral antennae, with or without tapered frontal fil- aments; without eyes. Modified first or tentacular segment more or less fused to prostomium; tentaculophores lateral to prostomium, each with acicular process of variable development, with or without few setae and pair of dorsal and ventral tentacular cirri. Facial tubercle anterior and ventral to prostomium, more or less developed. Second or buccal segment with first pair of elytra, biramous parapodia and long ventral or buccal cirri attached basally on neuropodia lateral to ventral mouth. Muscular pharynx eversible, with 2 pairs of jaws and up to 9 pairs of papillae around opening. Parapodia long, biramous, both notopodia and neuro- podia subconical with elongate acicular processes. Paired elytra and bulbous elyt- rophores relative few (7-12 pairs), on segments 2, 4, 5, 7, continuing on alternate segments to near end of body. Dorsal cirri with cylindrical cirrophores and distal styles, attached posterior to notopodia on segments 3, 6, 8, continuing on alternate and posterior segments lacking elytra. Dorsal tubercles on cirrigerous segments, corresponding in position to elytrophores, indistinct or more or less developed. Ventral cirri short, on ventral side of neuropodia on all segments. Pygidium variable in position and size, with dorsal anus and pair of anal cirri. Ventral nephridial or segmental papillae indistinct or variously developed on certain segments. Three genera, each with a single species, are covered in this report: Pelago- macellicephala iliffei, new genus, new species; Bathykermadeca turnerae, new species; Bathybahamas charleneae, new genus, new species. Pelagomacellicephala, new genus Type-species.— Pelagomacellicephala iliffei, new species. Gender: feminine. Diagnosis.—Segments up to 21. Elytra and elytrophores 9 pairs, on segments 2, 4,5, 7,9, 11, 13, 15, and 17. Bilobed prostomium with rounded anterior lobes, without frontal filaments, with median antenna and palps, without lateral antennae or eyes. Tentaculophores of tentacular segment with 2 pairs of tentacular cirri, without projecting acicular process or setae. Dorsal cirri on non-elytra-bearing segments, with long styles. Dorsal tubercles on cirrigerous segments short, nodular. Parapodia long, biramous, both rami with projecting acicular processes, notopodia much shorter and smaller than neuropodia. Notosetae few, shorter and subequal in width to neurosetae, with double rows of close-set thickened serrations along one side. Neurosetae numerous, delicate, flattened, with serrate lateral borders. Ventral cirri short, but long on segment 2. Pharynx with pair of extra long mid- VOLUME 98, NUMBER 1 131 dorsal and midventral papillae and 4 pairs of short lateral papillae; 2 pairs of jaws with basal teeth. Etymology. —Pelago, from pelagos, Greek, sea, plus macellicephala, indicating the close relationship of this sea worm to the type-genus of the subfamily. Gender: feminine. Pelagomacellicephala iliffei, new species Fig. 1 Material examined.—BAHAMAS: Turks and Caicos Islands, Middle Caicos Island, Conch Bar Cave, 28 Oct 1982, collected by hand with suction bottle while snorkeling in large 1 to 1.5 m deep tidal pool in totally dark section of cave, Thomas M. Iliffe and Jill Yager, collectors; holotype, USNM 96262, and 3 para- types, USNM 96263-5. Description.—Length of holotype (largest specimen) 9 mm, width 5 mm in- cluding setae, segments 21, last one small. Length of smallest young paratype (USNM 96264) 5 mm, width 2.5 mm, segments 19. Body flattened, tapered anteriorly and posteriorly, with long parapodia angled downward. Color: white, translucent, giving general appearance of pelagic polychaete. Elytra and bulbous elytrophores 9 pairs, on segments 2, 4, 5, 7, 9, 11, 13, 15, and 17 (Fig. 1A, C). Elytra large, covering dorsum, delicate, transparent, showing “veins,” without tubercles or papillae (Fig. 1G). Dorsal cirri on non-elytra-bearing segments; dorsal tubercles on cirrigerous segments from segment 8, short, nodular (Fig. 1B, D). Prostomium deeply bilobed, with anterior lobes rounded; median antenna with small ceratophore in middle of prostomium, with very long, slender, smooth style; palps stout, long, tapered (Fig. 1A). Tentacular segment not distinct dorsally; tentaculophores lateral to prostomium, with 2 pairs of long dorsal and ventral tentacular cirri, those dorsal longer than ventral ones; bilobed facial tubercle anterior to prostomium (Fig. 1A). Second or buccal segment with first pair of large elytrophores, well-developed biramous parapodia and ventral or buccal cirri similar to tentacular cirri, attached to basal part of neuropodia lateral to ventral mouth (Fig. 1A). Pharynx with especially long middorsal and midventral papillae and 4 pairs of small papillae near lateral borders (Fig. 1H); 2 pairs of jaws with basal teeth (11-15) (Fig. 11). Parapodia biramous, as long as body width (Fig. 1A—D). Notopodia much shorter than neuropodia, subconical, with projecting acicular process on lower side. Notosetae few (up to 6), delicate, slightly curved, with 2 rows of close-set thickened serrations along one side and blunt tips (Fig. 1E). Neuropodia large, with projecting triangular acicular process, diagonally truncate supraacicular area and slightly rounded, truncate, subacicular area. Neurosetae numerous, forming fan-shaped bundle, delicate, fragile (many broken), flattened, serrate along lateral borders (Fig. 1F). Dorsal cirri with cylindrical cirrophores attached to postero- dorsal side of notopodia; styles delicate, transparent, smooth, variable in length, some extending to tips of setae or far beyond (on small paratype, some dorsal cirri up to half length of animal) (Fig. 1D). Ventral cirri short, tapered, attached close to distal end of neuropodia (Fig. 1C, D). Pygidium bulbous, enclosed by parapodia of posterior segments, with dorsal 132 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Pelagomacellicephala iliffei, A-G, holotype, USNM 96262; H, I, paratype, USNM 96265: A, Anterior end, dorsal view; B, Posterior end, dorsal view, including cirrigerous segments 19-21; styles of dorsal cirri missing on segment 19; C, Right elytrigerous parapodium segment 9, anterior view, acicula dotted; D, Right cirrigerous parapodium segment 10, posterior view; E, Notoseta; F, Neuroseta; G, Two left elytra; H, Dorsal view of extended pharynx, turned slightly to nght; I, Jaw. Scales: = 0.5 mm for A, B, G, H; 0.5 mm for C, D; 0.1 mm for E, F, I. VOLUME 98, NUMBER | 133 anus and rounded ventral lobe, anal cirri missing (Fig. 1B). Ventral segmental or nephridial papillae indistinct, none enlarged. Distribution. — Known only from anchialine habitat of Conch Bar Cave, Middle Caicos, Turks and Caicos Islands, Bahamas. (See Introduction.) Etymology.—The species is named for Thomas M. Iliffe, one of the collectors, in recognition of his interesting studies on cave faunas. Comparisons. — The pharynx of Pelagomacellicephala differs from all other gen- era in the Macellicephalinae in having especially long middorsal and midventral papillae, small lateral papillae and jaws with a row of basal teeth. Pelagomacel- licephala iliffei may be separated from the other two species of Macellicephalinae described herein according to the Key on page 142. Pelagomacellicephala iliffei agrees in a number of respects with Gesiella ja- meensis, found in a lava tunnel cave in the Canary Islands: both are short-bodied, translucent, with 9 pairs of delicate elytra, similar parapodia, and lack eyes. The pharynx of G. jameensis is of the usual polynoid type, with 9 pairs of papillae and 2 pairs of jaws without basal teeth; the pharynx of P. i/iffei is unique, as indicated above. The prostomium of P. i/iffei lacks lateral antennae, placing it in the Macellicephalinae; G. jameensis has small lateral antennae inserted ventrally, placing it in the Harmothoinae (Pettibone 1976:60) or in Gesiellinae, based on the unique accessory filamentous organs on the cirrophores of the dorsal cirri (Muir 1982:173). Bathykermadeca Pettibone, 1976 Type-species.— Macellicephala hadalis Kirkegaard, 1956. Gender: feminine. Diagnosis.—Segments 21. Elytra and elytrophores 9 pairs, on segments 2, 4, 5, 7, 9, 11, 13, 15, and 17. Prostomium deeply bilobed, conical anterior lobes with frontal filaments; ceratophore of median antenna inserted in anterior notch, with paired palps; without lateral antennae or eyes. Tentaculophores of tentacular segment lateral to prostomium, with 2 pairs of tentacular cirri and projecting acicular process, without setae. Dorsal cirri on non-elytra-bearing segments except segment 19 (lacks both elytra and dorsal cirri). Dorsal tubercles on cirrigerous segments inconspicuous or nodular. Parapodia biramous, both rami with elongate acicular processes, subequal in length. Notosetae subequal in length to and stouter than neurosetae, with 2 rows of spines and blunt tips. Neurosetae of 1 or 2 types. Posterior 4 segments somewhat modified and compressed. Pharynx with 7 pairs of dorsal and ventral papillae; 2 pairs of jaws with or without basal teeth. A single species, B. hadalis, was previously referred to Bathykermadeca, as Macellicephala hadalis Kirkegaard, 1956, from South Pacific in the Kermadec Trench, in 6600-6720 meters. Additional records of this hadal species have been added by Levenstein (1978:165) from the Yap Trench in 8560-8720 meters, and by Levenstein (1982a:59) from the Japan Trench in 7350-7370 meters. A new species from the North Atlantic is here added, B. turnerae. Bathykermadeca turnerae, new species Figs. 2, 3 Material examined.—North Atlantic off St. Croix, Virgin Islands, 17°57'36’N, 64°48'36”W, 3995 m, R. D. Turner Panel Study. Panels submerged on A/vin dive 134 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Bathykermadeca turnerae, A, paratype, USNM 96246; B-G, holotype, USNM 96245; H, paratype, USNM 96247: A, Anterior end, dorsal view; B, Right cirrigerous parapodium, posterior view; C, Right elytrigerous parapodium, anterior view, acicula dotted; D, Right half of segments 7 and 8, dorsal view; E, Short and long notosetae; F, Supra-acicular neuroseta; G, Subacicular neuroseta; H, Jaw. Scales: = 1.0 mm for A; 0.5 mm for B, C; 2.0 mm for D; 0.1 mm for E-G; 0.1 mm for H. 873, 17 Dec 1978: P-4 removed on dive 1080, 7 Dec 1980, holotype (USNM 96245); P-5, P-13 removed dive 1079, 6 Dec 1980, 22 paratypes (USNM 96248); P-12 removed dive 1083, 7 Dec 1980, paratype (USNM 96246), 28 paratypes (USNM 96247). North Atlantic, Bahamas, Tongue of the Ocean (TOTO), 24°53'12’N, VOLUME 98, NUMBER 1 135 77°40'12” W, 2066 m, R. D. Turner Panel Study: Panel T-A submerged on A/vin dives 492, 493, 20, 21 Jan 1974; removed dive 751, 8 May 1977, paratype (USNM 69249).—Panel T-1 submerged dive 552, 19 Apr 1975; removed dive 753, 10 May 1977, paratype (USNM 96250).— Panel 4A/T86 submerged dive 563, 6 May 1975; removed dive 753, 10 May 1977, paratype (USNM 96251).—Panel T-16 submerged dive 564, 8 May 1975; removed dive 755, 12 May 1977, 4 paratypes (USNM 96253).—Panels T-17, T-18 submerged dive 564, 8 May 1975; removed dive 752, 9 May 1977, 8 paratypes (USNM 96254).— Panel T-20 submerged dive 564, 8 May 1975; removed dive 755, 11 May 1977, 4 paratypes (USNM 96255).— Panels T-37, T-56 submerged dives 752, 755, 9, 12 May 1977; removed dive 851, 11 Nov 1978, 8 paratypes (USNM 96256).—Panels T-15, T-36 submerged dives 752, 755, 9, 12 May 1977; removed dive 852, 12 Nov 1978, 5 paratypes (USNM 96252).— Panel T-85, no data, paratype (96257). North Atlantic Deep Ocean Stations (DOS), R. D. Turner Panel Study: DOS- 1, 110 miles south of Woods Hole, Massachusetts, 39°46'N, 70°41'W, 1830 m: Panel N-34 submerged on Alvin dive 597, 30 Aug 1975; removed dive 773, 29 Jul 1977, 2 paratypes (USNM 96258).—DOS-2, 190 miles SE of Woods Hole, 38°18'24"N, 69°35'36”W, 3506 m: Panel N-42 submerged dive 602, 6 Sep 1975; removed dive 792, 26 Sep 1977, 4 small paratypes (USNM 96259).— Panels N-45, N-62 submerged dives 657, 681, 10 Jun, 12 Aug 1976; removed dive 790, 23 Sep 1977, 5 small paratypes (USNM 96260: Panel N-95 submerged dive 817, 29 Jun 1978; removed dive 1026, 27 Jun 1980, paratype (USNM 96261). Description. — Length of holotype (USNM 96245), largest specimen from off St. Croix, 17 mm, width including setae 11 mm, segments 21, last 2 very small. Largest paratype from Deep Ocean Station south of Woods Hole (USNM 96258) 14 mm long, 10 mm wide, with 21 segments. Smallest young paratype from off Bahamas (USNM 96255) 1 mm long, 1.2 mm wide, with 12 segments plus bulbous growing zone. Body oval, greatly flattened, tapering anteriorly and posteriorly, with very long parapodia and setae projecting posteriorly. No color. Dorsum with transverse ciliated bands, up to 4 per segment, continuing on bases of elytrophores and dorsal tubercles (Fig. 2D). Elytrophores 9 pairs, large, bulbous, on segments 2, 4, 5, 7, 9, 11, 13, 15, and 17 (Figs. 2A, C, D; 3B). Elytra large, oval, covering dorsum, opaque white, without tubercles or papillae, sometimes with slightly undulate posterior border; first elytra with small anterior notch (Fig. 3G, H). Dorsal cirri and nodular dorsal tubercles on segments 3, 6, 8, 10, 12, 14, 16 (Fig. 2B, D). Parapodia of posterior segments (18-21) somewhat modified and compressed (Fig. 3A—F; see below). Prostomium deeply bilobed, anterior lobes conical, with frontal filaments; me- dian antenna with short cylindrical ceratophore and short tapered style; palps long, stout, tapered (Fig. 2A). Tentacular segment not distinct dorsally; tentacu- lophores with delicate acicular process on inner side; pair of long dorsal and ventral tentacular cirri, ventral ones shorter than dorsal; prominent bilobed facial tubercle anterior to prostomium (Fig. 2A). Second or buccal segment with first pair of large elytrophores, well-developed biramous parapodia and ventral or buccal cirri similar to tentacular cirri, with short cirrophores on basal part of neuropodia lateral to ventral mouth. Pharynx with 7 pairs of subequal papillae around opening; 2 pairs of dark reddish-striped jaws with 5 teeth on edge (Fig. 2H). 136 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. Bathykermadeca turnerae, A-F, holotype, USNM 96246; G-I, paratype, USNM 96247: A, Ventral view posterior end including segments 18—21 (parapodia of 20 and 21 mostly hidden from view); B, Dorsal view posterior end including segments 17-21; C, Left cirrigerous parapodium of segment 18, posterior view; D, Left parapodium of segment 19, anterior view, acicula dotted; no elytrophore or dorsal cirrus present; E, Left parapodium of segment 20, anterior view, acicula dotted; F, Left parapodium of segment 21, posterior view, acicula dotted; G, First right elytron; H, Middle right elytron; I, Ventral view of left half of segments 12 and 13 showing segmental papillae. Scales: = 1.0 mm for A, B; 0.5 mm for C—F; 1.0 mm for G, H; 2.0 mm for I. Parapodia long (Fig. 2B, C). Notopodia well developed, conical, with long projecting acicular process on lower side, almost as long as neuropodia. Notosetae numerous, forming radiating bundle, short to long, as long as and much stouter than neurosetae, acicular, nearly straight, with blunt bare tips and 2 rows of widely- spaced spines on distal part (Fig. 2E). Neuropodia subconical, with projecting acicular process on anterior lobe; posterior lobe shorter, rounded. Neurosetae very numerous, long, slender, of 2 types: supra-acicular ones with widely-spaced spines VOLUME 98, NUMBER 1 137 along lateral borders and tapered pointed tips (Fig. 2F); subacicular ones with numerous close-set spines along basal two-thirds and slightly hooked bare tips (Fig. 2G). Dorsal cirri with elongate cylindrical cirrophores attached on posterior sides of notopodia; styles long, slender, smooth, with tapered tips, extending beyond tips of setae (Fig. 2B). Ventral cirri short, tapered, attached on middle of neuropodia (Fig. 2B, C). Segments posterior to elytrigerous segment 17 somewhat modified and com- pressed (Fig. 3A, B). Parapodia of segment 18 similar to more anterior cirrigerous parapodia with long dorsal cirri, except for absence of dorsal tubercles; parapodia long, directed posteriorly and enclosing shorter parapodia of segments 19-21 (Fig. 3A-C). Segment 19 lacking both elytrophores and dorsal cirri; small biramous parapodia enclosing still smaller parapodia of segments 20 and 21 (Fig. 3A, B, D). Parapodia of segment 20 with cirrophore of dorsal cirrus with short style, fused to projecting notopodial acicular process, without notosetae; neuropodium with tongue-like acicular lobe, rounded distally, with small bundle of slender neurosetae and small ventral cirrus (Fig. 3E). Notopodia of segment 21 similar to segment 20, neuropodia represented by small acicular lobe, lacking neurosetae and ventral cirrus (Fig. 3F). Anal ridge visible on dorsal side of segments 17 to 19, with anal opening on segment 18 (Fig. 3B). Pygidium small squarish lobe between bases of parapodia 19 and ventral to parapodia of 20 and 21, with pair of long anal cirri (Fig. 3A). Four pairs of small segmental or nephridial papillae on ventral side of segments 12-15 (Fig. 3]). Distribution.—North Atlantic, off St. Croix, Virgin Islands, off the Bahamas, and south and southeast of Woods Hole, Massachusetts, in depths from 1830 to 3995 meters, associated with wood panels. See Introduction. Etymology.—The species is named for Dr. Ruth D. Turner, whose interesting Alvin Deep-sea Wood and Panel Experiments furnished the specimens on which this study is based. Comparisons. — Bathykermadeca turnerae agrees in most respects with B. had- alis (Kirkegaard), including the presence of the four posteriormost segments being somewhat modified and compressed, and with both elytrophores and dorsal cirri lacking on segment 19. They may be separated, however, on the following char- acters: B. hadalis (Kirkegaard) B. turnerae, n. sp. Dorsal tubercles Inconspicuous Nodular Neurosetae One type Two types Jaws Without basal teeth With basal teeth Nephridial papillae 6 large pairs on segments 4 small pairs on segments 12-17 12-15 Bathykermadeca turnerae may be separated from the other two species of Ma- cellicephalinae described herein according to the Key on page 142. Bathybahamas, new genus Type-species.— Bathybahamas charleneae, new species. Gender: feminine. Diagnosis.—Segments 18. Elytra and elytrophores 8 pairs, on segments 2, 4, 5, 138 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 7, 9, 11, 13, and 15. Bilobed prostomium with blunt anterior peaks, without frontal filaments, with median antenna and palps, without lateral antennae or eyes. Tentaculophores of tentacular segment with 2 pairs of tentacular cirri and projecting acicular process, without setae. Facial tubercle trilobed. Dorsal cirri on non-elytra-bearing segments, with long styles; dorsal tubercles on cirrigerous seg- ments with digitiform ciliated extensions. Parapodia long, biramous, both rami with projecting acicular processes, subequal in length. Notosetae numerous, sub- equal in length to and stouter than neurosetae, with single row of teeth. Neurosetae numerous, slender, of 2 types. Ventral cirri short, long on segment 2. Pharynx with 7 pairs of papillae; 2 pairs of jaws entire, without basal teeth. Etymology.—Bathy, from bathys, Greek, deep, plus bahamas, referring to the locality in the deep Atlantic off the Bahama Islands plus s ending. Gender: fem- inine. Bathybahamas charleneae, new species Fig. 4 Material examined.—North Atlantic, Tongue of the Ocean (TOTO) off the Bahama Islands, 24°53'12”N, 77°40'12”W, 2066 m, R. D. Turner Panel Study: Panel T-1 submerged A/vin dive 552, 19 Apr 1975; removed dive 753, 10 May 1977, paratype, USNM 96244.—Panel T-16 submerged dive 564, 8 May 1975; removed dive 755, 12 May 1977, holotype, USNM 96243. Description. —Length of holotype 8 mm, width 6 mm including setae, segments 18. Paratype 9 mm long, 7 mm wide. Body oval, flattened dorsoventrally, with very long parapodia and setae extending laterally and posteriorly. Color: Anterior third of dorsum brownish; pharynx with dark brown lining. Dorsum with trans- verse ciliated bands, 2 per segment, extending onto elytrophores and dorsal tu- bercles. Elytrophores large, bulbous, 8 pairs, on segments 2, 4, 5, 7, 9, 11, 12, and 15, with dorsal cirri on 3 posterior smaller segments (Fig. 4A, B). Elytra all missing. Dorsal tubercles on cirrigerous segments with digitiform ciliated extensions pro- jecting laterally (Fig. 4C). Prostomium deeply bilobed, with blunt anterior peaks; median antenna with large ceratophore in middle of prostomium, with long style; palps stout, long, tapered (Fig. 4A, H). Tentacular segment not distinct dorsally; tentaculophores lateral to prostomium, with prominent acicular process on inner side, without setae, with pair of long dorsal and ventral tentacular cirri; trilobed facial tubercle anterior to prostomium (Fig. 4A, H). Second or buccal segment with first pair of large elytrophores, well-developed biramous parapodia and ventral or buccal cirri similar to tentacular cirri, with short cirrophores on basal part of neuropodia lateral to ventral mouth (Fig. 4A). Pharynx with 7 pairs of subequal papillae around opening; 2 pairs of jaws without extra basal teeth (Fig. 4D). Parapodia long, biramous (Fig. 4B, C). Notopodia round, with long projecting acicular process on lower side, almost as long as neuropodia. Notosetae numerous, forming radiating bundle, short to long, nearly as long as and stouter than neu- rosetae, slightly curved, with blunt tips and single row of widely-spaced teeth along one side (Fig. 4E). Neuropodia conical, with long projecting acicular process on anterior lobe; posterior lobe slightly shorter, rounded. Neurosetae numerous, long, slender, of 2 types: supraacicular ones slightly stouter, flattened, with widely- VOLUME 98, NUMBER 1 139 Fig. 4. Bathybahamas charleneae, A—G, holotype, USNM 96243; H, I, paratype, USNM 96244: A, Dorsal view anterior end, pharynx completely extended, only basal part shown; styles of median antenna and left dorsal tentacular cirrus missing; B, Right elytrigerous parapodium, anterior view, acicula dotted; C, Right cirrigerous parapodium, posterior view; D, Isolated jaw; E, Short and long notosetae; F, Supra-acicular neuroseta; G, Subacicular neuroseta; H, Dorsal view anterior end, pharynx partially extended; right ventral tentacular cirrus missing; parapodia of segment 2 not shown; I, Ventral view of segments 11-14 showing elongate papillae on 12 and 13. Scales: = 1.0 mm for A, H; 0.5 mm for B, C; 0.1 mm for D-G; 1.0 mm for I. 140 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON spaced serrations on lateral sides and pointed bare tips (Fig. 4F); subacicular neurosetae slender, with numerous faint, fine, close-set hairs or spines along one side, tapering distally with delicate hairs (Fig. 4G). Dorsal cirri with elongate cylindrical cirrophores on posterior side of notopodia; styles long, slender, ex- tending beyond setae (Fig. 4C). Ventral cirri short, tapered, in middle of neuro- podia (Fig. 4B, C). Pygidium small lobe nearly hidden by parapodia and setae of posterior seg- ments. Small ventral segmental papillae on segments 6-15 on holotype. Paratype with 2 pairs of elongate ventral papillae on segments 12 and 13 (Fig. 41). Distribution. — North Atlantic off the Bahamas, in 2066 meters depth, associated with wood panels and with much more abundant polynoid, Bathykermadeca turnerae. Etymology.—The species is named for Charlene D. Long, who made prelimi- nary identifications of this and many other polynoid polychaetes. Comparisons. — Bathybahamas differs from the four previously described genera of the Macellicephalinae also possessing 8 pairs of elytra; i.e. Bruunilla Hartman, 1971 (emended Pettibone 1979), Bathyeliasona Pettibone, 1976, Bathyvitiazia Pettibone, 1976, and Bathypolaria Levenstein, 1981, in having neurosetae of 2 types, instead of a single type, in having notopodia of the biramous parapodia subequal in length to the neuropodia, instead of considerably shorter, and in having dorsal tubercles with digitiform branchial processes, instead of lacking them. This latter feature is present also in Bathyfauvelia Pettibone, 1976 (with 9 pairs of elytra) and Bathycatalina Pettibone, 1976 (with 12 pairs of elytra) (see Key below). Bathybahamas lacks (1), the characteristic wing-like structure on the ventral side of the lower lip of Bruunilla, (2), the keel-like pygidium of Bathy- polaria, (3), the jaws with basal teeth of Bathyvitiazia and Bruunilla and (4), the different type of notosetae of Bathyeliasona. The five genera with 8 pairs of elytra can be separated according to the following Key. Key to the Genera of Macellicephalinae with 8 Pairs of Elytra 1. With pair of wing-like structures on ventral side of segments 2 and 3 (lower lip); trilobed facial tubercle (fig. 1b, in Pettibone 1979). Prostomium rounded, without anterior projections and frontal filaments (fig. la, in Pettibone 1979). Jaws with basal teeth (fig. 2c, in Pettibone 1979). No- topodia much shorter than neuropodia (fig. 2a, b, in Pettibone 1979) .. BS Bae Oi oe a arg et Me pe a Bruunilla Hartman, 1971 (emend. Pettibone, 1979) B. natalensis Hartman, 1971 (Mozambique Basin, 4886-5068 m depth) — Without pair of wing-like structures on ventral side on lower lip ...... 2 2. Dorsal tubercles with lateral digitiform branchial extensions (Fig. 4C). Notopodia subequal in length to neuropodia (Fig. 4B, C). Neurosetae of 2 types (Fig. 4F, G). Prostomium with projecting blunt peaks, without frontal filaments; facial tubercle trilobed; tentaculophores with prominent acicular process, without setae (Fig. 4A, H). Jaws of pharynx without basal teeth (hig: 41D) steep eee oe eee Bathybahamas, new genus B. charleneae, new species (Off Bahamas, 2066 m depth) VOLUME 98, NUMBER 1 141 — Dorsal tubercles indistinct, not forming digitiform branchial structures. Notopodia much shorter than neuropodia. Neurosetae of single type... 3 3. Prostomium. with projecting anterior lobes and frontal filaments; tenta- culophores with transverse row of setae (fig. 14a, b, in Pettibone 1976). Jaws of pharynx without basal teeth. Neurosetae flattened, greatly ex- panded, with close-set lateral serrations (fig. 13e, in Pettibone 1976 .... 5 00-0 Hb OE SUE FAR oe 055 Se RR Real Bathyeliasona Pettibone, 1976 B. abyssicola (Fauvel, 1913) (Bay of Biscay, 4380 m depth) — Prostomium rounded, without projecting anterior lobes and frontal fila- ments; tentaculophores without setae (fig. 20a in Pettibone 1976; fig. la in Levenstein 1981). Neurosetae not greatly expanded, with widely spaced spines along lateral sides (fig. 20h in Pettibone 1976; fig. le in Levenstein IDES). args Catia Rates iis cee Scene et eae ak OO SRO aan sn ee 5 ee 4 4. Jaws of pharynx with numerous teeth (fig. 20c in Pettibone 1976). Seg- ments 17. Notosetae slender, delicate, flattened, with serrations along lateral borders (fig. 20g in Pettibone 1976). Pygidium oval (fig. 20b in REtHbONe WOM O)ies sc ke keke Se nme: Bathyvitiazia Pettibone, 1976 B. pallida (Levenstein, 1971) (Kamchatka Trench, 3816 m depth) — Jaws of pharynx without teeth. Segments 15. Notosetae thick, toothed along one side (fig. 1d in Levenstein 1981). Pygidium prominent, keel- hike((ie libanyevenstem 1981) 2. ....-.- Bathypolaria Levenstein, 1981 B. carinata Levenstein, 1981 (Canadian Arctic, 2750-3920 m depth) Key to Macellicephalinae with Dorsal Tubercles Forming Cirriform Ciliated Branchial Structures 1. Notopodia shorter than neuropodia (fig. 21e in Pettibone 1976). Elytra 9 pairs, segments 19-21. Neurosetae of single type, flattened, paddle-like (fig. 22e in Pettibone 1976). Notosetae with numerous spinous rows (fig. PA Gainer ttibone 9716) ) 6254. are See Bathyfauvelia Pettibone 1976 B. affinis (Fauvel, 1914) (Off Madeira, 0—2380 m depth) — Notopodia subequal in length to neuropodia (Fig. 4B, C; fig. 23c in Pet- IONE PION G) eee I Rea. eet aS eH ENA ViOinT. Furevei teat. ois Re Sees: 2 2. Elytra 12 pairs, segments 24. Neurosetae of single type, flattened, paddle- like (fig. 23e in Pettibone 1976). Notosetae with numerous spinous rows (fig. 23d in Pettibone 1976) ............... Bathycatalina Pettibone, 1976 B. filamentosa (Moore, 1910) (Off Santa Catalina Is., 611-1097 m depth) — Elytra 8 pairs, segments 18. Neurosetae of 2 types, not flattened, paddle- like (Fig. 4F, G). Notosetae serrated along edge (Fig. 4E) ............. ee: feere ete reel fo 0 Aces. ol ites! Bathybahamas, new genus B. charleneae, new species (Off Bahamas, 2066 m depth) 142 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Key to Three Species of Macellicephalinae from North Atlantic Reported Herein 1. Elytra 8 pairs. Segments 18. Dorsal tubercles with lateral digitiform ciliated extensions (Fig. 4C). Prostomium with blunt anterior peaks, without fron- tal filaments, with long median antenna (Fig. 4H). Tentaculophores with projecting acicular process, without setae; facial tubercle trilobed (Fig. 4A, H). Parapodia with notopodia and neuropodia subequal in length (Fig. 4C). Pharynx with 7 pairs of subequal papillae; jaws without basal teeth. Without long ventral segmental papillae or with 2 pairs of elongate papillae on segments 12 and 13 (Fig. 41) .. Bathybahamas charleneae, n. gen., n. sp. — Elytra 9 pairs. Segments 19-21. Dorsal tubercles short, nodular (Figs. 1D, 2B, D). Facial tubercle bilobed (Figs. 1A, 2A). Jaws of pharynx with basal tee the (SST DM) era i io at ectre ey ae ees Sactagt tn, 26 Aha ge 2 2. Prostomium with conical anterior lobes with frontal filaments, with short median antenna (Fig. 2A). Tentaculophores with small acicular process, without setae (Fig. 2A). Parapodia with notopodia and neuropodia sub- equal in length (Fig. 2C). Pharynx with 7 pairs of subequal papillae; jaws with basal teeth (Fig. 2H). Posterior segments (18-21) modified and com- pressed, without elytra or dorsal cirri on segment 19 (Fig. 3B, D). Four pairs of ventral segmental papillae on segments 12-15 (Fig. 3I)........ SSR Le NT eR NI ode AO aN Bathykermadeca turnerae, n. sp. — Prostomium with rounded anterior lobes, without frontal filaments, with long median antenna (Fig. 1A). Tentaculophores without acicular process or setae (Fig. 1A). Parapodia with notopodia much shorter than neuro- podia (Fig. 1C). Pharynx with 5 pairs of papillae, median ones extra long, lateral ones small; jaws with more numerous basal teeth (1 1-15) (Fig. 1H, I). Posterior segments not modified, with dorsal cirri on segment 19 (Fig. 1B). Ventral segmental papillae indistinct, none enlarged ............. ee re ot 2 RY Pelagomacellicephala iliffei, n. gen., n. sp. Subfamily Harmothoinae Horst, 1917 Genus Harmothoe Kinberg, 1856 Type-species.— Harmothoe spinosa Kinberg, 1856. Gender: feminine. Diagnosis.—Segments up to 45. Elytra and elytrophores 15 pairs, on segments 2, 4, 5, 7, alternate segments to 23, 26, 29, and 32. Dorsal cirri on non-elytra- bearing segments. Nodular dorsal tubercles on cirrigerous segments. Elytra with microtubercles, with or without macrotubercles. Bilobed prostomium with ce- phalic peaks, 3 antennae and paired palps; lateral antennae with distinct cerato- phores, inserted ventral to ceratophore of median antenna; usually with 2 pairs of eyes. Tentaculophores of tentacular segment lateral to prostomium, with small acicular process, with or without setae, with 2 pairs of tentacular cirri. Parapodia biramous, both rami with elongate acicular processes; neuropodia usually with supra-acicular digitiform process. Notosetae as stout as or stouter than neurosetae, with well-marked spinous rows and blunt to pointed tips. Neurosetae with elongate spinous regions, with tips slightly hooked, at least some with secondary subter- minal tooth. Pharynx with 9 pairs of papillae and 2 pairs of hooked jaws, without basal teeth. VOLUME 98, NUMBER 1 Table 2.—Records of Harmothoe ingolfiana from Deep Ocean Stations (DOS). 143 Panel Submerged Alvin dive Removed A/vin dive Number specimens USNM cat. no. DOS-1 N24, N27 N26, N28 N28 N34 N35 N47 N54 N74 N54 N65 N67 N72 N99 DOS-2 N31, N39 N37 N40 N42 N44 N45 N62 N50 N63 N60 N82, N83 N95 597 30 Aug 75 597 30 Aug 75 597 30 Aug 75 597 30 Aug 75 597 30 Aug 75 658 15 Jun 76 658 15 Jun 76 774 30 Jul 76 658 15 Jun 76 685 17 Aug 76 685 17 Aug 76 773 29 Jul 77 834 18 Sep 78 601 5 Sep 75 601 5 Sep 75 601 5 Sep 75 602 6 Sep 75 657 10 Jun 76 657 10 Jun 76 681 13 Aug 76 657 10 Jun 76 681 13 Aug 76 681 13 Aug 76 790 23 Sep 77 817 29 Jun 78 658 15 Jun 76 685 17 Aug 76 794 28 Sep 77 773 29 Jun 77 834 18 Sep 78 774 30 Jul 77 794 28 Sep 77 794 28 Sep 77 773 29 Jul 77 776 1 Aug 77 834 18 Sep 78 1026 27 Jul 80 777 3 Aug 77 681 13 Aug 76 657 10 Aug 76 792 26 Sep 77 790 23 Sep 77 790 23 Sep 77 778 4 Aug 77 817 29 Jun 78 1031 3 Aug 80 1026 17 Jul 80 20 14 11 39 9 18 81248 81231 81232 81250 81233 81234 81235 81236 31237 81238 81251 81239 81249 81240 81241 81242 81243 81244 81245 81246 81252 81247 Two species of Harmothoe are covered in this report: H. ingolfiana Ditlevsen, 1917, and H. vagabunda, new species. . Harmothoe ingolfiana Ditlevsen Fig. 5 Harmothoe ingolfiana Ditlevsen, 1917:32, pl. 1: figs. 12, 14, pl. 2: fig. 15, text- fig. 15a—d.— Wesenberg-Lund, 1950:8.— Wolff, 1979:117, 120, 123-124. Material examined. —North Atlantic, south of Iceland, 61°30'N, 22°30’W, 1836 m, Ingolf-Exped. sta 67, 10 syntypes (ZMC). North Atlantic Deep Ocean Stations (DOS), R. D. Turner Panel Study: DOS- 1, 110 miles south of Woods Hole, Massachusetts, 39°46'N, 70’41'W, 1830 m, 372 specimens, including many young (USNM; see Table 2). DOS-2, 190 miles southeast of Woods Hole, 38°18'24”N, 69°35'36”W, 3506 m, 184 specimens, including many young (USNM;; see Table 2). Description.—Length of complete syntype 12 mm, width with setae 5 mm, segments 36. Length of incomplete syntype of 16 segments 9 mm, width 8 mm. Three specimens from DOS-1 (USNM 81248) 32-33 mm long, 13-15 mm wide, segments 38-39. No color. 144 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON GS ee es = => = Harmothoe ingolfiana, USNM 81248: A, Anterior end, dorsal view; B, Right elytrigerous Figs 5: parapodium, anterior view, acicula dotted; C, Right cirrigerous parapodium, posterior view; D, Short and long notosetae; E, Upper, middle and lower neurosetae; F, First elytron, with detail of microtu- bercles, macrotubercles and papillae; G, Middle left elytron with detail of same. Scales: = 1.0 mm for A; 1.0 mm for B, C; 0.1 mm for D, E; 2.0 mm for F, G. Body flattened, subrectangular, tapered slightly anteriorly and posteriorly. Elytra 15 pairs, large, covering dorsum, circular, oval to subreniform in shape, densely covered with small, conical microtubercles, becoming larger near lateral and pos- terior borders, with multispinous tips; some elytra with few macrotubercles near VOLUME 98, NUMBER 1 145 posterior border, cylindrical to oval, smooth or spinous distally (Fig. 5F, G; pl. 1: figs. 12, 14, in Ditlevsen 1917). Elytrophores large, bulbous (Fig. 5A, B). Dorsal cirri on non-elytra-bearing segments; dorsal tubercles on cirrigerous segments nodular (Fig. 5C). | Prostomium bilobed, with cephalic peaks; median antenna with stout cerato- phore in anterior notch, with long style, minutely papillate; lateral antennae with short stout ceratophores inserted ventrally, with short styles, minutely papillate; palps stout, tapered, smooth; without eyes (Fig. 5A; pl. 2: fig. 15, in Ditlevsen 1917). Tentaculophores lateral to prostomium, with small acicular process, 2-3 setae and dorsal and ventral tentacular cirri similar to median antenna (Fig. 5A). Facial tubercle not distinct, with facial ridge. Second or buccal segment with first pair of large elytrophores, biramous parapodia and long ventral buccal cirri in- serted basally on neuropodia lateral to ventral mouth, similar to tentacular cirri (Fig. SA). Parapodia biramous (Fig. 5B, C; text-fig. 14, in Ditlevsen 1917). Notopodia smaller than neuropodia, rounded, with acicular process on lower side. Notosetae numerous, forming radiating bundle, short to long, stouter and as long or longer than neurosetae, straight, sword-shaped, reddish amber-colored, with prominent transverse spinous rows and rather long bare tapered tips (Fig. 5D; text-fig. 15a, in Ditlevsen 1917). Dorsal cirri with cylindrical cirrophores on dorsoposterior side of notopodia; styles slender, extending beyond setae, with scattered micro- papillae (Fig. 5C). Neuropodia conical, with presetal acicular process, with only small supra-acicular extension (without digitiform supra-acicular process, as usual in Harmothoe). Upper neurosetae more slender, with longer spinous regions and bifid tips; middle and lower neurosetae stouter, with entire bare tips and shorter spinous regions (Fig. 5E). All neurosetae with bifid tips in smaller specimens (text- fig. 15b-d, in Ditlevsen 1917). Ventral cirri short, subulate, with micropapillae on upper part (Fig. 5B, C). Pharynx with 9 pairs of papillae and 2 pairs of hooked jaws. Nephridial papillae small, beginning on segment 6. Pygidium small, bulbous, with pair of anal cirri. Distribution.— North Atlantic, south of Iceland to off New England, in 1830 to 3506 meters depth. In burrows of wood-boring bivalves (Xylophagainae). Remarks.—The original material of H. ingolfiana, described by Ditlevsen (1917), was collected from /ngolf-Expedition station 67 south of Iceland in 1783 meters (975 fms), corrected by Wesenberg-Lund (1950:8) to 1836 meters. Ditlevsen (1917:33) remarked “‘that it is a pronounced deep-sea form, is beyond doubt.” In connection with his study on “Macrofaunal Utilization of Plant Remains in the Deep Sea,”” Wolff (1979:117, 120, 123-124) reported that “‘A log of pine wood, several meters long, was collected at 1800 m in the North Atlantic (K. W. Ock- elmann, pers. comm.)” and “A Xylophaga-bored piece of a pine log, c. 15 x 35 cm, from Jngolf st. 67 in the North Atlantic (1836 m)” contained a number of polychaetes with preliminary identifications by the late A. Eliason. Among them was Harmothoe ingolfiana, described by Ditlevsen from the same Ingo/f station. Thus it appears safe to assume that the syntypes of H. ingolfiana were associated with the boring bivalves (Xylophaga) in wood. The numerous specimens, from young to adult, identified as H. ingo/fiana in the wood panel study by Ruth Turner in the North Atlantic, indicate that it is an opportunistic species, using the wood and animal associates as a substrate, as well as a source of food. Numerous 146 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON specimens of the ampharetid polychaete, Decemunciger apalea, were described by Zottoli (1982) from the same wood panels. Harmothoe vagabunda, new species Fig. 6 Material examined. — North Atlantic, Tongue of the Ocean (TOTO), off Bahama Islands, 24°53'12”N, 77°40'12”W, 2066 m, R. D. Turner Panel Study, submerged on Alvin dive 564, 8 May 75; panels T-16, T-17, T-20 removed on dives 752, 754, 9-12 May 77, 7 paratypes, USNM 81227-81229. Off St. Croix, Virgin Islands, 17°57'36"N, 64°48’36"W, 3995 m, R. D. Turner Panel Study, submerged on Alvin dive 873, 17 Dec 78; P-3 removed on dive 1080, 7 Dec 80, paratype (USNM 81226); P-5, P-13 removed on dive 1079, 6 Dec 80, 20 small paratypes, USNM 81225. Alvin dives 875, 876, 19, 20 Dec 78, picked up pieces of ““wild’’ wood, about 5 and 6 feet long, holotype, USNM 81223 and 15 paratypes, USNM 81224. Description.— Length of holotype from off St. Croix (USNM 81223) 13 mm, width with setae 6 mm, segments 35; paratype (USNM 81224) 9 mm long, 4 mm wide, with 34 segments. Length of paratype from off Bahamas (USNM 81228) 10 mm, width 5 mm, segments 31, last 4 smaller, regenerating. No color. Body flattened, slightly tapered anteriorly and posteriorly. Elytra 15 pairs, large, covering dorsum, oval, soft, opaque white, nearly covered with conical micro- tubercles, low, oval on anterior part; some scattered cylindrical papillae on surface and on lateral borders but without long fringe of papillae (Fig. 6F). Elytrophores large, bulbous (Fig. 6A, B). Dorsal cirri on non-elytra-bearing segments; dorsal tubercles nodular (Fig. 6C). Prostomium bilobed, rounded, with cephalic peaks very small or indistinct; median antenna with large ceratophore in anterior notch, with style about 2 times longer than prostomium having scattered short papillae and slender tip; lateral antennae with rather large bulbous ceratophores inserted ventrally, with styles short and subulate; palps stout, tapered, slightly longer than median antenna; without eyes (Fig. 6A). Tentaculophores lateral to prostomium, with small acicular process and 2-4 setae on inner side; dorsal tentacular cirri about same length as median antenna, ventral ones shorter; raised facial ridge but without distinct facial tubercle. Second or buccal segment with pair of large bulbous elytrophores, bira- mous parapodia and long ventral buccal cirri inserted basally on neuropodia lateral to ventral mouth, similar to tentacular cirri (Fig. 6A). Parapodia biramous (Fig. 6B, C). Notopodia smaller than neuropodia, rounded, with acicular process on lower side. Notosetae numerous, forming radiating bun- dle, short to long, much stouter than and slightly shorter than neurosetae, slightly curved, with spinous rows and tapered pointed bare tips (Fig. 6D). Dorsal cirri with cylindrical cirrophores on dorsoposterior side of notopodia; styles slender, extending beyond setae, with scattered short papillae and slender terminal filament (Fig. 6C). Neuropodia conical with longer presetal acicular process; postsetal lobe shorter, rounded. Neurosetae numerous, forming fan-shaped bundle. Upper neu- rosetae longer, more slender, with longer spinous regions and bifid split tips; lower neurosetae shorter, stouter, with tips slightly curved, with slender secondary tooth (Fig. 6E). Ventral cirri short, tapered (Fig. 6B, C). VOLUME 98, NUMBER 1 147 —S> C <3. Fig. 6. Harmothoe vagabunda, holotype, USNM 81223: A, Dorsal view anterior end, pharynx partially extended; B, Right elytrigerous parapodium, anterior view, acicula dotted; C, Right cirrigerous parapodium, posterior view; D, Short and long notosetae; E, Upper, middle and lower neurosetae; F, Right elytron, with detail of some microtubercles and papillae. Scales: = 1.0 mm for A; 0.5 mm for B, C; 0.1 mm for D, E; 1.0 mm for F. Pharynx with 9 pairs of black papillae and 2 pairs of hooked jaws. Ventral nephridial papillae small, beginning on segment 6. Pygidium small, with pair of anal cirri. Distribution.—North Atlantic off Bahamas in 2066 meters depth and off St. 148 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Croix, Virgin Islands, in 3995 meters depth. In burrows of marine wood-boring bivalves (Xylophagainae), along with more numerous polynoids Bathykermadeca turnerae. The ampharetid polychaete Endecamera palea was described by Zottoli (1982) from the same ‘“‘wild”’ wood at the St. Croix station. Etymology.—vagabunda, from Latin vagabundus, wandering, referring to its association with sunken pieces of wood. Comparisons.— Both the deep-sea H. vagabunda and H. ingolfiana lack eyes and the neuropodia lack distinct digitiform supra-acicular processes, contrary to the usual condition in species of Harmothoe. Harmothoe ingolfiana has distinct cephalic peaks; they are very small or lacking in H. vagabunda. The elytra of the former have microtubercles that are spinous-tipped, with some macrotubercles present; the microtubercles are only conical in H. vagabunda. Acknowledgments I appreciate the opportunity to work up the cave polynoids received from Dr. Thomas M. Iliffe, as a part of his study on cave faunas, and the detailed infor- mation he provided on the cave habitat. I also appreciate being a part of the interesting wood and panel experiments being carried on by Dr. Ruth D. Turner in the deep North Atlantic. The collection of specimens from caves in the Turks and Caicos islands was supported in part by a National Science Foundation Grant (BSR 8215672) to Thomas M. Iliffe. The study conducted on A/vin “wood islands” by Ruth Turner was supported by an Office of Naval Research grant to Harvard University. My thanks go to Dr. J. B. Kirkegaard of the Zoologisk Museum, Copenhagen, for the loan of types. The manuscript benefited from the reviews of Thomas E. Bowman and Meredith L. Jones. Literature Cited Bowman, T. E., T. M. Iliffe, and J. Yager. 1984. New records of the troglobitic mysid genus Sty- giomysis: S. clarkei, n. sp. from the Caicos Islands, and S. holthuisi (Gordon) from Grand Bahama Island (Crustacea: Mysidacea)— Proceedings of the Biological Society of Washington 97(3):639-646. Dietz, R. S., J. C. Holden, and W. P. Sproll. 1970. Geotectonic evolution and subsidence of Bahama Platform.— Geological Society of America Bulletin 81:1915-1928. Dill, R. F. 1977. The Blue Holes-geologically significant submerged sink holes and caves off British Honduras and Andros, Bahama Islands.— Proceedings of Third International Coral Reef Sym- posium, Miami, Florida 2:238-242. Ditlevsen, Hj. 1917. Annelides I.—The Danish Ingolf-Expedition, Copenhagen 4(4):1-—71. Fauvel, P. 1913. Quatriéme note préliminaire sur les Polychetes provenant des campagnes de /’Hi- rondelle et de la Princess-Alice, ou déposées dans le Musée Océanographique de Monaco. — Bulletin de l’Institut Océanographique de Monaco 270:1-80. 1914. Aphroditiens pélagiques des campagnes de /’Hirondelle, de la Princess-Alice et de l’'Hirondelle IT. (Note préliminaire).— Bulletin de |’Institut Océanographique de Monaco 287: 1-8. Gregor, V. A. 1981. Karst and caves in the Turks and Caicos Islands, B.W.I.— Proceedings of the Eighth International Congress of Speleology, Bowling Green, Kentucky, U.S.A.:805-807. Hartman, O. 1971. Abyssal polychaetous annelids from the Mozambique Basin off southeast Africa, with a compendium of abyssal polychaetous annelids from world-wide areas.—Journal of the Fisheries Board of Canada 28(10):1407-1428. Hartmann-Schréder, G. 1974. Die Unterfamilie Macellicephalinae Hartmann-Schréder, 1971 (Pol- ynoidae, Polychaeta). Mit Beschriebung einer neuer Art, Macellicephala jameensis, n. sp., aus VOLUME 98, NUMBER 1 149 einen Héhlengewdsser von Lanzarote (Kanarische Inseln).— Mitteilungen aus dem Hambur- gischen Zoologischen Museum und Institut 71:75-85. lliffe, T. M., H. Wilkens, J. Parzefall, and D. Williams. 1984. Marine lava cave fauna: composition, biogeography, and origins. —Science 225(4659):309-311. Kirkegaard, J. B. 1956. Benthic Polychaeta from depths exceeding 6000 meters.—Galathea Report 2:63-78. Levenstein, R. J. 1971. [Polychaete worms of the genus Macellicephala and Macellicephaloides (Family Aphroditidae) from the Pacific Ocean. Jn Fauna of the Kurile-Kamchatka Trench.]— Trudy Institut Okeanologii P.P. Shirshov Academiia Nauk SSSR 92:18-35 [in Russian, English summary]. 1978. [Polychaetes of the family Polynoidae (Polychaeta) from the deep-water trenches of the western part of the Pacific.]—Trudy Institut Okeanologii P.P. Shirshov Academiia Nauk SSSR 112:162—173 [in Russian, English summary]. . 1981. [Some peculiarities in the distribution of the polychaetes of the family Polynoidae in the Canadian basin of the Arctic Ocean.]— Trudy Institut Okeanologii P.P. Shirshov Academiia Nauk SSSR 115:26-36 [in Russian, English summary]. 1982a. [On the fauna of the family Polynoidae from the trench of Japan.]—Trudy Institut Okeanologii P.P. Shirshov Academiia Nauk SSSR 117:59-62 [in Russian, English summary]. 1982b. [New genera of the subfamily Macellicephalinae (Polychaeta, Polynoidae) from the Tasman trench.]— Zoologichesky Zhurnal 61(9):1291-1296 [in Russian, English summary]. Moore, J: P. 1910. The Polychaetous annelids dredged by the U.S.S. “‘Albatross”’ off the coast of Southern California in 1904: II. Polynoidae, Aphroditidae and Segaleonidae.— Proceedings of the Academy of Natural Sciences of Philadelphia 62:328-—402. Muir, A. I. 1982. Generic characters in the Polynoidae (Annelida, Polychaeta), with notes on the higher classification of scale-worms (Aphroditacea).— Bulletin of the British Museum Natural History (Zoology) 43(3):153-177. Pettibone, M. H. 1976. Revision of the genus Macellicephala McIntosh and the subfamily Macel- licephalinae Hartmann-Schréder (Polychaeta: Polynoidae).— Smithsonian Contributions to Zo- ology 229:1-71. 1979. Redescription of Bruunilla natalensis Hartman (Polychaeta: Polynoidae), originally referred to Fauveliopsidae.— Proceedings of the Biological Society of Washington 92(2):384— 388. Turner, R. D. 1973. Wood-boring bivalves, opportunistic species in the deep sea.—Science 180: 1377-1379. 1977. Wood, mollusks, and deep-sea food chains.— Bulletin of the American Malacological Union for 1976:13-19. 1981. “Wood islands” and “thermal vents” as centers of diverse communities in the deep sea.— The Soviet Journal of Marine Biology 7(1):1-9. Wesenberg-Lund, E. 1950. Polychaeta.—Danish Ingolf-Expedition 4(14):1-92. Wilkens, H., and J. Parzefall. 1974. Die oekologie der Jameos del Agua (Lanzarote). Zur Entwicklung limnischer Hoehlentiere aus marinen Vorfahren.— Annales de Spéléologie 29(3):419—434. Wolff, T. 1979. Macrofaunal utilization of plant remains in the deep-sea. —Sarsia 64(1—2):117-136. Zottoli, R. 1982. Two new genera of deep-sea polychaete worms of the family Ampharetidae and the role of one species in deep-sea ecosystems. — Proceedings of the Biological Society of Wash- ington 95(1):48-57. Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 150-157 AN ADDITIONAL NEW SCALE WORM (POLYCHAETA: POLYNOIDAE) FROM THE HYDROTHERMAL RIFT AREA OFF WESTERN MEXICO AT 21°N Marian H. Pettibone Abstract.—A unique polychaete, Branchiplicatus cupreus, n. gen., n. sp., from the hydrothermal vent area at 21°N off western Mexico is described and referred to anew subfamily of Polynoidae, Branchiplicatinae. Branchiae are well developed and of an unusual type. The prostomium is truncate, rather than bilobed, as is usual in the Polynoidae. This is the fourth contribution dealing with the unusual polynoid polychaetes collected from hydrothermal vents off the Galapagos and on the East Pacific Rise off western Mexico at 21°N (Pettibone 1983, 1984a, b). The polynoids described herein were collected by the OASIS group of Scripps Institution of Oceanography during four A/vin dives in April-May 1982 in 2612-2633 meters at 21°N. They were a part of the numerous polynoid polychaetes sent to me by J. F. Grassle and I. Williams of the Woods Hole Oceanographic Institution (WHOI). The specimens were retrieved from washes of other invertebrates, including the vestimentiferans Riftia pachytila Jones, 1981, the giant clams Calyptogena magnifica Boss and Turner, 1980, and the ampharetid polychaetes 4/vinella pompejana Desbruyéres and Laubier, 1980. Other described polynoids collected from some of the same dives include Lepidonotopodium fimbriatum Pettibone, 1983, and L. riftensis Pettibone, 1984b. This is the second branchiate species of Polynoidae to be de- scribed from the hydrothermal vent areas, the first being Branchipolynoe sym- mytilida Pettibone, 1984a, commensal in the mantle cavities of the Galapagos deep-sea vent mussels. The types are deposited in the National Museum of Natural History, Smith- sonian Institution (USNM). This is OASIS Expedition Contribution number 25. Family Polynoidae Branchiplicatinae, new subfamily Branchiplicatus, new genus Type-species.— Branchiplicatus cupreus, new species. Gender: masculine. Diagnosis.— Body elongate, flattened, tapered posteriorly, segments up to 35, first achaetous. Elytra and elytrophores 12 pairs on segments 2, 4, 5, 7, 9, 11, 13, 15, 17, 19, 21, and 23. Dorsal cirri on non-elytra-bearing segments, with short cirrophores and long styles. Branchiae well developed on segments posterior to segment 2, of unique type formed of elongate folded sacs with ciliated bands and attached by broad base to lateral sides of flattened elytrophores and of dorsal tubercles on cirrigerous segments. Prostomium truncate anteriorly, not bilobed, fused to tentaculophores of first or tentacular segment, with long median antenna VOLUME 98, NUMBER 1 151 and pair of long palps; without lateral antennae or eyes. Tentaculophores without setae, with 2 pairs of tentacular cirri; without facial tubercle. Second or buccal segment with first pair of elytra, biramous parapodia and long ventral buccal cirri. Parapodia biramous, both rami with projecting acicular processes and numerous long, copper-colored setae. Notosetae much stouter than neurosetae, smooth or with double rows of spines. Neurosetae of 2 types. Ventral cirri short. Pharynx with 5 pairs of unequal papillae; 2 pairs of jaws with edges minutely denticled. Without distinct segmental ventral papillae. Etymology.—Branchi, from branchia (Greek) gills, plus plicatus, from plicatus (Latin) fold, referring to the unusual type of folded branchiae. The specific name cupreus, from cuprum (Latin) copper, referring to striking copper-colored setae. Branchiplicatus cupreus, new species Figs. 1-4 Material examined.—Pacific Ocean off western Mexico, 20°50’N, 109°06’W, OASIS Alvin dives in 1982: Dive 1214, 20 Apr, 2633 m, vestimentiferan wash, holotype (USNM 97219) and 21 paratypes (11 young; USNM 97220-2, 97226).— Dive 1219-10A & B, 25 Apr, 2612 m, Riftia and Calyptogena wash, fine fraction, 2 paratypes (USNM 97223).—Dive 1221-15, 4 May, 2618 m, Riftia and Calyp- togena wash, coarse fraction, 1 young paratype (USNM 97224).—Dive 1226-7, 10 May, 2616 m, Riftia, Calyptogena and Alvinella wash, paratype (USNM 97225). Description.— Length of holotype (USNM 97219) 50 mm, width with setae 15 mm, segments 35, last one small. Length of paratype (USNM 97221) 45 mm, width 19 mm, segments 34. Length of small paratype (USNM 97223) 13 mm, width 6 mm, segments 31. Body elongate, flattened ventrally, slightly arched dorsally, tapering slightly anteriorly and more so posteriorly (Fig. 1). No color but with striking long copper-colored setae. Prostomium truncate anteriorly, extending posterolaterally and fused to ten- taculophores of first or tentacular segment (Fig. 2A). Median antenna with cer- atophore attached to middle of prostomium, with long tapered style; palps stout, long, tapered; without lateral antenna or eyes. Tentaculophores without setae, with 2 pairs of long slender tentacular cirri, dorsal ones subequal in length to palps, ventral ones slightly shorter (Fig. 2A). Tentacular segment not visible dor- sally, forming upper and lateral lips of ventral mouth (Fig. 2A, B). Second or buccal segment with first pair of bulbous elytrophores, elytra, biramous parapodia, and long ventral buccal cirri similar to tentacular cirri and attached on bulbous cirrophores on bases of neuropodia lateral to lower lip and ventral mouth (Fig. 2A, B, G; 3A). Large muscular eversible pharynx with 5 pairs of papillae around opening, median pair longer with tapered tips, adjacent lateral pairs similar in shape but shorter, and lateral pairs shorter, rounded (Fig. 2D, E). Two pairs of large dark jaws with edges serrated or minutely denticled (about 20 teeth; Fig. 2E, F). Subdistally, pharynx with diagonal rows of small papillae on dorsal, ventral and lateral surfaces. Elytra 12 pairs, on segments 2, 4, 5, 7, 9, 11, 13, 15, 17, 19, 21 and 23, with dorsal cirri on non-elytra-bearing segments, including up to 12 posterior segments. Elytra (Figs. 1; 2G, H; 4C) large, oval, covering dorsum except for posterior segments. Elytra stiff, smooth, opaque, with “‘veins’’; first pair of elytra on segment 152 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Branchiplicatus cupreus, paratype, USNM 97220: Dorsal view, left; ventral view, right; only 2 elytra remain. 2x. 2 with anterior notch (Fig. 2G); posterior 12th pair on segment 23 elongate-oval, covering several posterior smaller segments (Fig. 4C). Posterior border of elytra with variable number of small round tubercles not sharply set off surface (Fig. 2G, H); elytral surface with minute sensory papillae. On small paratype (USNM 97223), tubercles scarcely demarcated and lacking on first and last pairs (Fig. 4C). Elytrophores, from segment 4 on, flattened, directed posterolaterally, with flat- tened extensions on anterior and posterior sides, and with branchiae attached on lateral sides (Figs. 2C; 3D). Dorsal tubercles, corresponding in position to elytro- phores on cirrigerous segments, also flattened, with anterior and posterior exten- sions, and with branchiae attached on lateral sides (Figs. 2C; 3B, C; 4A). Branchiae begin on segment 3 and continue posteriorly (Figs. 2A, C; 3B—D; 4A). Branchiae of unique type, formed of flattened elongate sacs, deeply folded and convoluted, attached by broad bases to lateral sides of elytrophores and dorsal tubercles and to dorsal sides of notopodia and dorsal cirrophores; surfaces of branchiae with transverse ciliated bands. Branchiae on elytrigerous segments with extra lobe on posterior side; branchiae on cirrigerous segments formed of equal- sized anterior and posterior flattened sacs fused basally. Parapodia biramous (Figs. 3A—D; 4A). Notopodia shorter than neuropodia, rounded, with projecting acicular process on lower side and with short bract on anterior side. Cirrophores of dorsal cirri cylindrical and fused on lower side to anterior notopodial bract; styles of dorsal cirri long, slender, extending far beyond setae (Figs. 3B, C; 4A). Notosetae numerous, forming radiating bundles, short to VOLUME 98, NUMBER 1 153 = Ss SSS SS. ~ Fig. 2. Branchiplicatus cupreus, A—C, holotype, USNM 97219; D-H, paratype, USNM 97221: A, Dorsal view anterior end; B, Ventral view anterior end, bases of palps and cirri only shown; C, Dorsal view right parapodia of segments 11 and 12; distal end of style of dorsal cirrus not shown; D, Dorsal view anterior end of extended pharynx; E, Ventral view of same; F, Dorsal jaws; G, First left elytron (segment 2); H, Left elytron 6 (segment 11). Scales = 2.0 mm for A, D; 1.0 mm for B, C, E; 0.5 mm for F; 2.0 mm for G, H. ~ Fig. 3. Branchiplicatus cupreus, paratype, USNM 97221: A, Right elytrigerous parapodium seg- ment 2, anterior view; B, Right cirrigerous parapodium segment 3, anterior view, ventral cirrus with 154 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON se G abnormal split tip; C, Right cirrigerous parapodium segment 14, posterior view; D, Right elytrigerous parapodium segment 15, anterior view, acicula dotted; E, Four short and long notosetae; F, Supra- acicular neuroseta; G, Middle and lower subacicular neurosetae. Scales = 1.0 mm for A—D; 0.1 mm for E-G. VOLUME 98, NUMBER 1 155 ‘@) SS ie y oe Fig. 4. Branchiplicatus cupreus, A, B, paratype, USNM 97221; C, small paratype of 31 segments, 13 mm long, USNM 97223; D-I, young paratype of 15 segments, 2.5 mm long, USNM 97222: A, Right cirrigerous parapodium segment 25, anterior view; B, Ventral view of left parapodium segment 20 showing nephridial opening; C, Right Ist, 2nd, 6th and 12th elytra (segments 2, 4, 11, 23); D, Dorsal view anterior end, elytra missing; E, Dorsal view posterior end; F, Right cirrigerous parapodium segment 8, anterior view; G, Right elytrigerous parapodium segment 9, anterior view, acicula dotted; H, Three short and long notosetae; I, Upper and lower neurosetae. Scale = 1.0 mm for A, C; 1.0 mm for B; 0.5 mm for D, E; 0.2 mm for F, G; 0.1 mm for H, I. 156 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON long, some as long as neurosetae. Notosetae numerous, forming radiating bundles, short to long, some as long as neurosetae. Notosetae much stouter than neurosetae, acicular, smooth or with double rows of spines near distal bare tips (Fig. 3E). Neuropodial presetal lobes conical, with projecting acicular process; postsetal lobes shorter, rounded. Neurosetae very numerous, slender, forming fan-shaped bundles. Supra-acicular neurosetae with 2 rows of prominent spines, tapering to long bare capillary tips (Fig. 3F). Subacicular neurosetae more slender, with shorter spines and slightly hooked bare tips (Fig. 3G). Ventral cirri short, tapered (Fig. 3B-D). Dorsal anal ridge and anus enclosed in branchiae and parapodia of few posterior segments (32-35). Small rectangular ventral pygidium between parapodia of seg- ments 34 and 35, with pair of long anal cirri (sometimes one short or both missing). No distinct nephridial or segmental papillae but ventral bases of neuropodia somewhat inflated with slightly projecting opening (not always visible; Fig. 4B). Young paratypes (USNM 97226) ranging in size from 5 mm in length, 3 mm in width, with 19 segments, last 2 very small, to 1 mm in length, 1 mm in width, with 10 segments plus growing region. Young paratype (USNM 97222; Fig. 4D- I) of 15 segments plus growing zone, 2.5 mm long, 2 mm wide, with 8 pairs of elytrophores (elytra missing). Developing flattened branchiae beginning on seg- ment 3 (Fig. 4D-G). Prostomium with slight anterior notch, instead of truncate; palps cylindrical, with terminal filament, instead of tapered; median antenna closer to anterior end (Fig. 4D). Parapodia similar in shape to adults (Fig. 4F, G). Notosetae of same type but more slender (Fig. 4H). Neurosetae also more slender; lower ones with more prominent spines (Fig. 41). Remarks.—The unique folded type of branchiae in Branchiplicatus separates the genus and subfamily from the other subfamilies of Polynoidae. The presence of branchiae is an unusual feature in the Polynoidae, as mentioned in the de- scription of Branchipolynoe symmytilida in Branchipolynoinae (Pettibone 1984a). In the latter group the well-developed branchiae are arborescent, rather than of the unique type in Branchiplicatus. The truncate, rather than bilobed prostomium also sets Branchiplicatus apart in the Polynoidae. There appears to be a greater fusion of the prostomium with the first or tentacular segment. The pharynx also differs from the other subfamilies in having five pairs of unequal papillae and two pairs of large serrated jaws. Branchiplicatinae agrees with the subfamilies Ma- cellicephalinae, Lepidonotopodinae, and Branchipolynoinae (Pettibone 1976; 1983; 1984a) in having prostomia with median antenna only and lacking lateral anten- nae, but in these subfamilies, the prostomium is bilobed, rather than truncate. Acknowledgments I wish to thank J. F. Grassle and I. Williams of the Woods Hole Oceanographic Institution for the material on which this study is based, as well as the members of the OASIS Expedition in 1982 of the Scripps Institution of Oceanography. The manuscript benefited from the reviews of Fenner A. Chace, Jr., Nancy J. Maciolek, and James A. Blake. Mr. Michael Carpenter kindly photographed the paratype. Literature Cited Boss, K. J.. and R. D. Turner. 1980. The giant white clam from the Galapagos rift, Calyptogena magnifica species novum.—Malacologia 20(1):161-194. VOLUME 98, NUMBER 1 157 Desbruyéres, D., and L. Laubier. 1980. Alvinella pompejana gen. sp. nov., Ampharetidae aberrant des sources hydrothermales de la ride Est-Pacific.—Oceanologica Acta 3(3):267—273. Jones, M. L. 1981. . Riftia pachytila, new genus, new species, the vestimentiferan worm from the Galapagos rift geothermal vents (Pogonophora). — Proceedings of the Biological Society of Wash- ington 93(4):1295-1313. Pettibone, M. H. 1976. Revision of the genus Macellicephala McIntosh and the subfamily Macel- licephalinae Hartmann-Schréder (Polychaeta: Polynoidae). —Smithsonian Contributions to Zo- ology 229:1-71. . 1983. Anewscale worm (Polychaeta: Polynoidae) from the hydrothermal rift-area off Western Mexico at 21°N.—Proceedings of the Biological Society of Washington 96(3):392-399. . 1984a. A new scale worm commensal with deep-sea mussels on the Galapagos hydrothermal vent (Polychaeta: Polynoidae).— Proceedings of the Biological Society of Washington 97(1): 226-239. . 1984b. Two new species of Lepidonotopodium (Polychaeta: Polynoidae: Lepidonotopodinae) from hydrothermal vents off the Galapagos and East Pacific Rise at 21°N.— Proceedings of the Biological Society of Washington 97(4):849-863. Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 158-176 THE OLIGOCHAETA OF GEORGES BANK (NW ATLANTIC), WITH DESCRIPTIONS OF FOUR NEW SPECIES Dale Davis Abstract.— Thirty-two oligochaete species are recorded from benthic grab sam- ples taken southeast of Massachusetts (Georges Bank). Taxonomic notes are pro- vided for all species not previously dealt with in earlier papers (Erséus and Davis 1984; Erséus 1984a; Davis 1984). Four of the species, Phallodrilus dorsosper- matheca, Phallodrilus christeri, Adelodrilus bacrionis, and Uniporodrilus vestigium are new to science and described herein. Some aspects of sediment-species rela- tionships, life histories, and geographical distributions are discussed. During the Georges Bank Benthic Infauna Monitoring Program, 32 species of marine Oligochaeta were encountered (Table 1). Of these 32, 11 are known to be new species. All members of the genus Adelodrilus, except A. bacrionis, n.sp., from Georges Bank were previously treated by Erséus and Davis (1984). Several species of Phallodrilus have been dealt with by Erséus (1984a), and Olavius corn- uatus has been described by Davis (1984). Cook (1969, 1971) studied the Oligochaeta of Cape Cod Bay, Massachusetts. His work was very thorough and is commonly used for reference when studying oligochaetes in shallow waters along the northeast coast of the U.S. Cook (1970a, b) also made significant contributions to the knowledge of deep-sea oligochaetes from the northwest Atlantic. Several other publications (Brinkhurst 1965, 1966; Brinkhurst and Jamieson 1971; Brinkhurst and Baker 1979; Lasserre 1971; Cook and Brinkhurst 1973) are useful for identification of marine oligochaetes from the northwest Atlantic, but are concerned again mostly with shallow-water forms. Only in the last few years have publications (Erséus 1979a, b, 1983a; Baker and Erséus 1979) dealt with the oligochaetes from the continental shelf of the northeast U.S. The material from the Georges Bank Monitoring Program provides a significant contribution to the knowledge of marine oligochaetes from this area. Materials and Methods All samples were taken in the course of the Georges Bank Benthic Infauna Monitoring Program performed by Battelle New England Marine Research Lab- oratory for the U.S. Department of the Interior, Minerals Management Service (Final Report for year one available through the National Technical Information Service). Starting in July 1981, samples were taken in February, May, July, and November of each year through May of 1984. Samples were taken with a 0.04 m? Van Veen grab. The samples were subsequently sieved through a 300 um screen and fixed in 10% formalin. In the laboratory, the samples were resieved through a stack of 500 um and 300 um screens and transferred into 70% isopro- panol for processing and storage. Oligochaete specimens were stained in Harris haematoxylon or paracarmine and mounted in permanent mounting media on VOLUME 98, NUMBER 1 159 Table 1.—List of species from Georges Bank. Tubificidae Uniporodrilus vestigium, n. sp. Adelodrilus anisosetosus Cook, 1969* Bathydrilus longus Erséus, 1979 A. bacrionis, n. sp. Heterodrilus occidentalis Erséus, 1981 A. correptus Erséus and Davis, 1984* Limnodriloides barnardi Cook, 1974 A. cristatus Erséus, 1983* L. medioporus Cook, 1969 A. inopinatus Erséus and Davis, 1984* Tubificoides intermedius (Cook, 1969) A. multispinosus Erséus, 1979* Tubificoides, sp. A A. pilatus Erséus and Davis, 1984* Tubificoides, sp. B Phallodrilus biprostatus (Baker and Erséus, 1979)* P. boeschi Erséus, 1984* Enchytraeidae P. coeloprostatus Cook, 1969* P davisi Erséus. 1984* Marionina welchi Lasserre, 1971 Grania, sp. A 3 alae Nn. sp. Granite, 60. B . christeri, n. sp. : P. flabellifera Erséus, 1984* ae a = P. parviatriatus Cook, 1971 Phallodrilus, sp. A Family unknown Olavius cornuatus Davis, 1984* Oligochaeta, n. fam., sp. A O. tenuissimus (Erséus, 1979) * Denotes species which have been dealt with in previous publications. glass slides. The type-series of the new species are deposited at the United States National Museum of Natural History (USNM), Washington, D.C. Area Studied Figure | is a map of the area southeast of Massachusetts showing Georges Bank and the surrounding area where the samples were taken. Station 5 was the location of a drilling platform and was designated as the “site-specific station.” Station 5 consisted of an array of 29 stations (5-1 to 5-29) located around the drill site. The remainder of the stations were located at key sites on Georges Bank and were designated as “regional stations.” Systematics Adelodrilus bacrionis, new species Figs. 2, 3 Holotype.—USNM 97236. Type-locality.— Georges Bank, SE of Massachusetts, USA, sta. 5-28, 40°39. 5'N, 67°41.9'W, 75 m, medium to coarse sand, Nov 1983. Paratypes.—USNM 97237-97238. Two specimens from off Massachusetts (Georges Bank), sta. 2, 40°59.0’N, 66°55.8'W, 71 m, medium to coarse sand, Nov 1983. Other material examined.—Five specimens from off Massachusetts (Georges Bank): 4 from type-locality, one from sta. 5-3, 40°39.8'N, 67°46.1'W, 78 m, medium to coarse sand, Nov 1981. Etymology.—The name bacrionis is Latin for “ladle,” referring to the shape of the giant penial setae of this species. 160 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Martha's 6 Vineyard Block Island @Regional Sampling MSite—Specific Sampling (0) 25 50 eed Mile 0 25 50 PS Km Bathymetry in Meters 72 7e 70° 69° 68° 67° 66° Fig. 1. Map showing positions of sampling stations. Description. — Length 1.9-3.0 mm, 26-33 segments; diameter: 90-151 wm an- teriorly, 116-148 wm at segment XI, 62-142 um posteriorly. Clitellum not ob- served in these specimens. Secondary annulation weak, 3—4 annuli per segment. Prostomium conical and rounded. Somatic setae (Fig. 3C) similar throughout except ventral setae of segments IX and X in mature specimens. Somatic setae slightly sigmoid; bifid with small, short teeth; upper tooth smaller than lower tooth; 41-59 um long, 1.5—1.8 um thick, 2—4 per bundle anteriorly; 32-48 wm long, 1.4—1.8 um thick, 2-3 per bundle except in specimens in transition stage where normal somatic setae may still be present to give 3—5 per bundle; one or two setae of a modified bundle bifid, slightly enlarged, 41-63 um long, 1.8—2.5 um thick; one seta of modified bundle sharply single pointed, larger than bifid setae of same bundle, 48-68 wm long, 2.7—3.2 wm thick. Ventral setae of XI highly modified into penial bundles, each containing: (1) one giant seta (Figs. 2:gs, 3A) with long, slightly sigmoid shaft much widened and cupped ectally to form ladle shape; 120-146 um long, 10.0—10.8 um thick at middle (much wider ectally); (2) one medium-sized, simple-pointed seta (Figs. 2:is, 3B) with 90 degree bend in middle and near ental end, slightly curved ectally, 86-98 um long, 3.6—4.5 um thick at middle; (3) 4—7 thin ectally hooked, small setae (Fig. 2:ss) 29-33 um long, 0.9-1.1 um thick. Giant and intermediate penial setae parallel and small penial setae perpendicular to long axis of worm. Ectal ends of penial setae located within deep, folded copulatory sacs. VOLUME 98, NUMBER 1 161 50pm (ered Seedbass ceal| Fig. 2. Adelodrilus bacrionis, lateral view of male genitalia in segment XI. a, atrium; cs, copulatory sac; gs, giant penial seta; is, intermediate penial seta; pr1, anterior prostate gland; pr2, posterior prostate gland; ss, small penial setae; vd, vas deferens. Pharyngeal glands small and lobed, located in V and anterior part of VI. Male genitalia (Fig. 2) paired in XI. Vas deferens thin-walled and broadly dilated; 5— 18 wm wide, about 150 um long; filled with random sperm; entering apex of atrium. Atrium elongate-oval; 54-75 um long, 23-34 um wide; thin outer lining; thick, granulated and ciliated inner epithelium; opening into inner end of copu- ow 50um [at et a Ff Fig. 3. Adelodrilus bacrionis: A, Giant penial seta; B, Intermediate penial seta; C, Somatic seta; D, Lateral view of segment X. s, spermatheca; se, ventral setae of segments IX and X. 162 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 4. Phallodrilus dorsospermatheca: A, Lateral view of male genitalia in segment XI, with pseudopenis retracted; B, Lateral view of male genitalia in segment XI, with pseudopenis everted. pp, pseudopenis; ps, penial setae; other abbreviations as for Fig. 2. latory sac. Anterior prostate gland large, somewhat lobed; positioned anterior and dorsal to atrium; attached to ental end of atrium, near entrance of vas deferens. Posterior prostate smaller than anterior prostate; also somewhat lobed; positioned dorsal and posterior to atrium, attached by long stalk to middle to ectal part of atrium. Male pores paired; located in line with ventral setae in posterior part of XI. Spermathecae (Fig. 3D) long and narrow, with 90 degree bend at point where duct enters ampullae; ducts long and hollow, 41-59 um long, 18-23 um wide; ampullae thin-walled and elongate, 33-59 um long, 14-23 um wide. Sperm in random masses. Spermathecal pores paired, in line with ventral setae in X near intersegmental furrow IX/X. Remarks.—This species is closely related to Adelodrilus cooki Erséus, 1978, and A. pilatus Erséus and Davis, 1984. All three species share the following characteristics: single-pointed setae in the ventral bundles of IX and X, spoon- shaped giant penial setae accompanied by intermediate penial setae, and small penial setae. It is also likely that A. pusillus Erséus, 1978, A. kiselevi (Finogenova, 1972), and A. correptus Erséus and Davis, 1984, are closely related to this group but these species do not have the modified setae in the ventral bundles of LX and X. Adelodrilus bacrionis is most easily distinguished from the other species in this group by the geniculate shape of both the intermediate penial setae and the sper- mathecae. Habitat.—Sublittoral, medium to coarse sand, 71—78 m. Distribution.— Northeast coast of the U.S.: off Massachusetts (Georges Bank). Phallodrilus dorsospermatheca, new species Figs. 4, 5 Holotype,—USNM 97233. Type-locality.— Georges Bank, SE of Massachusetts, USA, sta. 2, 40°59.0'N, 66°55.8’W, 71 m, medium sand, Feb 1984. VOLUME 98, NUMBER 1 163 3oum H 50um Fig. 5. Phallodrilus dorsospermatheca: A, Somatic seta; B, Penial seta; C, Lateral view of sper- matheca in segment X; D, Pseudopenis partially everted. Paratypes.—USNM 97234-97235. Two specimens from off Massachusetts (Georges Bank), both from type-locality. Other material examined.— Three specimens from off Massachusetts (Georges Bank), all from type-locality. Etymology.—The name dorsospermatheca refers to the dorsal position of the spermatheca. Description. — Length 3.8-5.3 mm, 34-38 segments; 66-99 um wide anteriorly, 84-99 um wide at segment XI, 55-90 wm wide posteriorly. Clitellum not observed in specimens examined. Secondary annulation very weak or not present. Prosto- mium conical and rounded, 1'/ times long as wide. Somatic setae (Fig. 5A) similar in shape throughout, sigmoid, bifid, upper tooth thinner and shorter than lower tooth; 30-36 wm long, 1.5-1.8 wm thick, 3—4 per bundle anteriorly; 28-33 um long, 1.5—1.8 wm thick, (1)—2—(3) per bundle posteriorly. Ventral setae of segment XI modified into penial bundles (Fig. 4:ps), each containing 4 setae. Penial setae (Fig. 5B) straight, single pointed, ectally hooked and clubbed, about 22 um long, 1 wm thick (very difficult to establish exact size); situated perpendicular to long axis of worm. Pharyngeal glands located in segments V and VI and anterior part of VII. Male genitalia (Fig. 4) paired in segment XI. Vas deferens (Fig. 4:vd) short, thin-walled, ciliated; 70-75 um long, 4.5—6.3 wm wide; entering atrium apically. Atrium (Fig. 4:a) straight and elongated, nearly erect; moderately thick outer lining; thick, granulated and ciliated inner epithelium; opening into small, simple copulatory sac (Fig. 4:cs); ectal tip capable of being protruded to form short, thick pseudopenis (Fig. 4:pp). Anterior prostate gland (Fig. 4:pr1) small, compact; situated ventral to atrium; attached to ental end of atrium, near entrance of vas deferens. Posterior prostate (Fig. 4:pr2) small, compact; situated directly posterior to atrium; attached to atrium ectally by short stalk. Male pores paired; in line with ventral setae in posterior part of segment XI. Spermathecae (Fig. 5C) small, elongated (probably not completely formed); duct 5—6 um long, 6—7 wm wide; ampullae 18-36 um 164 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 6. Phallodrilus christeri, lateral view of male genitalia in segment XI. Abbreviations as for Fig. 4. long, 12—14 um wide. No sperm observed. Spermathecal pores paired, in line with dorsal setae in anterior part of segment X. Remarks.—As discussed by Erséus (1984a), many species of Phallodrilus (P. coeloprostatus Cook, 1969; P. biprostatus (Baker & Erséus, 1979); P. parviatriatus Cook, 1971; and others) from the northwest Atlantic share the feature of small, ectally hooked and clubbed penial setae. Phalladrilus dorsospermatheca, which also shares this feature, further supports the view of Erséus that these species are a monophyletic group which has evolved and radiated in the northwest Atlantic. Phallodrilus dorsospermatheca is also similar to P. minutus Hrabé, 1973. Both species are small overall with small genitalia, similar penial setae, and spermathe- cal pores which are more dorsally positioned than those of most other members of the genus Phallodrilus. The shape of the protruded pseudopenis shown in the redescription of P. minutus (Erséus and Kossmagk-Stephan 1983; Fig. 1:pp) very closely resembles the partly protruded pseudopenis (Fig. 5D) of P. dorsosper- matheca. Phallodrilus dorsopermatheca differs from P. minutus in having smaller and more sharply hooked penial setae and spermathecal pores situated in line with the dorsal setae instead of more laterally. Habitat. —Sublittoral, medium sand, 71 m. Distribution. —Northeast coast of U.S.: off Massachusetts (Georges Bank). Phallodrilus christeri, new species Figs. 6, 7 Holotype.—USNM 97239. Type-locality.— Georges Bank, SE of Massachusetts, sta. 5-1, USA, 40°39.5’N, 67°46.2’W, 79 m, medium to coarse sand, Nov 1983. Paratypes.—USNM 97240-97241. Two specimens from off Massachusetts VOLUME 98, NUMBER 1 165 [5 yum 50m [Rsuinn ances # Fig. 7. Phallodrilus christeri: A, Somatic seta; B, Lateral view of spermatheca in segment X. (Georges Bank): one from type-locality; one from sta. 5-18, 40°39.6’N, 67°47.6'W, 80 m, medium to coarse sand, Nov 1984. Other material examined.— Four specimens from off Massachusetts (Georges Bank): three from sta. 5-4, 40°39.5'N, 67°46.5’W, 79 m, medium to coarse sand, May 1983; one from sta. 5-18, 40°39.6'N, 67°47.6’W, 80 m, medium to coarse sand, Nov 1983. Etymology.—This species is named for Dr. Christer Erséus (University of G6- teborg, Sweden) who has made many important contributions to the study of marine oligochaetes. Description. —Length 4.0-6.9 mm, 44—50 segments; 67—132 wm wide anteriorly, 134-142 um wide at segment XI, 55—143 wm wide posteriorly. Clitellum extending over X and XI. Secondary annulation strongly developed, 4-6 annuli per segment. Prostomium conical and rounded, 1'/ times longer than wide. Somatic setae (Fig. 7A) similar throughout; bifid, slightly sigmoid, with upper tooth smaller than lower tooth; 38—42 wm long, 1.0—1.5 wm thick, 2-3 per bundle anteriorly; 38-41 um long, 1.0—1.5 um thick, 2 per bundle posteriorly. Ventral setae of XI modified into penial bundles (Fig. 6:ps) with each containing 2 setae. Penial setae bifid; long, thin teeth equal in length; upper tooth thicker than lower tooth; shaft slightly sigmoid, sharply curved ectally; 38-48 um long, 1.5—1.8 um thick. Penial setae perpendicular to long axis of worm. Pharyngeal glands small, compact; located in posterior part of III. Male genitalia (Fig. 6) paired in XI. Vas deferens (Fig. 6:vd) moderately thick-walled and ciliated; 150-180 um long, 6.5-9.0 wm wide; entering atrium ectally. Atrium (Fig. 6:a) small, oval to comma-shaped; moderately thick outer lining; thick, granulated and ciliated inner epithelium; 41-62 um long, 15-17 wm wide; opening directly to outside through male pore; ectal tip slightly protrusible to form small pseu- dopenis. Anterior prostate gland (Fig. 6:prl) small and compact; positioned an- terior to atrium; entering atrium entally near entrance of vas deferens. Posterior 166 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON prostate gland (Fig. 6:pr2) also small and compact; positioned posterior to atrium; entering atrium ectally. Male pores paired; located in line with ventral setae in middle of segment XI. Spermathecae (Fig. 7B) large and oval; duct very short, 9-14 um long, 9-10 um wide; ampullae with very thick and glandular walls, 54— 128 wm long, 23-37 um wide, actual lumen small, filled with random sperm. Spermathecal pores paired, in line with ventral setae in anterior part of X. Remarks.—Phallodrilus christeri does not seem to be closely related to any single species, but instead, shares features of several species. The basic shape and structure of the atria are common in this genus. The bifid penial setae are found in only three other species (P. aquaedulcis Hrabé, 1960; P. vulnus Erséus, 1983b; and P. cristolatus Erséus, 1983b), but this is the only feature these species share with P. christeri. The male genitalia and the histology of the spermathecae of P. postspermathecatus Erséus, 1980, are similar to the new species, but the sper- mathecae are located in segment XII and the penial setae are not bifid. Habitat.—Sublittoral, medium to coarse sand, 79-80 m. Distribution. — Northeast coast of the U.S.: off Massachusetts (Georges Bank). Phallodrilus parviatriatus Cook, 1971 Phallodrilus parviatriatus Cook, 1971:204—207, fig. 1.—Erséus, 1979c:190. Type-material.—USNM 42015-42017; National Museum of Natural Sciences, Ottawa, Canada, 3413. Type-locality.—Cape Cod Bay, Massachusetts, USA, 41°54.0’N, 70°8.6’W, 17.1 m. New material examined.— Four specimens from off Massachusetts (Georges Bank): one from sta. 5-4, 40°39.5'N, 67°46.5'W, 79 m, medium to coarse sand, Jul 1981; one from sta. 5-22, 40°39.5’N, 67°43.3'W, 77 m, medium to coarse sand, Jul 1981; two from sta. 5-28, 40°39.5’N, 67°41.9’'W, 75 m, medium to coarse sand, Feb 1983. Remarks.—The new material conforms in all ways to the original description (Cook 1971) and represents a slight depth extension from 41 m to 79 m. Habitat.—Subtidal, medium to coarse sand, 17-79 m. Distribution.— Northeast coast of the U.S.; New Jersey, Massachusetts (Cape Cod Bay and Georges Bank). Olavius tenuissimus (Erséus, 1979c) Phallodrilus tenuissimus Erséus, 1979c:199-200, fig. 20. Olavius tenuissimus.—Erséus, 1984a. Type-material.—USNM 56208-56212. Type-locality. —Off Miami, Florida, USA, 25°43.0’N, 80°10.2’W, 3 m. New material examined.—Seven specimens from off Massachusetts (Georges Bank): five from sta. 9, 40°26.7'N, 68°09.8’W, 141 m, fine to medium sand, three collected Feb 1982, two collected Nov 1983; two from sta. 12, 40°22.2’N, 68°30.2’W, 103 m, fine to coarse sand, Nov 1983. Remarks. —Phallodrilus tenuissimus was thoroughly described by Erséus (1979c). Although the new material examined conforms to the original description, nearly all measurements tended to be in the low end of the size range reported by Erséus. VOLUME 98, NUMBER 1 167 Fig. 8. Uniporodrilus vestigium, lateral view of male genitalia in segment XI. sa, styliform atrium; other abbreviations as for Fig. 4. In the original description, Erséus (1979c) noted that the size of the penial setae varied between material from different localities. As for the other characteristics in the new material, the penial setae tended to be shorter than average (38-51 pm as compared to 41-74 um in the original description). All specimens in the new material have only one penial seta per ““bundle.”’ All of the material from the original description was found at shallow depths (less than 10 m) while the new material was found only at depths greater than 100 m. The new material represents a considerable range extension (and depth exten- sion) northward from North Carolina to off Massachusetts (Georges Bank). Habitat.—Subtidal, fine to coarse sand, 0.5-141 m. Distribution.—East coast of U.S.: Florida, North Carolina, Massachusetts (Georges Bank), Bermuda. Uniporodrilus vestigium, new species Figs. 8, 9, 10 Holotype.—USNM 97247. Whole mounted specimen. Type-locality.— Georges Bank, SE of Massachusetts, USA, sta. 5-1, 40°39.5'N, 67°46.2'W, 79 m, medium to coarse sand, Jul 1983. Paratypes.—USNM 97248-97250. One sectioned specimen, two whole mount- ed specimens from off Massachusetts (Georges Bank): sectioned specimen from sta. 5-11, 40°39.2’N, 67°46.6'W, 80 m, medium to coarse sand, May 1983; whole mounted specimens from type-locality, one collected May 1983. Other material examined.—Two sectioned specimens, three whole-mounted specimens from off Massachusetts (Georges Bank): one sectioned specimen from type-locality; one sectioned specimen from sta. 5-11, 40°39.2'N, 67°46.6’W, 80 m, medium to coarse sand, Jul 1983; one whole mounted specimen from type- locality; one from sta. 5-3, 40°39.8’N, 67°46.1'W, 78 m, medium to coarse sand, May 1983; one from sta. 5-20, 40°38.5'N, 67°46.1’W, 78 m, medium to coarse sand, Nov 1982. 168 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON — ; ( (4 We Sn pe CWRU (LALZGE CCE — 1ooym (eer | Fig. 9. Uniporodrilus vestigium, ventral view of spermatheca and male genitalia in segments X and XI. s, spermatheca; other abbreviations as for Fig. 8. Etymology.—The name vestigium is Latin for “‘footprint,’’ or something left behind. This refers to one bundle of penial setae which is “left behind” after the atrium from that side of the worm has disappeared. Description.—Length 5.5-7.1 mm, 48-54 segments; nearly uniform width throughout, 205-223 um wide at segment XI. Clitellum extending over XI and XII. Prostomium conical and rounded, slightly longer than wide. Secondary an- nulation strongly developed, 6-7 annuli per segment. Somatic setae (Fig. 10A) similar throughout, slightly sigmoid, bifid with upper tooth smaller than lower tooth; 35-51 um long, 1.8—2.0 wm thick, 2-3 per bundle anteriorly; 33-42 wm long, 1.5-1.8 um thick, 2 per bundle posteriorly. Ventral setae of segment XI modified into penial bundles (Figs. 8—9:ps), each containing 4—6 setae. Penial setae slightly sigmoid, single pointed and ectally hooked; 70-78 um long, 2.0—2.5 um thick; situated slightly oblique to parallel to long axis of worm. Pharyngeal glands small, located in segment V and anterior part of segment VI. Male genitalia (Figs. 8—9): sperm funnels, vasa deferentia, and anterior prostate glands paired; atrium and posterior prostate gland unpaired; in segment XI. Vasa deferentia (Figs. 8—9:vd) highly coiled, with thick muscular walls; vas deferens of side lacking atrium longer than other side, running under nerve cord and entering atrium apically, approximately 380 wm long; vas deferens of atrium side entering atrium apically near entrance of other vas deferens, approximately 270 um long. Atrium (Figs. 8—9:a) large, horseshoe-shaped, with thick, very muscular outer lining and thick, granulated inner epithelium; 250-345 um long, 72-77 um wide. Ectal tip of atrium styliform (Figs. 8—9:sa) and capable of protruding through body wall. Anterior prostate glands (Figs. 8—9:prl) small; one on atrium side of worm, entering atrium entally; one on side lacking atrium, possibly vestigial, attachment to atrium not seen. Posterior prostate gland (Figs. 8—9:pr2) very large, extending over both sides of worm, attached to ectal end of atrium. Male pore unpaired, located in line with ventral setae of atrium side of worm in posterior part of VOLUME 98, NUMBER 1 169 loym Fig. 10. Uniporodrilus vestigium: A, Somatic seta; B, Spermatheca. segment XI. Spermatheca (Figs. 9:s, 10B) unpaired, very large; duct moderately long and hollow, with first % having wide lumen to act as vagina to receive styliform atrium, last '4 constricted, 50-55 um long, 23-32 um wide; ampullae elongate-oval, sometimes slightly curved, with walls very thick containing large glandular spheres and smaller, yellowish spheres probably containing lipids, 145— 230 um long, 62—80 wm wide, walls up to 20 wm thick. Sperm in random masses. Spermathecal pore unpaired, mid-ventral or slightly to one side; in segment X near intersegmental furrow of IX/X. Remarks.— Uniporodrilus vestigium is closely related to U. granulothecus Er- séus, 1979b. The new species differs in having only one posterior prostate gland, the male pore being in line with the ventral setae instead of being mid-ventral, and the spermathecal pore in segment X instead of segment IX. The new species also has a more muscular atrium and vasa deferentia and fewer penial setae. With the addition of this species, the generic description of Uniporodrilus Erséus, 1979b, is slightly modified to include species with the spermathecal pore in seg- ment X, the male pore in line with the ventral setae, and one posterior prostate gland. If other species of this genus are found, it is likely that these characteristics will be highly variable. The most important generic characteristics are the unpaired atrium and spermatheca. Habitat. —Sublittoral, medium to coarse sand, 78-80 m. Distribution.— Northeast coast of the U.S.: off Massachusetts (Georges Bank). Bathydrilus longus Erséus, 1979d Bathydrilus longus Erséus, 1979d:144—145, fig. 9. Type-material.—USNM 55693-55700. Type-locality.—Continental shelf off the coast of New Jersey, USA, NW At- lantic, 30°06.6'N, 72°59.0'W, 70 m. New material examined (new records). Four specimens from off Massachusetts (Georges Bank): two from sta. 16, 40°34.2'’N, 67°12.3’W, 138 m, medium to coarse sand, Nov 1983; one from sta. 12, 40°22.2’N, 68°30.2’W, 103 m, medium to 170 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON coarse sand, Nov 1983; one from sta. 7, 40°28'N, 67°43.2'W, 165 m, medium to coarse sand, Jul 1981. Remarks. — Bathydrilus longus was thoroughly described by Erséus (1979d). The new specimens conform in most respects to the original description, with two notable exceptions. In two of the specimens examined, the male pores appeared to be located in the anterior part of segment XII. It was very difficult to establish the exact position of the septum between segments XI and XII but it was clear in these two specimens that the male pores are situated more posteriorly than depicted in the illustration for the original description (Erséus 1979d, fig. 9). The other exception to the original description is the length of the atrium. Three of the specimens examined have atria with lengths of 270-320 wm, which is considerably longer than the originally reported lengths of 185-250 um long. This material represents a slight range extension northward from off New Jersey to off Massachusetts. This material was also found at slightly deeper depths of 103-165 m as compared to 70-91 m for the type material. Habitat. —Sublittoral, medium to coarse sand, 70-165 m. Distribution.— Northeast coast of the U.S.: off New Jersey and Massachusetts (Georges Bank). Heterodrilus occidentalis Erséus, 1981 Heterodrilus occidentalis Erséus, 1981:121, fig. 11. Type-material.—USNM 60606-60610. Type-locality. —Near Beaufort, North Carolina, USA, 34°41'58”N, 76°37'20’W, 5 m. New material examined (new records).— Three specimens from off Massachu- setts (Georges Bank): two from sta. 15, 41°27.5'N, 68°00.7'W, 38 m, fine to medium sand, Jul 1981; one from sta. 16, 40°34.2'’N, 67°12.3’W, 138 m, medium to coarse sand, Nov 1981. Remarks.—The new material conforms in all ways to the original description (Erséus 1981). These records represent a slight range extension northward from New Jersey to off Massachusetts. One of the new specimens was found at 138 m, which is a considerable depth extension from 24 m. Habitat.—Subtidal, fine to coarse sand, 4.5—138 m. Distribution. —East coast of U.S.; North Carolina, New Jersey, and Massachu- setts (Georges Bank). Limnodriloides medioporus Cook, 1969 Fig. 11 Limnodriloides medioporus Cook, 1969:21—22, fig. 7; 1970b:980; 1971:212; 1974, 131.—Brinkhurst and Jamieson, 1971:530, fig. 8.24 A—C.—Erséus, 1982:225- 226, fig. 9. Type-locality.—Cape Cod Bay, Massachusetts, USA, 41°54.9'N, 70°15.12'W, 36.5 m. New material examined. —Six specimens from off Massachusetts (Georges Bank): four from sta. 13, 40°29.5'N, 70°12.6’W, 67 m, silt to very fine sand, Nov 1983; one from sta. 13A, 40°30.0’N, 71°00.5'W, 78 m, silt, Nov 1983; one from sta. TA, 40°32.15’N, 67°44.2'W, 165 m, silt to very fine sand, Feb 1984. VOLUME 98, NUMBER 1 171 10oym Fig. 11. Limnodriloides medioporus, lateral view of male genitalia in segment XI, with pseudopenis everted. pr, prostate gland; ad, atrial duct; other abbreviations as for Fig. 4. Remarks.— The specimens examined here conform in every way to the original description with one minor exception. The original description (Cook 1969) states that there are two somatic setae per bundle posteriorly. In the new material, there were occasionally only one or up to three per bundle. The median pore is often not visible, depending on whether the penes are everted or not. These can vary from the relaxed state, as in the original description (Cook 1969, fig. 7), to the fully everted position (Fig. 11). In many cases, the penes will be only partially everted so the median pore disappears, which is confusing for identification pur- poses. Habitat.—Subtidal silt to medium sand, 18-97 m. Distribution.— Northeast coast of U.S., Massachusetts through New Jersey. Limnodriloides barnardi Cook, 1974 Limnodriloides barnardi Cook, 1974:134-135, fig. 5.—Erséus 1976:32-33, fig. 3; 1982:232—234, not fig. 13. Limnodriloides winckelmanni Michaelsen, 1914:— Jamieson, 1977:338, not fig. 2. Type-material.—USNM 48730-48731. Type-locality.— Bahia de San Quintin, Pacific coast of Mexico, less than 2 m. New material examined (new records).—Three specimens from off Massachu- setts (Georges Bank); all from sta. 12, 40°22.2’N, 68°30.2’W, 103 m, medium to coarse sand, Nov 1983. Remarks.—Limnodriloides barnardi was thoroughly described by Cook (1974) and subsequently reviewed by Erséus (1982). The new material examined con- formed closely to the material examined by Erséus from the northern localities along the U.S. east coast. In the specimens studied here, the male and spermathecal pores were located very close together. The male pores were so close together in one specimen that they formed an I-shaped common median bursa. The spermathecal setae were 172 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON also smaller than those of the type-material (65-77 wm compared to 110-120 um), but were all located posterior to the spermathecal pores. This material represents a slight northward range extension on the east coast of the U.S. from New Jersey to off Massachusetts (Georges Bank). Habitat. —Subtidal, silt to coarse sand, 0.5-150 m. Distribution. — Massachusetts, New Jersey, Maryland, Virginia, North Carolina, East coast of Florida, Bahamas, Bermuda, and Pacific coast of Mexico. Tubificoides intermedius (Cook, 1969) Peloscolex intermedius Cook, 1969:11-12, fig. 2.—Brinkhurst and Jamieson, 1971: 512-513, fig. 8.22 K, L.—Holmquist, 1978: fig. 6B. Tubificoides intermedius Brinkhurst and Baker, 1979:1559, fig. 7. Type-material.—USNM 38259-38262. Type-locality.—Cape Cod Bay, Massachusetts, USA, 41°55.75'N, 70°21.07'W, 42.6 m. New material examined.— Three specimens from off Massachusetts (Georges Bank); all from sta. 13, 40°29.5'N, 70°12.6'W, 67 m, silt to very fine sand, one collected Nov 1982, two collected Nov 1981. Remarks.—The new material examined here conforms in most ways to the original description (Cook 1969); however, as mentioned by Brinkhurst and Baker (1979), the illustration in the original description (Cook 1969, fig. 2) is not accurate. The illustration by Holmquist (1978, fig. 6B) very accurately depicts the male genitalia as seen in a whole mount in the material examined here. Habitat.—Subtidal, silt to very fine sand, 7-67 m. Distribution.—Northeast coast of U.S.; Massachusetts (Cape Cod Bay and Georges Bank), Marionina welchi Lasserre, 1971 Marionina welchi Lasserre, 1971:453-454, fig. 2. Type-material.—USNM 43479-43481. Type-locality.—Cape Cod Bay, Massachusetts, USA, 41°54.00’N, 70°08.40'W, 17 m. New material examined (new records).—Six specimens from off Massachusetts (Georges Bank): one from sta. 5-10, 40°39.4'N, 67°46.9’W, 80 m, medium to coarse sand, Jul 19-14, 40°39.5'N, 67°44.7’'W, 79 m, medium to coarse sand, Jul 1981; one from sta. 5-2, 40°39.6'N, 67°45.8’W, 78 m, medium to coarse sand, May 1982; one from sta. 5-1, 40°39.5'N, 67°46.2'W, 79 m, medium to coarse sand, Nov 1983; two from sta. 5-18 40°39.6’N, 67°47.6'W, 80 m, medium to coarse sand, Nov 1983. Remarks.—The new material conforms in every way to the original description (Lasserre 1971). Marionina welchi was previously known only from Cape Cod Bay, Massachusetts. Although the new material does not represent a geographical range extension, it certainly represents a considerable habitat extension. Habitat.—Subtidal, medium to coarse sand, 14.6—80 m. Distribution. —Northeast coast of U.S.; Massachusetts (Cape Cod Bay, Georges Bank). VOLUME 98, NUMBER 1 173 Unidentified Material This group, with the exception of Phallodrilus sp. A, represents species which have not been identified because of problems within their respective taxonomic groups. Phallodrilus, sp. A has been confirmed to be a new species by Erséus (pers. comm.), but there was only one specimen found during the program. This species will be described by Erséus if it is determined that a complete description can be obtained from this one specimen. The genus Jubificoides is in dire need of revision, which often makes identi- fications questionable. This is the case for Tubificoides, sp. A and B. Species A is similar to 7. maureri, but the opening in the penial sheath seems to be more lateral in 7. maureri. Species B keys out to be 7. apectinatus using the key to the species of Tubificoides by Brinkhurst and Baker (1979). The problem lies in the difference between the original description (Brinkhurst 1965:133-134, fig. 5, O- T) and the description by Brinkhurst and Baker (1979:1559, fig. 9). There is a considerable difference in the shape of the penial sheath in these two descriptions, making it difficult to know which is correct. These problems will probably not be remedied until the revision of this genus is complete. Four new species of Grania were found from Georges Bank and confirmed by Erséus (pers. comm.). The descriptions of these species will be included in a revision of the genus Grania, which is currently being undertaken by Erséus and Coates (Erséus, pers. comm.). The species designated as Oligochaeta, n. fam., sp. A, represents what is probably a new family, which has been found in other material from along the east and west coasts of the U.S. and the Caribbean (Erséus, pers. comm.). The material from Georges Bank, along with material from the other locations, is currently being worked out by Erséus (pers. comm.). Discussion No specific data were recorded on life history or sediment-species relationships during the Georges Bank Monitoring Program; however, many obvious trends were observed which were consistent throughout the three-year program and should be noted. It should be emphasized that these observations are not based on quantitative data. Sediment-Species Relationships The sediment at Station 5 consisted mainly of medium to coarse sand, a feature shared by only two other stations (2 and 16). All other stations had a higher percentage of fine sand, except Stations 13, 13A and 14A, which were predomi- nantly silt. This ““medium to coarse sand” habitat is obviously preferred by most oligochaetes. In the three stations with this sediment type, there were generally about 50 to 75 specimens per sample with up to 20 species present. In the stations with higher percentages of fine sand, there were rarely more than 25 specimens per sample and generally not more than five species present. In the stations with high amounts of silt, there were large numbers of only one to three species. The medium to coarse sand sediments were dominated by members of the subfamily Phallodrilinae; Phallodrilus, Adelodrilus, and Uniporodrilus. Bathydri- lus is the only member of this subfamily that was not found in the coarse sediment, 174 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON which may be an indication of its relationship to the Phallodrilinae. All species of the family Enchytraeidae reported herein were also found in the coarse sedi- ments. The finer sediments were dominated by Limnodriloides medioporus and Tu- bificoides, sp. A. Bathydrilus longus, Heterodrilus occidentalis, Limnodriloides barnardi, and the gutless Phallodrilinae occurred repeatedly at specific stations without any apparent relationship to sediment type. The stations with silty sediments always had large numbers (up to 300 per sample) of Limnodriloides medioporus which were usually accompanied by a few specimens of Tubificoides intermedius and/or Tubificoides, sp. B. Life Histories Sexually mature specimens of many of the species were found in all samples suggesting that they are capable of reproducing year-round. It was quite apparent that the number of sexually mature specimens for all species was higher in the November samples; however, Adelodrilus pilatus seemed to be the only species that was sexually mature exclusively in November. There are obviously some species with a much lower number of sexually mature individuals than juveniles at any time of the year when compared to the “usual number,” which can be shown by example. Tubificoides, sp. B was relatively common in nearly all samples from Station 13A, but only three sexually mature specimens were found during the entire program. For most species, there were usually at least a few sexually mature individuals in every sample. Geographical Distribution Nearly all of the species from Georges Bank are limited in their distribution to the northeast coast of the United States. The distribution of Olavius tenuissimus extends south to Florida and Bermuda and Limnodriloides barnardi may be cos- mopolitan in its distribution, being found from the east and west coasts of North America. The unidentified species such as Tubificoides, sp. B and the Grania complex may be more widely distributed if they turn out to be known species. Acknowledgments I am indebted to Dr. Christer Erséus (University of G6teborg, Sweden) for his technical assistance and encouragement; to Dr. James A. Blake and Dr. Nancy J. Maciolek (Battelle New England Marine Research Laboratory) for their valuable criticism of the manuscript; and to Battelle New England Marine Research Lab- oratory for financial support. Specimens from Georges Bank were collected as part of the Georges Bank Benthic Infauna Monitoring Program, supported by contract no. 14-12-0001-2912 from the U.S. Department of the Interior Minerals Management Service to Battelle New England Marine Research Laboratory. Literature Cited Baker, H. R., and C. Erséus. 1979. Peosidrilus biprostatus n. g., n. sp., a marine tubificid (Oligochaeta) from the eastern United States.—Proceedings of the Biological Society of Washington 92(3): 505-509. VOLUME 98, NUMBER 1 175 Brinkhurst, R. O. 1965. Studies of the North American aquatic Oligochaeta II. Tubificidae.—Pro- ceedings of the Academy of Natural Sciences of Philadelphia 117:117-172. 1966. A contribution to the systematics of the marine Tubificidae.— Biological Bulletin 130(3):297-303. ,and H.R. Baker. 1979. A review of the marine Tubificidae (Oligochaeta) of North America. — Canadian Journal of Zoology 57(8):1553-1569. , and B. G. M. Jamieson. 1971. Aquatic Oligochaeta of the world.—Oliver and Boyd, Edin- burgh, 860 pp. Cook, D. G. 1969. The Tubificidae (Annelida, Oligochaeta) of Cape Cod Bay with a taxonomic revision of the genera Phallodrilus Pierantoni, 1902, Limnodriloides Pierantoni, 1903 and Spiridon Knoller, 1935.—Biological Bulletin 136:9—27. 1970a. Peloscolex dukei n. sp. and P. aculeatus n. sp. (Oligochaeta, Tubificidae) from the Northwest Atlantic, the latter being from abyssal depths.— Transactions of the American Mi- croscopical Society 88(4):492—497. . 1970b. Bathyaland abyssal Tubificidae (Annelida, Oligochaeta) from the Gay Head-Bermuda transect, with descriptions of new genera and species.— Deep-Sea Research 17:973-98 1. . 1971. The Tubificidae (Annelida, Oligochaeta) of Cape Cod Bay. II: Ecology and systematics, with the description of Phallodrilus parviatriatus nov. sp.— Biological Bulletin 141(2):203-221. . 1974. The systematics and distribution of marine Tubificidae (Annelida, Oligochaeta) in the Bahia de San Quintin, Baha California, with descriptions of five new species. — Bulletin of the Southern California Academy of Sciences 73(3):126—140. , and R. O. Brinkhurst. 1973. Marine flora and fauna of the northeastern United States. Annelida: Oligochaeta.— NOAA Technical Report NMFS CIRC-374. Davis, D. A. 1984. Olavius cornuatus sp. n. (Oligochaeta, Tubificidae) from Georges Bank (NW Atlantic). — Zoologica Scripta 13 [in press]. Erséus, C. 1976. Marine subtidal Tubificidae and Enchytraeidae (Oligochaeta) of the Bergen area, Western Norway.—Sarsia 62:25—40. . 1978. New species of Adelodrilus and a revision of the genera Adelodrilus and Adelodriloides (Oligochaeta, Tubificidae).—Sarsia 63:135-144. 1979a. Bermudrilus peniatus n. g., n. sp. (Oligochaeta, Tubificidae) and two new species of Adelodrilus from the Northwest Atlantic. — Transactions of the American Microscopical Society 98(3):418—427. . 1979b. Uniporodrilus granulothecus n. g., n. sp., a marine tubificid (Oligochaeta) from eastern United States.— Transactions of the American Microscopical Society 98(3):414—418. —. 1979c. Taxonomic revision of the marine genus Phallodrilus Pierantoni (Oligochaeta, Tu- bificidae), with descriptions of thirteen new species. — Zoologica Scripta 8:187—208. 1979d. Taxonomic revision of the marine genera Bathydrilus Cook and Macroseta Erséus (Oligochaeta, Tubificidae), with descriptions of six new species and subspecies.— Zoologica Scripta 8:139-151. —. 1980. New species of Phallodrilus (Oligochaeta, Tubificidae) from Arctic deep-sea and Norwegian fjords.—Sarsia 65:57-60. 1981. Taxonomic revision of the marine genus Heterodrilus Pierantoni (Oligochaeta, Tu- bificidae).— Zoologica Scripta 10:111-132. . 1982. Taxonomic revision of the marine genus Limnodriloides (Oligochaeta, Tubificidae). — Verhandlungen des Naturwissenschaftlichen Vereins in Hamburg (Neue Folge) 25:207-277. 1983a. New records of Adelodrilus (Oligochaeta, Tubificidae), with descriptions of two new species from the Northwest Atlantic.—Hydrobiologia 106:73-83. 1983b. Deep-sea Phallodrilus and Bathydrilus (Oligochaeta, Tubificidae) from the Atlantic ocean, with descriptions of ten new species.— Cahiers de Biologie Marine 24:125-146. . 1984a. Taxonomy of some species of Phallodrilus (Oligochaeta, Tubificidae) from the North- west Atlantic, with description of four new species.— Proceedings of the Biological Society of Washington 97:812-826. 1984b. Taxonomy and phylogeny of the gutless Phallodrilinae (Oligochaeta, Tubificidae) with descriptions of one new genus and twenty-two species. — Zoologica Scripta 13 [in press]. ,and D. Davis. 1984. Three new species of Ade/odrilus (Oligochaeta, Tubificidae) from Georges Bank (NW Atlantic).— Proceedings of the Biological Society of Washington 97:834-843. , and K. J. Kossmagk-Stephan. 1983. Redescription of Phallodrilus minutus (Oligochaeta, 176 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Tubificidae) based on new material from the North Sea coast of Germany.—Sarsia 68:229-— 231. Finogenova, N. P. 1972. New species of Oligochaeta from Dniepr and Bug Firth and Black Sea and revision of some species. — Trudy Zoologicheskogo Instituta. Akademiya Nauk SSSR, Leningrad 52:94—-116 [in Russian]. Holmquist, C. 1978. Revision of the genus Peloscolex (Oligochaeta, Tubificidae). — Zoologica Scripta 7:187-208. Hrabé, S. 1960. Oligochaeta Limicola from the collection of Dr. S. Husmann.—Spisy Prirodovedecke Fakulty University v Brne 415:245-277. . 1973. A contribution to the knowledge of marine Oligochaeta, mainly from the Black Sea.— Travaux du Muséum d’Histoire Naturelle ‘Grigore Antipa’ 13:27-38. Jamieson, B. G. M. 1977. Marine meiobenthic Oligochaeta from Heron and Wistari Reefs (Great Barrier Reef) of the genera Clitellio, Limnodriloides and Phallodrilus (Tubificidae) and Grania (Enchytraeidae). — Zoological Journal of the Linnean Society 61:329-349. Lasserre, P. 1971. The marine Enchytraeidae (Annelida, Oligochaeta) of the eastern coast of North America with notes on their geographical distribution and habitat. — Biological Bulletin 140(3): 440-460. Battelle New England Marine Research Laboratory, 397 Washington Street, Duxbury, Massachusetts 02332. Present address: University of Hawaii at Manoa, Department of Zoology, Edmondson Hall, 2538 The Mall, Honolulu, Hawaii 96822. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 177-190 PENAEOID SHRIMP FAUNA FROM TROPICAL SEAGRASS MEADOWS: SPECIES COMPOSITION, DIURNAL, AND SEASONAL VARIATION IN ABUNDANCE Raymond T. Bauer Abstract.—The motile invertebrate epifauna of seagrass (Thalassia testudinum) meadows at Dorado, north coast of Puerto Rico, was sampled monthly during the day and at night for a year. The penaeoid shrimp component of the mobile epifauna was dominated by two small sicyoniids, Sicyonia parri (65% of N) and S. laevigata (21% of N). Metapenaeopsis goodei, M. martinella, and M. smithi were much less numerous. All penaeoid shrimps were collected in significantly higher numbers at night. Laboratory observations indicate that all species burrow just under the bottom during the day but are active at night. The nocturnal emergence of these penaeoids and their increased susceptibility to capture at and after dusk was documented by sampling which began before and ended after sunset; numbers of shrimp taken increased dramatically with increasing darkness. Sicyonia parri and S. laevigata showed significantly higher abundances in spring and summer months at one of two replicate sampling sites while Metapenaeopsis juveniles exhibited a similar pattern at both sites. There was no evidence of seasonality in Metapenaeopsis adults. In recent years various investigators have conducted sampling programs in subtropical and tropical seagrass meadows dominated by turtlegrass, Thalassia testudinum (Bauer, in press; Greening and Livingston 1982; Gore et al. 1981; Heck 1976, 1977, 1979; Thorhaug and Roessler 1977; Hooks et al. 1976). The results of these studies on community structure show that, as in seagrass meadows in temperate areas (Heck and Orth 1980a, b; Kikuchi and Pérés 1977; Kikuchi 1966), the motile invertebrate epifauna sampled by pushnets, epibenthic dredges, otter trawls, and drop net techniques is dominated numerically by decapod crus- taceans such as caridean shrimps, penaeoid shrimps, paguroid crabs, and brachy- uran crabs. Kikuchi (1966, 1974) and Reid (1954) reported that decapods, e.g., shrimps, are preferred food items of fishes foraging over seagrass beds. Initially, analyses of community structure, 1.e., species composition and relative abundance, were carried out on collections taken during daylight hours. More recently, Bauer (in press), Leber and Greening (ms), and Greening and Livingston (1982) have demonstrated that the Thalassia epifaunal community is ‘“‘awake”’ at night; more species are collected at night and individual species abundances are significantly higher in night samples. Penaeoid shrimps are often a numerically important component of the motile epifauna in Thalassia meadows (Greening and Livingston 1982; Gore et al. 1981; Heck 1976, 1977). Ina year-long monthly sampling program conducted in seagrass beds at Dorado, north coast of Puerto Rico, penaeoids frequently comprised 10- 15% of the total number of individuals in a monthly night sample (range: 1—41%). 178 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON The purpose of this report is to describe the species composition of the penaeoid fauna from these seagrass meadows, to compare estimates of abundance based on day and night sampling, and to describe seasonal variations in abundance of the numerically dominant species. Methods The seagrass meadows sampled were located in a cove just east of the Dorado Balneario (public beach) near Dorado (18°29'N, 66°15’W), on the north coast of Puerto Rico. These grassbeds are described by Bauer (in press); further details on study areas, methods used, and sampling information can be obtained from that report. A shallow rocky reef protects the beds from the normal 2-6’ (0.6—1.8 m) ocean swells characteristic of the north coast. The Dorado grassbeds have the form of a terrace raised approximately | m above the surrounding sand bottom. Two species of seagrasses, Thalassia testudinum and Syringodium filiforme, were the primary vegetation; 7. testudinum was dominant in blade density throughout the investigation. A variety of species of attached benthic algae occurred in the grassbeds but were never extremely abundant; drift algae (e.g., Greening and Livingston 1982; Gore et al. 1981) were not present in noticeable quantity during the study period. Two areas within the Dorado seagrass meadows were delimited as replicate sampling sites. The two sites were 40 m apart and almost separated by the sur- rounding sand bottom. The median water depth at Site 1 varied from 0.7—1.0 m (depending on tidal height) while Site 2 was somewhat shallower (0.4—0.7 m). Sediments under Site 1 were muddy sand; those at Site 2 were similar but scattered coral rubble also occurred there. These grassbeds were relatively level and free of holes so that uninterrupted runs of the pushnet could be taken. A 0.5-m wide pushnet with a 1.0-mm mesh liner sewn into the net bag was used to sample the mobile invertebrate epifauna. A sample unit was a 10 m run so that each sample covered an area of 5 m?. Collecting took place when the tidal level was lower than 0.2 m. A small part of Site 2 was exposed by the lowest tides; samples were not taken when this area was exposed. Night collections were usually made at new moon to first quarter or before moonrise at other lunar phases; the only quantitative field work reported here done under the light of full moon was in March 1982. Monthly day and night sampling was conducted from February 1982 to Feb- ruary 1983. Each month, 10 day and 10 night samples were taken at each of the two sites. A map of each site was divided into areas the size of a sample unit, 10 m X 0.5 m. These units were numbered and sample locations were chosen by using a random numbers table. In the field, one end of these randomly selected units was located by measurements from reference markers. The median speed at which the net was pushed varied from 0.7—0.9 m/sec. After a pushnet run, all material was removed from the net and placed in a plastic bag with 37—40% formaldehyde added to make an approximately 10% formalin solution. In the laboratory, animals were sorted out and placed in 70% ethanol for permanent storage. Day samples were those taken between sunrise and sunset; night collections were conducted between sunset and sunrise. The time of day or night field work VOLUME 98, NUMBER 1 72) varied with low tide periods during which such work was carried out. Because of the timing of tides, three sets of samples took place across the night—day or day— night transition; since I report variations in individual sample abundance with time for these particular samples, time of field work will be given in greater detail for them. May Site 1 day samples began at 1725 Atlantic Standard Time and ended at 1925; sunset was at 1833 and darkness (when flashlight became necessary to read and record data) fell at approximately 1900. The May Site 2 night col- lections were from 1830 to 2000; the time of sunset and darkness was the same as for May Site 1 day. The June Site 1 night samples were from 0440 to 0555; sunrise was at 0522. In June at Site 2 there were two day (just after sunrise and during the afternoon) and no night samples. Bad weather prevented field studies in February 1982 at Site 1 and caused termination of work in July at Site 1 after only 3 samples. April Site 1 day and Site 2 night collections had to be discarded because of poor preservation. Water temperature varied from 26-30.5°C and salinity from 34-36%o during the study period (measured monthly during field work). Observations on day-night activity were carried out on captive animals in recirculating aquaria with sand bottom in which Thalassia plants were imbedded. Shrimps were maintained under either a variable day-night light cycle which coincided with working hours or a 12 hr day:12 hr night cycle controlled by a timer. ““Daylight”’ was fluorescent light; night observations were made under constant red light, with flashlight with and without red filter, by flash photography, or by turning on day lights during a dark cycle. The classification of dendrobranchiate shrimps given by Pérez Farfante (1977, 1978) is followed in this report. Results Species composition.—Approximately 7500 dendrobranchiate shrimps were captured (Table 1). Almost all individuals were species in the superfamily Pen- aeoidea. Two members of the family Sicyontidae, Sicyonia parri (Burkenroad) and S. /aevigata Stimpson comprised 85.9% of the total collected (Table 1). Three Metapenaeopsis species (Penaeidae), M. smithi Schmitt, M. martinella Pérez Far- fante, and M. goodei (Smith) were much less abundant (Table 1). I grouped all juvenile (without the well developed petasma or thelycum needed for species identification) Metapenaeopsis together; these juveniles accounted for 9.3% of the total number of shrimps. The genus Penaeus was represented by only 31 late postlarvae. Only four specimens of the superfamily Sergestoidea, family Serges- tidae (Lucifer faxoni Borradaile) were taken in the samples. Diel variation in abundance.— All penaeoid species were much more abundant in night samples than in day collections. At both sites, the mean number of Sicyonia parri per square meter of sampling effort was significantly higher than the day mean (95% confidence limits of day and night means do not overlap) (Fig. 1). In 3 of 13 months, no individuals of this species were captured during the day at Site 1. Estimates of monthly abundance based on night samples ranged from 1-14/m? (monthly means). Sicyonia laevigata followed a similar pattern of con- sistently higher mean numbers of individuals in samples taken at night (Fig. 2). On six of 27 occasions, there were no S. /aevigata taken during daylight hours. 180 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Species composition and relative abundance of dendrobranchiate shrimps from Dorado, Puerto Rico, seagrass meadows (totals of 13 months, both sites, day + night). Rank Species Number collected (and % of total) 1. Sicyonia parri 4827 (64.7%) 2. S. laevigata 1582 (21.2%) 3. Metapenaeopsis juveniles 693 (9.3%) 4. M. smithi 210 (2.8%) 5. M. martinella 70 (0.9%) 6. M. goodei 42 (0.6%) 7. Penaeus sp. (late postlarvae) 31 (0.4%) 8. Lucifer faxoni 4 (0.1%) Estimates of mean density based on nocturnal samples ranged from 0.3-3.0/m7?. Relatively few adult Metapenaeopsis goodei, M. smithi, and M. martinella oc- curred in the samples (Table 1). To analyze day—night variation in Metapenaeopsis, adults of all three species were grouped (Fig. 3). Metapenaeopsis adults were taken in only 1 of 27 day sampling periods. Mean abundance in night samples varied from 0.08—0.80/m?. Metapenaeopsis juveniles were also primarily night collect- able (Fig. 4); the mean number/m? in monthly night collections ranged from O- 2.4. Three sets of samples were taken during the day to night or night to day transition; the changes in numbers of shrimps taken with increasing darkness or increasing light gives another view of diel variation in their collectability. Two day to night sample sets taken in May 1982, demonstrate the dramatic increase in numbers of Sicyonia parri and S. laevigata with increasing darkness (Fig. 5). The positive correlation between shrimps/sample and sample number (increasing sample number = increasing darkness) is statistically significant for both species (Table 2). Numbers of Metapenaeopsis adults + juveniles) per sample were pos- itively correlated with higher sample numbers at Site 2 but not at Site 1 (Table 2). In June at Site 1, night to day collections were made and a decrease in numbers of Sicyonia parri and S. laevigata with time (higher sample numbers = increasing light) was demonstrated (Fig. 6). The negative correlation between shrimp abun- dance and sample number was significant for S. parri but not for S. laevigata (Table 2). Metapenaopsis juveniles and adults occurred in low densities during the first five samples taken before light and only 1 individual was collected in the last five pushnet runs near or after sunrise. This decline in Metapenaeopsis num- bers was significant (Table 2). Seasonal Variation in Abundance (Night Samples) To look at possible differences in monthly abundances, a one-way ANOVA, using the log (x + 1) transformation, was done on monthly night mean densities for Sicyonia parri, S. laevigata, Metapenaeopsis adults, and M. juveniles for each site. Differences between individual monthly means were determined with the Student-Newman-Keuls test, using the P = 0.05 level of significance (Table 3). VOLUME 98, NUMBER 1 181 NO. of INDIVIDUALS /M* 0:5 03 ll O'F RH fin -m | re MO th F M A M J J A S O N D J B Fig. 1. Monthly day and night abundances of Sicyonia parri. Bars represent mean number of individuals captured per square meter of sampling effort; vertical lines are the 95% confidence limits on the means. A dash (—) signifies no sampling for that period; a zero (0) means that no individuals were taken. Black bars are night means; clear bars represent day means. For S. parri at Site 1, abundances in April, May, June, and July were significantly higher than in the remaining months; at site 2, a similar pattern was not obvious (Fig. 1, Table 3). The trends in abundance of S. /aevigata were similar to S. parri at Site 1, with May, June, and July means forming a group distinctly greater than the other months. Although there were significant differences at site 2 in S. /ae- vigata (P < 0.001), groups of similar means were highly overlapping, and a pattern of highs and lows similar to Site 1 is not apparent (Table 3). Mean abundances of Metapenaeopsis adults were highly overlapping at Site 1 (Fig. 3, Table 3) and monthly means were not significantly different at Site 2 (P > 0.10). However, seasonal differences were present in Metapenaeopsis juveniles (Fig. 4, Table 3). 182 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON N_ 40 = ~ (op) =i 3.0 : Site | = Night = Day OH = 20 oO Z | oA 1.0 | | : | a e) o" -2o- Ei -fis La ot 0 FHM A Mod "3 “a” S "0° N “Dee I982 1983 MONTH 3.0 Site 2 20 F M A M J A OO kal Da le Fig. 2. Monthly day and night abundances of Sicyonia laevigata. Symbols same as Fig. 1. At Site 1, May, August, and September means were significantly higher than all others; at Site 2, August, September, October were months of peak abundance. Laboratory observations on diel activity.—Both Sicyonia species were noctur- nally active in laboratory aquaria with sand bottom and Thalassia. During the daylight cycle, these shrimps remained buried just under the sandy surface. At night (complete darkness or red light) the sicyoniids emerged from the sand onto the surface of the aquarium bed; some individuals walked over the sand while many crawled up or clung to various parts of seagrass leaves. Sicyonia spp. could be made to burrow during a night cycle simply by turning on day (fluorescent) lights; emergence could be evoked during a day cycle within a short time by turning off lights. The few observations made on Metapenaopsis spp. indicated night emergence-day burrowing behavior similar to Sicyonia species. Metapen- aeopsis individuals were not seen climbing seagrass leaves as did the sicyoniids. VOLUME 98, NUMBER 1 183 S) wS SO 25 NO. of INDIVIDUALS / M2 Go) Site 2 00 - > Mo —o Ilo no Lo Ilo Bo fo fo Ec 0 Ih Rome Me PRM eed eA US a Oc ANG Spy Ph. F Fig. 3. Monthly day and night abundances of Metapenaeopsis adults (M. goodei, M. smithi and M. martinella grouped together). Symbols same as Fig. 1. Discussion The penaeoid shrimp fauna collected from seagrass beds at Dorado, north coast of Puerto Rico, was dominated numerically by Sicyonia parri and S. laevigata. Conspicuous by their scarcity were members of the genus Penaeus; of nearly 7500 penaeoid shrimps collected, only 31 Penaeus (late postlarvae) occurred in the samples. In other studies on seagrass mobile invertebrate epifauna, sicyoniids were quite rare or absent while either Penaeus duorarum (Florida) or P. notialis (Caribbean) was one of ten most abundant invertebrate species (Greening and Livingston 1982; Gore et al. 1981; Heck 1976, 1977). If appearance in local fisheries is some indication of population abundances, then Penaeus spp. may be rare not only on the north coast but in other areas of Puerto Rico as well. Although Suarez Caabro (1979) does include ““Penaeus spp.” 184 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 40 ON (e) 1.0 NO. of INDIVIDUALS /M* ine) O Site 2 05 = O O_O F UM WA OM edo a J [A (SO AN 4D eee Fig. 4. Monthly day and night abundances of Metapenaeopsis juveniles. Symbols same as Fig. 1. in the list of 130 species taken in local fisheries of Puerto Rico, no further infor- mation is given by him. Furthermore, Weiler and Suarez Caabro (1980) do not list penaeoids in their report on species composition and catch records of Puer- torican fisheries. An occasional local fishery for Penaeus spp. is known from southwest Puerto Rico (Laguna Joyuda, Boquerén) (Roger Zimmerman, pers. comm.). Three species, P. schmitti, P. subtilis, and P. notialis, contribute to this fishery (Allan Stoner, pers. comm.). I have collected only a very few specimens of Penaeus spp. in other seagrass beds on the north coast (Luquillo platform, Condado Lagoon). Schmitt (1935) collected a few specimens of Penaeus brasi- liensis on the north coast. In summary, shrimps of the genus Penaeus do occur in Puerto Rico but there is no indication of large populations. Brackish water mangrove areas and seagrass meadows are available as juvenile habitats on the north coast and the rest of the island. The factor or combination of factors which VOLUME 98, NUMBER 1 185 Site | S. parri Bs S. laevigata wm 40 cad B 20 Z Gem fo) fo) Zz l 5 0) I 5 10 SAMPLE NUMBER | ) fe) | ) @) Fig. 5. Increase in numbers of Sicyonia parri and S. laevigata taken with increasing darkness (higher sample numbers) in collections beginning before and ending after sunset (May 1982). prevent the occurrence of large populations of Penaeus spp. around Puerto Rico are not known; on the north coast, the narrowness of the insular shelf (a few kilometers) could be one possible factor. Both the Sicyonia and the Metapenaeopsis species from the Dorado seagrass beds were nocturnally active. The laboratory observations indicated that, in the presence of daylight, S. parri, S. laevigata, and Metapenaeopsis species burrow just under the bottom sediments. In darkness they emerge from daytime hiding places to walk over the bottom and, in the species of Sicyonia, to climb up the seagrass blades. Field evidence confirms the nocturnal activity of these species. The numbers of Sicyonia spp. taken in night collections were always much higher than in the day; the nighttime presence of these shrimps on the seagrass beds and especially their habit of climbing seagrass leaves makes them susceptible to push- net capture at night. Their emergence from day burial at dusk is confirmed by samples begun before and terminating after sunset; the increase in numbers of 186 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON S. parri S. laevigata @ 20 NO. of INDIVIDUALS | S) lO SAMPLE NUMBER Fig. 6. Decline in numbers of Sicyonia parri and S. laevigata taken with increasing light (higher samples numbers) in collections beginning before and ending after sunrise (June 1982, Site 1). sicyoniids per sample with increasing darkness was dramatic. Their return to beneath the surface at dawn was documented by the capture of fewer and fewer individuals with increasing light in collections that took place over the night—day transition. That the sicyoniids were actually burying themselves in the seagrass meadow sediments (instead of migrating elsewhere or avoiding the net during the day) is confirmed by some day observations in which I dug and screened bottom sediments from the seagrass beds; sicyoniids occurred beneath the surface in these sediments. Both laboratory and field evidence show that Metapenaeopsis spp. are active at night and burrow during the day. Several investigators have made observations on the nocturnal behavior of Table 2.—Correlations of number of individuals/sample and sample number (=increasing time) in collections beginning before and ending after sunset or sunrise. The Spearman rank correlation coef- ficient (r,) is calculated for the possible correlation for each species or group. The probability is given for the one-sided hypothesis: no positive correlation (day to night samples) or no negative correlation (night to day). The null hypothesis is rejected when P < 0.05. (S) = significant test; (NS) = nonsig- nificant. Species re Probability May Site 1 (day to night) Sicyonia parri +0.918 <0.001 (S) S. laevigata +0.797 0.005 > P > 0.001 (S) Metapenaeopsis spp. +0.312 >0.10 (NS) May Site 2 (day to night) Sicyonia parri +0.912 <0.001 (S) S. laevigata +0.788 0.005 > P > 0.001 (S) Metapenaeopsis spp. +0.670 OLOWS SIP S OLOI CS) June Site 1 (night to day) Sicyonia parri —0.770 0.01 > P > 0.005 (S) S. laevigata —0.360 >0.10 (NS) Metapenaeopsis spp. —0.684 0.025 > P> 0.01 VOLUME 98, NUMBER 1 187 Table 3.—Comparison of monthly mean abundances (night samples). Months are listed in order of increasing means. Vertical lines join months whose means are not significantly different (Student- Newman-Keuls test, -P > 0.05). There were no significant differences between means for Metapen- aeoposis adults at Site 1 (one-way ANOVA, P > 0.10). Months of zero abundance are not listed below. (S. = Sicyonia; M. = Metapenaeopsis). S. parri S. laevigata M. juveniles M. adults Site 1 Sept Dec Feb 83 = Oct Nov April Aug Jan June Dec Aug Dec Nov March July Jan Sept Jan Feb 83 | Feb 83 Nov March Oct Oct April April May June | June Sept July July Aug May May Site 2 Sept May Feb 83 May May Jan Dec Feb 83 Feb 83 Sept May Jan Oct Feb 82 Jan July Nov Dec Nov Feb 82 Dec Feb 83 July March Jan March Sept Dec Aug Nov Aug Aug March Aug Oct Nov Feb 82 July Sept July Oct Oct various penaeoid species. Cobb et al. (1973) reported that Sicyonia brevirostris were much more abundant at night and that gut content analyses also indicated nocturnal activity. Pérez Farfante (1971) noted that the few data available indi- cated that Metapenaeopsis goodei, M. smithi, and M. martinella were night active. Several Penaeus spp., e.g., P. aztecus and especially P. duorarum make shallow day burrows which they leave at night (Wickham and Minkler 1975; Pérez Farfante 1969; Fuss 1964; Williams 1958). Light intensity has been shown to be the most important factor mediating activity in these Penaeus species (Bishop and Herrn- kind 1976; Wickham and Minkler 1975; Fuss and Ogren 1966). In the laboratory, I could cause emergence of Sicyonia spp. during the day by simply turning off the lights for 20-30 minutes. When lights were turned on again, the shrimps began burrowing within a very few minutes. Seasonal variations in abundance of Sicyonia and Metapenaeopsis spp. were not as notable as those of the nine most numerous caridean species from the same meadows (Bauer, in press). The carideans had marked population highs in late spring and summer with a smaller peak in December and January; abundance peaks and troughs were very highly correlated statistically. Seasonal differences in abundance were found in Sicyonia parri and S. laevigata at Site 1, with sig- 188 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON nificantly higher densities in late spring and summer months; this pattern was not apparent at Site 2. When sorting the samples from the field, it was obvious that large numbers of Sicyonia juveniles arrived at the seagrass meadows in May and June, perhaps accounting for higher Sicyonia abundances in that period. These observations need to be verified by size-frequency analysis of monthly collections (now in progress). Metapenaopsis adults showed little seasonal variation in num- bers. However, recruitment of Metapenaeopsis juveniles was not continuous; sharp increases in numbers occurred from August and September or October at both sites. Penaeoid species were less abundant than caridean shrimps at the Dorado seagrass meadows. Sicyonia parri total abundance was a little higher than that of Hippolyte curacaoensis, the fifth ranked caridean; S. /aevigata was intermediate in total captured between the seventh and eight ranked carideans, Processa ber- mudensis and P. riveroi, respectively (Bauer, in press). Metapenaeopsis spp. were comparatively rare. I consider the Sicyonia night abundances reported here to be good estimates of population densities. Sicyonia spp. are heavily armored, robust, benthic species; I know of no reports of night swimming away from the bottom for Sicyonia spp. In addition, the small mesh (1 mm) used in the pushnet assured that the small adults and juveniles were taken. Sicyonia parri and S. laevigata are small sicyoniids (maximum sizes given in Williams 1984). Juveniles and small males with 2—4 mm carapace length were at times quite numerous; shrimps of this size can easily slip through the 6-7 mm mesh of trawls and scrapes generally used in seagrass sampling studies (e.g., Greening and Livingston 1982; Heck 1976, 1977, 1979). Bauer (in press) also found that densities of carideans (similar in size or smaller than sicyoniids) estimated by fine mesh pushnet samples were much higher than in studies using trawls or scrapes with larger mesh and equal to caridean densities taken by drop net (Gore et al. 1981). Metapenaeopsis abundances might be underestimated in pushnet samples, even those taken at night. Wheeler (1937, cited in Pérez Farfante 1971) reported that M. goodei was collected near the water surface at night. If this is normal behavior (i.e., not a case of shrimps being attracted by a bright artificial light) and members of these species do swim off the bottom at night, then estimates of abundance would be in error. In summary, penaeoid shrimps of the genera Sicyonia and Metapenaeopsis are very nocturnal and therefore susceptible to pushnet capture at night. Quantitative sampling for analysis of community structure or life history must take place at night to include these species and to estimate their relative abundances and den- sities. In addition, a small mesh net should be used in collecting to include the juveniles and small adults of these penaeoids. Acknowledgments I am grateful to the many University of Puerto Rico students who have helped in field sampling, sorting of samples, and preparation of illustrations. Financial support for students and for equipment was provided by University of Puerto Rico OCEGI grants. I thank Dr. Isabel Pérez Farfante for her help in species identifications and for providing me with valuable literature references. The ed- itorial suggestions of Drs. Isabel Pérez Farfante, Fenner A. Chace, Jr., and Roger Zimmerman were quite helpful. VOLUME 98, NUMBER 1 189 Literature Cited Bauer, R. T. Diel and seasonal variation in species composition and abundance of caridean shrimps (Crustacea, Decapoda) from seagrass meadows on the north coast of Puerto Rico.— Bulletin of Marine Science 35 (2). [In press]. Bishop, J. M., and W. F. Herrnkind. 1976. Burying and molting of pink shrimp, Penaeus durorarum (Crustacea: Penaeidae), under selected photoperiods of white light and UV-light.— Biological Bulletin 150:163-182. Cobb, S. P., C. R. Futch, and D. K. Camp. 1973. Memoirs of the Hourglass Cruises: the rock shrimp, Sicyonia brevirostris Stimpson, 1871 (Decapoda, Penaeidae).— Florida Department of Natural Resources Marine Research Laboratory Volume 3, Part 1, 38 pp. Fuss, C. M., Jr. 1964. Observations on burrowing behavior of the pink shrimp, Penaeus duorarum Burkenroad.— Bulletin of Marine Science, Gulf and Caribbean 14:62-73. , and L. H. Ogren. 1966. Factors affecting activity and burrowing habits of the pink shrimp, Penaeus duorarum Burkenroad.— Biological Bulletin 130:170-191. Gore, R. H., E. E. Gallaher, L. E. Cotto, and K. A. Wilson. 1981. Studies on Decapod Crustacea from the Indian River region of Florida. XI. Community composition, structure, biomass, and species-areal relationships of seagrass and drift algae-associated macrocrustaceans. — Estuarine, Coastal and Shelf Science 12:485-508. Greening, H. S., and R. J. Livingston. 1982. Diel variation in the structure of seagrass-associated macroinvertebrate communities.— Marine Ecology Progress Series 7:147-156. Heck, K. L., Jr. 1976. Community structure and the effects of pollution in seagrass meadows and adjacent habitats.— Marine Biology 35:345-375. 1977. Comparative species richness, composition, and abundance of invertebrates in Ca- ribbean seagrass (Thalassia testudinum) meadows (Panama).— Marine Biology 42:335-348. 1979. Some determinants of the composition and abundance of motile macroinvertebrate species in tropical and temperate turtlegrass (Thalassia testudinum) meadows.—Journal of Biogeography 6:183-—200. , and R. J. Orth. 1980a. Seagrass habitats: the roles of habitat complexity, competition and predation in structuring associated fish and motile macroinvertebrate assemblages. Jn V. S. Kennedy, ed., Estuarine Perspectives.— Academic Press, New York, pp. 449-464. , and 1980b. Structural components of eelgrass (Zostera marina) meadows in lower Chesapeake Bay. Decapod Crustacea. — Estuaries 3:289-295. Hooks, T. A., K. L. Heck, Jr., and R. J. Livingston. 1976. An inshore marine invertebrate community: structure and habitat association in the northeastern Gulf of Mexico.—Bulletin of Marine Science 26:99-109. Kikuchi, T. 1966. An ecological study on animal communities of the Zostera marina belt in Tomioka Bay, Amakusa, Kyushu.— Publications from the Amakusa Marine Biological Laboratory, Kyu- shu University 1:1—106. —. 1974. Japanese contributions on consumer ecology in eelgrass (Zostera marina L.) beds, with special reference to trophic relationships and resources in inshore fisheries. — Aquaculture 4:145-160. , and J. M. Pérés. 1977. Consumer ecology of seagrass beds. Jn C. P. McRoy and C. Helfferich, eds., Seagrass ecosystems. — Marcel Dekker, Inc., New York, pp. 147-193. Pérez Farfante, I. 1969. Western Atlantic shrimps of the genus Penaeus.— Fishery Bulletin 67:46 1- 591. 1971. Western Atlantic shrimps of the genus Metapenaeopsis (Crustacea, Decapoda, Pen- aeidae), with description of three new species. —Smithsonian Contributions to Zoology 79:1- 37. . 1977. American solenocerid shrimps of the genera Hymenopenaeus, Haliporoides, Pleoticus, Hadropenaeus new genus, and Mesopenaeus new genus.—Fishery Bulletin 75:261-346. 1978. Families Hippolytidae, Palaemonidae (Caridea), and Penaeidae, Sicyoniidae and So- lenoceridae (Penaeoidea). Jn W. Fischer, ed., FAO species identification sheets for fishery purposes, Western Central Atlantic (Fishing Area 31). Vol. VI. Food and Agriculture Organi- zation.— United Nations, Rome, Italy. [unpaginated]. Reid, G. K. 1954. An ecological study of the Gulf of Mexico fishes in the vicinity of Cedar Key, Florida.— Bulletin of Marine Science, Gulf and Caribbean 4:1—94. 190 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Schmitt, W. L. 1935. Crustacea Macrura and Anomura of Porto Rico and the Virgin Islands. — New York Academy of Sciences, Scientific Survey of Puerto Rico and the Virgin Islands 15:125-— 262. Suarez Caabro, J. A. 1979. El mar de Puerto Rico: una introduccion a las pesquerias de la Isla.— Editorial Universitaria, Universidad de Puerto Rico, Rio Piedras, 259 pp. Thorhaug, A., and M. A. Roessler. 1977. Seagrass community dynamics in a subtropical estuarine lagoon.— Aquaculture 12:253-277. Weiler, D., and J. A. Suarez Caabro. 1980. Overview of Puerto Rico’s small-scale fisheries statistics 1972-1978.—Commonwealth of Puerto Rico CODREMAR Publications 1:1—27. Wheeler, J. F.G. 1937. Further observations on lunar periodicity.— Journal of the Linnean Society of London 40:325-345. Wickham, D. A., and F. C. Minkler, III. 1975. Laboratory observations on daily patterns of burrowing and locomotor activity of pink shrimp, Penaeus duorarum, brown shrimp, Penaeus aztecus, and white shrimp, Penaeus setiferus.—Contributions in Marine Science 19:21-35. Williams, A. B. 1958. Substrates as a factor in shrimp distribution.— Limnology and Oceanography 3:283-290. 1984. Shrimps, lobsters, and crabs of the Atlantic coast of the eastern United States, Maine to Florida.—Smithsonian Institution Press, Washington, 550 pp. Department of Biology, University of Puerto Rico, Rio Piedras, Puerto Rico 00931. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 191-203 TWO NEW SPECIES OF TWO NEW GAMMARIDAN GENERA (CRUSTACEA: AMPHIPODA) FROM THE FLORIDA KEYS James Darwin Thomas and J. Laurens Barnard Abstract.—Two new species of the new genera Anamaera (A. hixoni) and Spa- thiopus (S. looensis), both belinging to the section Gammarida of Amphipoda, are described from the Florida Keys. Anamaera is a new genus close to Maera, Ceradocus, and Ceradomaera, but has an unique combination of minor character expressions. Males of Spathiopus have an unusual paddle-shaped antenna 2 but otherwise Spathiopus appears to be an apomorph of Elasmopus. The new taxa described herein were collected at Looe Key Reef, off Big Pine Key in the Florida Keys. Looe Key Reef is now a National Marine Sanctuary of the United States of America and the amphipods described herein were collected while compiling a faunal list of the Sanctuary. Section Gammarida The reader is referred to Barnard and Barnard (1983) for a discussion of clas- sification in this group and the advisability of omitting commitment to superfam- ilies and families until more studies on morphology, behavior, and genetics have been completed. These genera would probably be assignable to the hadzioid (=melitoid) family group depicted by Bousfield (1978, 1983) but we cannot find any characters to separate that group from others, except Anisogammaridae, in the greater Gammarida (section) group of Gammaridea. Anisogammaridae have accessory gills and are quite distinctive.-Crangonyctoids (with or without sternal gills and with or without paddle calceoli), melitoids-hadzioids (without gill 7), gammaroids (with or without gill 7) and other groups in Gammarida are as yet not well described and are poorly defined. Legends.—Capital letters denote main parts in the following list; lower case letters to left of capital letters or in body of figure indicate modifications as follows; lower case letters to right of capital letters indicate specimens described in captions: A, antenna; B, body; C, coxa; D, dactyl; F, accessory flagellum; G, gnathopad; H, head; I, inner plate or ramus; J, prebuccal; K, lacinia mobilis; L, labium; M, mandible; O, outer plate or ramus; P, pereopod; Q, incisor; R, uropod; S, max- illiped; T, telson; U, labrum; V, palp; W, pleon; X, maxilla; Z, gill; a, aberrant; f, flat; left, lateral; m, medial; n, dissected; 0, opposite; r, right; s, setae removed. Family Gammaridae (classical sense) Spathiopus, new genus Diagnosis.— Body lacking dorsal teeth. Rostrum obsolete. Antenna 1 longer than antenna 2, article 2 longer than 1, article 3 short, secondary flagellum 192 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 2-articulate, calceoli absent. Article 5 of antenna 2 shorter than article 4, flagellum in male flat from side view but articles expanded into broad paddle from dor- soventral views. Mandibular incisors weakly toothed, laciniae mobilae toothed, rakers present, molars triturative, molar surface with large accessory knobs or “choppers,” palp article 3 falcate. Inner lobes of lower lip fleshy. Medial margins of inner plates on maxillae almost naked, apex of inner plate on maxilla 1 with about 4 setae, palp 2-articulate. Maxillipeds ordinary but inner plates lacking stout thick spines, dactyl weakly unguiform, with large nail. Anterior coxae of medium length, none strongly reduced in size, coxa 4 weakly excavate posteriorly. Gnath- opod 1 of melitid form (see Barnard & Barnard 1982 for definition), thus wrist elongate, setose and poorly lobed, hand shorter, subrectangular, palm short and weakly oblique; gnathopod 2 of male enlarged, subchelate, wrist short and lobate, hand elongate, large, palm oblique. Pereopods short. Epimera and pleopods well defined and ordinary. Uropods 1-3 short, stout, spinose, outer ramus of uropod 3 with minute second article. Telson fully cleft, all major spines apical. Gills present on coxae 2-6, thin oostegites present on coxae 2-5. Type-species.— Spathiopus looensis, new species. Etymology.—Named for the spatulate condition of the flagellum of antenna 2. This condition is present in both sexes, but is best developed in adult males. Relationship.—This genus appears to be an apomorph of various species in Elasmopus Costa as it differs substantially only in the expansion of the articles on the flagellum of antenna 2. This unusual character is also accorded generic status in the Gammaropsis-Audulla pair of genera in the Corophioidea and prob- ably indicates a special function associated with the structure. Spathiopus looensis, new species Figs. 1-3 6699 Description: Holotype male “‘g” 4.31 mm.—Eyes ovate, small, brown in alcohol, with halo of free ommatidial elements around pigment core. Lateral cephalic lobes with small notch below eyes. Article 1 of peduncle of antenna 1 with line of 3 ventral spines. Flagellum shorter than peduncle, aesthetascs absent. Labrum broadly heart-shaped, epistome unproduced. Right and left rakers 3 and 3 (first left raker complex), each molar with spinose protrusion and major seta, palp article 3 with DE setae. Each outer lobe of lower lip with cone. Inner plate of maxilla 1 with 2 large and 3 small apical or outer setae, outer plate narrow, with 7 spines, palp broader than outer plate and with densely armed apex, palps symmetrical. Inner plate of maxilla 2 narrower than outer, with 1—2 apicomedial setae. Inner plate of maxilliped with falcate ventral coupling hook, outer plate with medial and apical spines, article 3 with serrate apical hook. Coxa 1 extended forward anteroventrally. Gnathopod 1 as illustrated, article 4 of gnathopod 2 with lobe underriding article 5, latter also with weak apical lobe pointing distad, palm or hand densely setose and smooth, merging evenly with posterior margin of hand, dactyl short, medial surface with longitudinal ridge confining dactyl override, ridge sinuous and unevenly serrate. Pereopods 3-4 of similar structure but pereopod 4 smaller, article 5 with pair of large apicoposterior spines, apicalmost pair of spines on article 6 slightly smaller than preceding spine pairs, dactyl with 3 setules at declivity. Article 2 of pereopods 5-7 pyriform, VOLUME 98, NUMBER 1 193 BE ae = ype SN le et ae ieee SSN \N Fig. 1. Spathiopus looensis. Unattributed figures = holotype male “‘g” 4.31 mm; i = male “1” 4.77 mm; k = female “k’”’ 4.47 mm. 194 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Spathiopus looensis. Unattributed figures = holotype male “g” 4.31 mm; k = female “k” 4.47 mm. VOLUME 98, NUMBER 1 195 Fig. 3. Spathiopus looensis. Unattributed figures = holotype male “‘g’”’ 4.31 mm; k = female ‘“‘k” 4.47 mm. tapering apically, weakly lobate, face with outer ridge (pereopod 5) or ridges (pereopods 6-7). Gills on coxae 2-6, that on coxa 3 longest, those on coxae 2—4 subequal, club-shaped, flat, that of coxa 5 similar but smaller, that of coxa 6 smallest and almost orbicular. Epimera 1-3 with small sharp posteroventral tooth, margin above with 1-2 setule notches, ventral spine formula of epimeron 1 = 1-1-1, on face of eipmeron 2 = 1-1-1, on epimeron 3 = 1-1-2-3-2-1-1. Urosomite | with ventral spine, basal face of uropod 1 with pair of spines, dorsal margin of peduncle with 7 short spines, 196 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON apex with a single large spine, similar large spine medially, peduncle of uropod 2 with 4 short dorsolateral spines, rami of uropods spinose like Elasmopus. Uro- pod 3 short, peduncle expanded and plate-like, inner ramus slightly smaller than outer, with 2 medial spines, outer ramus lateral spine formula = 3-3, apical spines =4, apical spines on inner ramus = 3; article 2 on outer ramus with 2 setae. Telson short, broad, lobes tapering but apically notched, inner wing sharp, each notch with stout (abnormally stunted) spine, lateral margins each with pair (or abnor- mally one seta) of setules, each lateral apical wing with setule. Female “k’’ 4.47 mm.—Like male but flagellum of antenna 2 only weakly expanded and only scarcely longer than article 4 of peduncle, article 5 of peduncle not shortened; gnathopod 2 much smaller than male and hand lacking medial ridge. Armament formula on epimeron 2 = 1-2-1-2, epimeron 3 = 1-2-3-3-2-1. Ventral facial spines on outer ramus of uropod 3 reduced to 1. Oostegites thin, setal formula of oostegites 1-4 = 2-2-3-3, anterior, 1-2-2-1 distal, and 1-2-1-2 posterior. Variations. — Adults relatively uniform, male “i” like holotype but telsonic spines not stunted and normally long as shown for female telson. Male “‘h” telson also normal, spine count on epimeron 2 = 1-3-1-0, epimeron 3 = 1-3-4-2-3-0. Telson normal (with lateral setae in pairs but many specimens with setae reduced to one on one side or the other). Etymology. — Named for the type-locality. Holotype. —USNM 195127, male “g’’ 4.31 mm, illustrated. Type-locality.— Florida Keys, Looe Key Reef, 9 Oct 1983, in backreef area, in algae-covered rubble in sand, 1 m, station LKR4H, coll. J. D. Thomas. Material.—The type-locality, male “h’’ 3.96 mm (observed), male “1” 4.77 mm (observed), female “‘k”’ ovigerous 4.47 mm (illustrated) and 8 other specimens. Same area, 7 m, 8 Oct 1983, station LKFR4C, coll. J. D. Thomas (4). Biscayne Bay, Ragged Keys, 1200 feet west of marker no. 5 in channel between two keys, 2m, 8 Apr 1982, coll. Iver M. Brook and J. D. Thomas (10); same area, Oct 1982, coll. Biosystems, Inc. Station # 16 (30). Distribution. — Florida, Biscayne Bay south to Looe Key Reef, 1-7 m, occurring primarily in backreef areas of coral reefs in rubble, and in other shallow protected areas adjacent to deeper water where algae covered rocky substrates are available. Anamaera, new genus Diagnosis.— Body lacking dorsal teeth. Rostrum weak. Antenna 1 longer than antenna 2, article 2 longer than 1, article 3 short, secondary flagellum 3—5-artic- ulate, calceoli absent. Article 5 of antenna 2 shorter than article 4, flagellum cylindrical. Mandibular incisors strongly toothed, laciniae mobiles toothed, rakers present, molars triturative, palp article 3 linear, about as long as article 2, with A, D and E setae sparse. Inner lobes of lower lip fleshy. Medial margins of inner plates on maxillae naked, at best bearing hair-like armaments. Apex of inner plate on maxilla | with about 5 setae, palp 2-articulate. Maxillipeds ordinary but inner plates lacking stout thick spines, dactyl weakly unguiform, with large nail. Anterior coxae of medium length, none strongly reduced in size, coxa 2 largest, coxae 1—4 weakly excavate posteriorly, coxa 1 with anteroventral tooth and pos- teroventral cusp. Gnathopod 1 of melitid form (see Barnard and Barnard 1983, VOLUME 98, NUMBER 1 197 Fig. 4. Anamaera hixoni. Unattributed figures = male holotype “‘a”’ 5.22 mm; c = female “c” 4.04 mm; d = female “d”’ 3.10 mm. 198 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON for definition), thus wrist elongate, setose and poorly lobed, hand shorter, sub- rectangular, palm short and oblique; gnathopod 2 of male asymmetrical, one side enlarged, subchelate, wrist short and lobate, hand elongate, large, palm oblique and continuous with hind margin of hand; other side small like gnathopod | and similar, wrist elongate, unlobed, hand somewhat broader and longer than on gnathopod 1, palm oblique, anterior margin of hand more setose, posterior setae more curved apically. Pereopods short. Epimera serrate behind. Pleopods well defined and ordinary. Uropods 1-3 long, slender, spinose, basal face of uropod | with spine, peduncle of uropod 3 scarcely elongate, rami equiramous though outer ramus with minute second article. Telson fully cleft, lobes pointed, cuspidate, with apical and dorsofacial spines. Gills pres- ent on coxae 2-6, thin oostegites present on coxae 2-5. Type-species. —Anamaera hixoni, new species. Etymology.— Named for the concept as “‘variant Maera,”’ thus “‘ana”’ reflecting Greek “again.” Relationship. — Differing from Maera Leach in the asymmetry of male gnatho- pod 2. Differing from Ceradocus Costa in the long article 3 of the mandibular palp, lack of tooth on article 1, the poorly setose medial margins of maxillae 1—2, and the absence of dorsal body serrations. Differing from Ceradomaera Ledoyer, 1973 in the lack of dorsal teeth on the pleon and the long article 3 of the mandibular palp. Anamaera hixoni, new species Figs. 4-7 Description: Holotype male “‘a.”»—Body generally similar to E/asmopus and Maera, sparsely setose dorsally, with long antenna | and short antenna 2. Head with anteroventral notch and sinuous tooth below, eye of medium size, with brownish purple core surrounded by clear apices of ommatidia. Article 1 of an- tenna | about as long as head, flagellum shorter than peduncle. Gland cone of antenna 2 very long (aberrantly short and regenerant on left side in holotype and other specimens but normally long on both sides), flagellum about as long as article 4 of peduncle. Epistome produced upward and forward slightly, upper lip from anterior view with truncate, almost smooth ventral margin. Mandibles bulky, with large inner vertical keel bearing small but strongly triturative molar ventrally, each molar with plumose seta, incisors toothed, right lacinia mobilis bifid and complexly mutltitoothed, left lacinia mobilis flattened and 4-toothed, rakers about 7, palp article 1 scarcely elongate, article 2 of medium length, article 3 linear, about as long as article 2, with 1 A seta, 4 D setae and 2 E setae. Mandibular lobes of lower lip weak, broad, obtuse. Inner plate of maxilla 1 small, longer than broad, with 2-3 long apical and 3-2 small apicolateral setae; outer plate narrow, with 11 spines (some hidden in illustrations), palp broad, 2-articulate, with 2 ranks of apical and subapical armaments, right and left sides symmetrical. Plates of maxilla 2 erect, long, slender, inner narrower, armed medially only with thin hair-like armaments, no orthodox setae medially. Inner plate of maxillipeds excavate, with long lateral and small medial cusp, no stout apical spines, with several medial VOLUME 98, NUMBER 1 199 ZGS 4, LEZ LA Fig. 5. Anamaera hixoni. All figures = male holotype “a” 5.22 mm. 200 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 6. Anamaera hixoni. Unattributed figures = male holotype “a” 5.22 mm; c = female “‘c” 4.04 mm. VOLUME 98, NUMBER 1 201 Fig. 7. Anamaera hixoni. Unattributed figures = male holotype “a” 5.22 mm; d = female “d” 3.10 mm. and apical setae, outer plate large, with blunt medial spines and many apical spines becoming thinner apicad, palp slender, elongate, dactyl stubby-unguiform, nail strong, with 1+ accessory nail(s). Coxa | with anterior excavation and anteroventral sharp cusp pointing forward, coxa 2 much larger, anterior margin straight, but posterior excavate, thus ventral half of coxa appearing swollen, coxa 3 similar but much smaller, coxa 4 somewhat more quadrate, coxae 5—7 scarcely but successively shorter, ordinarily lobate, coxae 1—4 with short ventral setae, coxae 2—3 with 4 and 3 posterior spines. Article 5 of gnathopod 1 with very dense medial comb of spines. Article 2 of larger male gnathopod bearing anteroapical pair of cusps representing terminus of inner and outer anterior keels, article 3 of gnathopod 2 slightly elongate, article 4 with sharp posterodistal cusp pointing distally; palm of larger male gnathopod with distal spinose margin bearing inner and outer rows of 5 and 4 spines, then adjacent acclivity with flat topped smooth proximal lobe bearing few largely facial setae, dactyl simple (with marginal setae), extending halfway along palmar-hand margin, proximal part of this margin with setal tufts. Palm of smaller gnathopod 202 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 2 short, oblique, with sparse short spines and defined by row of 4 long and short setal-spines in tandem cluster. Coupling spines of pereopods 3-7 small, slightly uneven, simple, dactyls with acclivity marked by whip, bearing 2 other setules, one marginal, one facial; article 2 of pereopods 5—7 weakly expanded, with one facial ridge, posteroventrally lobate subsharply, posterior margins with medium serrations, some spines of articles 4— 6 as long as or longer than article 5. Epimera 1-3 each with lateral ridge, formula of ventrofacial spines on epimera 1-3 = 1-3-5, formula of posterior serrations (counted as points) = 2-4-7. Pleonites dorsally smooth, pleonite 4 weakly humped dorsally. Basofacial spine of peduncle on uropod | attached to raised lateral plaque, each peduncular apex with large spine, medial margin with 8 smaller spines (otherwise as illustrated), medial margin of peduncle on uropod 2 with 4 spines in tandem but apically with row of 4 or 5 spines forming fan and attached in dorsoventral row (otherwise as illustrated), apices of outer rami on uropods 1-2 alike, with 4 large and one tiny spines, of inner rami with 5 large spines. Uropod 3 overextending uropods 1-2, peduncle complexly spinose, outer ramus slightly shorter than inner, both rami complexly spinose. Telson with middorsolateral pair of spines, each obliquely excavate, apex with pair of long spines, sharp apex with lateral setule, also pair of setules proximal and lateral to apex. Female “‘c”’ 4.04 mm.—Generally like male but both sides of gnathopod 2 small, like small version of male gnathopod 2 (not therefore reillustrated). Antennae, gnathopod | and epimera like male. Some spines on coxae 2-3, uropod 3 and telson either longer or shorter than shown for male but these differences proved not to be of sexual value, merely individual variation in both sexes. (Telson illustration therefore to be discounted.) Oostegites long and thin, on coxae 2—S, apical setal formula = 3-3-3-4, anterior setae = 2-1-1-0, posterior setae = 1-1-1- 0, posterior setules = 1-1-1-6. Variations. —Either right or left gnathopod 2 of male enlarged. Males and fe- males found to vary in spine lengths especially on telson and uropod 3. Young female “d’’ with 3 articles in accessory flagellum, 4 in flagellum of antenna 2. Spine count on coxae 2-3 = 4-3. Spine count on epimera 1,2,3 = 1-3- 4, posterior points on epimera 1,2,3 = 2-2-5. Apex of each telsonic lobe with 2 spines, outer long, inner short; each dorsal spine pair with outer longer than inner. Male “‘e”’ also with left gland cone stunted and probably regenerant as in ho- lotype. Gnathopod 2 apical spine area of palm with 4 and 4 spines; coxa 2 with 5 posterior spines, coxa 3 with 3. Male “‘f’ with apical spine area on male gnathopod 2 palm bearing 8 and 6 spines. Spine counts on epimera 1,2,3 = 1-3-5. Tllustrations.— Views of holotype gland cone on antenna 2 showing left views on left antenna 2 (gland cone stunted) and right antenna 2 (from medial view) intact and then a left dissected view to show dorsal tooth of article 2 seen only medially. Color.— Body white and wine-rose, rose color forming diffused blotches, bands and spots as follows: core of eye; base of rostrum between eyes; scattered dorsal diffusion on pereonites 1-7, on pereonites 1-2 occurring middorsally and ex- tending laterally as blotches, on pereonites 3—7 occurring as band at posterior margin; oblique slashed from belly of pleonites 1-3 showing through epimera; VOLUME 98, NUMBER 1 203 light dorsal transverse band each on pleonites 1-3; basal blotch each in peduncles of uropods 2 and 3 and each lobe of telson. Food.—Stomach contents include substantial volumes of sand grains and bro- ken pieces of filamentous algae and other brown, green and ochraceous minute particles of organic matter. Etymology. — Named in honor of Ray F. Hixon (10 June 1947 to 19 Mar 1984), a rare and extraordinary person who spent many hours in the waters of Biscayne Bay and the Florida Keys. Holotype.—USNM 195126, male, ‘“‘a’’ 5.22 mm (illustrated). Type-locality.—JDT LKR 4H, Florida Keys, Looe Key, 1 m, 9 Oct 1983, algae covered rubble on sand, coll. J. D. Thomas. Material.—Type-locality, female ““b’’ 3.72 mm, female “‘c’’ 4.04 mm, young female ‘“‘d’’ 3.10 mm. Biscayne Bay, Florida, Ragged Keys, in channel between two keys, 2 m, coll. Iver M. Brook and J. D. Thomas, male “e’’ 4.28 mm, male “f? 4.50 mm. Distribution. — Florida from Biscayne Bay south to the Lower Florida Keys, 1— 2 m. Acknowledgments We thank Dr. Pat McLaughlin, whose specimens were collected by Biosystems Research Inc., as part of the Biscayne Bay Restoration and Enhancement Program under the direction of the Dade County Department of Environmental Resources Management, and funded by the Florida Department of Environmental Regu- lation. The first author was also supported by grants from the National Science Foundation (DEB8 121128) anda contract from the Sanctuary Programs Division, Office of Ocean and Coastal Resource Management, NOAA. Both authors received support from the Scholarly Studies Program of the Smithsonian Institution, for which we thank Dr. David Challinor. We thank Carolyn Cox Lyons of New York City for inking our plates. Literature Cited Barnard, J. L., and C. M. Barnard. 1983. Freshwater Amphipoda of the world. Vol. I, Evolutionary patterns:i—vill, 1-358, i—xvil; vol. II, Handbook and bibliography:359-830.— Hayfield Asso- ciates, Mount Vernon, Virginia. Bousfield, E.L. 1978. A revised classification and phylogeny of amphipod crustaceans. — Transactions of the Royal Society of Canada, Series 4,16:343-390. 1983. An updated phyletic classification and palaeohistory of the Amphipoda. Jn F. R. Schram, ed., Crustacean issues 1, Crustacean phylogeny: XI + 372 pp.—A. A. Balkema, Neth- erlands, pp. 257-277. Ledoyer, M. 1973. Etude des amphipodes gammariens des biotopes de substrats sableux et sablo- vaseux de la région de Tuléar et de Nosy-Bé (Madagascar).—Tethys, Supplement 5:51-94, plates 1-30. (JDT) Newfound Harbor Marine Institute, Rt. 3, Box 170, Big Pine Key, Florida 33043. (JLB) Department of Invertebrate Zoology, NHB-163, Smithsonian In- stitution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 204-220 THREE NEW SPECIES OF THREAD SNAKES (SERPENTES: LEPTOTYPHLOPIDAE) FROM HISPANIOLA Richard Thomas, Roy W. McDiarmid, and Fred G. Thompson Abstract.—Three new species of thread snakes of the genus Leptotyphlops are described from Hispaniola as: L. calypso from the Samana Peninsula, and L. asbolepis from the Sierra Martin Garcia, Dominican Republic, and L. leptepileptus from the Massif de la Selle, Haiti. These three species together with L. bilineatus and L. pyrites form a distinct group of Leptotyphlops that is restricted to the West Indies. All five species are compared and a key to the seven species of the genus known from the West Indies is presented. The first collection of a leptotyphlopid from Hispaniola (Thomas 1965) resulted in increased field effort to secure additional specimens of these secretive snakes. Further collecting yielded species of this genus different from Leptotyphlops pyrites Thomas from three widely scattered points on the island (Fig. 1): The Samana Peninsula and the Sierra Martin Garcia in the Dominican Republic and the north slopes of the La Selle Massif in Haiti. The snakes from each of these localities not only are distinct from L. pyrites but also from one another. Field work during the past few years also has extended the known range of L. pyrites from the vicinity of the type-locality near Pedernales, Dominican Republic, west into Haiti along the south coast and north into the Valle de Neiba. Thomas (1965) considered Leptotyphlops bilineatus Schlegel of the Lesser An- tilles and L. pyrites to be the only known members of a distinct Antillean group. The defining feature of this “‘bi/ineatus group” was the presence of two subocular supralabial scales that prevent the ocular scale from extending to the labial margin. In all other members of the family, a single scale called the oculolabial (ocular of Klauber 1940) covers the eye and extends to the labial border. The two original species in the group also had a similarly striped color pattern. The three new species described herein are members of the bilineatus group that depart signif- icantly in certain features from L. bilineatus and L. pyrites. We continue to use the term ‘“‘bilineatus group” as a convenient means of designating those species of Leptotyphlops having the subocular supralabial scales. That all of the known members are restricted to the West Indies suggests that we may be dealing with a monophyletic radiation. However, it is also possible that the group is non-monophyletic and represents remnants of an old, formerly more widespread group within the genus, whose only relicts happen to be West Indian. Following this interpretation, the presence of subocular supralabial scales could well be a plesiomorphous character. An osteological study under way (Thomas) may clarify relationships within the genus and shed light on the nature of the bilineatus group. Methods and Terminology We use certain conventions of description and measurement that should be noted. VOLUME 98, NUMBER 1 205 @ asbolepis 4 calypso @ leptepileptus B pyrites Fig. 1. Map of Hispaniola showing localities for Leptotyphlops species. (1) With reference to most scales, length refers to the greatest anterior-posterior measurement, and width refers to the greatest transverse measurement, even if the transverse dimension is the greater. When this convention is not used, as with a diagnonally placed scale, the “length”’ is measured along the major axis and is so stated. Height is the greatest vertical measurement when the surface of the scale is largely lateral. All measurements were made with dial calipers unless otherwise noted. (2) Supranasal and infranasal scales are equivalent to upper and lower nasals as used by Klauber (1940). Prefrontal, frontal, interparietal, and interoccipital in some instances are designated PF, F, IP, and IO, respectively. (3) When the rostral is described as protuberant, it has a distinct central bulge, 1.€., In transverse section the edges of the rostral scale would be seen to lie flat against adjacent scales and the central part to arch outward (Fig. 2). The resulting dorsal outline of the head may be almost trilobed. An extreme of protuberance is seen in the ogival outline of the head of Rhinoleptus (Orejas-Miranda et al. 1970, figs. 2 and 3). The protuberance we describe does not result from a pre- shedding condition such as has been seen in some typhlopids (Richmond 1961). (4) A decurved snout describes a condition in which the ventral surface of the snout is straight (horizontal), or even slightly concave, and the bulge in the rostral drops slightly below this plane. (5) Scale row reductions are described in two ways. First, the number of mid- ventral scales anterior to the anal scale was recorded at the point at which scale rows fuse. Scale rows were counted left to right and ventral to dorsal with the midventral row being 0; thus one could have a reduction formula of 20 (2 + 3)/ 16 (2 + 3). Second, the distance in millimeters (Z) anterior to the vent was mea- sured at the last reduction step (13 rows to 12 rows at ventral 16). The point of reduction is then expressed as a percentage of snout-vent length (SVL) computed by [1 — (Z/SVL)] x 100. This is much more informative than standard scale row reduction formulae, which are of dubious comparative value when longitudinal counts differ. 206 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON (6) Head width was measured at the level of the parietals using an ocular micrometer mounted in a dissecting microscope. (7) Rostral width was measured at the widest point on the snout and length from the supranasal-infranasal suture to the midpoint on the posterior edge. (8) Midbody diameter (MBD) was measured to the nearest 0.1 mm transversely so that the relatively rigid axial muscle mass and rib cylinder rather than the softer visceral mass defined the body diameter. (9) Total length was measured along a ruler to the nearest millimeter (mm), and tail length to the nearest 0.1 mm from the posterior cloacal lip to the tip of the caudal spine. SVL was obtained by subtracting tail length from total length and rounding to the nearest mm. (10) A major dichotomy in eye size and morphology exists in this group of leptotyphlopids and is not the result of preservation techniques, age in preser- vative, or pre-molting opacity. This character is used in the diagnoses. In large- eyed species the eyes are equal to about 4 the distance from the anterior border of the naris to the posterior margin of the eye; they are close to the surface and are surrounded by a distinct, clear orbital space. The ocular scale bulges slightly outward over the orbit, and that area lacks scale organs. In the small-eyed species the eye is about 4 the naris-to-eye distance and visible only as a small black dot well beneath the scale surface. There is no evident orbital space nor brille-like differentiation of the ocular. Scale organs are randomly distributed over the surface of the ocular. (11) All scale organs that we discerned in these leptotyphlopids are small tu- bercles, some more flattened than others (flattening is probably an artifact of preservation). Orejas-Miranda et al. (1977) showed that differences in the density of scale organs on the heads of some leptotyphlopids may be of taxonomic value. We did not count scale organs because among the new species no great variation in scale organ density was observed. The species we describe have scale organs scattered over the head. They are concentrated on the snout and infralabials, become sparse on the posterior head region, and are largely absent behind the parietals. The scale organs appear to be more numerous in the large-eyed, heavily pigmented members of the group (bilineatus and pyrites) than in the new species. (12) All specimens were sexed; if hemipenes were not everted, sex was deter- mined by dissection of hemipenes or dissection of gonads. (13) Data for holotypes listed in [ ] in description sections for each species. (14) Statistics were done on an Apple computer using the program “‘Quickstat” by C. Richard Tracy. Specimen citations reference the following museums: UF— University of Flor- ida, Florida State Museum; USNM—National Museum of Natural History. Some additional specimens in the collection of Richard Thomas (RT) and in the Albert Schwartz Field Series (ASFS) will be deposited in other museums. Leptotyphlops calypso, new species igs, 2.5 Holotype.—USNM 236659, adult male, taken 6.5 km S Las Galeras, Provincia de Samana, Dominican Republic, on 22 Feb 1975, by Roy W. McDiarmid. Paratypes.—(all from Provincia de Samana, Dominican Republic). RT 8859, VOLUME 98, NUMBER 1 207 Fig. 2. Anterolateral view of the heads of A, Leptotyphlops calypso (USNM 236658, paratype) and B, L. leptepileptus (ASFS V49850, paratype) showing differences in snout and rostral shape. Line = 1 mm. USNM 236658, adult males, ca. 4 km S Las Galeras, 8 Aug 1981, R. Thomas. — RT 8883, juvenile male, ca. 5 km S Las Galeras, 12 Aug 1981, Sra. Matias. Diagnosis.—A relatively slender (SVL/MBD 73-87), unpigmented (pink in life), small-eyed Leptotyphlops of the bilineatus group having 4 supra- and 4 infralabials, third supralabial in subocular position; high number of middorsal scales (370-— 380); far posterior reduction from 14 to 12 scale rows (96-97% SVL) by fusion of rows 2 and 3; rostral moderate; snout broadly rounded, not protuberant; su- pranasal rhomboidal; ocular small, hexagonal; temporal-parietal suture length equal to '/ or less the parietal-occipital suture; anal and ventral tubercles in males; external anal spurs in at least some males; pelvic girdle including ilium, ischium, pubis and femur present. Distribution.—Known only from the area between 4 and 6.5 km S Las Galeras on the Samana Peninsula of the Dominican Republic (Fig. 1). Description. —(all specimens male, N = 4) (Figs. 2 and 3, Tables 1 and 2). SVL 124-190 [166] mm (X = 166.5 mm); tail 5.8-8.9 [7.7] mm (X = 7.8 mm); MBD 1.7-2.2 [1.9] mm (x = 2.0 mm); SVL/MBD 73-87 [87] (x = 82.8). Head parallel- sided, tapering anterior to slight temporal bulge (head width 1.88—2.14 mm; X< = 2.001 mm); snout somewhat truncate in dorsal aspect, broadly rounded in lateral aspect with nearly vertical anterior face, not protuberant or decurved. Eye small (equal to % to % distance from anterior border of naris to posterior border of eye), deeply embedded, no clear orbital space. Rostral moderate in width, parallel- 208 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. Dorsal and lateral views of the head of Leptotyphlops calypso (RT 8859, paratype). Line = 1 mm. sided ventrally, expanding slightly on tip of snout and tapering posteriorly to a narrow, truncate margin; virtually all of rostral visible in frontal view. Prefrontal large, hexagonal, slightly broader than long; frontal smaller, hexagonal, distinctly broader than long; interparietal larger (PF > IP > IO > F). Supranasal roughly rhomboidal, narrowest dorsomedially, broadly curved on posteroventral edge, not angled; ventralmost point a broad wedge between infranasal and first supra- labial. Infranasal large, mostly visible in lateral view, extending dorsally to a level just below eye; free edge extending from widepoint of rostral posteroventrally, suturing with supranasal and first supralabial; naris under edge of infranasal just anterodorsal to supranasal-first supralabial suture. Ocular small, hexagonal, about 1.3 times higher than long, with a short supranasal suture. Supraocular large, elongate, about twice as long (major axis) as wide, pentagonal, almost a parallel- ogram, extending ventrally to just above eye. Parietal and occipital large and blocklike, occipital larger than parietal; both about 1.6 times wider than long and spanning two paramedian scale rows; occipital slightly emarginate on distal free edge. Temporal inserting between parietal and occipital for distance equal to 4 or less the length of parietal-occipital suture. Four supralabials, first suturing dorsally with supranasal, second picketlike with dorsal apex inserted between supranasal and ocular, third abutting dorsally on ocular, fourth large and sub- triangular, most of its area posterior to ocular, in contact with posteroventral edge of ocular, parietal, temporal, and first scale of dorsal row 3. Mental scale with median ventral notch, each winglike lobe extending posterolaterally inside labial margin along posterior median edge of first infralabial; postmental cycloid. In- fralabials 4, the fourth large, oval, platelike. Middorsal scales 375-380 [379] (x = 377.5); subcaudal scales 19-20 [20] (< = 19.5). Scale rows 14, reducing to 12 at VOLUME 98, NUMBER 1 209 96-97% [96] SVL by fusion of scale rows 2 and 3 (aberrantly 3 and 4 on right side of RT 8883) [20 (2 + 3) and 16 (2 + 3) midventral scales anterior to vent]; caudal scale rows 12. Anal scale roughly pentagonal with posterior median apex. Prominent spur visible externally on each side of vent beneath scale in two specimens (the holotype and USNM 236658). Tubercular scale organs present (except in RT 8883, a juvenile) around cloaca and anteriorly on 3 ventralmost scale rows for distance of up to nearly '4 SVL. Pigmentation lacking (pink in life). Variation.—As all specimens are males, no sexual dimorphism is evident. The tuberculation of the scales around the vent and along the ventral surface is most extensive in RT 8859 and USNM 236658, somewhat less so in the holotype, and absent in the juvenile (RT 8883). The anal tubercles and claws likely are secondary sexual characteristics. The juvenile and one adult (RT 8859) lacked externally visible spurs, but the adult, which was cleared and stained, has internal spurs. During removal of the skin an opening to the exterior was evident. An examination of radiographs of the juvenile (RT 8883) reveals a much less developed pelvic girdle rudiment (only ilial and ischial elements ossified) and no obvious internal spurs. Some abnormal fusions of head scales were noted. In the holotype the fourth supralabial on the left is partly fused with the temporal. In RT 8883, supralabials 1 and 2 on the right are fused, as are supralabials 3, 4, and the temporal; the left side of the head is damaged, and the supralabial condition cannot be ascertained. Remarks.— All specimens were collected along the road south of Las Galeras that parallels a prominent limestone ridge. The area is a mixture of open pasture and mixed mesic cultivation (bananas, yams, coffee, corn, coconuts, papayas) interspersed with some scrubby to semi-wooded habitat. The holotype and two other specimens (RT 8859, USNM 236658) were taken from beneath very large limestone rocks in an open pasture; a third specimen was beneath the same rock with one of the paratypes but escaped. The juvenile was found crawling on the floor of an outdoor kitchen. Etymology.— Calypso is a proper noun that derives from the Greek verb “to hide” (kalypto, “‘I hide’’); this new species is certainly well hidden in nature, as those of us who have looked for it can attest. Furthermore, Calypso, the nymph who sequestered Odysseus on Ogygia, was in island creature; and calypso, as a music form, has West Indian associations, even if not in the Hispaniolan tradi- tions. Leptotyphlops asbolepis, new species Fig. 4 Holotype.—UF 54802, adult female, taken on the west slope of Loma del Aguacate, 350 m, Sierra Martin Garcia, Provincia de Barahona, Dominican Re- public, on 29 Jan 1976, by Fred G. Thompson. Paratype.—USNM 236660, adult male, same data as holotype. Diagnosis.—A relatively stout (SVL/MBD 56-60), small-eyed, uniformly pig- mented Leptotyphlops of the bilineatus group having 4 supra- and 4 infralabials, third supralabial in subocular position; relatively low middorsal scale number (302-342); far posterior reduction from 14 to 12 scale rows (98-99% SVL) by fusion of rows 2 and 3; rostral moderate in size; snout slightly decurved and protuberant; rhomboidal supranasal; small to large, hexagonal ocular; temporal- 210 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON parietal suture length equal to or slightly less than length of parietal-occipital suture; no anal or ventral tubercles; no anal spurs; no pelvic girdle. Distribution.— Known only from the type locality (Fig. 1). Description. —(Fig. 4, Tables 1 and 2). SVL 135-[156] mm; tail 6.6-[6.8] mm; MBD 2.3-[2.5] mm; SVL/MBD 56-[60]. Head slightly broader than neck (1.97- 2.01 mm; x = 1.99 mm), tapering towards snout from slightly swollen temporal region; snout somewhat protuberant, rounded in lateral aspect, slightly decurved. Eye small (equal to ca. 4 distance from anterior edge of naris to posterior margin of eye), deeply embedded, no clear orbital space. Rostral moderately wide, parallel- sided ventrally, expanding on tip of snout, where slightly protuberant, and tapering to truncate posterodorsal margin; ventral portion nearly horizontal. Prefrontal, frontal, and interparietal subhexagonal, wider than long, increasing in size in that order; interoccipital smaller than interparietal, cycloid, isomorphic with succeed- ing middorsal scales. Supranasal roughly rhomboidal, narrowest dorsomedially, broadly curved on posteroventral edge, not angled, ventralmost point a broad wedge between infranasal and first supralabial. Infranasal large, mostly visible in lateral aspect, extending dorsally to point just above level of eye; posterior edge extending from wide point of rostral posteroventrally, contacting supranasal and first supralabial; naris under edge of infranasal about midway along infranasal- supranasal suture. Ocular small to large, about 1.2 to 1.6 times higher than long, hexagonal, with long supranasal suture. Supraocular small, a short pentagon, about 1.5 times longer (major axis) than wide, extending ventrally to point well above eye. Parietal and occipital large, about twice as wide as long, each spanning 2 dorsal scale rows; parietal somewhat emarginate on distal free edge; occipital markedly to moderately emarginate. Temporal inserted between parietal and oc- cipital a distance greater than '4 parietal-occipital suture. Four supralabials, second picketlike with dorsal apex inserted between supranasal and ocular, third abutting dorsally on ocular, fourth large and subtriangular, most of its area posterior to ocular, in contact with posteroventral edge of ocular, parietal, temporal, and first scale of row 3. Mental with median ventral notch, each winglike lobe extending posterolaterally inside labial margin along posterior median edge of first infra- labial; postmental cycloid. Infralabials 4, fourth large, oval and platelike. Mid- dorsal scales 302-[342]; subcaudals [18]—19. Scale rows 14, reducing to 12 at 8 and 6 and [7 and 4] midventral scales anterior to vent at 98-[99]% SVL by fusion of rows 2 and 3; caudal scale rows [12]. Anal scale roughly pentagonal with posterior median apex. No anal spurs. Tubercular scale organs scattered over head, largely absent posterior to occipitals, concentrated on rostral and infrala- bials; no scale organs evident around vent. Pigmentation relatively uniform and dense over body but with some unpigmented patches; coloration faintly lineate due to slightly denser melanophores at centers of scales; head unpigmented, pig- mentation beginning about level of occipitals. Variation.—The difference of 40 middorsal scales between the type and the paratype is probably the result of sexual dimorphism, as is also the relative difference in tail length (see following description for evidence on sexual dimor- phism in these snakes). The other main difference between the type and the paratype is in the size of the ocular scale. Because sexual dimorphism in this character is unknown in other species of Leptotyphlops, most likely this represents extremes of variation. VOLUME 98, NUMBER 1 211 Fig. 4. Dorsal and lateral views of the head of Leptotyphlops asbolepis (UF 54802, holotype). Line = 1 mm. Remarks.— The two specimens of L. asbolepis were collected under limestone boulders in a mesic forest zone reached by trail up the mountainside from La Salina (Puerto Alejandro) on the east side of the Bahia de Neiba. Etymology.— From the Greek, asbolos, “‘soot,”’ and J/epis, ‘“‘scale,” in reference to the darker, more uniform coloration of this species. Leptotyphlops leptepileptus, new species Figss2, 5 Holotype.—USNM 236661, adult female, taken at Soliette, 5 km airline NW Fond Verettes, 366 m, Département de |’Ouest, Haiti, one of series collected on 19 Jul 1978, by native collectors and Richard Thomas. Paratypes.—(all same locality as holotype) RT 5596, juvenile male, 5614, fe- male, 19 Jul 1978, native collectors and Richard Thomas.—RT 5682-5685, 5696- 5715, USNM 236662-71, 19 males, 15 females, native collectors, 23 Jul 1978.— ASFS V49834-70, 13 males, 23 females, 1 undetermined, native collectors, 13 Jul 1979. Diagnosis.— Relatively slender (SVL/MBD 72-94), small-eyed, silvery tan or piebald Leptotyphlops of the bilineatus group having 3 supra- and 3 infralabials, second supralabial in subocular position; high number of middorsal scales (377- 212 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 5. Dorsal and lateral views of the head of Leptotyphlops leptepileptus (USNM 236661, ho- lotype). Line = 1 mm. 414): far posterior reduction from 14 to 12 scale rows (84-95% SVL) by fusion of rows O and 1; large, protuberant rostral; slightly decurved snout; triangular supranasal; small, pentagonal ocular; temporal-parietal suture length equal to 4 or less length of parietal-occipital suture; no anal or ventral tubercles; no anal spurs; pelvic vestiges usually absent. Distribution.—Known only from the type-locality, an intermontane valley on the north slopes of the Massif de la Selle of Haiti (Fig. 1). Description. —(Figs. 2 and 5, Tables 1 and 2). SVL: males (juveniles of 102, 105 mm excluded) 144-174 mm, x = 156.03 mm, SE = 2.548 mm (N = 29); females 140-198 [184] mm, * = 172.88 mm, SE = 4.275 mm (N = 41); tail length: males 5.0—7.4 mm, X = 6.80 mm, SE = 0.013 mm (N = 32); females 5.8-7.4 [7.1] mm, Xx = 6.87 mm, SE = 0.008 mm (N = 40); MBD 1.4—2.3 [2.1] mm, x = 1.99 mm, SE = 0.176 mm (N = 74); SVL/MBD 72-94 [87.6]. Head narrow (1.45- 1.68 mm, x= 1.60 mm, SE = 2.121 mm, N =72), parallel-sided with slight temporal bulge, tapering anterior of eyes to somewhat protuberant, decurved snout. Eye small ({ to '% distance from naris to eye), deeply embedded, no clear orbital space. Rostral large, covering about '2 snout anterior to eye in dorsal aspect, tapering posteriorly to truncate margin, protuberant on snout tip; parallel- sided ventrally, widening gradually onto front of snout; ventral portion horizontal, somewhat concave, not completely visible in frontal view. Prefrontal very large, subhexagonal to nearly cycloid; frontal smaller (shorter), hexagonal; interparietal VOLUME 98, NUMBER 1 213 and interoccipital larger than frontal, smaller than prefrontal, cycloid (in general PF = = IP > IO 2 = PF). Supranasal large, subtriangular, posterior edge forming nearly right angle with nearly horizontal ventral edge; ventralmost point at apex of broad angle between nasal and first supralabial. Infranasal small, surface largely ventral, mostly not visible in lateral aspect; dorsal tip extending to just below level of eye; posterior edge extending from widepoint of rostral posteroventrally, contacting supranasal-first labial suture. Supraocular an elongate, irregular pen- tagon (almost a parallelogram), about twice as wide as long (major axis), ventral end inserted between supranasal and ocular and extending to point above eye for distance equal to about '4 eye diameter. Parietal and occipital large, less than twice as wide as long, each spanning two paramedian rows of dorsal scales; oc- cipitals slightly smaller than parietals, emarginate on distal half of free edge. Temporal inserting between parietal and occipital a distance 4 to 4 length of parietal-occipital suture. Three supralabials, surface of first nearly ventral (trans- verse); dorsal edges of second and third partially abutting ventral edge of ocular, both occluding ocular from labial border; second in short contact with supranasal and third with parietal, temporal, and first scale of row 3. Infralabials 3, third large, oval, platelike. Middorsal scales 377-414 (males 377-395, x = 385.22, SE = 0.866, N = 32; females 393-414 [411], x = 404.95, SE = 0.773, N = 41). Sub- caudals 17-22 (males 18-22, mode 20; females 17-21 [19], mode 20). Reduction from 14 to 12 scale rows occurring at 84-97% [92] SVL by fusion of scale rows 0 and 1 [31 (0 + 1)/31 (0 + 1) midventral scales anterior to vent]; scale rows of tail 12. Anal scale roughly pentagonal with median posterior:apex. No anal spurs; no pelvic girdle, pelvic vestiges occasionally present. Head unpigmented; brown (silvery in life) body pigmentation beginning on neck and becoming uniform over all of body except anal scale; variant pigmentation (20%) with irregular unpig- mented and more darkly pigmented blotches randomly distributed over body. Hemipenes simple, everted organs expanded basally, tapering towards tip, no ornamentation and no complex structures; size minute, about 1 mm long in largest specimens. Sulcus spermaticus entering organ on medial surface, proceeding distad about ¥4 length, then spiralling counterclockwise , turn (apical aspect) and con- tinuing to tip of organ. Variation.— Pronounced sexual dimorphism exists in middorsal counts, SVL, tail length as a percentage of body length, and reduction level (% SVL). Differences between means of these characters were all significant at P < 0.000001 when tested with the t-test. Subcaudal counts have the same mode in both sexes, but the range of counts for males was higher than that for females. The piebald color morph occurs in both sexes. Variants from the standard configuration of head scales include a small, supernumerary scale separating the third supralabial from contact with the parietal (USNM 236662 bilateral; RT 5705, right side), the wedge- insertion of the second supralabial between the ocular and supranasal (ASFS V49854, left side), and 4 supralabials (bilateral) in USNM 236662, although in this specimen the second supralabial does not insert between the supranasal and the ocular, as it does in the species of this group for which four supralabials is the normal condition. Only one of 35 x-rayed and two cleared and stained spec- imens had a trace of a pelvic rudiment. In that male (RT 5713) a pair of small, opaque elements (ischial remnants?) lying lateroventrally below the second ver- tebrae anterior to the cloaca is obvious in the radiograph. None of the hemipenes 214 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON appears completely everted, although many are nearly so; they are minute, none measuring more than | mm in length. Remarks.—The type-locality is a settlement along the valley of the Riviére Soliette, tree-lined and shady compared to the more open cultivation-scrub mosaic of the surrounding limestone hills. Some specimens were collected from piles of stream-worn cobbles in a shady (mango) rest area along the road. Unfortunately, we have no way of knowing how far away and into what different habitats, if any, the Haitian collectors ranged to collect the balance of the specimens. Etymology. —Leptepileptus is from the Greek meaning extremely thin, literally ““thin-upon-thin.” Comparisons and Discussion.—The three species we describe obviously are more closely related to one another than to the other two species within the bilineatus group. They are small-eyed, small-headed, relatively long-snouted, slen- der, lightly (or not at all) pigmented snakes with high numbers of middorsal scales and scale row reduction occurring on the body. In contrast, Leptotyphlops bili- neatus and L. pyrites are shorter, stouter, larger-headed, larger-eyed, shorter- snouted, boldly patterned snakes with lower middorsal scale counts and no scale row reduction on the body. Among our trio of new species, the differences are nevertheless pronounced. Leptotyphlops leptepileptus has a strikingly narrower head (Fig. 6) and broader rostral scale (Fig. 7); it also is unique among the three in having three supra- and infralabial scales and a scale row reduction by fusion of rows O and 1. The large, triangular, last supralabial (3 in /eptepileptus, 4 in calypso and asbolepis) partly extends beneath the ocular in L. /eptepileptus, where- as its area is largely posterior to the ocular in L. calypso and L. asbolepis. In L. leptepileptus the rostral is larger and more protuberant, and the snout is more decurved and depressed (more transversely oval in cross section); as a reflection of this, the infranasal and first supralabial are more nearly transverse in position than they are in the other two species. The ocular distance is notably shorter in L. leptepileptus (Fig. 8). The lateral head scale differences between L. leptepileptus and the other species are largely attributable to the lack of equivalent supralabials. The second supralabial in the 4-labial species inserts wedge-like between the supranasal and the ocular and accounts for the more rhomboidal shape of the supranasal by putting an extra facet on the posteroventral margin of the supranasal. Likewise the ocular becomes more hexagonal by the second supralabial insertion. The large, triangular, last supralabial is largely excluded from the subocular space in the 4-labial species but occupies part of the sub-ocular space in L. /eptepileptus. Therefore, the differences are not easily viewed as the result of simple fusion of one supralabial with another to get from the 4- to the 3-supralabial condition (or simple splitting, if the reverse was the sequence). Other aspects of shape differences among the species are obvious by comparison of Figs. 3, 4, 5, and 9 and Table 1. The absence of pelvic vestiges in L. asbolepis and most L. leptepileptus is a feature undocumented in other species of the genus (List 1966), although Tihen (1945) reported a personal communication from Leonard Laufe that some (un- specified) species lack them. Examination of three specimens of L. bilineatus (USNM 119168 and USNM 222954, radiographs; USNM 236657, cleared and stained) and one specimen of L. pyrites (RT 7600, cleared and stained) failed to reveal pelvic vestiges in these species as well. Based on our preliminary findings, it appears that the bilineatus group has species which clearly document an evo- VOLUME 98, NUMBER 1 ANS) 2.5 ® © asbolepis ® bilineatus A calypso O leptepileptus 0 pyrites HEAD WIDTH (mm) 60 80 100 120 140 160 180 200 SNOUT-VENT LENGTH (mm) Fig. 6. Scatter diagram of head width versus snout—vent length for bilineatus group Leptotyphlops. Numbers indicate symbols representing more than one specimen. lutionary transition including forms with well developed pelvic girdles and ex- ternal anal spurs (L. calypso), species in which a pelvic vestige is only rarely present (L. /eptepileptus), and species which seemingly lack pelvic vestiges com- pletely (L. asbolepis, L. bilineatus, L. pyrites). A comparison of the single juvenile of L. calypso to adults of that species indicates a sequential pattern of ossification of the pelvic girdle with posterior elements (ilium and ischium) appearing before the anterior and lateral components. Thus, one can envision a reduction and ultimate loss of pelvic girdle components in West Indian species of Leptotyphlops through modification of the developmental process. This interpretation is strengthened by the detection of an ossified ischial vestige in only one specimen of L. leptepileptus whereas all others (36) examined have lost the girdle completely. These findings suggest the value of a detailed examination of the sexual and ontogenetic changes in pelvic girdle components during development and growth of Leptotyphlops. Leptotyphlops calypso has a more rounded and swollen snout with an almost flat, ramlike anterior surface but narrow rostral. The prominent perianal and ventral tubercles in L. calypso, if consistent (presumably in adult males only), are 216 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON © asbolepis ® bilineatus A calypso O leptepileptus O pyrites oO ROSTRAL WIDTH (mm) ° N a 0.5 0.25 0.5 0.75 ROSTRAL LENGTH (mm) Fig. 7. Scatter diagram of rostral scale width versus rostral scale length for the bilineatus group Leptotyphlops. Numbers indicate symbols representing more than one specimen. probably unique (the small sample of L. asbolepis precludes our being sure). Since we have but one male L. asbolepis, we also cannot be sure that anal spurs do not exist in that species. However, the absence of a pelvic vestige in L. asbolepis suggests that anal spurs may be absent. The claws reported for L. humilis and L. nigricans by List (1955) seem much less well developed than those of L. calypso. List noted the possibility that claws emerge to the surface during the breeding season only. Leptotyphlops asbolepis, although amply distinct in combination of characters (Tables 1 and 2), lacks strikingly unique features; it is the most darkly pigmented of the three and is intermediate in rostral size and snout shape between L. calypso and L. leptepileptus. The extent to which the temporal inserts between the parietal and occipital is very distinctive, and the supraocular is also small compared to that of the other species. One result of our collecting has been the acquisition of more material of Lep- totyphlops pyrites. Originally known only from the xeric to semixeric lowlands of the western Barahona Peninsula of the Dominican Republic, we now have taken it in the southeastern coastal plain of Haiti. The habitat at the localities east of Belle-Anse is xeric limestone scrub and remnant woods, similar to some of the area near the type-locality. At Mare Geoffrey, 19 km W Thiote, the habitat is more mesic. At this locality the road crosses a dry (no doubt intermittent) river VOLUME 98, NUMBER 1 217 Table 1.—Comparison of the head shape and scalation among the five species of the bilineatus group of Leptotyphlops. calypso asbolepis leptepileptus pyrites bilineatus Rostral narrow, non- intermediate, broad, protu- narrow, non- narrow, non- protuber- slightly berant protuber- protuberant ant protuber- ant ant Snout swollen, intermediate depressed, short, blunt short, blunt blunt decurved Supraocular moderate small large moderate moderate Ocular hexagonal, hexagonal, pentagonal, hexagonal, hexagonal, short ante- short ante- long ante- short ante- short ante- rior suture rior suture rior suture rior suture rior suture Infranasal large, high; large, high; small, low; large, high; large, high; surface lat- surface lat- surface surface lat- surface later- eral eral largely eral al ventral First labial surface large- surface large- surface large- surface large- _ surface largely ly lateral ly lateral ly ventral ly lateral lateral Last labial largely poste- largely poste- _— partly be- largely poste- _ largely poste- rior to rior to neath ocu- rior to rior to ocu- ocular ocular lar ocular lar Parietal-oc- long short long long long cipital su- ture bed with steep banks of river cobble substratum and sparse, low, scrubby growth with some trees. We found seven L. pyrites together in loose soil and gravel around the roots of a small leguminous tree; two others were found in somewhat more exposed situations, one under a rock and one in a piece of abandoned termite Table 2.—Comparison of major diagnostic characteristics among the five species of the bilineatus group of Leptotyphlops. calypso asbolepis leptepileptus pyrites' bilineatus SVL-maximum 190 (167) 156 (146) 198 (164) 138 (115) 108 (90) (mean) SVL/MBD 73-87 56-60 72-94 43-64 35-41 Middorsals 370-380 302-342 377-414 262-287 170-189 Reduction level 96-97 98-99 84-95 —? —? (% SVL) Rows fused in Dar 8 2+3 0+ 1 —? —? reduction Labials 4 4 3 4 4 Eye size small small small large large Color unpigmented uniform except uniform or piebald dark with dark with head except head stripes stripes Anal spurs + — - = — Anal tubercles =r - — = = 1 Data in part from Thomas (1965). 2 Reduction occurs posterior to vent. 218 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON © asbolepis ® bilineatus A calypso O leptepileptus O pyrites 0.25 OCULAR-SUPRANASAL SUTURE LENGTH (mm) 0.25 0.5 0.75 OCULAR SCALE LENGTH (mm) Fig. 8. Scatter diagram of length of the ocular-supranasal suture versus maximum length of ocular scale for the bilineatus group Leptotyphlops. Numbers indicate symbols representing more than one specimen. Fig. 9. Dorsal and lateral head views of Leptotyphlops bilineatus (A and B, USNM 236657) and L. pyrites (C and D, RT 7607). Line = 1 mm. VOLUME 98, NUMBER 1 219 nest under a rock. Three specimens from 6 km NW Duvergé in the Valle de Neiba (Dominican Republic) extends the known range about 50 km to the northeast of the type-locality and across the Sierra de Baoruco. At this locality the habitat was extremely xeric; the snakes were found under palm trunk cuttings. In meristic characters none of the new L. pyrites material departs significantly from those of the hypodigm, although one Duverge specimen (RT 4423) is at the upper extreme in middorsal scales (287). The Haitian specimens are darker than the Dominican specimens, the bold dorsal-zone striping being much obscured. In contrast, the three Valle de Neiba snakes lack the median and paramedian dorsal stripes, having only a pale median dorsal band. These individuals also appear to differ from topotypical L. pyrites in the shape and proportional rela- tionships of certain head scales. Without more specimens the significance of this variation is difficult to assess. Specimens examined. —Leptotyphlops bilineatus, Martinique: USNM 119168, Martinique: USNM 236657, Plage du Diamant.—St. Lucia: Anse-La-Raye: USNM 222954, 0.1 mi E of Anse Galet River. Leptotyphlops pyrites, Haiti: Département de Ouest: RT 7201, 9.6 km E Belle-Anse; RT 7222, 11.2 km E Belle-Anse; RT 7600-7608, RT 7692, 19.5 km W Thiote, 600’.— Dominican Republic, Provincia de Independencia: RT 4423 9125, 9126, 6 km W Duvergé. Key to West Indian Speices of Leptotyphlops 1. Ocular scale excluded from labial border by supralabials' ............. 7) — Ocular (oculolabial) extends to labial border ........................ 6 2. Middorsal scales fewer than 290; eye large, obvious, equal to '4 distance from anterior border of naris to posterior margin of eye; striped color DARKS TMM SE Ar ates error eis: ON kN.. sees Lb puteh serait yield (Bh, Sb oe earl eae. 3 — Middorsal scales more than 300; eye small, indistinct, equal to 4 the distance from anterior border of naris to posterior margin of eye; generally WMD OMMIECOl OSMAN ac. heute Se Pen. cent a cesar ate eS Uae aga 4 3. Middorsal scales 170-189; small size, maximum snout—vent length 108 mm; known from Barbados, Martinique and St. Lucia .......... bilineatus — Middorsal scales 262-287; medium size, maximum snout-vent length 138 mm; known from several localities on southern coastal plain of Hispaniola 50 0.5 6 Gy blpedh ee eee tial ate ie a ene lira arin Se an aR Na NOE Ne eet COP pyrites 4. Three labial scales; middorsal scales 377-414; scale row reduction by fusion of rows 0 + 1; known only from the Massif de la Selle in Haiti ree ree Ge th Pt RE LO ive as a ne bo ae Ae ers 8 tp leptepileptus — Four labial scales; middorsal scales fewer than 380; scale row reduction DYerUSIOMEOMROWSI2 tape Shek. Wiss Ocoee eM ee hase Deis aie Besos 5 5 5. Middorsal scales 302-342; body uniformly pigmented; anal tubercles and spurs absent; known only from the Sierra Martin Garcia, Dominican ING pul iGreen. Serer res fe ey Gert atin. GH RS Ble m eels oi asbolepis — Middorsal scales 370-380; body unpigmented; anal tubercles and spurs present in most males; known only from the tip of the Samana Peninsula, DOnTia cansRe publican, paren erties teers LE Wek Soe: calypso 6. Uniformly dark above, no light spot on snout or tail tip; known only from Watling Island (=San Salvador), Bahamas ...................... columbi 220 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON — Usually with longitudinal dark stripes on each dorsal scale row, distinct light spot on snout and tail tip; known from Bay Islands of Honduras, Swan Islands, Providence and San Andres Islands ............... goudotii Acknowledgments The work in Haiti was supported by NSF grant SER 77-04629 to Thomas. We thank Luis Rivera Cruz for his very capable field assistance in Haiti. The field work by McDiarmid was made possible in part by funds from the U.S. Fish and Wildlife Service and the help of the late Howard W. Campbell. George Gorman and Albert Bennett provided the opportunity to visit the Dominican Republic that led to one Valle de Neiba specimen of L. pyrites. Field work by Thompson was supported by a grant from the National Geographic Society. We thank Albert Schwartz (ASFS) for greatly augmenting the sample size of L. /eptepileptus and William P. McLean for kindly donating a specimen of L. bilineatus. The Matias family south of Las Galeras was most helpful and hospitable during a visit by Thomas to look for L. calypso. Keith Christian and Terry Hazen helpfully pro- vided computer facilities and advice and Claudia Angle prepared some of the figures. Ron Crombie and George Zug provided useful comments on the manu- script. Literature Cited Klauber, L.M. 1940. The worm snakes of the genus Leptotyphlops in the United States and northern Mexico.— Transactions of the San Diego Society of Natural History 9(18):87-161. List, J.C. 1955. External limb vestiges in Leptotyphlops.—Herpetologica 11(1):15-16. . 1966. Comparative osteology of the snake families Typhlopidae and Leptotyphlopidae. — Illinois Biological Monograph 36:1-112. Orejas-Miranda, B. R., R. Roux-Esteve, and J. Guibe. 1970. Un nouveau genre de Leptotyphlopides (Ophidia) Rhinoleptus koniagui (Villiers).—Comunicaciones Zoologicas del Museo de Historia Natural de Montevideo 10(127):1-4. ——,, G. R. Zug, D. Y. E. Garcia, and F. Achaval. 1977. Scale organs on the head of Leptotyphlops (Reptilia, Serpentes): a variational study. — Proceedings of the Biological Society of Washington 90(2):209-213. Richmond, N. D. 1961. The status of Typhlops silus Legler.—Copeia 1961 (2):221-222. Thomas, R. 1965. The genus Leptotyphlops in the West Indies with description of a new species from Hispaniola (Serpentes, Leptotyphlopidae).—Breviora 222:1-12. Tihen, J. A. 1945. Notes on the osteology of typhlopid snakes.—Copeia 1945(4):204—210. (RT) Department of Biology, University of Puerto Rico, Rio Piedras, Puerto Rico 00931; (RWM) U.S. Fish and Wildlife Service, National Museum of Natural History, Smithsonion Institution, Washington, D.C. 20560; (FGT) Florida State Museum, University of Florida, Gainesville, Florida 32611. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 221-231 A NEW SKINK (REPTILIA: SAURIA: LETOLOPISMA) FROM FIJI George R. Zug Abstract.—A new lizard, Leiolopisma alazon, is described from the southern- most island cluster, Ono-i-Lau, of Fiji. This species represents the first record of Leiolopisma from the Fijian Islands. Leiolopisma alazon shows no evidence of close relationship with its nearest geographic congeners on New Caledonia. In- stead, L. alazon does show a relationship to the New Zealand “‘leiolopisma”’ skinks and is most similar to the species of Cyclodina, except by its possession of a transparent palpebral disc in each lower eyelid. To facilitate comparison with New Caledonian Leiolopisma, the latter were studied and five species groups are tentatively recognized. The skinks of the Fiji Islands have long been known to represent three genera: Cryptoblepharus, Emoia, and Lipinia. A fourth genus, Eugonglylus, may occur in the Fijian group, but its presence remains unconfirmed. Thus, the discovery of a different lygosomine skink in the Ono-i-Lau group was a great surprise, particularly since the new species was found on only the smallest island of the three searched in this group. The new skink possesses an alpha palate, 11 premaxillary teeth, and toes covered dorsally by a single row of scales, characters of Greer’s (1974) group II skinks. It can further be recognized as a member of the genus Leiolopisma by moveable lower eyelids with a large transparent palpebral disc in each lid and well developed prefrontal scales. Leiolopisma is predominantly a Southwest Pacific group with species on Lord Howe Island, New Caledonia, New Hebrides, New Zealand and Chatham Islands, Australia, and Tasmania (Greer 1979). Because the specimens from Ono-i-Lau differ in several features from the other species of Leiolopisma, the Ono-i-Lau population is described as: Leiolopisma alazon, new species Figs. 1, 2 Holotype.—USNM 230000, adult male; Fiji, Ono-i-Lau, Yanuya Island, 20°37’S 178°41'W, coll. George R. Zug, 29 Apr 1982. Paratopotypes.—USNM 229989 (cleared and stained), USNM 229990-9999 (alcoholics), juveniles, females, and males with same collection data as holotype. Diagnosis.— This skink is a small Leiolopisma with an adult snout-vent length (SVL) of 45-65 mm, a robust elongate body, short but well developed limbs, a pair of frontoparietal scales, prefrontals not in contact, smooth dorsal scales, 34— 37 scale rows around midbody, and an orange to red tail in adults. These features, singly or in combination, distinguish this species from all currently recognized Leiolopisma species. The Australian species L. duperreyi, L. greeni, L. metalli- cum, L. ocellatum, L. palfreymani, L. platynotum, L. pretiosum, and L. trilinea- tum possess a single (fused) frontoparietal scale; L. baudini, L. coventryi, L. en- 222 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Dorsal and lateral view of the head of Leiolopisma alazon. Based on the holotype, USNM 230000. Inset 1 mm. trecasteauxil, and L. zia, have fewer than 34 scale rows around midbody; L. spenceri has supernasals; adult L. lichenigerum has SVL greater than 75 mm. No New Zealand Leiolopisma possesses a uniform reddish or orangish tail dorsally. The majority, L. acrinasum, L. chloronoton, L. fallai, L. gracilicorpus, L. grande, L. homalonotum, L. infrapunctatum, L. lineoocellatum, L. nigriplantare, L. ota- gense, and L. suteri also have adult SVL greater than 75 mm. Leiolopisma fas- ciolare, known only from the original description, has 27 scale rows around mid- body. The majority of the New Caledonian species, L. austrocaledonicum, L. deplanchei, L. euryotis, L. novacaledonicum, L. steindachneri, L. tricolor, and L. variabile, have a single frontoparietal. Leiolopisma greeri and L. nigrofasciolatum have large prefrontals broadly in contact medially. Description of holotype.— Adult male of 60.4 mm SVL and 46 mm tail length (regenerated); general habitus of elongate and robust body, head triangular in outline, depressed and rounded snout, limbs short but well formed, robust and subcylindrical tail gradually tapering to point. Head length (tip of snout to ear opening) 12.4 mm, head width (at angle of jaw) 9.0 mm, snout to naris distance 1.1 mm, axilla to inguen (trunk) length 34.8 mm, hindlimb length 16.0 mm. Forelimb (length, 12.3 mm) shorter than hindlimb; no overlap when adpressed. Rostral wider than high, in broad contact with frontonasal; prefrontals mod- erate-sized, paired, and widely separated by frontal-frontonasal contact; fronto- nasal octagonal, width and length subequal, in broad contact with frontal; frontal VOLUME 98, NUMBER 1 223 Fig. 2. Dorsal view of Leiolopisma alazon (USNM 230000). truncated rhomboidal, length twice width, in broad contact with frontoparietals; frontoparietals paired, somewhat rhomboidal, length slightly greater than width, bordering interparietal and parietals posteriorly; interparietal large, approximately half area of frontoparietal, rhomboidal, width approximately two-thirds length; parietals paired, trapezoidal, length two to three times width, bordered posteriorly by pair of large nuchal scales and upper secondary temporal scales. Nasal large, obovate, grooved, naris circular and nearly centered in scale; anterior loreal higher than wide; posterior loreal wider than high; upper and lower preoculars large, contiguous with row of small, distinct suboculars separating labials from lower eyelid scales. Four large supraoculars on each side; 9 supraciliaries; 10 upper eyelid scales; anteriorly an accessory supraciliary between second supraciliary, eyelid, and upper preocular; 3-4 postoculars; lower eyelid with large palpebral disc, 1.1 mm in maximum length; primary temporal and pair of secondary tem- porals on each side; upper secondary temporal largest and on posterior border of parietal. Supralabials 7, fifth beneath eye, sixth highest; infralabials 6; mental semicircular; postmental large and pentagonal; 3 pairs of chinshields, first pair largest and only pair in contact medially. Ear openings vertically elliptical, 0.6 x 1.3 mm. Dorsal scales smooth, in 70 transverse rows from base of tail to nuchals; 36 rows of scales around midbody; ventral scales slightly larger than dorsals and laterals; preanals slightly larger than preceding scales; 14 smooth lamellae beneath fourth finger; 21 smooth lamellae beneath fourth toe. Finger lengths, 3 = 4 > Ose letoes4i 6) eo 1. 224 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON In life, dorsum with brownish olive ground color, spotted with light golden or beige and dark brown scales; labial and mental area with dark brown streaks or spots; venter golden beige; tail rufous orange above and below. In alcohol, dorsum brown with scattering of dark brown and beige spots; dark brown spots concen- trated dorsolaterally to give impression of stripe from posterior edge of eye to inguen; sides light brown, rapidly grading to cream colored venter; head brown dorsally with few dark specks, upper and lower lips with dark brown bars that extend onto chinshields; limbs beige with numerous dark brown spots dorsally; tail orangish beige dorsally and laterally, cream ventrally. Variation. — The paratopotypic specimens are all smaller than the holotype. The two females (USNM 229990, 229998) are 48.1 and 42.7 mm SVL, respectively; both appear mature or are maturing because each contains small, but vitellogenic, follicles. The remaining specimens are males, ranging from 26.9 to 57.7 mm SVL. The smallest may be a hatchling/newborn, since it possesses a ventral scale ab- erration which may be a yolk-sac scar. Two males (51.3 and 57.7 mm SVL) have large testes and epididymides and are presumed mature. The other males (33-49 mm SVL) are immature with small testes. In all specimens, original or completely regenerated tails are equal to body length or slightly longer, robust, cylindrical with gradual taper to the tip. The head is triangular on all specimens with a head width to head length proportion of 0.67—0.74. The limbs are well formed and short. The hindlimb and forelimb when adpressed to trunk remain widely sep- arated. Trunk length (measured between fore- and hindlimbs) is approximately half the snout—vent length (48-58%, mean 53%). Dorsal head scale pattern is relatively invariable; the differences are minor shape and positional alterations. Only the supraciliaries vary in number, 8—9; 8 supra- ciliaries is the modal number with 9 resulting from a division of the second supraciliary. The lateral head scales are somewhat more variable in size, shape, and position. The upper eyelid bears 8-11 scales, mode 10. The nasal retains the obovate to parallelogram shape; naris is typically central, but in some individuals, the naris is shifted anteroventrally toward inferior border; posterodorsal and medioventral grooves extend from the naris in a few individuals, never creating a distinct division of the nasal, ventral groove most prominent when present. Anterior and posterior loreals are higher than wide and wider than high, respec- tively, as in the holotype, although the width of the anterior loreal varies from approximately two-fifths to half of the height and, in the posterior loreal, the shape is rectangular to weakly trapezoidal. The preoculars are paired in all spec- imens, the upper usually equal in size to lower, occasionally smaller. The subocular row is complete in all individuals; although in two individuals, the suboculars above the fifth supralabial are only slightly larger than the scales covering the lower eyelid. The suboculars grade imperceptibly into the postoculars; one post- ocular is always wedged between the last two supraciliaries. Similarly, there are one or two small scales wedged between the first and second supraciliaries and the anterior upper eyelid scales. Of the temporals the size ranking of small to large is invariably primary, lower secondary, and upper secondary. There are 7 supralabials, fifth beneath orbit, and 5-7 infralabials, mode 6, in all specimens. Ear opening is elliptical and subequal to palpebral disc of lower eyelid in all. Low rounded auricular scales are barely evident along anterior margin, 2 or 3 in number. VOLUME 98, NUMBER 1 DES All body scales are smooth. The dorsal scales lie in 71-77 transverse rows, mean 73.8, from base of tail to nuchals. There are 34-37, mode 34, scales around midbody. The fourth finger bears 12-15 lamellae, mode 15, the fourth toe 20- 23 lamellae, mode 22. In life, all individuals had the same coloration and general pattern as the ho- lotype; however, there was a single striking difference in tail color. USNM 22989- 95 had tails in shades of salmon, 22999699 in shades of beiges; these colors were as evident ventrally as dorsally. Tail color differences have no association with size (age), sex, or reproductive condition. In alcohol, dorsal ground colors range from tan to brown with a various density of dark brown spotting. In all, there is a concentration of these spots into an interrrupted dorsolateral stripe on each side. In individuals less than 55 mm SVL, these dark spots are dense and arranged into ragged-edged, irregular transverse bars. Etymology.— The name “alazon”’ is Greek (neuter) for wanderer or rover. It is used in allusion to this species wandering from the common geographic track (New Caledonia through New Zealand to Australia) of Pacific Leiolopisma as well as to its dispersal to the remote Ono-i-Lau group. Ecological Observations Yanuya Island is a small coral island, probably less than a hectare in area, in the Ono-i-Lau cluster. It is a fairly flat island with only scattered, irregularly spaced limestone outcrops, of 2—4 m in height. The entire island is covered by forest; the canopy is nearly closed at approximately 10 m. The island serves as a rookery area for brown boobies and sooty terns. Presumably, the rookery accounts for the presence of a sandy humus-like soil and the reported presence of “plenty snakes,” presumably the boid Candoia bibroni. The forest floor had a light scat- tering of leaves and fallen limbs. All but one of the Leiolopisma alazon were in or under rotten logs; the single exception was a juvenile caught resting on a blade of grass. The smallest specimen (26.9 mm SVL) appears to have a yolk-sac scar and, hence, may have been recently hatched or born. However, in “leiolopisma” of similar adult size (e.g., Lampropholis delicata & quichenoti, Clarke 1965), one week old juveniles have 17-20 mm SVL. Since none of the females were gravid or pregnant, it is impossible to declare this species as ovi- or viviparous. Both females (42.7 and 48.1 mm SVL) possess small, but clearly vitellogenic follicles. The smallest male with well developed testes and epididymides is 51.3 mm SVL; a male of 49.3 mm SVL has small testes and is likely immature. These data suggest a sexual dimorphism in body size with females maturing at a smaller size than males. Three other lizard species were seen or captured on Yanuya: Cyrtodactylus pelagicus, Emoia cyanura, and Emoia cf. samoensis complex. Emoia cyanura was the most abundant of the Yanuya lizards. Cyrtodactylus is the only species sharing the same microhabitat as Leiolopisma, but the two were not found under the same logs. Leiolopisma was not found on either of the two other larger islands visited in the Ono-i-Lau cluster. Davora Island lacks a humus-like soil, having only a fine coral rubble with a dead leaf cover; Lipinia noctua, Cryptoblepharus, and Emoia 226 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON cyanura were present. Estad Island is a sand island with partially humus-like soil; only Emoia cyanura was observed. Distribution and Relationships With Other Oceanic Leiolopisma The occurrence of a Leiolopisma in the southern Fijian Lau group seems anom- alous, but then the distribution of the genus Leiolopisma is a peculiar one. The occurrence of the type-species of Leiolopisma (L. telfairi) on Round Island of the Mauritius group in the western Indian Ocean with all other representatives in the Southwest Pacific makes little zoogeographic sense, no matter how ardent a sup- porter one is of long distance dispersal or vicariance. On zoogeographic grounds, Leiolopisma currently appears to be a polyphyletic group and with further study is likely to be divided into the West Indian Ocean Leiolopisma and a Southwest Pacific group of one or more genera!; however, such a taxonomic reshuffling is well beyond the scope of the present study. Considering only the distribution of the Pacific Leiolopisma, the pattern still remains unique among the Pacific her- petofauna, since it is J-shaped incorporating the New Caledonian cluster south- ward to New Zealand and westward to center on southeastern Australia and Tasmania. Within each of the aforementioned areas, the genus has undergone great diversity. The origin of this pattern has been variously explained. Although Brown (1956) does not address this pattern directly, he does suggest that the Loyalty and New Caledonian fauna derive largely from the Australian region. Towns (1974) treats the origin of the New Zealand skink fauna and recognizes four possible dispersal routes: 1) Australia-Tasmania track; 2) Fiji-Tonga-Ker- madec track; 3) New Caledonia-Norfolk track; 4) Australia-Lord Howe track. Towns’ analysis does not eliminate any of these tracks as a possible dispersal route and, although he seems to favor the two Australian tracks, he also advocates the Fiji-Kermadec track for the Leiolopisma suteri-lichenigerum group. Greer (1974) labels the Leiolopisma pattern as relictual; he further suggests that the diversity of Leiolopisma at the periphery of the major radiation of skinks may result from Leiolopisma being at a competitive disadvantage with the more ad- vanced skink genera. However, he also suggests Australia as the source area for the eastward dispersal of Leiolopisma into the Southwest Pacific. Hardy (1977) proposes the Sulawesi-New Guinea region as the source area for the ancestral Leiolopisma stock which, with multiple invasions, entered Australia and New Caledonia; New Zealand in turn received multiple invasions from these two source areas. Only Hardy’s interpretation is based on hypotheses of interspecific rela- tionships and, hence provides rational estimates of vicariance tracks or dispersal pathways. The presumed close relationship of the Australian L. coventryi and L. entrecasteauxii with the New Zealand Leiolopisma species and with the New ' Recently Greer (1979) split Pacific Leiolopisma into two genera, the alpha palate Leiolopisma and the beta palate Lampropholis. Earlier Hardy (1977) divided the New Zealand Leiolopisma species into the scaly eyelid Cyclodina and the palpebral disc eyelid Leiolopisma. Other genera are likely to be recognized in the future, because, even now, the division of “‘leiolopisma” skinks is uncertain. Greer (1979) does not recognize Cyclodina or Pseudemoia. My usage of Leiolopisma, thus, straddles the recommendations of Hardy and Greer. I anticipate that in the future a/azon will be re-assigned to a currently unrecognized genus, such as Oligosoma. VOLUME 98, NUMBER 1 DAT AUSTROCALEDONICUM HE! DERLANE ANTER. LOREAL HIGHER THAN WIDE POSTER. Coe WIDER ANTER. & POSTER. LOREALS HAN HIGH NIGROFASCIOLATUM HIGHER THAN WIDE LIMBS MV LIMBS MEDIUM VARIABILE TOE LAMELLAE MANY QU Se L ANTER. & POSTER. LOREALS DERATE TO LARGE WIDER THAN HIGH TOE LAMELLAE FEW SUPRANABIALSHS De NOVACALEDONICUM SUPRALABIALS 7 GREERI FRONTOPARIETALS PAIRED PREFRONTALS IN CONTACT ANTER. LOREAL REDUCED Cee oie ne SINGLE PREFRONTALS NO MAUS Ce ANTER. LOREAL EXCLUDED FROM SUPRALABIALS ANTER. LOREAL TOUCHES SUPRALABIALS FRONTOPARIETALS PAIRED SUBOCULAR ROW PREFRONTALS NO CONTACT INCOMPLETE COMPLETE EURYOTIS ae STEINDACHNERI TRICOLOR Fig. 3. Taxon and character phenogram of the major species/forms of New Caledonian Leiolopisma and L. alazon. The resulting clusters of species/forms likely indicate some degree of relationship; however, hypotheses of relationships require the establishment of character state polarities and an analysis of a wider array of morphological characters. Caledonian L. nigrofasciolatum suggested to Hardy that the Southwest Pacific Leiolopisma had their initial center of divergence in New Caledonia and one stock reached Australia via the Eastern Australian Current, and another stock reached New Zealand by the tropical cyclone trackway. If dispersal of Leiolopisma is aided by oceanic currents and/or cyclonic storms, only the latter seems likely to have placed Leiolopisma in the southern Lau group of Fiji, since the general flow of ocean currents is westward and southwestward through the Fijian islands. The storm track would have to have been a highly irregular one to have crossed the New Caledonian and Loyalty region and then traveled 1300 km east-northeast to the Lau islands. Such an interpretation of the presence of Leiolopisma alazon in Ono-i-Lau is highly speculative, but seems more reasonable than an upcurrent dispersal of 1450 km from New Zealand. If this interpretation is correct, the closest relatives of L. alazon should be with the New Caledonian species; however, such a relationship is not strongly supported by the evidence (external morphology) currently avail- able to me. From a strictly phenetic viewpoint, I recognize five species groups of New Caledonian Leiolopisma (nigrofasciolatum, variabile, euryotis, deplanchei, and austrocaledonicum groups; Fig. 3 and Tables 1 and 2). Note that this arrangement is very tentative, since I have not examined representatives of all species and have, of necessity, relied on literature descriptions. The nigrofasciolatum group shares with alazon paired frontoparietal scales and high numbers of dorsal trunk and midbody scale rows; otherwise, they are very dissimilar owing to the strikingly different body form of a long thin trunk, pointed head, and long slender limbs of 228 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Character matrices for Fijian and New Caledonian Leiolopisma. Body measurements and proportions (means). Abbreviations: BL, body length; HbL, hindlimb length; HL, head length. Minimum adult Taxon SVL (mm) HL/SVL HbL/SVL HbL/BL alazon 45-50 0.21 0.29 0.56 austrocaledonicum 40 0.20 0.43 0.86 deplanchei 35 (?) 0.20 0.29 0.60 euryotis 35 (?) 0.21 0.42 0.82 greeri* 60 ? ? zy nigrofasciolatum 80 0.20 0.41 0.82 novaecaledonicum* 60 0.21 0.39 ? steindachneri 50 (?) 0.25 0.47 1.08 tricolor 45-50 0.22 0.44 1.01 variabile 80 0.23 0.57 1.33 * Data derived from original description. nigrofasciolatum group members. Leiolopisma variabile shows no similarity to L. alazon; in fact, L. variabile is unlike any other New Caledonian Leiolopisma. Leiolopisma variabile shares body size and shape with L. nigrofasciolatum, but its meristic characters are strikingly different, particularly such features as the granular-scale patch on the posterior surface of the thigh and the high number of fourth toe lamellae. The euryotis group is a monothetic group based on the re- duction of the anterior loreal, and as such, L. a/azon does not belong with this group. It is further excluded from the euryotis group by differences in habitus and associated scale characters, e.g., dorsal scale rows and fourth toe lamellae (see Tables 1 and 2). Leiolopisma deplanchei? and L. alazon have little in common aside from body shape and relative limb length. The austrocaledonicum group is a diverse complex of species and populations; they share general features such as loreal shape and number of supralabials with L. alazon, but differ in many other features including habitus. Since Leiolopisma alazon has no obvious affinities with the New Caledonian Leiolopisma, a brief comparison with the New Zealand leiolopismine skinks is appropriate. Considering New Zealand as a “source”’ area, however, assumes a much more ancient arrival of Leiolopisma into the Lau group and a biogeograph- ical explanation based on vicariance rather than long-distance dispersal (Springer 1982). Such an explanation suggests an arrival (in the Fiji area, not the Lau group specifically) time of 20-50 million years ago during the formation of the Mel- anesian marginal sea (Crook and Belbin 1978). This interpretation would indeed make Leiolopisma an old lineage, but this is not discordant with its presumed primitiveness relative to other lygosomine skinks (Greer 1974) or the age of many extant genera of lizards (Estes 1983). The New Zealand “‘leiolopisma”’ skinks possess primitive traits relative to most of their Australian and New Caledonian congeners (Hardy 1977). Cyclodina has 2 Leiolopisma deplanchei Bavay (1869:23) is a small lizard with an elongated cylindrical trunk and short limbs, and it is the L. deplanchei referred to herein. Leiolopisma deplanchei Bocage (1873:229) is a junior homonym and refers to a large lizard with long limbs; the original description suggests that Bocage’s name is a junior synonym of L. nigrofasciolatum Peters (1869). Roux (1913:115) similarly identified deplanchei Bocage as a synonym of nigrofasciolatum Peters. 229 VOLUME 98, NUMBER 1 = = = = = aL = = + + posiejus S c-£ 0 CY-LE €£-67¢ 6£-LE SC-I7 Oc-LI CG 6E-SE Ch-8E Ce-vEe pS-SP 99-£9 cc-IS id ouou id 8-9 L-9 9 IA A A 8 L lL G 6 6 I I I 8 6-8 8 oul duwio9 oul (4 (4 C MA=Y poonpol - + + + + — T T T I 4 C I I I C I IeA uy 1s Aou du oid Ina dop sne eye s19j9eIeyO SE 0 0 € (46 ce-0E EG (43 é SEMG é 81 Ge 6€-PE ce vE é £6-82 é Os id id id id 9 OI-8 é 9) A IIA IA A L 6 8 L ¢ €< é G I €-C é I 8 OI-6 é 8 oul oul oul dwos G onal Cc G yY 70 mm SVL) and incomplete subocular rows; most tend to have slender bodies and longer limbs. Nonetheless, some species, e.g., L. lineoocellatum and L. suteri do share similar scale counts and habitus with L. alazon. Thus, the data suggest the derivation of L. alazon from a New Zealand “‘lei1olopisma”’ stock (Fig. 4). The greater similarity of L. alazon to the New Zealand Cyclodina than to the New Zealand Leiolopisma emphasizes the need for a re-evaluation of the rela- tionships of the “leiolopisma”’ skinks. The diversity of the New Caledonian Leio- lopisma and the apparent structural divergence of the different species and species groups similarly encourages such a re-evaluation. Acknowledgments The specimens of the new species were collected during a herpetological and ichthyological survey of Fiji sponsored by the Max and Victoria Dreyfus Foun- dation and the Smithsonian Scholarly Studies Program, with logistic support from the Marine Laboratory of the Fijian Ministry of Agriculture and Fisheries and the School of Natural Resources, The University of the South Pacific. The support from these institutions and associated individuals is much appreciated. I also wish to thank my colleagues on the “Blue Dolphin” for their companionship and VOLUME 98, NUMBER 1 Ap)il cooperation during the Lau voyage, C. A. Ross for his assistance and encourage- ment in the analysis of the Leiolopisma, and W. C. Brown, R. I. Crombie, G. S. Hardy, W. R. Heyer, R. Sadlier, and V. G. Springer for their constructive reviews of the manuscript. Colleagues in other museums have been most helpful in loaning specimens; I wish to thank E. N. Arnold (British Museum (Natural History)), E. J. Censky and C. J. McCoy, Jr. (Carnegie Museum of Natural History), H. K. Voris (Field Museum of Natural History), P. Alberch and J. P. Rosado (Museum of Comparative Zoology, Harvard University), E. R. Brygoo (Muséum National d’Histoire Naturelle), and D. M. Harris and A. G. Kluge (University of Michigan, Museum of Zoology). Literature Cited Bavay, A. 1869. Catalogue des reptiles de la Nouvelle-Caledonie et description d’especies nouvelles. — Memoires de la Societe Linneenne de Normandie 15:1-37. Bocage, J. V. Barboza du. 1873. Sur quelques sauriens nouveaux de la Nouvelle Caledonie et de l’Australie.—Jornal de Sciencias Mathematicas, Physicas e Naturae da Academia Real das Sciencias de Lisboa 15:228-232. Brown, Walter C. 1956. The distribution of terrestrial reptiles in the islands of the Pacific Basin. — Proceedings of the 8th Pacific Science Congress 34:1479-1491. Clarke, Carina J. 1965. A comparison between some Australian five-fingered lizards of the genus Leiolopisma Dumeril & Bibron (Lacertilia: Scincidae).— Australian Journal Zoology 13(4):577- 592. Crook, K. A. W., and L. Belbin. 1978. The Southwest Pacific area during the last 90 million years. — Journal of the Geological Society of Australia 25(1):23-40. Estes, Richard. 1983. Sauria terrestria, Amphisbaenia.— Handbuch der Palaoherpetologie 10:1-249. Greer, Allen E., Jr. 1974. The generic relationships of the scincid lizard genus Leiolopisma and its relatives.— Australian Journal of Zoology, Suppl. Ser., (31):1-67. . 1979. A pnylogenetic subdivision of Australian skinks.— Records of the Australian Museum 32(8):339-371. Hardy, Graham S. 1977. The New Zealand Scincidae (Reptilia: Lacertilia); a taxonomic and zoo- geographic study.— New Zealand Journal of Zoology 4:221-325. 1979. The karyotype of two scincid lizards, and their bearing on relationships in genus Leiolopisma and its relatives (Scincidae; Lygosominae).— New Zealand Journal of Zoology 6: 609-612. Peters, W. 1869. Neue Gattungen und neue oder weinger bekannte Arten von Amphibien (Eremias, Dicrodon, Euprepes, Lygosoma, Typhlops, Eryx, Rhynchonyx, Elapomorphus, Achalinus, Co- ronella, Dromicus, Xenopholis, Anoplodipsas, Spilotes, Tropidonotus).—Monatsberichte der Koniglich-Preussichen Akademie der Wissenschaften zu Berlin 1869:432—447. Roux, Jean. 1913. Les reptiles de la Nouvelle-Calédonie et des iles Loyalty. Jn Fritz Sarasin and Jean Roux, eds., Nova Caledonia.—A. Zoologie, Vol. 1, pp. 79-160. Springer, Victor G. 1982. Pacific plate biogeography, with special reference to shorefishes.—Smith- sonian Contributions to Zoology (367):1-182. Towns, D. R. 1974. Zoogeography and the New Zealand Scincidae.— Journal of the Royal Society of New Zealand 4(2):217-226. Department of Vertebrate Zoology (Amphibians & Reptiles), National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 232-236 HALOPHILA DECIPIENS, AN UNREPORTED SEAGRASS FROM THE PHILIPPINES Ernani G. Menez and Hilconida P. Calumpong Abstract.— Halophila decipiens Ostenfeld is reported for the first time from the Philippines. This report and a previous one from the Gulf of Siam represent the only distribution records of the species in the northern Pacific. On July 29, 1983 and April 11, 1984, the authors visited an artificial reef site at Sumariling Beach, Siaton, Negros Oriental, Philippines (Fig. 1B) for purposes of collecting seaweeds and seagrasses for systematic study. Among their collections were fertile specimens of an unreported seagrass, Halophila decipiens Ostenfeld (Hydrocharitaceae). The discovery of H. decipiens in the Philippines is a new distribution record. Additionally, this record and a previous one from the Gulf of Siam represent the only locations in the northern Pacific. _ The specimens are cited according to the senior author’s field number and are deposited in the U.S. National Herbarium, Smithsonian Institution, Washington, D.C., in the Silliman University Herbarium, Central Philippines, in the Rijks- herbarium, Leiden, Netherlands, and in the Philippine National Herbarium. Halophila decipiens Ostenfeld, Bot. Tidsskr. 24:260. 1902. Figs. 2, 3 Description. — Plants monoecious, pale green, with long, thin rhizomes not more than 1 mm in diameter; internodes 5—40 mm long; usually one long root present below each erect shoot. Erect shoot short, bearing a pair of leaves borne on each node. Leaves petiolate, the petioles triquetrous, 5-10 mm long, enveloped by a pair of transparent, orbicular to ovate-elliptic stipules having only the dorsal surface hairy and the apex emarginate. Leaf blades oblong, oval or elliptic, round at the apex, the base shortly attenuate or cuneate, 6-10 mm wide, up to 22 mm long, the margins minutely serrate; lateral veins 6-9 pairs, a few occasionally forked, the midrib connected to the intramarginal vein at the top; both surfaces of leaf blade beset with minute, unicellular hairs. Spathes arising between a pair of leaves, ovate, obovate, or elliptic, occasionally slightly apiculate, transparent and scarious, up to 7 mm long and 6 mm wide, with an uneven margin, only the dorsal surface densely hairy, enclosing | male and 1 female flower. Hooded peri- anth segments 3, enclosing a long-pedicelled staminate flower up to 3.5 mm long and 0.3 mm in diameter consisting of 3 oblong, sessile anthers, each with a thin membranous cover. Pistillate flower subsessile, with an elliptic to ovoid ovary, up to 2.5 mm long and | mm in diameter; hypanthium 2—4 mm long; styles 3, up to 12 mm long. Fruits ovoid to subglobose, 4 mm long and 2.5 mm in diameter, beaked. Natural history.—Specimens of Halophila decipiens from the Philippines were collected by SCUBA diving in a bay, 1.5 km from a river outlet, at depths of 11 to 23 meters. During the rainy season, in June to August, the water in the bay VOLUME 98, NUMBER 1 233 Fig. 1. Halophila decipiens. A, World distribution; B, Philippine distribution. becomes turbid due to agitation of the bottom; otherwise it is clear. The plants grew in soft mud mixed with fine sand. They were observed growing with fertile Halophila minor and H. spinulosa at 11 meters, but with H. minor only at 13, 17, and 20 meters. Pure stands of H. decipiens were found at 23 meters depth. The plants thrive in sheltered areas on soft mud, sand-mud, and sandy substrates, usually in deep waters. They have been collected from an estuary in New Cale- donia, mangrove-swamps on Guadeloupe and Puerto Rico, at the entrance of a river in Queensland, in a creek at Kingston Harbor in Jamaica, and at reef sites 234 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 10 mm Y/ i 2mm 2mm 290 p Fig. 2. Halophila decipiens. A, Habit of sterile specimen; B—C, Stipules, with hairs on dorsal surface; D, Cross-section of leaf showing hairs on both surfaces; E, Magnified hair on leaf surface; F, Leaf, showing lateral veins and serrate margins; G, Magnified serrate margin of a leaf. in Queensland and Puerto Rico. Den Hartog (1970) considers H. decipiens pan- tropical. Range.— Halophila decipiens occurs in the Seychelles, Cargados Carajos, India, Sri Lanka, Thailand, Indonesia, Queensland, New South Wales, New Caledonia, Tahiti, Jamaica, Puerto Rico, Virgin Islands, Guadeloupe, St. Vincent, Marti- VOLUME 98, NUMBER 1 235 Fig. 3. Halophila decipiens. A, Habit of a fertile specimen with male and female flowers; B, Male and female flowers without the spathe; C, Magnified portion of a style with papillae; D, Male and female flowers enclosed by spathe; E-F, Spathes with keels and hairs; G, Mature flowering stage, showing female flower after the styles have fallen off and a male flower; H, Male flower with three anthers enclosed by three perianth segments; I, Spathe enclosing beaked fruit and male flower after anthesis, the latter showing long pedicel, perianth segments and persistent connective tissues; J, Beaked fruit, showing subglobose seeds. 236 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON nique, Barbados, Tobago, Trinidad, Curacao, Venezuela, Colombia, Panama, and Costa Rica. These distribution data are from den Hartog’s (1970) publication. Remarks.—There are four species of Halophila previously recorded (Menez, Phillips and Calumpong 1983) from the Philippines. Halophila ovalis, H. minor, and H. decipiens, which constitute 50% of the known taxa in section Halophila, occur in the Philippines. Since these three species are similar in having erect shoots with one pair of leaves, their taxonomic distinctions are based primarily on leaf morphology and whether they are monoecious or dioecious. Halophila minor, probably the smallest of the three species, and H. ovalis, the largest, are characterized by having entire leaf margins and no hairs on the leaf surfaces. Halophila decipiens has serrate leaf margins and hairs on both surfaces of the leaf blade. In addition, the latter has 6—9 pairs of lateral veins; H. minor has 4-7, and H. ovalis has 12-22. Halophila decipiens is monoecious and occurs in deeper waters, while H. minor and H. ovalis are dioecious and are found in shallower waters. Due to the great variability in the morphology of taxa in section Halophila (i.e., leaf size and shape, plant size, and venation), which is probably influenced by the environment, taxonomic confusion apparently still exists. McMillan and Williams (1980), in their study of isozymes, demonstrated that leaf shape and venation patterns of taxa in section Halophila are correlated with their isozyme variations. Specimens studied. —83 EM-1, Sumariling Beach, Siaton, Negros Oriental, Phil- ippines, soft mud mixed with fine sand, silty bottom, 23 m deep, 29 Jul 1983; 84 EM-1, Sumariling Beach, Siaton, Negros Oriental, Philippines, soft mud mixed with fine sand, silty bottom, 11-20 m deep, 11 Apr 1984. Acknowledgments The senior author acknowledges the financial support of the Smithsonian In- stitution Fluid Research Fund for travel and research in the Philippines. The authors appreciate the assistance provided by Mr. Lawton Alcala, Ms. Janet Estacion, and Mr. Daniel Catada of the Silliman University in collecting research materials of marine plants at Sumariling Beach. We wish to thank Dr. and Mrs. Fred Van deVusse for their hospitality and logistic assistance. The loan of Hal- ophila decipiens specimens from the U.S. National Herbarium, Smithsonian In- stitution, is gratefully acknowledged. We would like to express our gratitude to Dr. Ronald Phillips, Seattle Pacific University, and Dr. Calvin McMillan, Uni- versity of Texas at Austin, for their encouragement and critical review of this paper. Lastly, useful discussions and comments from Dr. Dieter Wasshausen, Smithsonian Institution regarding the staminate flower (Fig. 31) are appreciated. Literature Cited Hartog, C. den. 1970. Seagrasses of the world. Amsterdam, London.—North-Holland Publishing Company. 275 pages, 63 figures. McMillan, C., and S. Williams. 1980. Systematic implications of isozymes in Halophila section Halophila.— Aquatic Botany 9:21-31. Menez, E., R. Phillips, and H. Calumpong. 1983. Seagrasses from the Philippines.—Smithsonian Contributions to the Marine Sciences, Number 21, 40 pages, 26 figures. (EGM) Smithsonian Oceanographic Sorting Center, Museum of Natural His- tory, Smithsonian Institution, Washington, D.C. 20560. (HPC) Silliman Univer- sity Marine Laboratory and Department of Biology, Silliman University, Du- maguete City 6501, Philippines. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 237-242 REDESCRIPTION OF HETEROCARPUS LAEVIS A. MILNE EDWARDS (CRUSTACEA: DECAPODA: PANDALIDAE) Brian Kensley and William Tobias Abstract. — Heterocarpus laevis was figured by A. Milne Edwards in 1883, based on material from Martinique. One other specimen has been recorded from off the Cayman islands. No written description was provided, and the species has not been recorded since. Based on fresh material from St. Croix, U.S. Virgin Islands, the species is figured and redescribed. Heterocarpus laevis is unique in this genus, for lacking lateral carapace carinae. A deepwater shrimp trapping survey was conducted by the second author at selected sites off St. Croix, U.S. Virgin Islands, in August 1982, as part of the Virgin Islands Fishery Development and Demonstration Project. Twenty-four hour-sets were made with polyethylene traps (Fathoms Plus, San Diego, California), 73 cm wide by 87 cm long by 30 cm high, baited with blue runner (Caranx crysos). Trap mesh size was reduced to 1.3 cm with an internal plastic netting. Each set consisted of six traps spaced at 20 m intervals with 5 kg weights before the first and after the last trap. Several deepwater shrimps and other invertebrate species were obtained from a depth of 460 m in the Salt River Canyon, off the north coast of St. Croix. These specimens were submitted to the first author for identification. Among these were six specimens which proved to be Heterocarpus laevis. In April 1883, Alphonse Milne Edwards published a set of 44 plates of new or rare crustaceans from various sources, including the Blake, and Travailleur and Talisman expeditions. Among these was Heterocarpus laevis, taken by the Blake Expedition in 169 fathoms (309 m) off Martinique in the Caribbean Sea. The plate included a lateral view of an entire animal, plus enlarged figures of the first antenna, the antennal scaphocerite, and the shorter pereopod 2. No written de- scription was published. Faxon (1896) recorded a second specimen, taken by the Blake Expedition from 297 fathoms (543 m) off Cayman Brac in the Cayman islands. This lack of description and rarity of records prompted the present paper. Family Pandalidae Genus Heterocarpus A. Milne Edwards Heterocarpus laevis A. Milne Edwards, 1883 Figs. 1-3 Heterocarpus laevis A. Milne Edwards, 1883 [unnumbered plate, no. 28 in se- quence].— Faxon, 1896:161.—De Man, 1920:109. Material.—Salt River Canyon, St. Croix, U.S. Virgin Islands, 460 m, coll. W. J. Tobias, 19 Aug 1982, sample #004. 238 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Be oe ae = \ 10 1 } J Fig. 1. Heterocarpus laevis: Male in lateral view. Scale in mm. Carapace Rostral Abdominal Rostral length length length formula 6) 31.4 B)5)2) 54.5 13/13 ) S18) 3)3)55) 54.0 et ce) BS)) - SED) - 2 30.2 37.0 56.0 12/2 2 30.6 38.8 55:0 yl 2 338s 38)2) Sy) 12/13 Description.—Rostrum 4-4 longer than carapace, directed oblique-dorsally, with lateral carina in basal '4—'4, with 12-13 ventral teeth almost reaching apex; with 11-13 dorsal teeth extending almost to apex, including 6 or 7 teeth posterior to orbital margin. Carapace with middorsal carina becoming obsolete in posterior ’%; strong antennal spine barely overreaching strong branchiostegal spine, both spines with obscure rounded buttress; carapace lacking lateral carinae; rostral base and anterior carapace around antennal and branchiostegal spines finely pubescent, hairs short, almost scale-like. Abdominal somites 1 and 2 dorsally rounded; somite 3 with middorsal elongate raised area, but not forming true carina; somites 4-6 dorsally rounded. Pleura 1- 3 ventrally rounded; pleuron 4 with small posteroventral tooth; pleuron 5 pos- teroventrally produced into acute spine; somite 6% length of telson. Telson (Fig. 3b) with broad shallow middorsal groove becoming obsolete pos- VOLUME 98, NUMBER 1 289 Aloeceln — 8 ef 2x5 ARES = Fig. 2. Heterocarpus laevis: a, Mandible; b, Mandibular palp; c, Maxilla 1; d, Maxilla 2; e, Antenna 1 peduncle; f, Antennal scaphocerite; g, Maxilliped 1; h, Maxilliped 2. Scales in mm. teriorly; 5 pairs dorsal spinules; posterior margin acutely triangular, with 2 pairs elongate spines internal to last of dorsal spinules. Eyes with spherical cornea much broader than stalk; ocellus lacking. Antenna 1, (Fig. 2e) stylocerite elongate-slender, reaching at least to midlength of external margin of 3rd peduncle article. Antenna 2, scaphocerite blade (Fig. 2f) just over- reaching distolateral spine; distal margin broadly rounded. Mouthparts as figured. 240 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. Heterocarpus laevis: a, Abdominal somite 3 in dorsal view; b, Telson and left uropod; c, Male, pleopod 1; d, Male, pleopod 2; e, Appendix masculina and appendix interna, further enlarged; f, Chela of pereopod 1; g, Longer pereopod 2; h, Shorter pereopod 2. Scales in mm. Mandibular palp, (Fig. 2a, b) basal article lobed at inner distal angle; 2nd article broad, about *% length of 3rd article. Maxilliped 2, (Fig. 2h) terminal article of endopod twice wider than long, set obliquely on penultimate article. Maxilliped 3 overreaching scaphocerite by half of terminal article; latter armed with rows of short spines, tipped with small corneous spine. VOLUME 98, NUMBER 1 241 Pereopods 1—4 with strap-shaped epipods. Pereopod 1 reaching just beyond scaphocerite; dactylus (Fig. 3f) less than 1 mm in length, set obliquely on propodal apex, with single apical spine; propodus basally broader than distally, with rows of setae on proximomesial surface, 74 length of carpus; latter only slightly shorter than merus; ischium with row of 8 or 9 short spines on posterodistal margin. Pereopod 2, shorter leg reaching distal 74 of scaphocerite; chela subequal to carpus in length, fingers with entire cutting edges, subequal to palm in length; carpus of 6 or 7 articles, proximal and distal articles longer than intervening 4 or 5 subequal articles; ischium with carinate posterior margin; longer leg overreaching scapho- cerite by chela and distal '4 of carpus; latter of 20—23 articles, subequal in length to merus and ischium together. Pereopod 3 extending beyond scaphocerite by length of dactylus and propodus; dactylus 4 length of propodus, with 5 spines on posterior margin; propodus bearing short scattered spines on posterior surface; carpus with strong spine near laterodistal margin, another just distal to midlength, 14-16 short irregularly spaced spines on mesial and posterior surfaces; merus with 8 or 9 large lateral spines, 7 large mesial spines; ischium with 2 spines on posterior surface. Pereopod 4 overreaching scaphocerite by dactylus and % of propodus; dactylus '4 length of propodus; carpus with 1 or 2 strong posterolateral spines, several small scattered spines on posteromesial surface; merus with 8 spines on lateral surface, 7 on mesial surface; ischium with 2 spines on posterior surface. Pereopod 5 overreaching scaphocerite by dactylus and *% of propodus; dactylus ¥% length of propodus; carpus with single lateral spine at about midlength; merus with 8 posterolateral spines, and single mediodistal spine; ischium unarmed. Pleopod 1 in male, (Fig. 3c) endopod about '2 length of exopod, distally broad- ened, with low mesial lobe bearing hooks; distal and lateral margins setose. Pleo- pod 2 (Fig. 3d) in male, appendix interna and appendix masculina subequal in length; appendix interna with distal triangular area bearing hooks; appendix mas- culina bearing about 25 spines of varying lengths on distal and lateral margins. Outer ramus of uropod (Fig. 3b) with groove on outer surface proximally well defined, becoming obsolete distally; distolateral fixed spine at distal 34; both rami just falling short of telsonic apex. Remarks. —Several differences between Milne Edwards’ figure (1883) and the Virgin Islands material are noted: The stylocerite is shorter in the original figure, than in the present material; no spines are shown on the merus of pereopod 5; the uropodal rami overreach the telsonic apex by a short distance; the scaphocerite is shown as being parallel-sided for most of its length. In spite of these differences, the overall closeness in the general form, and the agreement in the number of articles in the carpi of the second pereopods, make the identity of the present material unmistakable. Heterocarpus laevis is unusual in being the only species in the genus lacking lateral carapace carinae. The present material was captured along with several specimens of Plesionika edwardsii and Plesionika sp. Acknowledgments Funds for the Virgin Islands Fishery Development and Demonstration Project were provided by the National Marine Fisheries Service under the Saltonstall- Kennedy Act. 242 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON We are grateful to Dr. F. A. Chace, Jr., Smithsonian Institution, for reading and commenting on a draft of the manuscript. Literature Cited Faxon, W. 1896. Reports on the results of dredging, under the supervision of Alexander Agassiz, in the Gulf of Mexico and the Caribbean Sea, and on the east coast of the United States 1877 to 1880, by the U.S. Coast Survey Steamer “Blake,” Lieut.-Commander C. D. Sigsbee, U.S.N., and Commander J. R. Bartlett, U.S.N., commanding. 37. Supplementary notes on the Crus- tacea.— Bulletin of the Museum of Comparative Zoology at Harvard College 30(3):153-166. Man, J. G. de. 1920. The Decapoda of the Siboga Expedition. Part 4. Families Pasiphaeidae, Stylodactylidae, Hoplophoridae, Nematocarcinidae, Thalassocarididae, Pandalidae, Psalido- podidae, Gnathophyllidae, Processidae, Glyphocrangonidae, and Crangonidae.—Siboga-Ex- peditie, monographie 39a3:1-318. Milne Edwards, A. 1883. Recueil de figures de Crustacés nouveaux ou peu connus.—Paris, April 1883. 44 Pls. (BK) Department of Invertebrate Zoology, National Museum of Natural His- tory, Smithsonian Institution, Washington, D.C. 20560; (WT) Department of Conservation and Cultural Affairs, Division of Fish and Wildlife, Box 1878, Frederiksted, St. Croix, U.S. Virgin Islands 00840. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 243-254 LUCAYARINA CATACUMBA, NEW GENUS, NEW SPECIES, A BAHAMIAN SEA-CAVE AMPHIPOD (CRUSTACEA: AMPHIPODA: LYSIANASSIDAE) Janice Clark and J. L. Barnard Abstract.—A new genus of lysianassid amphipod, analogue of the Indo-Pacific Glycerina, is described from 2 sea-caves (blue holes) in the Bahamas Islands. Lucayarina differs fromG/lycerina in having article 2 of pereopods 5-7 indentured (in contrast to pereopod 5 only in Glycerina), the short article 3 and absence of dactylar shroud on gnathopod 1. Legend Capital letters denote main parts in following list; lower case letters to left of capital letters or in body of figure indicate modifications as per following list; lower case letters to right of capital letters indicate specimens described in captions: A, peduncle; B, body; C, coxa; D, dactyl; E, male duct; G, gnathopod; H, head; I, inner plate or ramus; J, prebuccal; K, male pores sternite 7; L, labium; M, mandible; N, pereon; O, outer plate or ramus; P, pereopod; R, uropod; S, max- illiped; T, telson; V, palp; W, pleon; X, maxilla; Y, gill; Z, pleopod; a, aberrant; f, flat; 1, left; o, opposite; r, right; s, setae removed. Family Lysianassidae Lucayarina, new genus Type-species. —Lucayarina catacumba, new species. Diagnosis.—Head deeper than long, rostrum short, ocular lobe produced and gently rounded, lower antennal sinus weak. Eyes bilateral. Antenna | slender, rather long; first article of peduncle slightly inflated, lacking tooth, longer than 2 and 3 combined; first flagellar article elongate, in female half, in male as long as article 1 of peduncle, with dense cover of aesthetascs in trans- verse rows on lateral surface; calceoli absent in material at hand. Accessory fla- gellum well developed, with 5 or 6 articles, article 1 as long as 2 and 3 combined. Antenna 2 with article 4 slightly longer than 5, flagellum shorter than peduncle. Epistome slightly concave, distinctly separated from evenly rounded upper lip. Mandible with rakers but no lacinia mobilis on either side, incisor with only one weak inner hump, remaining edge straight and smooth; molar prominent, slightly rounded, triturative, lacking major seta; palp moderately strong, attached very slightly proximal to base of molar, article 1 short, article 2 setose distally, article 3 shorter than 2, weakly falciform, with D and E setae. Lower lip with simple, broadly rounded outer lobes, inner lobes absent. Maxilla 1 with well developed inner plate bearing 1-2 major apical setae; outer plate with oblique apical margin bearing 11 denticulated spines weakly divided into 3 sets of 5, 5, plus 1 weakly discontiguous medial subapical spine; palp 2- articulate, almost reaching apex of outer plate, article 2 with 4 spine-teeth and several midapical serrations. Maxilla 2 with inner plate slightly shorter, more 244 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON pointed and much narrower than outer plate, apical armaments stout, inner plate with setae on apicomedial margin. Maxillipeds with inner plate subrectangular, bearing 3 short apical spines and plumose setae apically, with plumose setae medially; outer plate squamiform, with 2 rows of spines ventrally, inner marginal row being shorter and truncated (with ragged, possibly worn ends); palp 4-articulate, article 2 reaching to apex of outer plate, dactyl long, unguiform, with outer long and inner short setules, short apical nail barely distinct. Coxae 2—4 of same length, coxa 1 shorter and weakly bent forward; coxa 4 with ordinary posterior excavation and strong ventral lobe. Oostegites and gills on legs 2-6; brood plates of female slender, with setae apically (generally setae rudimen- tary). Gills saclike, not pleated, with much smaller auxiliary lobe attached proxi- mally. Gnathopods ordinary; gnathopod | simple, subequal in length to gnathopod 2, article 3 short, article 5 slightly shorter than 6, hand spinose, dactyl ordinary, not shrouded in setae; gnathopod 2 minutely chelate (or parachelate). Pereopods 3— 4 article 6 with short spines posteriorly, locking spines large. Article 2 of pereopods 5-7 deeply serrate posteriorly; no pereopods prehensile. Pleopods normal, peduncles poorly setose, with rami of subequal length; pe- duncle with 2 mediodistal denticulate coupling-hooks. Epimeron 3 dominant, only epimeron 2 ventroapical corner somewhat sharply produced. Uropods | and 2 well developed, styliform, no ramus with special notch; uropod 3 biramous, peduncle box-like, both rami broadly lanceolate and shorter than peduncle, outer ramus 2-articulate, inner ramus shorter than outer. Telson slightly longer than broad, deeply cleft. Description.— Antennal article 3 with lateral-distal row of spines. All coxae moderately setose. Only gill 6 subordinate (crenulation distally). Relationship.— This genus closely resembles G/lycerina Haswell, 1882, (=Glyc- era Haswell, 1879) which has two known species, the type-species G. tenuicornis (Haswell, 1879), (? = G. affinis Chilton, 1885, said to be Amaryllis macrophthal- mus by Stebbing, 1910), and G. teretis Andres, 1981. Another taxon, Lysianassa woodmasoni Giles, 1890, was made a synonym of G. tenuicornis by Pirlot (1936). If this composition is accepted, G. tenuicornis would have a distribution from southeastern Australia to India and Indonesia in 0-141 m; (G. affinis occurs southeastern Australia in shallow water), and G. teretis from the Red Sea in 1869 m. One must note that no modern record of Glycerina from India has been published and G/ycerina from Australia has never been properly described, and therefore identifications in Indonesia and India are suspect. Despite this, the known characters of Glycerina are adequate to distinguish our new genus. Lucayarina differs from Glycerina in the short article 3 and the absence of the inner dactylar setal shroud on gnathopod | and the deep tooth-like serrations of article 2 on pereopods 6-7; these teeth are present in Glycerina only on pereopod 5. If we can take Pirlot’s (1936) and Andres’ (1981) accounts as typically descrip- tive of Glycerina (note that the Australian specimens have not been described properly) then Lucayarina also differs from Glycerina in the unpleated gills, the lack of special notch on the inner ramus of uropod 2, the strongly setose article 2 of the mandibular palp, the thinner outer plate of the maxilliped, the contiguous and apical placement of the spines on the outer plate of maxilla 1, and the absence VOLUME 98, NUMBER 1 245 of a tooth on article 1 of antenna |. Dr. Andres believes the short article 3 of gnathopod 1 prevents any close affinity to be ascribed between Glycerina and Lucayrina: Dr. Lowry notes the different kind of spines present on the outer plate of maxilla 1 compared with Glycerina, differences in the mandibular molar, and strong differences in the setal shroud of gnathopod 1. J. L. Barnard (1969) did not notice the weak similarity of Glycerina to the scopelocheirid genera typified by Scopelocheirus Bate, although he did note this for a similar genus [chnopus Costa. Scopelocheirids are typified by a shroud of propodal and/or dactylar setae hiding the dactyl of gnathopod 1; in Jchnopus and Glycerina these setae are weakly developed on the posterior margin of the dactyl. In Lucayarina these setae are absent or very weakly represented (no more so than in non-scopelocheirid members of Lysianassidae). [chnopus and Glycerina closely resemble each other in this character but Lucayarina lacks these setae. We leave the problem of the significance of this character and its higher taxonomic value to our colleagues now studying the internal divisions of Lysianassidae. Lucayarina otherwise differs from Jchnopus in the unpleated gills, the presence of deep teeth on article 2 of pereopods 5-7 (although the serrations of Ichnopus pelagicus Schellenberg, 1926, are said to be slightly enlarged), in the lack of a tooth on article 1 of antenna 1, the contiguous and apically placed spines on the outer plate of maxilla 1, the presence of rakers on both left and right mandibles (which needs study in Jchnopus), and the non-attenuate, non-hooked apex of the mandibular palp. Dr. Andres believes that the dense medial setation on the inner plate of maxilla 1 prevents any close relationship to be ascribed between Lucayarina and, for example, Scopelocheirus. . Lucayarina differs from Aroui Chevreux (1911) in the presence of pereopodal teeth, the poorly setose inner plate of maxilla 1, the non-paddle-shaped plates of maxilla 2, and the unpleated gills. Dr. Andres believes the slightly elongate article 3 on gnathopod 1 and the presence of a lacinia mobilis in Aroui prevent any affinity being ascribed between Aroui and Lucayarina. Lucayarina differs from Menigratopsis Dahl (see Just 1976 for careful rede- scription) in the toothed pereopods, the absence of left lacinia mobilis, the well armed article 6 of pereopods 3-7, with locking spines, and the diverse extension of plates on maxilla 2; characters also noted by Drs. Andres and Lowry include the triturative molar, lack of calceoli, lack of accessory lobes on gills, and ar- mament differences on palp of maxilla 1 and outer plate of maxilliped. Etymology.—Word combination of Lucaya from the main tribe of Arawak Indians inhabiting the Bahamas before being exported and extirpated; and part of Glycerina; feminine. Lucayarina catacumba, new species Figs. 1-5 Description of female ‘“‘a”’ 4.35 mm.— Head about 60% as long as wide, rostrum about 20% as long as remainder of head; ocular lobes mammilliform; eyes long, oval, capsule absent, pigment absent in alcohol. Antenna 1 as long as antenna 2; second and third articles short, article 3 with 8 medium and 1 long aesthetascs, primary flagellum slender, longer than peduncle, with 11 articles, article 1 almost 1.4 times as long as article 3 of peduncle, formula 246 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON LL—_——__ ~— Fig. 1. ; r = female “‘r.”’ VOLUME 98, NUMBER 1 247 Fig. 2. Lucayarina catacumba, all figures = female “‘a.”” Views of outer plate of maxilla 1 based on both sides of appendage from various aspects. 248 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ; m = male “ VOLUME 98, NUMBER 1 249 Fig. 4. Lucayarina catacumba, all figures = female “a.” 250 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ct ee ees Par 4 Fig. 5. Lucayarina catacumba, unattributed figures = female “a”; m = male ““m”’; p = female “‘p.” VOLUME 98, NUMBER 1 251 of long aesthetascs = 9-2-2-2-1-0-0-0-0-0-0, some aesthetascs on article 1 emerg- ing from telescoped part, accessory flagellum 5-articulate. Article 3 of antenna 2 with 3 ventral setules; flagellum with 8—9 articles (left and right respectively). Upper lip and epistome both protuberant and separated by line of articulation. Mandible with raker row composed of 3 strongly curved spines; incisor with protrusion on lateral corner and notch on medial corner; palp article 2 moderately thin with mediodistal row of 6 setae; article 3 much shorter than 2, distally tapering, with comblike row of 4 D setae on medial edge and 4 apical E setae. Inner plate of maxilla 1 moderately long, thin, apex rounded with 2 apical setae and one setule; outer plate with 11 apical spines, 5 set facially and 1 offset medially; article 1 of palp % length of article 2, palp broad, long, apex of article 2 with 4 lateral spine teeth and 8 medial cusps. Inner plate of maxilla 2 thinner and shorter than outer, with 3 thick medial setae, 3 thin medial setae, 3 apical setae (2 plain, 1 toothed) and 1 facial toothed seta (R = 4 medial setae and one facial bifid seta). Maxilliped with inner plate reaching just beyond middle of outer plate, with 3 short stout apical spines; outer plate with row of short blunt spines on medial margin, plus ventrofacial row of short pointed spines and 3 basomedial longer setae; lateral face with series of ‘ghost’ setae (2 uneven rows = 11 setae), palp article 2 longer than others, article 3 with facial pubescence towards apex, dactyl digitiform, *4 as long as article 3, faintly curved, with stout nail inserted apically and with 3 long outer and 2 short inner accessory setae. Coxae 2—4 extending subequally; coxa 1 quadrate, ventral margin with 3 setules, one setule set in notch at apicoventral corner, 2 (right side = 1) posterior facial setae and one facial seta toward anterodorsal corner; coxa 2 rectangular, expanding distalwards, anterior facial margin with 3 setae (right with 2), ventrally with 6 setae, one seta set in notch at apicoventral corner, one seta set facially at posterior margin; coxa 3 similar to 2, but apicoventral corner produced, 3 (right side = 2) anterior margin setae set facially, 4 along ventral margin, one setule set in notch at apicoventral corner, one posterior facial marginal seta, one setule set facially in anterodorsal corner; coxa 4 produced posterodistally, excavation above lobe large, anterior margin with 5 (R = 4) facial setae, ventral with 8, 5 facial setae scattered on produced posteroventral lobe. Oostegites with rudimentary apical setae. Gnathopod | article 6 tapering distally, with short spines and medium setae on posterior margin; dactyl with 3 subapical setules; gnathopod 2 stouter than 1, article 5 nearly twice as long as 6, subquadrate, almost twice as long as broad. Pereopods 3 and 4 article 2 slender, as long as 3, 4 and 5 combined; article 4 shorter than 6, distally wider than 5, with anterodistal apex greatly produced; dactyl strong, curved, with weak nail and 2 setules; locking spines weakly S- shaped. Pereopod 5 shorter than 6 and 7; pereopod 6 slightly smaller than 7; article 2 produced posterodistally; article 4 broader than 5; article 6 longer than 3 and 4 combined. Each epimeron with one setule on posterior edge. Each pleopod with pair of locking spines, articles of outer and inner rami on pleopods 1-3 = 11 and 9, 11 and 9, 11 and 10, basal setal formulas on rami (running from lateral to medial on each ramus) = 6-2-1-2, 4-1-1-3, 5-1-1-2. Urosomite 3 with winglike plaque surrounding base of telson on each side, with 1 or 2 small spines at the corners of joint between telson and urosomite. Uropod 252 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 1 rami of equal length and as long as outer ramus of uropod 2; rami slightly shorter than peduncle, outer with 3, inner with 2 marginal spines. Uropod 2 rami of subequal length, both longer than peduncle, outer with 3, inner with 2 marginal spines; [left uropod 2 aberrant, see Fig. 5]. Uropod 3 (aberrant in female “‘a,”’ see Fig. 5, following description from normal female) peduncle with 2-3 spines, rami lanceolate, outer ramus not as long as peduncle, article 2 spiniform, about 70% as long as article 1, medioapical corner of article 1 sharply produced; inner ramus slightly exceeding apex of article 1 of outer ramus; only armament of rami formed by one seta at medial base of inner ramus. Telson with slightly convex lateral margins, tapering towards notched apices, each lobe bearing short terminal spine and setule set in notch, and pair of dorsal penicillate setules near midlateral margins. Male ““m’’—3.87 mm. (Figs. 1, 3, 5): Antenna 1 flagellum with 8 articles, accessory flagellum with 6; formula of aesthetascs = (5 groups), 2,2,2,2,0,0,0. Uropod 3 peduncle with 2 dorsal spines (right with 3); pereonite 7 with 2 sternal pores and conspicuous duct running from base of coxa 7 through body to per- eonite 5. Female “r’’ 3.71 mm (Andros): Description based only on differences from holotype; parentheses show comparison to holotype when item is more, otherwise all statements concern “fewer”: Unlike holotype, eyes with ommatidia (facets) mostly solid, not divided, inner ends not atrophied nor vermiform. Antenna | article 3 with 7 medium and | long aesthetascs; primary flagellum with 9 articles; formula of long aesthetascs = 8-2-2-2-2-2-1-1-0-1-0 (more than holotype). Antenna 2 article 4 with 1 dorsal seta, article 5 with 1 ventral seta (more than holotype); flagellum on both right and left sides with 8 articles (similar to holotype). Mandible palp article 2 with mediodistal row of 5 setae; article 3 with comblike row of 3 D setae on medial edge. Maxilliped outer plate ventrofacial row with 4 short, pointed spines and 2 basomedial longer setae; lateral face with 14 “‘ghost”’ setae. Coxa 2 lacking facial seta toward anterodorsal corner; coxa 3 with 3 setae along ventral margin and no facial setae on posterior margin or anterodorsal corner; coxa 4 with 3 facial setae on anterior margin and 2 facial setae on posteroventral lobe. Gnathopod | article 6 with 2 anterior setae. Pereopod 3 articles 4 and 5 posterior margins each with | long seta; pereopod 4 articles 4 and 5 posterior margins each with 2 long setae; article 6 anterior margin naked, posterior margin with 3 medium setae and 1 short spine; pereopod 5 article 4 anterior margin naked; pereopod 6 article 6 posterior margin with 2 setae, anterior margin with 3 pairs of short spines and a single proximal spine. Articles of outer and inner rami on pleopods 1-3 = 10 and 9, 9 and 8, 9 and 8. Uropod 2 peduncle with 3 dorsomedial spines (more than holotype). Holotype female “n’’ 3.90 mm.—Uropods 1-3 normal. Female “p” 4.0 mm (Fig. 5). Epimera 2 and 3 with single setule on posterior edge. Remarks.—The Andros material thus appears to differ only slightly from the Grand Bahama material in the eyes and variations in spines and setae. Holotype.—USNM 195132, female ‘“‘n’”’ 3.90 mm. VOLUME 98, NUMBER 1 253 Type-locality.—Grand Bahama Island, Cemetery Cave (a blue hole offshore 30+ m), 25 Oct 1982, 15 m depth, “red amphipods,” on lobster carcass, coll. Thomas M. Iliffe and Dennis Williams. Other material.—Type-locality, female “a’’ 4.35 mm (main illustrations), fe- male ‘“‘g” 5.37 mm (whole mount), male “‘m”’ 3.87 mm (illustrated), female “‘p”’ 4.0 mm (illustrated), female “‘q’” 4.12 mm, male “‘s” 3.84 mm, female “t” 3.12 mm, and 250 other specimens, no apparent terminal males present; Andros Island, British Blue Holes Expedition, 1981-82, blue hole no. 31, from lobster carcass 100 m horizontal locus, depth of 15 m, Dr. George F. Warner, University of Reading, England, female “‘r’’ 3.71 mm (compared fully to holotype, head illus- trated), and 30 other specimens. Etymology.—Catacumba from the Latin for “underground burial gallery” or ‘tad catacumbas” “at the canyon” “at which one of the longest catacombs is situated”’; feminine. Distribution.—Sea caves in the Bahamas Islands. 99 66 Acknowledgments We thank the collectors of the material and especially Jill Yager, of Grand Bahama Island, who sent us the Cemetary Cave material and Dr. Thomas M. lliffe, of Bermuda Biological Station, for information about the localities. Dr. J. K. Lowry of the Australian Museum, Dr. Jean Just of the Zoological Museum, University of Copenhagen, and Dr. H. G. Andres of Hamburg University kindly offered comments on the manuscript. Linda B. Lutz of Mobile, Alabama, inked our drawings and prepared the plates for publication. We thank Patricia Bell Crowe for laboratory assistance. Literature Cited Andres, H. F. 1981. Lysianassidae aus dem Abyssal des Roten Meeres. Bearbeitung der Koderfange von FS “Sonne” —MESEDA I (1977) (Crustacea: Amphipoda: Gammaridea).—Senckenber- giana Biologia 61:429-443. Barnard, J. L. 1964. Deep-sea Amphipoda (Crustacea) collected by the R/V ““Vema”’ in the Eastern Pacific and the Caribbean and Mediterranean Seas.—Bulletin of the American Museum of Natural History 127:3—46, figs. 1-33. . 1969. The families and genera of marine gammaridean Amphipoda. — Bulletin of the United States National Museum 271:1-535, figs. 1-173. Chevreux, E., and L. Fage. 1925. Amphipodes.—Faune de France 9:1—488, figs. 1-438. Chilton, C. 1885. Notes on a few Australian Edriophthalmata.— Proceedings of the Linnean Society of New South Wales 9:1035-1044. Della Valle, A. 1893. Gammarini del Golfo di Napoli.—Fauna und Flora des Golfes von Neapel und der angrenzenden Meeres-Abschnitte, Monographie 20:i—xi + 1-948, atlas of 61 pls. Giles, G. M. 1890. Natural history notes from H.M.S. Indian marine survey steamer ‘Investigator’, Commander Alfred Carpenter, R.N., D.S.O., commanding. No. 15. Description of seven ad- ditional new Indian amphipods.—Journal of the Asiatic Society of Bengal 59:63-74. Haswell, W. A. 1879. On Australian Amphipoda.— Proceedings of the Linnean Society of New South Wales 4:245-279, pls. 7-12. . 1880. Preliminary report on the Australian Amphipoda.— Annals and Magazine of Natural History (5)5:30-34. 1882. Catalogue of the Australian stalk- and sessile-eyed Crustacea.— Australian Museum, Sydney: i-xxiv, 327 pp., 4 pls. Just, J. 1976. On the marine genus Menigratopsis Dahl, 1945 from North Atlantic and Arctic waters (Crustacea, Amphipoda, Lysianssidae).— Astarte 9:1—12, figs. 1-9. 254 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Pirlot, J. M. 1936. Les Amphipodes de expedition du Siboga, deuxiéme partie: Les Amphipodes gammarides, I: Les Amphipodes fouisseurs: Phoxocephalidae, Oedicerotidae. —Siboga-Expe- ditie, Monographie 33b:57-1 13, figs. 12-34. Sars, G.O. 1895. Amphipoda. An account of the Crustacea of Norway with short descriptions and figures of all the species 1, i—viui, 711 pp., 240 pls., 8 suppl. pls. Schellenberg, A. 1926. Amphipoda 3: Die Gammariden der Deutschen Tiefsee-Expedition. — Wis- senschaften Ergebnisse Deutschen Tiefsee-Expedition “Valdivia” 1898-1899 23:195—243, 28 figs., pl. 5. Stebbing, T. R. R. 1906. Amphipoda, I. Gammaridea.—Das Tierreich 21:1-806. . 1910. Crustacea. Part 5. Amphipoda. Scientific results of the trawling expedition of H.M.C:S. Thetis. —Memoirs of the Australian Museum 4(2):565-658, pls. 47-60. NHB-163, Division of Crustacea, Department of Invertebrate Zoology, Smith- sonian Institution, Washington, D.C., 20560. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 255-280 NEW SPECIES OF ISOPODA FROM THE FLORIDA MIDDLEGROUNDS (CRUSTACEA: PERACARIDA) Allan Hooker Abstract. —Six new species of isopods, five asellotes and one anthurid, are de- scribed from the Florida Middlegrounds in the northeastern Gulf of Mexico. One is described as a new genus, viz. Mexicope kensleyi. The other five species de- scribed are: Pleurocope floridensis, Munnogonium wilsoni, Janira biunguicula, Gnathostenetrioides pugio, and Mesanthura hopkinsi. The specimens were all collected from artificial habitats of mostly man-made materials. Isopod crustaceans may be an important part of the macro-epifauna and macro- infauna of various marine habitats. The state of our knowledge of these links in the food-web of the macro- and megafauna of the western Atlantic, Caribbean, and Gulf of Mexico waters is generally limited to faunal checklists and inventories which usually deal with specific localities, e.g., Menzies and Frankenberg 1966, Hudson et al. 1970, Rouse 1970. There are also several regional reports and monographic accounts. Of these, the most comprehensive is that from Puerto Rico (Menzies and Glynn 1968). As part of a survey of the biota of the Florida Middlegrounds, members of the Dauphin Island Sea Lab participated in an inventory of the area’s epibenthic fauna between June 1978 and January 1981. A large majority of the diminutive asellote isopods of this study were collected in artificial habitats placed on and retrieved from a hermatypic coral reef at a depth of approximately 30 meters. Habitat placement and retrieval was accomplished by means of SCUBA. The anthurid species was found in vacuum samples collected by the submersible research vessel Diaphus. The six new isopods are described and illustrated. Where practical, artificial dichotomous keys are presented. Materials and Methods All the specimens of this study were collected from the Florida Middlegrounds at 28°35’N, 84°16’W (see Fig. 1). A component of the West Florida shelf, 150 Km south of the north Florida coast and 160 km northwest of Tampa Bay, the Middlegrounds are influenced by the Caribbean-derived Loop Current, Florida Bay waters, and the West Florida Estuarine Gyre (Austin and Jones 1974). The area is characterised by coral reefs, projecting discontinuously from the bottom, not exceeding depths of 55 m (Jordan 1952). An abundant invertebrate community is supported by the area’s reefs (Hopkins et al. 1977) among which are this study’s isopods. Nearly all the new species were obtained from artificial habitat cryptofaunal samples; however, selected sponges (especially Age/as dispar) and corals (especially Madracis decactis) played host to several of the collected specimens. Agelas dispar was found to host up to ten species of isopods, while Madracis decactis up to 14 256 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON "& ange sTETSON SONNIER BANK cid FLORIDA y, ~ 5 e ~ WEST \ _EWING BANK” MIDDLEGROUND 28° FLOWER GARDEN, .-AS! FLOWER GARDEN : f att BRIGHT BANK : @ 26 AM) aS @ Wo . 0 0 Ss a sé { é DRY TORTUGAS S-- EUORIDATKENS pal Wg : ~"" ALACRAN REEF C See , NUEVQ -- «CAYO ARENAS F271 \\ dvanauita REEF REEF -. BANCO INGLESES ISLANDS S y: ‘O >, : AND REEFS ISLAS DE LOBOS , : a: TRIANGULOS § pet e@ REEFS _ OBISPO SHOALS s ye @ SUPPOSED AREA 20° 2 ARCAS CAYS.~ “ ~ COZUMEL OF SCATTERED CORAL HEADS _ENMEDIO REEF <: VERACRUZ REEFS ~ oe 600 GRAND 40077.” ““ MISTERIOSA CAYMAN BANK 98° 96° 94° 92" 90° 88° 86 84 g2° 80° Fig. 1. The Gulf of Mexico, showing the Florida Middlegrounds and other known reef locations (after Bright and Pequegnat 1974). species (see Table 1). The unexpectedly large numbers of specimens found in the artificial habitats may be due to the extra effort expended in picking and sorting as against the same effort involved in picking and sorting naturally occurring samples. Attempting to maximize internal heterogeneity, artificial habitats contained predetermined quantities of PVC tubing of two diameters, oyster shell, plastic straws, trawl netting, plastic swizzle sticks, fibreglass insulation, and plastic meshed “‘scrubbies,” enclosed in vinyl-clad hardware cloth (1 cm x 1 cm mesh), struc- turally strengthened externally by a plastic milk case. Figure 2 is a cross-sectional view of such an artificial habitat. Habitats were placed on the reef’s rubble, and anchored by railroad iron sections. Habitat retrieval occurred at regular intervals corresponding to scientific cruises to the Middlegrounds. Retrieval consisted of enclosing the habitats in cloth bags, floating them to the surface by means of lift bags, and surface recovery. Both habitat placement and retrieval were accomplished by means of SCUBA. On retrieval, intact individual habitats were fixed in 10% formalin, and trans- ported to the laboratory. All the habitat substrates were rinsed over a 0.5 mm mesh sieve, the resulting animals and rubble stored in 70% anhydrous alcohol. The reef rubble found in nearly all the habitats was probably a result of habitat settling and storm-induced disturbance. No attempt was made to determine which of the internal microhabitat substrates was favored by the asellotes. The single VOLUME 98, NUMBER 1 257 Table 1.—Distinguishing characters of Pleurocopidae, Mexicope, Abyssianiridae, Janira, and J. biunguiculus. Explanation of abbreviations: a, absent; am, ambulatory; b, biramous; c, covered; dil, dilated; e, expanded; ep, exposed; 1, lacking; m, modified; n, non-expanded; nr, normal; ns, non- stalked; p, present; ps, parallel-sided; s, stalked; sag, sagittate; su, subchelate; u, uniramous; v, varied; ?, unknown or uncertain. Pleuro- Kupho- Abyssi- Mexicope cope Prethura Santia munna anira Janira Janira (unas- (Pleuro- (Pleuro- (Pleuro- (Pleuro- (Abyssi- (Janiri- _ biunguic- signed) copidae) copidae) copidae) copidae) aniridae) dae) ulus Ant. I peduncular segments D; 2 2; 1,2,3 ) 4(?) 3,4 1 Ant. I flagellar articles 7 4 3 3,4,5 3 2(?) Vv 5 Ant. I aesthetascs p p,1 l 1-3 1 qv 0,2(?) 2) Ant. II peduncular segments 6 6 6 Vv 5 B 6 ? Ant. II flagellar articles 13 6,7 8 Vv 15 ? Vv ? Ant. II aesthetascs 0 1 0 0 0 2 0 ? Ant. II peduncular scale p 1 1 1 ] l p p Length of Ant. I vs Ant. II II>I I>II II>I II>I II>I ? II>I I>) Maxillipedal palp seg. 1-3 n n n n n e e e Maxilla I b u b b b b b b Mandibular palp ] 1 1 p,1 l p p p Molar process m m nr nr nr nr nr nr Eyes p p p p p a p p Eye location s s S s s - ns ns Uropods b b m b,u b b b b Pereopod I am su su su su su am,su am Claws on Prp. II-VII 2 1 yD) D} 2) 2 3,2(?) 2 Epimera (dorsally) 1-7 2-3,5-6 0 0 5-7 1-7 =1-7,2-7 1-7 Pleonal segments 2 12 2, 1,2 1 2) 2 2 Male Pip. I distally ps ps ps ps dil sag Vv ps Male Plp. II nr nr e nr nr nr nr nr Anus Cc ep ? ep ep Cc 2 ep species of anthurid was found in bottom vacuum samples provided by a minisub- mersible research vessel. Systematics Infraorder Asellota Family Pleurocopidae Pleurocope floridensis, new species Figs. 3, 4 Description.—Body broad, ovate, nearly twice longer than wide; broadest at pereonites 3 and 4. Dorsum covered with numerous minute tubercles. Cephalon about 12 times wider than long, widest at eyestalk origin. Eyestalks projecting laterally nearly to spine of second pereonite, with 3 light-brown ommatidia; preoc- ular lobes absent. Frontal margin slightly convex. Coxal plates dorsally visible on pereonites 2-3 and 4—7. Pereonites 2—4 subequal in length; slightly longer than subequal pereonites 1, 5—7; pereonites 5—7 narrower than preceding segments; coxal plates of pereonites 2, 3, 5 and 6 each with single lateral spine, tipped with 2 elongate simple setae. Pleotelson twice longer than broad, slightly raised at uropodal insertion, posteriorly tapering, elongate; 2 spines on each ventrolateral 258 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Plastic Coated Retaining Wire (O.5"' x 0.5" mesh size) 4 PVC Pipes (6" x 1") 12 PVC Pipes (6" x .5!") 1O Chore Ready Cleaning Puffs (plastic) 0.000400 Boodooooccccc000d | Plastic Mesh Bundle, G ‘ 100 each: Straws (7.75"' x .2' and Stirrers (5" x .13!' Pa O-0-0-O70-0-0-0-0 Used Trawl Net (600 g. ) ist Whole Oyster Shells (10.5 Ibs.) Milk Crate bi ‘ Meiofauna Habitat Fiberglass Insulation (3"' x 6" x 12") Fig. 2. Schematic cross-sectional view of artificial habitat utilized in the Florida Middlegrounds. margin; 10-12 elongate articulate setae arising from each distal ventrolateral margin, several extending beyond pleotelsonic apex. No anterior pleonite visible. Antennular peduncle 2-segmented, first segment slightly longer than second; flagellum of 4 articles, third longest, bearing 2 terminal aesthetascs; fourth article with | aesthetasc and elongate seta. Antenna reaching to midpoint of antennular flagellum, with 6-segmented peduncle, third segment strongly produced mesially, forming lobe tipped with row of 5 robust setae; fifth and sixth segments subequal in length, together equal to previous 4 segments together; flagellum half length of peduncle, of 6 articles, terminal article bearing 2 setae and | apical aesthetasc. Mandible lacking palp; rudimentary lacinia mobilis on left mandible; molar process highly modified, conical, directed distally, incisor also modified, an elon- gate extension with several clefts. Inner lobe of maxilla 1 rudimentary; outer lobe with 8 or 9 spines. Maxilla 2 biramous, inner ramus supporting 4 terminal setae; both lobes of outer ramus tipped with 2 setae. Maxilliped with narrow 5-segmented palp; epipodite distally rounded; broad endite with 2 coupling hooks, and setules fringing entire concave distal margin. VOLUME 98, NUMBER 1 259 Fig. 3. Pleurocope floridensis: A, Holotype ¢, dorsal view; B, Antenna 2; C, Antenna 1; D, Pereopod 2; E, Pereopod 1; F, Pereopod 7; G, Maxilliped. Pereopod 1 prehensile, dactyl with 2 elongate spines distally and lanceolate process on proximal inner margin; propod nearly oval; merus and carpus shortest segments, subequal in length. Dactyl of pereopod 2 with single claw, bearing setules on penultimate and ultimate articles. 260 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON D 1 ( | 1 1 \ \ \ Fig. 4. Pleurocope floridensis: A, Right mandible; B, Left mandible; C, Maxilla 1; D, Maxilla 2; E, Operculum 9; F, Pleopods 4 and 5 6; G, Uropod; H, Pleopod 1 4; I, Pleopod 2 4; J, Pleopod 3 3. Uropods pedunculate, inserted dorsolaterally; peduncle expanded distally; lengths of peduncle, exopod, and endopod subequal. Rami of pleopod 1 fused proximally, elongate, widest proximally, with single spine on outer distal margin, apex bearing several setae. Pleopod 2 with long, slender endopod; exopod tapering distally, tipped with elongate spine. Endopod of pleopod 3 slightly longer than exopod, bearing 2 simple setae apically; outer branch tipped with 3 plumose setae. VOLUME 98, NUMBER 1 261 Female: Body broader than male. Sexual dimorphism not apparent in gnatho- pod. Brood pouch of 3 pairs of oostegites. Material.— Holotype male TL 1.15 mm, United States National Museum, cat- alog number 184943. Allotype female TL 0.96 mm, USNM 184944. Paratypes deposited in Dauphin Island Sea Lab Museum, catalog number 6183-1507. Etymology.—The specific epithet floridensis refers to the locality where the asellote was collected. Remarks. —See Remarks section for Mexicope kensleyi. Key to the Species of Pleurocope 1. Dorsum with long setae; pleon of two segments; uropodal rami longer (HIDE, FOCUS IS SOM act ain ate Seem aan Delp oi ieee Pen aR Ah ee P. dasyura — Dorsum without long setae; pleon of one segment; uropodal peduncle as I(SPDES GAGS STEN Ca, ake See ea a re A Ba eee Ace Rr a floridensis Family Incertae Sedis Mexicope, n. gen. Diagnosis.—Eyes present; frontal margin of head straight; antennulae normal; scale present on antennal peduncle. Pleon 2-segmented, second segment as wide as long, with lateral serrations. Epimera spiniform, present on pereonites 1-7. Mandibular palp lacking; molar process conical, setiferous; maxillipedal palp 5- segmented, all segments subequally wide. Pereopod 1 ambulatory; all pereopodal dactyli biunguiculate. Uropoda elongate, biramous. Male pleopod 1 distally bi- lobed, both lobes setiferous. Type-species.— Mexicope kensleyi, n. sp. Etmyology.—The generic epithet Mexicope is a combination of the prefix Mexi-, indicating the Gulf of Mexico, and the sufhix -cope derived from Pleurocopidae, the family to which Mexicope shows most similarities. Mexicope kensleyi, new species Figs. 5, 6 Description.— Female: Body nearly 3 times longer than wide, widest at pereo- nites 3 and 4. Head extending laterally, as wide as epimera of pereonite 1; preocular spines present; eyes dorsal, black, borne on small lobes. Pereonites 1-7 with spiniform coxae; coxae of pereonites 2—7 with recurved setae. Pleon with simple setae laterally and fringed setae apically; pleonite 1 free. Antennule extending to proximal end of antennal peduncle segment 6; distal segment of 2-jointed peduncle with several plumose setae; flagellum of 7 articles, article 5—7 each with single terminal aesthetasc. Antenna heavily setose; peduncle 6-segmented, third segment with seta-bearing squama, fifth segment slightly longer than preceding 4 segments together; flagellum slightly longer than peduncle, first article longest, remaining 12 articles subequally long, together about twice length of first article. Mandible lacking palp, with conical setiferous molar process; left mandible with quadridentate lacinia, quadridentate incisor molar process, row of 1 simple and 6 dentate spines. Maxilla 1 inner ramus short, outer ramus with 11 spines. Maxilla 262 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 5. Mexicope kensleyi: A, Holotype 2, dorsal view; B, Left mandible; C, Right mandible; D, Antenna 1; E, Maxilla 1; F, Maxilliped; G, Maxilla 2; H, Antenna 2; I, Pereopod 1; J, Pereopod 2; K, Pereopod 7. 2, inner ramus supporting 4 spines distally, inner lobe of outer ramus with 2 dentate spines and 2 fringed setae terminally, outer lobe with 4 fringed setae. Maxilliped with 5-segmented palp, segments subequally wide, terminal segment VOLUME 98, NUMBER 1 263 <= = <2 PPD?) \ = lly, MMLEL LLL Wr pedissss,, MMe leble LILLE, \ Aly fi M iN f G Z Zz CS iN 4 ie ey ree Wai astro EF Fig. 6. Mexicope kensleyi: A, Pleopod 1 6; B, Pleopod 2 4; C, Pleopod 3 4; D, Pleopod 4 6; E, Pleopod 5 4; F, Pleopod 1, immature 6; G, Pleopod 2, immature 46; H, Operculum 8; I, Uropod. bearing several setae distally. Endite broad, outer margin rounded; inner margin with 2 coupling hooks prominent tooth distally; bearing numerous setules on distal third. Epipodite small, narrowly rounded apically. Pereopod | ambulatory; subequally long basis and propodus longest segments; merus bulbous; dactyl with 2 claws. Pereopod 2 with biunguiculate dactyl, carpus, and propodus equally long. Pereopod 7 longest, similar to pereopod 2. 264 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Uropoda elongate, setiferous, biramous, peduncle and exopod subequal in length, endopod slightly longer. Operculum as broad as long; distal half of lateral margin somewhat excavate, proximal half bearing numerous narrow setae; 12 prominent feather setae along distal subacute margin. Male: Pleopod 1 rami fused proximally, bearing several short setae and single elongate seta laterally; bilobed distally, outer lobe with 3 setae and several setules terminally; inner lobe bearing several setules apically and 2 elongate setae me- dially. Pleopod 2 sympod with alternating long and short setae along outer margin, 6 plumose setae on outer distal third; endopod coiled distally, with flared stylet, exceeding length of sympod. Material.— Holotype female TL 2.95 mm, USNM 184940. Allotype male TL 1.66 mm, USNM 184941. Paratypes DISL Museum 6183-1506. Etymology.—The specific name kens/eyi is for Brian Kensley, of the National Museum of Natural History (Smithsonian Institution), for the assistance and guidance he has given me. Remarks (Pleurocope floridensis and Mexicope kensleyi).—Consistently chal- lenging to taxonomists, the genus Pleurocope has undergone five different family assignments since its inception by Walker in 1901. Describing P. dasyura from the Mediterranean Sea, Walker (1901) placed his genus in the family Munnidae. Wolff (1962:64) reassigned Pleurocope to the family Dendrotionidae with hesi- tancy and noted its exceptional characters in the family diagnosis. Facilitated by several fresh specimens of P. dasyura, Fresi and Schieke (1972) re-evaluated Pleurocope, and felt it necessary to erect a new family, Pleurocopidae. Wilson (1980) argued that the cephalic appendages Fresi and Schieke (1972) cite as basis for the erection of Pleurocopidae are “. . . specializations not of familial signifi- cance...’ and placed Pleurocope in the family Antiadidae. Finally, recent efforts (Kensley, 1982) to rectify nomenclatural and priority problems will place the genus in the family Pleurocopidae. After comparing Pleurocope and the other pleurocopid genera, Kuphomunna Barnard, 1914, Santia Sivertsen and Holthuis, 1980 (=Antias Richardson, 1906), and Prethura Kensley, 1982, the heterogeneity of this family becomes immediately apparent. Kensley (1982), citing the diverse nature of the family’s constituent genera, is admittedly hesitant in grouping this foursome, which he believes need to be studied in greater detail in order to establish taxonomic relationships. Pleurocope floridensis can be distinguished from P. dasyura Walker by its single segmented pleon, short dorsal setae, and subequally long uropodal peduncle and rami. A third member of the genus is presently being described by Wilson and Carter (Wilson, pers. comm.). P. floridensis differs from it by possessing only five robust setae along the mesial margin of the second antenna’s third segment. The new monotypic genus, Mexicope, also offers taxonomic difficulties. Its unique combination of diagnostic attributes does not fit readily into any existing family. Both Wilson (pers. comm.) and Kensley (pers. comm.) feel Mexicope is most closely related to the family Pleurocopidae, even though it shares several characters with the Janiridae. The new genus, however, differs significantly from its pleurocopid allies by possessing peduncular scales on its antennae, a conical setiferous molar process, and an ambulatory first pereopod. Mexicope can also be distinguished from the janirids by its nonexpanded maxillipedal palp segments VOLUME 98, NUMBER 1 265 1-3, absence of a mandibular palp, modified molar process, and stalked eyes. Table 2 compares the distinguishing characters of Mexicope, the genera of Pleu- rocopidae, and the genus Janira. Because of this overlapping of familial features and because of the present vagueness of family and generic diagnoses of the Janiridae and Pleurocopidae, Mexicope is not assigned a familial position. Family Paramunnidae Munnogonium wilsoni, new species Figs. 7, 8 Description. — Male: Body pear-shaped, widest at pereonite 3, nearly twice longer than wide. Frontal margin of head broadly rounded; black eyes on short lateral projections; coxal plates dorsally visible on pereonites 2—7. Pleotelson as long as broad, laterally dentate to point of uropodal insertion, posteriorly rounded. Antennular peduncle 2-segmented, second segment expanded distally; first and fourth articles of flagellum subequally long, latter bearing single elongate aesthetasc and several setae apically. Third segment of antennal peduncle elongate, with seta-bearing scale; sixth peduncular segment slightly longer than fifth segment, with several setae distally. Flagellum of 6 or 7 segments, each article slightly shorter than the preceding one. Mandibles with strong, truncate molar process; palp reduced to glabrous scale; left mandible with 4-toothed incisor, lacinia mobilis trifid, 4 elongate simple setae; incisor of right mandible 5-toothed, spine row of 2 serrate setae and 3 simple setae. Inner ramus of maxilla | with 1 plumose and 3 elongate simple setae; 8 terminal spines on outer ramus, single spine arising from body of outer ramus. Maxilla 2, inner ramus broad, bearing several elongate simple setae and 1 feather seta along distal margin, with several rows of setules of distal half; inner lobe of outer ramus with 3 elongate simple setae terminally, outer lobe carrying | feather seta and 2 elongate simple setae. Maxillipedal epipodite extending to first palp segment; palp 5-segmented, joints 1-3 expanded; endite setose along distal margin, 2 coupling hooks along inner margin. Pereopod 1 subchelate; dactyl with terminal and accessory claws; carpus with 2 prominent sensory spines inferiorly; basis longest article, subequal in length to ischium and merus together. Single claw on dactyl of pereopod uni-unguiculate; basis and ischium subequally long. Uropodal protopod not visible dorsally; endopod twice longer than exopod, both rami apically setose. Pleopod 1 sagittate, lateral corners cleft, supporting setae of varying lengths. Pelopod 2 laterally fringed with several elongate setae; endopod distally thin, elongate. Female: Slightly broader than male; operculum distally narrowing to rounded apex, bearing several slender setae along convex lateral margins. Ovigerous para- type with 9 eggs in brood pouch. Material.— Holotype male, TL 0.86 mm, USNM 184946. Allotype female, TL 0.98 mm, USNM 184947. Paratypes, DISL Museum 6183-1509. Etymology.—The specific name is in honor of George Wilson, University of 266 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 7. Munnogonium wilsoni: A, Holotype 4, dorsal view; B, Antenna 1; C, Antenna 2; D, Pereopod 1; E, Pereopod 2; F, Pereopod 7. California at San Diego, who has recently helped to rectify paramunnid-munnid inconsistencies, and has helped me immeasurably. Remarks.—A new genus and species, Munnogonium waldronense, was insti- tuted by George and Stromberg (1968) after collecting what was thought to be a previously undescribed asellote from the San Juan Archipelago, Washington. Although justified in their erection of a new genus, the asellote was not, in fact, a new species. Menzies and Barnard (1959) had described the same species, placing VOLUME 98, NUMBER 1 267 Fig. 8. Munnogonium wilsoni: A, Left mandible; B, Right mandible; C, Maxilla 1; D, Maxilla De E, Maxilliped; F, Uropod; G, Operculum 2; H, Pleopod 1 4; I, Pleopod 2 6; J, Pleopod 3 8; K, Pleopod 46. it in the genus Austrosignum Hodgson, 1910, naming it A. tillerae. Discovering the conspecificity of 4. waldronense and A. tillerae, Bowman and Schultz (1974) reviewed Munnogonium, comparing it to Munna Kroyer, 1838, Pleurogonium Sars, 1864, and Austrosignum. They found Munnogonium most closely related 268 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON to Austrosignum, but differing from its generic allies in the absence of a mandibular palp, *“‘. .. character of generic value . . .”, (Bowman and Schultz 1974). By trans- ferring A. tillerae (and all other previously described Austrosignum species lacking mandibular palps) to Munnogonium and applying the law of priority, the type- species became M. tillerae (Menzies and Barnard, 1959). Based on the position of the anus, shape of the male first pleopod, and pro- portions of the third antennal segment, Wilson (1980) transferred Munnogonium and several other genera from the family Munnidae to the resurrected family Pleurogonidae Wilson, 1980. Holthuis (pers. comm.), however, correctly indicates that the law of priority dictates the family name Paramunnidae Vanhoffen, 1914, takes precedence over Pleurogoniidae Nordenstam, 1933, if the genera Pleuro- gonium Sars, 1899, and Paramunna Sars, 1866, are considered to belong to the same family. Munnogonium wilsoni can be distinguished from all its congeners by the pres- ence of coxal plates on pereonites 2—7, and the unique combination of antennal peduncular scales and short eyestalks. Key to the Species of Munnogonium 1. Eyes not visible dorsally, lacking ocular peduncles ................... Ae Aten BES ie yee PPC ac 8 iene ed Or ae ae M. erratum Schultz, 1964 — Eyes visible dorsally, borne on short or long peduncles ............... D 2. Pereonites 3 and 4 laterally notched ............ M. subtilis Kensley, 1976* — Rereonitesss and: 4 notiaterallysnotched. es... 4.5 eee 3 32) Lateral pleonalsmarpingnon=semrate!: .255-5 054. 4005 940200400. ae 4 =) Lateralépleonal manompsenmatemen.o..4....50.... 560.4005) eee 5 4. Eyes borne on elongate peduncles; antennule of 6 articles ............. rok i ote aenanss Meeeags 0 7 * ect ek ry M. grande Hodgson, 1910 — Eyes bore on short stalks; antennule of 7 articles 3..............558e et ane ere, Pane xa 8 Ghose OS Ay ee M. globifrons Menzies, 1962 5. Antenna without peduncular scale WM. tillerae (Menzies and Barnard, 1959) =) Antenna with: peduncular scale wh... 5 oc. ook oh ake oe Cee 6 6. Eyes borne on elongate peduncles; male antenna with peduncular segment SESUDO Wala SWOllena ee era M. maltinii Schiecke and Fresi, 1972 — Eyes borne on short stalks; coxal plates dorsally visible on pereonites DEST FRG BOT, NORA TT Oe ie Ei aan ck LOS SL eee De ret eh, M. wilsoni, n. sp. Family Janiridae Janira biunguicula, new species Figs. 9, 10 Description.—Body about 3 times longer than wide, widest at pereonite 4. Cephalon wider than long, projecting strongly at anterolateral corners; prominent frontal margin slightly concave distally. Eyes of 3 ommatidia, brown, dorsolateral. Pereonites 1—4 bearing bilobed coxal plates; pereonites 5—7 with single epimera; lateral margins of pereonites 1-3 directed anteriorly. Pleon 2-segmented, pleonite 1 inconspicuous; pleotelson shield-shaped, as wide as long. * Based on single ovigerous female. VOLUME 98, NUMBER 1 269 Fig. 9. Janira biunguicula: A, Holotype 6, dorsal view; B, Antenna 1; C, Maxilliped; D, Maxilla 2; E, Maxilla 1; F, Right mandible; G, Left mandible. Antennule with 2-segmented peduncle; flagellum of 4 articles, terminal article elongate, with 2 aesthetascs. Antenna broken at scale-bearing peduncular segment. Mandibular palp 3-segmented, second segment bearing 2 prominent forked setae; strong molar process carrying 2 setae; scale present near base of molar 270 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 10. Janira biunguicula: A, Pleopod 1 6; B, Pleopod 2 4; C, Pleopod 3 6; D, Pleopod 4 4; E, Pleopod 5 4; F, Pereopod 1; G, Pereopod 2; H, Operculum 2. process; 6 dentate spines in spine row; lacinia mobilis of left mandible with 5 teeth incisor 5-toothed; incisor of right mandible with 6 teeth. Outer ramus of maxilla 1 with 6 plumose setae, 1 simple and 5 dentate spines on distal margin; inner ramus bearing several simple setae. Maxilla 2, inner ramus bearing numerous setae; inner lobe of outer ramus with elongate setae, outer lobe VOLUME 98, NUMBER 1 271 tipped with 4 elongate setae. Maxillipedal epipodite broad, distally subtriangular, apically narrowly rounded; palp 5-segmented, segments 1-3, and especially seg- ment 2 wider than segments 4 and 5; endite bearing several fringed setae-spines distally; 2 retinaculae on inner margin. Pereopod | ambulatory; dactyl shortest segment, biunguiculate; basis, propod, and carpus subequally long. Pereopod 2 with 2 claws on dactyl; carpus longest segment, slightly longer than propod; pereopod 7 shortest, dactyl biunguiculate. Pleopod 1, rami elongate, tapering for proximal two-thirds of length, distal margin of rami slightly concave, with 4 or 5 stout setae; outer margin of pleopod 2 rounded, sympodal apex rounded, slightly longer than basally robust endopod. Endopod of pleopod 3 bearing 3 strong feather setae terminally; exopodal outer margin setose, spine present distally; exopod of pleopod 4 with prominent feather seta. Uropod pedunculate, biramous; endopod twice length of exopod, nearly twice as wide. Female: Broader than male; operculum nearly circular, with 2 setae distally. Material.— Holotype male, TL 1.19 mm, USNM 184937. Allotype female, TL 1.08 mm, USNM 184938. Paratypes, DISL Museum 6183-1508. Etymology.—The specific name biunguicula refers to the two dactylar claws present on all seven pereopods. Remarks.—No other asellote family has demonstrated the heterogeneity of genera nor repeatedly presented taxonomic problems as has the Janiridae. Dis- regarding possible synonymies, Wolff (1962) listed 35 genera, and when combined with more recent generic designations, the number stands close to 40. Many of these genera are monotypic, casting doubt on their validity, in regard to characters traditionally considered generically diagnostic. Two major problems have con- tributed to the present taxonomic tangle. First, the family diagnosis is very broad, often serving as a catch-all for non-related genera. The disparate nature of janirid genera is exemplified by juxtaposing Caecianiropsis Menzies and Pettit, 1956, _ Vermectias Silvertsen and Holthuis, 1980, Katianira Hansen, 1916, Carpias Rich- ardson, 1902, and Neojaera Nordenstam, 1933. Second, as Wilson (pers. comm.) suggests, no one really knows what a janirid is. Only three recurring morphological traits have been found among the janirids: uropods with a peduncle, a scale present on the antennal peduncle (not true for /ais Bovallius, 1887), and expanded max- illipedal palp segments 1-3. The problems of systematics within the Janiridae are brought to light by Janira biunguicula, which falls into the same systematic category as many other newly described janirids, while all related genera (or perhaps, more appropriately, ‘forms’) can be precluded by their generic diagnoses. Janira’s most closely related genera, Taniropsis Sars, 1899, and Janiralata Menzies, 1951, can both be discounted by their distally dilate first male pleopods, /aniropsis by its elongate, prehensile first male pereopod, and Janiralata by its subchelate first pereopod in both sexes. The dorsal display and arrangement of epimera, and the distally non-dilate male pleopod 1 of J. biunguicula agree with the Janira diagnosis; however, the type-genus, Janira maculosa Leach, 1814, differs from J. biunguicula in two important aspects: J. maculosa possesses triunguiculate dactyli, whereas J. biun- guicula has biunguiculate dactyli, and pereopod 1 of J. maculosa is prehensile, while that of J. biunguicula is ambulatory. Ay) 24 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ae xp \, ie Fig. 11. Gnathostenetrioides pugio: A, Holotype 6, dorsal view; B, Antenna 1; C, Antenna 2; D, Left mandible °; E, Right mandible, immature 6; F, Left mandible, mature 6; G, Pleotelson and uropods. VOLUME 98, NUMBER 1 273 These differences may characterize a new genus, but J. biunguicula has been placed in Janira for two reasons. First for the aforementioned characters it shares with the generic diagnosis, and second, the Janiridae by sheer number of dissimilar genera, should not be burdened by yet another genus that may not be valid and which would only serve further to confuse the situation. Table 2 compares the distinguishing features of J. maculosa and J. biunguicula. That the Janiridae require an intensive study to resolve the present complexity is obvious. Schultz (1976) suggests that the number of dactylar claws may be of generic or higher significance. Unfortunately, not all janirid descriptions contain claw number information. Wolff (1962) has found J. alta Stimpson, 1853, with three claws rather than two as mentioned by Stimpson, and suspects both J. japonica Richardson, 1908, and J. tristani Beddard, 1886, are also triunguiculate, contrary to their original descriptions. . Superfamily Parastenetrioidea Family Parastenetriidae Gnathostenetrioides pugio, new species Figs. 11-13 Description.— Male: Body elongate, parallel-sided. Cephalon length and width nearly equal; frontal margin projecting between antennulae, concave anteriorly; mandibular prolongation extending to midpoint of antennular flagellum; eyes dorsal, of 5 light-brown ommatidia. Pereonite | longest; pereonites 2—4 with seta- bearing coxae; pereonites 5—7 slightly produced posterolaterally. Pleotelson as wide as long, with lateral spines anterior to uropodal insertion, free pleonal seg- ment lacking. Peduncle of antennule 3-segmented, first segment broadest, as long as following 2 segments together; flagellum of 2 articles, 2 subequally long aesthetascs on terminal article, aesthetascs nearly as long as flagellum. Antennal peduncle of 6 segments, third segment bearing setiferous squama, segment 5 as long as segments 1—4 together, segment 6 longest; flagellum of 18 setose articles. Mandibular palp 3-segmented, second segment with 2 fringed setae and 4 shorter simple setae, terminal segment with serrate seta distally; armed with elongate, slightly crenulate projection distolaterally, bearing truncate setiferous molar pro- cess; incisors with cusps; lacinia mobilis of left mandible bearing 4 teeth and single serrate seta. Outer ramus of maxilla | bearing 10 serrate spines distally; inner ramus slender, carrying several elongate simple setae terminally. Second maxilla, outer lobe of outer ramus with 4 elongate simple setae and 1 short simple seta. Maxillipedal endite narrow, with 3 coupling hooks, mediodistal margins setose; palp segment 2 widest, expanded distally; epipodite extending to third palp, api- cally subacute. Merus of pereopod | with seta-bearing projecting corner; propod, merus, and carpus setiferous posteriorly; propod with seta-bearing tooth and several combed, curved spines along palmar edge; dactyl bearing several pectinate curved spines and short terminal claw. Pereopod 2 with subequally long merus and carpus; dactyl shortest joint, bearing 2 claws. Basis of pereopod 7 nearly oval, longest joint; dactyl shortest segment, biunguiculate. Pleopod 1 large; sympods nearly 3 times broader than long; rami separate, non- 274 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON _~ NO LAN SS “ Ns oN be ON Be 7774 Fig. 12. Gnathostenetrioides pugio: A, Maxilla 1; B, Maxilla 2; C, Maxilliped; D, Pereopod 1 6; E, Pereopod 2; F, Pereopod 7. overlapping, bearing numerous setae distally. Endopod of pleopod 2 2-segmented, second joint truncate distally, slightly expanded medially; 2-segmented, second joint truncate distally, slightly expanded medially; 2-segmented exopod, pleopod 3 exopod of 2 joints, endopod with 2 plumose setae terminally. 276 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Uropod pedunculate; inner ramus longer and broader than outer ramus, both carrying numerous elongate setae terminally. Female: Similar to male, fewer setae on inner margin of pereopod 1 merus, carpus, and propodus; operculum with mesiodistal margin acutely cleft. Material.— Holotype male TL 3.20 mm, USNM 184934. Allotype female TL 1.60 mm, USNM 184935. Etymology.—The Latin pugio refers to the dagger-like process on the male mandible. Remarks. — Describing a new asellote from the Mediterranean Sea, Amar (1957) found it necessary to designate a fourth subtribe, Parastenetrioidea, a new family, Parastenetriidae, and a new genus, Guathostenetrioides to accommodate the un- usual features of the new species G. /aodicense. Amar felt the pleopodal structure, his basis for erecting a new subtribe, to be intermediate in form between the pleopodal morpohlogy of the subtribes Stenetrioidea Hansen, 1905, and Para- selloidea Hansen, 1905. Based upon features other than pleopods, Gnathostene- trioides is clearly more closely related to Stenetrioidea than to Paraselloidea. The parastenetrioideans and stenetrioideans have probably descended from a common ancestor with pleopodal differences evolving into the present dichotomy. Besides diagnostic pleopods, Gnathostenetrioides also bears anomalous man- dibular prolongations; an elongate, tusk-like protuberance in the male; a shorter, acute process in the female. The use of these remarkable processes is unknown, but they may be involved in mating behavior, which could account for their marked sexual dimorphism. Unlike the stenetrioideans, the two parastenetrioidean species do not demon- strate marked sexual dimorphism in their subchelate first pereopod. Male para- stenetrioidean gnathopods are, however, more hirsute on the inner margins of the merus, carpus, and propodus than in their female counterparts. Only the second recorded species of the subtribe, G. pugio differs from its congener, G. /aodicense by the absence of setae along the lateral margin of pleopod 1, an inner margin length of pleopod 1 less than twice the length of the palmar margin, and the female’s acute opercular mesiodistal incision. The close similarity between the two species is not totally surprising; the related stenetrioidean genus, Stenetrium Haswell, 1881, also demonstrates conservatism with regard to inter- specific differences among its approximately 40 members. Key to the species of Gnathostenetrioides 1. Pereopod 1 propodus bearing several setae on outer lateral margin, inner margin greater than twice the length of the palmar margin; female oper- Culummincision rounded 35. 5555555 558550550565 G. laodicense Amar, 1957 — Pereopod 1 propodus lacking outer lateral margin setae, inner margin less than twice the length of the palmar margin; female operculum distal in- CISIONMTACULC oe yd ue oe ce ds dM Ona s RAMS OP ce Re ee pugio, Nn. sp. Family Anthuridae Mesanthura hopkinsi, new species Figs. 14, 15 Description. — Female: Body proportions: C = 1 < 2>3<4=5 > 6 > 7. Pig- mentation pattern of cephalon dorsum triangular, originating anteriorly between VOLUME 98, NUMBER 1 277 Fig. 14. Mesanthura hopkinsi: A, Holotype 2, dorsal view; B, Antenna 2; C, Antenna 1; D, Max- illiped; E, Maxilla, apex enlarged; F, Maxilla; G, Mandible. 278 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 15. Mesanthura hopkinsi: A, Pereopod 1; B, Pereopod 2; C, Pereopod 7. black eyes. Pereonites 1-6 with non-pigmented central area surrounded by pig- ment in varying near-circular patterns; pereonite 7 with a U-shaped pigment. Telson with chromatophores, bearing 4 elongate setae distally. Pleonites 1-5 fused. Antennular flagellum 2-jointed, terminal article with several setae and 2 aes- thetascs; peduncle of 3 segments, second segment shortest. Antennal peduncle 5-segmented, segments 1 and 2 wider than long, flagellum uniarticulate, bearing numerous apical setae. Mandibular palp 3-segmented, setae-bearing distal segment shortest; second segment longest, carrying elongate setae distally; molar of 1 lobe; incisor of 3 cusps; lamina dentata with 5 serrations. Maxilla slender, with 8 spines distally. Maxilliped 5-segmented, terminal segment set obliquely on penultimate segment, with 2 simple setae, 2 fringed setae/spines and | plumose seta/spine. Unguis of pereopod 1 slightly longer than dactyl; dactyl with spine-bearing ventrolateral lobe, broadly rounded proximally; carpus rounded distally; merus with convex outer margin. Dactyl of pereopod 2 slightly longer than unguis, with strong ventrodistal spine; propodus with distal seta-bearing tooth; carpus trian- gular, with 3 simple setae and row of setules on inner margin. Carpus of pereopods VOLUME 98, NUMBER 1 279 4-7 with anterior margin shorter than posterior margin. Propod of pereopod 7 with row of robust setae along posterior margin, longest article slightly longer than the basis. Uropodal exopod heavily setose medially, notched apically, lacking pigment; endopod pigmented, rounded distal margin heavily setose, inner margin serrate. Male: Unknown. Material.— Holotype female TL 2.43 mm, USNM 184949. Etymology.—Iin honor of Thomas S. Hopkins, Dauphin Island Sea Lab, who provided all the specimens of this study, and for his patience and understanding. Remarks.—Species of the anthurid genus Mesanthura Barnard, 1914, are by definition, distinguished by the persistent chromatophoric pigmentation patterns displayed dorsally. The species-specific patterns vary widely in chromatophoric density, arrangement and location. One species, M. protei Kensley, 1980, (Kensley and Poore 1982) exhibits polychromatism, with three documented patterns and possibly more. Multiple intraspecific patterns would appear to be the exception since no other polychromatic members of this genus are known. Mesanthura paucidens Menzies and Glynn, 1968, and M. pulchra Barnard, 1925, are also found in the waters off Florida. Mesanthura hopkinsi can be dis- tinguished from these congeners by its pigmentation pattern, particularly that of the cephalon and pleon. Mesanthura hopkinsi’s cephalic pigment is triangular, whereas M. paucidens’s is a transverse band located just behind the eyes, and M. pulchra is a wide scattering nearly covering the entire head. Mesanthura hopkinsi’s pleonal pigmentation consists of three, laterally independent transverse bands on the first three pleonal segments, and a fourth, wide, transverse band on pleonal segments four and five, whereas M. pauciden’s is five, laterally connected, transverse bands and M. pulchra’s is a nondescript, broad scattering of chromato- phores. Acknowledgments I would like to express thanks to members of my graduate committee, Dr. Thomas Hopkins, University of Alabama; Dr. Brian Kensley, National Museum of Natural History; and Dr. Joseph Scheiring, University of Alabama, for all their guidance, assistance, and time. George Wilson, University of California at San Diego, provided expertise, insights, and encouragement. Richard Heard, Gulf Coast Research Laboratory, and David Camp, Florida Department of Natural Resources, also assisted in the taxonomic organization of the manuscript. Dawn Dardeau has my thanks and praise for her expert sorting. Finally, the entire faculty, staff, support personnel and graduate student body at Dauphin Island Sea Lab warrant thanks in appreciation for all their efforts. Funding was provided by BLM Contract AA550-CT7-34 to Dr. Thomas Hopkins. Literature Cited Amar, R. 1957. Gnathostenetrioides laodicense nov. gen., noy. sp. type nouveau d’Asellota et clas- sification des Isopodes Asellotes.— Bulletin de l’Institut Océanographique 1100:1-10. Austin, H. W., and M. L. Jones. 1974. Seasonal variations of physical oceanographic parameters of the Florida Middlegrounds and their relation to zooplankton biomass on the West Florida Shelf. —Florida Scientist 37:5-16. Bowman, T. E., and G. A. Schultz. 1974. The isopod crustacean genus Munnogonium George and 280 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Stromberg, 1968 (Munnidae, Asellota).— Proceedings of the Biological Society of Washington 87:265—272. Bright, T. J., and L. H. Pequegnat. 1974. Biota of the West Florida Garden Bank. Gulf Publications Co.: Houston, Texas. 453 pp. Fresi, E., and U. Schiecke. 1972. Pleurocope dasyura Walker, 1901 and the Pleurocopidae, new family (Isopoda, Asellota).—Crustaceana, supplement 3:207-213. George, R. Y., and J. O. Stromberg. 1968. Some new species and new records of marine isopods from San Juan Archipelago, Washington, U.S.A.—Crustaceana 14:226—230. Hopkins, T. S., D. R. Blizzard, S. A. Brawley, S. A. Erie, D. E. Grimm, D. K. Gilbert, P. G. Johnson, E. H. Livingston, C. H. Lutz, J. K. Shaw, and B. B. Shaw. 1977. A preliminary characterization of the biotic components of composite strip transects on the Florida Middlegrounds, North- eastern Gulf of Mexico. University of Miami, Proceedings of the Third International Coral Reef Symposium, pp. 31-37. Hudson, J. H., D. M. Allen, and J. Costello. 1970. The flora and fauna of a basin in Central Florida Bay.—U.S. Fish and Wildlife Service, Special Scientific Report 64, 14 pp. Jordan, G. F. 1952. Reef formation in the Gulf of Mexico off Apalachicola Bay, Florida.— Bulletin of the Geological Society of America 63:741-744. Kensley, B. 1982. Prethura hutchingsae, new genus, new species, an asellote isopod from the Great Barrier Reef, Australia (Crustacea: Isopoda: Pleurocopidae).— Journal of Crustacean Biology 2: 255-260. , and G. C. B. Poore. 1982. Anthurids from the Houtman Abrolhos Islands, Western Australia (Crustacea: Isopoda: Anthuridae).— Proceedings of the Biological Society of Washington 95: 625-636. Menzies, R. J., and J. L. Barnard. 1959. Marine Isopoda on coastal shelf bottoms of Southern California: systematics and ecology.— Pacific Naturalist 1:1—35. , and D. Frankenberg. 1966. Handbook on the Common Marine Isopod Crustacea of Georgia. University of Georgia Press: Athens, Georgia. 93 pp., 27 figs., 4 pls. , and R. W. Glynn. 1968. The common marine isopod Crustacea of Puerto Rico.—Studies on the Fauna of Curacao and Other Caribbean Islands 27:1-33. Rouse, W. L. 1969 (1970). Littoral crustaceans from southwest Florida.— Quarterly Journal of the Florida Academy of Science 32:127-152. Schultz, G. A. 1976. Species of Asellotes (Isopoda: Paraselloidea) from Anvers Island, Antarctica. — Antarctic Research Series 26:1-—35. Walker, A. O. 1901. Contributions to the malacostracan fauna of the Mediterranean.—Journal of the Linnean Society of London (Zoology) 28:290-307. Wilson, G.D. 1980. New insights into the colonization of the deep sea: systematics and zoogeography of the Munnidae and the Pleurocopidae comb. nov. (Isopoda, Janiridae). —Journal of Natural History 14:215-236. Wolff, T. 1962. The systematics and biology of bathyal and abyssal Isopoda Asellota.—Galathea Report 6:1-320. Department of Environmental Regulation, Southwest District, 7601 Highway 301 North, Tampa, Florida 33610. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 281-287 CRYPTOPENAEUS CROSNIERI, A NEW SPECIES OF SHRIMP, AND A NEW RECORD OF C. SINENSITS ; (PENAEOIDEA: SOLENOCERIDAE) FROM AUSTRALIAN WATERS Isabel Pérez Farfante and Brian Kensley Abstract.—Cryptopenaeus crosnieri, a new species of solenocerid shrimp, is described from four specimens collected in Australian waters. It differs from C. catherinae in exhibiting neither a notch nor a depression on the postrostral carina, in having a longer scaphocerite, and in lacking an ischial spine on the third pereopod. Also, the thelycum exhibits a pair of small, flat to weakly convex plates, instead of strong bosses, on the anterior part of sternite XIII. Cryptopenaeus crosnieri differs from C. sinensis (originally described as the type and only species of the new genus Crassipenaeus), in lacking the three rounded thelycal promi- nences on sternite XIV typical of the latter, and in the presence of a meral spine on the third pereopod. The solenocerid genus Cryptopenaeus was proposed by De Freitas (1979) for C. catherinae, a new shrimp taken off southern Mozambique at depths between 310 and 500 m. Recently, three additional members of the genus have been discovered, one of them, C. sinensis (Liu and Zhenru, 1983) from the South China Sea, another from the waters of Indonesia, which is being described by A. Crosnier, and the third, described herein, from a locality off eastern Australia. Four spec- imens of the latter species were collected off the northeast coast of New South Wales by the R/V Kapala of the New South Wales Fisheries. The new record of C. sinensis is based on a male obtained off the northwest coast of Australia, southwest of Cape Leveque, by the U.S.S.R. R/V Lira. This latter species is the type-species of a new genus, Crassipenaeus Liu and Zhenru, 1983, which we consider to be a synonym of the earlier Cryptopenaeus De Freitas, 1979. The terminology used in the descriptions has been discussed and illustrated by Pérez Farfante (1969, 1977). Cryptopenaeus crosnieri, new species Figs. 1, 2 Material.—Holotype, 2, Australian Museum, AM-P32481, carapace length 29.5 mm, rostral length 8.5 mm, total length about 102 mm. Type-locality: NE of North Solitary Island, New South Wales, Australia, 29°47—49'S, 153°41’E, 234 m, R/V Kapala sta 78-05-07, shrimp trawl.—Paratypes, 1 2, Australian Museum, 2 2, National Museum of Natural History, Smithsonian Institution, Washington, D.C., USNM 189097, collected with holotype. Description. — Body relatively robust (Fig. 1); carapace microscopically setose- punctate, abdomen glabrous. Rostrum horizontal, with dorsal margin straight and ventral margin convex, short, its length 0.30 to 0.18 that of carapace, decreasing 282 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ee Fig. 1. Cryptopenaeus crosnieri, holotype 2 29.5 mm carapace length, NE of North Solitary Island, New South Wales, Australia. Lateral view. Scale = 10 mm. proportionally with increasing size, and in adults reaching between distalmost part of first antennular article and midlength of second. Rostral plus epigastric teeth 7-9, fifth tooth situated at level of orbital margin, last tooth distinctly posterior to apex. Adrostral carina strong, separated from teeth by conspicuous groove, dorsal border of groove covered with narrow band of densely set short setae. Postrostral carina well marked, long, almost reaching posterior margin of carapace and lacking notch or depression. Orbital spine absent; postorbital spine slender, sharp, and long; antennal spine minute; pterygostomian spine strong, with broad base; hepatic spine slender, sharp, and as long as postorbital. Bran- chiostegal and suprahepatic spines lacking. Gastro-orbital sulcus lacking; cervical sulcus, accompanied by raised, sharp carina, gently sinuous, its dorsal extremity ending distinctly ventral to postrostral carina at about midlength of carapace; hepatic sulcus rather deep, subhorizontal posteriorly, anterior part of sulcus ac- companied by sharp carina; branchiocardiac carina almost indistinguishable; sub- marginal carina well defined. Antennular peduncle about 0.5 as long as carapace. Prosartema narrow, acute, long, attaining proximal 0.4 length of second antennular article, and bearing long, densely set marginal setae. Stylocerite ending in sharp spine and extending 0.6- 0.7 distance between its proximal extremity and mesial base of distolateral spine; latter spine sharp, long, reaching as far as proximal 0.3 of second article. Anten- nular flagella subequal in length, long, 2.2 times carapace length in shrimp 29.5 mm cl and 1.4 in shrimp 50 mm cl; mesial flagellum slender and subcylindrical throughout its length, lateral flagellum moderately depressed in proximal half where about twice as broad as mesial, then becoming filiform. Scaphocerite ex- ceeding antennular peduncle by as much as 0.25 of its own length; lateral rib VOLUME 98, NUMBER 1 283 (Ld Fig. 2. Cryptopenaeus crosnieri, holotype: A, Telson and right uropod, dorsal view; B, Thelycum. Scales: 5 mm. ending in long spine reaching distal margin of lamella. Antennal flagellum incom- plete in all specimens examined. Mandibular palp reaching between apex of ischiocerite and midlength of car- pocerite. Third maxilliped exceeding antennular peduncle by at least length of dactyl and distal extremity, as much as 0.60 length, of propodus. First pereopod extending at least to apex of ischiocerite or as far as distal end of carpocerite. Second pereopod reaching distal end of antennular peduncle or exceeding it by 0.6 length of dactyl. Third pereopod extending to distal end of antennular peduncle or surpassing it by as much as propodus and 0.15 length of carpus. Fourth pereopod Overreaching carpocerite by at least distal extremity, or by 0.7 length of dactyl. Fifth pereopod surpassing antennular peduncle by dactyl and at least 0.3 length of propodus or by as much as dactyl and 0.70 length of propodus. Order of pereopods in terms of their maximum anterior extensions: first (shortest), fourth, second, third, and fifth. Third maxilliped falling short of third pereopod. First pereopod armed with elongate acute spine on basis and ischium, and small spine on merus; second pereopod armed with small but sharp spine on basis. Coxa of fifth pereopod in female mesially produced into short plate bearing sharp spine anteromesially. Abdomen either lacking or bearing weak dorsomedian carina along posterior half of second somite; third through sixth somites with well marked carina, sharp except on third; sixth somite armed with moderately strong tooth at posterior end of carina and small spines posteroventrally. Telson (Fig. 2A) bearing small pair of fixed, posterolateral spines; terminal part short, length about twice basal width; median sulcus moderately deep, extending to about midlength. Rami of uropod subequal in length, falling slightly short or barely surpassing apex of telson; lateral ramus with lateral ridge ending in short spine. Thelycum (Fig. 2B) with subrectangular plate of sternite XIV bearing pair of deep, longitudinal furrows flanked laterally by strong ridges joined anteriorly by 284 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON transverse one; area between furrows raised into low anteromedian protuberance, apical portion of which sometimes produced into minute spine. Pair of small setose anterior plates, abutting transverse ridge, flat or very slightly convex, and with mesial margins diverging anteriorly. Anterior part of sternite XIII with high median ridge produced anteriorly into strong, anteriorly directed blunt tooth. Posterior thoracic ridge produced anteriorly in paired, thick, biconvex processes separated by median depression, ridge strongly inflated posteriorly. Size.—Females 25—30 mm carapace length, about 91-103 mm total length. Geographic and bathymetric ranges.—This shrimp has been found off eastern Australia (Fig. 4), at a depth of 234 m. Discussion. — Cryptopenaeus crosnieri can readily be separated from C. cath- erinae by the following characteristics: the lack of a notch or a depression on the postrostral carina posterior to the level of the cervical sulcus; the longer scapho- cerite, which overreaches the antennular peduncle by as much as 0.25 of its own length instead of reaching or barely overreaching the peduncle; and the lack of an ischial spine on the third pereopod. Moreover, the thelycum of C. crosnieri bears a pair of flat or very slightly convex plates, with divergent mesial margins, on the anterior part of the sternite XIV, rather than a pair of strong, suboval bosses, as it does in C. catherinae. The thelycum of C. crosnieri also differs strikingly from that of C. sinensis by the armature of sternite XIV which in the latter species consists of one anteromedian and two lateral low, rounded promi- nences. It seems worth mentioning that in C. catherinae, as in C. crosnieri, a very minute spine is present on the merus of the third pereopod, a character that is not mentioned by De Freitas in the original description. This meral spine is absent from the male of C. sinensis from Australia. Also, as in C. crosnieri, sternite XIII in females of C. catherinae is armed with a median carina produced anteriorly in a strong tooth, a feature not cited by De Freitas. Etymology.—It is a pleasure to name this shrimp in honor of Alain Crosnier of O.R.S.T.O.M. and the Muséum National d’Histoire Naturelle, Paris, who has contributed greatly to our knowledge of decapod Crustacea. Cryptopenaeus sinensis (Liu and Zhenru, 1983) Fig. 3 Crassipenaeus sinensis Liu and Zhenru, 1983:171, fig. 1. Material.—1 4, Zoological Museum of the Moscow State University, off NW Australia, SW of Cape Leveque, 17°35'48”S, 119°33’00”E, 320-335 m, 28 May 1973, R/V Lira haul 63, coll. O. Petrov. Supplementary description of male genitalia.—Petasma (Fig. 3C, D) cincinnu- late along proximal half of dorsomedian lobe. Ventromedian lobule with mem- branous proximal part bordered by lateral rib, latter expanding and then merging distally with heavy sclerotized elongate plate; latter with subcircular terminal flap bearing marginal row of minute spinules and produced proximolaterally in con- cave, subcylindrical process; dorsal rim of concavity armed with five teeth (in only male available). Dorsolateral lobule supported by heavy rib extending its length and terminating in subacute projection at medial base of terminal part. Ventrolateral lobule heavily sclerotized, with ventral costa produced distally in VOLUME 98, NUMBER 1 285 D Fig. 3. Cryptopenaeus sinensis, 6 40 mm carapace length, off NW Australia: A, Ventral view of left appendices masculina and interna, and distolateral spur; B, Dorsolateral view of left appendix masculina and distolateral spur; C, Dorsal view of left half of petasma; D, Ventral view of same. Scales: A, B= 2 mm, C, D = 3 mm. angular projection, its apex curving towards subcylindrical process and bearing marginal row of very minute spinules. Appendix masculina (Fig. 3A, B) roughly trapezoidal, convex dorsally, excavate ventrally, with distal margin oblique and bearing short, stout setae; appendix interna subequal in length to appendix masculina, with lateral base expanded, 286 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 4) C. crosnieri Xf C. sinensis Fig. 4. Australian localities from which Cryptopenaeus crosnieri and C. sinensis were collected. and rounded distal margin armed distally with brush of stout setae. Basal sclerite with lateral wall produced distolaterally in foliaceous projection terminating in blunt spur. Discussion. —Liu and Zhenru (1983) proposed the new genus Crassipenaeus for their new species C. sinensis, recorded from 261 m in the South China Sea. Having compared the description of this genus with that of Cryptopenaeus De Freitas, 1979, it has become obvious that there are no significant differences between the two taxa to justify their separation. Consequently, we regard Crassipenaeus as a synonym of the earlier Cryptopenaeus. Acknowledgments We thank H. H. Hobbs, Jr., and F. A. Chace, Jr., of the Smithsonian Institution, and B. B. Collette, of the National Marine Fisheries Service, Systematics Labo- ratory, for reviewing the manuscript and giving helpful comments. We are es- pecially grateful to D. J. G. Griffin, Director of the Australian Museum, Sydney, for making the types of C. crosnieri available, and to B. G. Ivanov, of the All- Union Research Institute of Marine Fisheries and Oceanography, VNIRO, Mos- cow, U.S.S.R., for the loan of the male specimen of C. sinensis. VOLUME 98, NUMBER 1 287 Literature Cited De Freitas, A. J. 1979. A new genus and species of the penaeoid family Solenoceridae (Crustacea, Decapoda) from south-east African waters.—Annals of the South African Museum 77:123- 131. Liu, J. Y., and Z. Zhenru. 1983. On a new genus and two new species of solenocerid shrimps (Crustacea, Penaeoides) from South China Sea.—Chinese Journal of Oceanography and Lim- nology 1:171-176. Pérez Farfante, I. 1969. Western Atlantic shrimps of the genus Penaeus.—U.S. Fishery Bulletin 67: 461-591. . 1977. American solenocerid shrimps of the genera Hymenopenaeus, Haliporoides, Pleoticus, Hadropenaeus new genus, and Mesopenaeus new genus.— U.S. Fishery Bulletin 75:261-346. (IPF) National Marine Fisheries Service, Systematics Laboratory, National Mu- seum of Natural History, Smithsonian Institution, Washington, D.C. 20560; (BK) Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 98(1), 1985, pp. 288-292 THE FAUNA OF ATLANTIC MARINE CAVES: EVIDENCE OF DISPERSAL BY SEA FLOOR SPREADING WHILE MAINTAINING TIES TO DEEP WATERS C. W. Hart, Jr., R. B. Manning, and T. M. Iliffe Abstract.— Evidence is presented that significant ties exist between the faunas of marine caves and those of the deep seas, that marine cave faunas may contain very old elements, and that marine caves have served as faunal refuges over very long periods of time. In addition, the term “‘crevicular’’ is introduced to designate those aquatic habitats formed by crevasses in and among rocks, as well as to describe the organisms that live in those habitats. We feel that data recently gathered in Bermuda, the Bahamas, the Turks and Caicos, Ascension, and the Canary Islands (Fig. 1) shed light on some of the puzzling distributional patterns noted for cavernicolous crustaceans found on oceanic islands of the Atlantic and the Pacific, and bear generally on sea floor spreading. In our work we have been dealing exclusively with (and drawn our conclusions from) invertebrates— primarily shrimps—that inhabit anchialine waters and ma- rine caves. That is, waters having no surface connection with the sea, but which nevertheless contain salt or brackish water the level of which fluctuates with the tides. Sket and Iliffe (1980) summarized information then available on the fauna of the caves of Bermuda, and reported that pools in the caves were inhabited by a wide variety of marine invertebrates—ranging from ciliates to tunicates. In dis- cussing the zoogeographical affinities of the cave fauna of the island, they noted striking zoogeographical connections between Bermuda and the east, and com- mented on four theories that had been proposed to explain the distribution of marine cave organisms: 1) Plate tectonics, as stated by Sterrer (1973) for interstitial fauna, was discussed and partially rejected because Bermuda had never been a part of the con- tinental plates. 2) Stranding on the shoreline of receding fossil seas, as suggested by Stock (1977), was ruled out because Bermuda was too young to have experienced significant shoreline changes. 3) Connections with the abyssal depths, as suggested by Webb (in Sterrer 1973), was discounted because it was felt that changes in meiofauna with depth and substrate indicated that such interconnections were unlikely. Also, at that time no abyssal species had been identified from the caves. Drifting on flotsam was deemed to be a possibility for short distances— such as from the Caribbean— but Bermuda was close enough to Africa for this to have been feasible only for a comparatively short time after the island was formed, about 110 million years ago. 4 — VOLUME 98, NUMBER 1 289 “Bahamas ny S75Turks & Caicos H € => s : iS we ous x . Ascension Fig. 1. Map showing Atlantic and Caribbean islands or island groups from which the organisms discussed in this paper were taken. Since the Sket and Iliffe paper appeared, only four years ago, additional data have been gathered on the caves of Bermuda, the Turks and Caicos, the Bahamas, and the Canary Islands that allow us to expand on these observations— particularly as they relate to shrimp distributions and to abyssal connections. With some of the cave organisms, it might be easy to believe that they could reach Bermuda through oceanic dispersal of pelagic larvae. The occurrence of species on Bermuda that are widespread in similar habitats in the Caribbean is such an example. However, the fact that we do not yet know if those shrimp produce pelagic larvae weakens such an hypothesis. With other cave shrimps, however, it is even more difficult to create a scenario for their dispersal. Typhlatya iliffei Hart and Manning, 1981, endemic to Bermuda, has as its closest relative Typhlatya rogersi Chace and Manning, 1972—a species that is endemic on Ascension Island in the Southern Atlantic. The remaining representatives of the genus occur in subterranean fresh waters of Caribbean islands, Mexico, Central America, and the Galapagos. The genus Procaris was described from Ascension Island by Chace and Manning (1972) and the following year Holthuis (1973) described another species belonging to the genus from Hawaii. Only recently a third species has been found on Bermuda (Hart, Manning, and Iliffe, in prep.). The distributions of these species, as well as that of the Ligur-Barbouria-So- mersiella complex spanning the Atlantic from the Mediterranean to Bermuda and 290 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON the Caribbean (Manning and Hart 1984), lead us to give added credence to the deep-sea ties of these species, as well as emphasizing the hypothesis that some of the island species may represent relicts of ancient stocks. Our investigations have revealed a number of examples leading to these con- clusions, and a survey of the literature has yielded others. With regard to deep- sea ties, the following seem relevant: 1) Material recently collected in a Bermuda cave by Iliffe contained a repre- 2) 3) 4) 5) 6 7 8 9 ) —S ) —_ sentative of a new order of Peracarida (Bowman and Iliffe 1985). Almost simultaneously, a closely related species was collected in the open ocean at a depth of 1000 meters. (Sanders, Hessler, and Garner 1985). A new halocyprid ostracod that has been found in a Bermuda cave is con- sidered by Martin Angel (personal letter to Iliffe) to be intermediate between the undoubtedly ancient Thaumatocyprididae and the Halocyprididae. An- gel suggests that this indicates the cave faunas to be important links in establishing the evolution of the present oceanic faunas. An undescribed polychaete from a cave in the Turks and Caicos, represen- tative of a group otherwise known from deep waters, is presently being described by Marian Pettibone. A new ostracod from the Turks and Caicos, with deep sea relatives, is being described by L. S. Kornicker. A new shrimp family from the Turks and Caicos, related to the bresiliid shrimp recently described from the thermal vents of the Galapagos Rift (Williams and Chace 1982), is being described by Hart, Manning, and Iliffe (in prep.). A species of Munidopsis (M. polymorpha Koelbel), a speciose genus oth- erwise known from shelf, slope, and abyssal depths (Doflein and Balss 1913), is among the inhabitants of a lava tube formed 3000-5000 years ago in Lanzarote, Canary Islands. Wilkens and Parzefall (1974) suspect that this shrimp and other inhabitants of the lava tube might be widespread in the neighboring Atlantic. Miyake and Baba (1970), in their list of the known West African species, gave no records for this species outside of caves. Another inhabitant of that lava tube is a species of the amphipod genus Spelaeonicippe, family Pardaliscidae, which, as pointed out by Stock and Vermuelen (1982:4), “are predominantly bathyal/abyssal/hadal, pelagic Amphipoda.”’ These same authors described a second species of this genus from waters of a limestone cave on Providenciales, Turks and Caicos. In addition, Pettibone (1976) discussed a polynoid polychaete, Gesiella ja- meensis, from the same lava tube. The polychaete belongs to a subfamily otherwise known only from deep water. And three other genera—Barbouria, Ligur, and Somersiella—known pri- marily from anchialine or cave habitats point in the same direction. Four of the five species assigned to these genera occur only in these habitats, but the type-species of Ligur is a deep-sea species, known only from shelf-slope depths. As for the possible antiquity of some crustacean species, J. Tuzo Wilson (in litt.) suggested that “‘it is just conceivable that forms of life might have survived on Ascension from the time when the Atlantic was very narrow and the forerunners of Ascension were in contact with Brazil and the Cameroons.” VOLUME 98, NUMBER 1 291 Following on this, Hobbs and Hart (1983), in a review of the genus Atya, found that two species—A. gabonensis and A. scabra—are identical on both sides of the Atlantic, and that Atya intermedia (inhabiting two islands in the Gulf of Guinea) is so similar to Atya innocous (inhabiting the Greater and Lesser Antilles and Central America) that they were reluctant to recognize them as separate species. They noted that these shrimps must be considered little, if at all, changed since the Africa-America continental masses were still approximate—suggesting that this group of shrimps is extremely old, dating from at least the Middle Jurassic, 175 million years ago. And finally, Dliffe, Hart, and Manning (1983) hypothesized that part of the cavernicolous invertebrate marine fauna of Bermuda represent groups that sur- vived on submerged or emergent sea mounts along the Mid-Atlantic Ridge since the early Mesozoic, and that geothermal activity may have maintained water temperatures in caves sufficiently high to protect certain species during periods of glaciation. Conclusions We believe that the evidence presented here supports the conclusions 1) that there are significant ties between the marine cave fauna and the fauna of the deep sea, 2) that the cave faunas may contain very old elements, and 3) that caves have served as refuges over very long periods of time. It seems likely that the subterranean habitats on ocean islands consist not only of caves, but of crevicular habitats in the rock—similar environments that are potentially available for colonization anywhere in the water column. Thus, while surface caves and pools in limestone may be relatively young, the actual habitat may be as old as the island on which it is found. We use the term “crevicular” to designate those aquatic habitats formed by crevasses in and among rocks. We also use it to describe the organisms that live in those habitats. For the purposes of this definition, a cave sensu stricto is simply a large crevass, or it may be merely the gateway through which smaller crevasses are made accessible to the student of crevicular organisms. The crevicular habitat differs from the interstitial habitat primarily in size, and the animals that live in both habitats are likely to be thigmotactic. Following on this, it seems probable that species, or species-groups, could form a continuum—reaching from the caves of one island into the deep waters via the natural crevasses among rocks and so on up the slopes of other islands or continental masses. One does not have to wander far from that idea to see that if crustacean species are aS conservative as some appear to be, then the actual spreading of the sea floor could be a means by which seme species have achieved their present day distributions. Indeed, Chace and Hobbs’ (1969:21) tongue-in-cheek proposal of a ““continuous spelean corridor’ between islands may not be so far-fetched after all. Acknowledgments This work was supported by a National Science Foundation Grant (BSR 8215672) to Thomas M. Iliffe, and by funds from the Smithsonian Institution’s Scholarly 292 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Studies Program granted to C. W. Hart, Jr., and R. B. Manning. This paper is Contribution No. 977 from the Bermuda Biological Station for Research, Inc. Literature Cited Bowman, Thomas E., and Thomas M. Iliffe. 1985. Mictocaris halope, a new unusual peracaridan crustacean from marine caves on Bermuda.—Journal of Crustacean Biology 5(1):58-73. Chace, Fenner A., Jr., and Horton H. Hobbs, Jr. 1969. The freshwater and terrestrial decapod crustaceans of the West Indies with special reference to Dominica.— United States National Museum Bulletin 292:1-258. , and Raymond B. Manning. 1972. Two new caridean shrimps, one representing a new family, from marine pools on Ascension Island (Crustacea: Decapoda: Natantia).— Smithsonian Con- tributions to Zoology 131:1—18. Doflein, Franz, and Heinrich Balss. 1913. Die Galatheiden der Deutschen Tiefsee-Expedition. — Wissenschaftliche Ergebnisse der deutschen Tiefsee-Expedition auf dem Dampfer “Valdivia”’ 1898-1899. 184 pp, 17 pls. Jena: Gustav Fisher. Hart, C. W., Jr., and Raymond B. Manning. 1981. The cavernicolous caridean shrimps of Bermuda (Alpheidae, Hippolytidae, and Atyidae).— Journal of Crustacean Biology 1(3):441—456. Hobbs, Horton H., Jr., and C. W. Hart, Jr. 1983. The shrimp genus Atya.—Smithsonian Contri- butions to Zoology 364:1-143. Holthuis, L. B. 1973. Caridean shrimps found in land-locked saltwater pools at four Indo-West Pacific localities (Sinai Peninsula, Funafuti Atoll, Maui, and Hawaii islands), with the descrip- tion of one new genus and four new species.— Zoologische Verhandelingen 128:1—53. lliffe, Thomas M., C. W. Hart, Jr., and Raymond B. Manning. 1983. Biogeography and the caves of Bermuda.— Nature 302(5904):141-142. Manning, Raymond B., and C. W. Hart, Jr. 1984. The status of the hippolytid shrimp genera Barbouria and Ligur: A reevaluation.— Proceedings of the Biological Society of Washington 97(3):655-665. Miyake, Sadayoshi, and Keiji Baba. 1970. The Crustacea Galatheidae from the tropical-subtropical region of West Africa, with a list of the known species.— Atlantide Report 11:61—97. Pettibone, Marian. 1976. Revision of the genus Macellicephala McIntosh and the subfamily Ma- cellicephalinae Hartman-Schréder (Polychaeta: Polynoidea).—Smithsonian Contributions to Zoology 229:iv + 71. Sanders, H. L., R. R. Hessler, and S. P. Garner. 1985. Hirsuta bathyalis, sp. et fam. nov., an unusual deep-sea benthic peracaridan crustacean from the tropical Atlantic.—Journal of Crustacean Biology 5(1):30-57. Sket, Boris, and Thomas M. Iliffe. 1980. Cave fauna of Bermuda.—International Revue der gesam- mten Hydrobiologie 85(6):87 1-882. Sterrer, W. 1973. Plate tectonics as a mechanism for dispersal and speciation in interstitial sand fauna.— Netherlands Journal of Sea Research 7:200—222. Stock, J. H. 1977. The taxonomy and zoogeography of the hadziid Amphipoda with emphasis on the West Indian taxa.— Studies on the Fauna Curacao, The Hague 55:1—130. ——, and Jan J. Vermeulen. 1982. A representative of the mainly abyssal family Pardaliscidae (Crustacea, Amphipoda) in cave waters of the Caicos Islands.—Bijdragen tot de Dierkunde 52(1):3-12. Wilkens, H., and J. Parzefall. 1974. Die Oekologie der Jameos del Aqua (Lanzarote). Zur Entwicklung limnischer Hohlentiere aus marinen Vorfahren.— Annales de Speleologie 29(3):419—434. Williams, Austin B., and Fenner A. Chace, Jr. 1982. A new caridean shrimp of the family Bresiliuidae from thermal vents of the Galapagos Rift.—Journal of Crustacean Biology 2(1):136—147. (CWH and RBM) Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560; (TMI) Ber- muda Biological Station for Research, Inc., Ferry Reach, St. George’s, Bermuda. INTERNATIONAL COMMISSION ON ZOOLOGICAL NOMENCLATURE % British Museum (Natural History) Cromwell Road London SW7 5BD 3 December 1984 The following Opinions have been published by the International Com- mission on Zoological Nomenclature in the Bulletin of Zoological Nomen- clature, volume 41, part 4, on 30 November 1984: Opinion No. 1277 (p. 212) 1278 (p. 215) 1279 (p. 218) 1280 (p. 221) 1281 (p. 223) 1282 (p. 225) 1283 (p. 227) 1284 (p. 231) 1285 (p. 233) 1286 (p. 235) 1287 (p. 238) Ptilium Gyllenhal, 1827 and Ptenidium Erichson, 1845 (In- secta, Coleoptera): conserved. The generic name Rhincodon A. Smith, 1829 (Pisces): con- served. Chrysolina Motschulsky, 1860 (Insecta, Coleoptera): con- served. Rafinesque, C.S., 1822 ‘On the Turtles of the United States’: suppressed. Acmaea limatula Carpenter, 1864 (Mollusca, Gastropoda): conserved. Panopeus H. Milne Edwards, 1834 (Crustacea, Decapoda): designation of type species. LYMANTRIIDAE Hampson, [1893] given nomenclatural precedence over ORGYIIDAE Wallengren, 1861 and DA- SYCHIRIDAE Packard, 1864 (Insecta, Lepidoptera). Peggichisme Kirkaldy, 1904 (Hemiptera, Heteroptera): des- ignation of type species. Barbus altianalis Boulenger, 1900 and B. rueppelli Boulen- ger, 1902 (Pisces, Cyprinidae): conserved. Chermes fusca Zetterstedt, 1828 (Insecta, Homoptera): con- served. Sesia andrenaeformis Laspeyres, 1801 (Insecta, Lepidoptera): conserved. The Commission regrets that it cannot supply separates of Opinions. R. V. MELVILLE Secretary 293 INTERNATIONAL COMMISSION ON ZOOLOGICAL NOMENCLATURE % British Museum (Natural History) Cromwell Road London SW7 5BD 3 December 1984 The Commission hereby gives six months notice of the possible use of its plenary powers in the following cases, published in the Bulletin of Zoo- logical Nomenclature, volume 41, part 4, on 30 November 1984 and would welcome comments and advice on them from interested zoologists. Correspondence should be addressed to the Secretary at the above address, if possible within six months of the date of publication of this notice. Case No. 2136 THRESKIORNITHIDAE Richmond, 1917 (Aves): application to place on Official List of Family-Group names in zoology and to give precedence over PLATALEINAE Bonaparte, 1838, and other com- peting Family-Group names. Cricetodon minus [sic] Lartet, 1851 (Mammalia, Rodentia): revised request for a ruling on interpretation. Report on Glyphipterix Hiibner, [1825] (Insecta, Lepidoptera). Octolasion Orley, 1885 (Annelida, Oligochaeta, Lumbricidae): rati- fication of the designation of the type species and the introduction of Octolasion (Octodrilus) by Omodeo, 1956 in accordance with usage, with the suppression of the designation of the type species and of the names Octolasion (Incolore) and Octolasion (Purpureum) by Om- odeo, 1952. Revised submission regarding the nominal genus Diplosoma Mac- Donald, 1859 (Ascidiacea), and proposed alternative designation of Leptoclinum fulgens Milne Edwards, 1841, as type species of Lepto- clinum Milne Edwards, 1841. Aphodius rufus Moll, 1782 and Aegialia rufa Fabricius, 1792 (Insecta, Coleoptera): proposed conservation under the plenary powers by suppression of Aphodius scybalarius Fabricius, 1792. Proposed use of plenary powers to conserve certain junior synonyms in the family PYGOPIDAE (Brachiopoda). Delphinus truncatus Montagu, 1821 (Mammalia, Cetacea): proposed conservation by suppression of Delphinus nesarnack Lacépéde, 1804. R. V. MELVILLE Secretary 294 pigs Vs i a We Fe i hag tS ps as | A 2 Mice he DOA : ‘ i! | + ike Jala , ‘ 7 i as ae a 7 > i ~ er we et ieee i nan Seo we f ; ina a vem ; q af 7 i ue (Ot 4% »*s ‘4 2 - }, re t * ts 7 wy % i 2 = , = a | *s ; ’ 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 will be published in English (except for Latin diagnosis/description of plant taxa which should not be duplicated by an English translation), with a summary in an alternate language when appropriate. Publication Charges.—Authors will be asked to assume publication costs of page-charges, tabular material, and figures, at the lowest possible rates. Submission of manuscripts.—Manuscripts should be sent to the Editor, Proceedings of the Biological Society of Washington, National Museum of Natural History, Smithsonian Insti- tution, Washington, D.C. 20560. Review.—One of the Society’s aims is to give its members an opportunity for prompt pub- lication of their shorter contributions. Manuscripts will be reviewed in order of receipt by a board of associate editors and appropriate referees. 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. Synonymy of abbreviated style (author, date, page) (except in botanical papers), with full citations of journals and books (no abbreviations) in the Literature Cited is required. The establishment of new taxa must conform with the requirements of the appropriate in- ternational codes of nomenclature. Authors are expected to be familiar with these codes and to comply with them. New species-group accounts must designate a type-specimen deposited in an institutional collection. The sequence of material should be: Title, Author(s), Abstract, Text, Acknowledgments, Literature Cited, Author’s (s’) Address(es), Appendix, List of Figures (entire figure legends), Figures (each numbered and identified), Tables (each table numbered with an Arabic numeral and heading provided). Manuscripts should be typed, double-spaced throughout (including tables, legends, and foot- notes) on one side of 812 xX 11 inch sheets, with at least one inch of margin all around. Submit a facsimile with the original, and retain an author’s copy. Pages must be numbered on top. One manuscript page = approximately 0.5 printed page. Underline singly scientific names of genera and lower categories; leave other indications to the editor. Figures and tables with their legends and headings should be self-explanatory, not requiring reference to the text. Indicate their approximate placement by a pencil mark in the margin of the manuscript. Illustrations should be planned in proportions that will efficiently use space on the type bed of the Proceedings (12.5 x 20 cm) and should not exceed 24 x 15 inches. Figures requiring solid black backgrounds should be indicated as such, but not masked. Art work will be returned only on request. Proofs.—Galley proofs will be submitted to authors for correction and approval. Reprint orders will be taken with returned proof. Costs.—Page charges @ $60.00, figures @ $10.00, tabular material $3.00 per printed inch. All authors are expected to pay the charges for figures, tables, changes at proof stage, and reprints. Payment of full costs will probably facilitate speedy publication. CONTENTS New kangaroo mice, genus Microdipodops (Rodentia: Heteromyidae), from Idaho and Nevada John C. Hafner Mississippiellidae, a new eulobosinid (““Thecamoebinid’’) family (Protozoa) Richard W. Huddleston and Drew Haman First record of the genus Zebrias (Pisces: Pleuronectiformes: Soleidae) from the Philippine Islands, with the description of a new species Jeffrey A. Seige] and Thomas A. Adamson The correct name for the Palearctic Brown, or Flat-skulled, Shrew is Sorex roboratus Robert S. Hoffmann Taxonomic status and relationships of the Swan Island hutia, Geocapromys thoracatus (Mam- malia: Rodentia: Capromyidae), and the zoogeography of the Swan Islands vertebrate fauna Gary S. Morgan Lolliguncula argus, a new species of loliginid squid (Cephalopoda: Myopsida) from the tropical eastern Pacific Thomas F. Brakoniecki and Clyde F. E. Roper Description of a wood dwelling sipunculan, Phascolosoma turnerae, new species Mary E. Rice A synopsis of the Laccornis difformis species group with a revised key to North American species of Laccornis Des Gozis (Coleoptera: Dytiscidae) G. W. Wolfe and P. J. Spangler Two new species of Protura (Insecta) from North America Ernest C. Bernard A new member of the genus Distocambarus (Decapoda: Cambaridae) from the Saluda Basin, South Carolina Horton H. Hobbs, Jr. and Paul H. Carlson Oman ypsilon, a new genus and species of blenniid fish from the Indian Ocean Victor G. Springer Perioculodes cerasinus, n. sp., the first record of the genus from the Caribbean Sea (Amphipoda: Oedicerotidae) James Darwin Thomas and J. L. Barnard Zenarchopterus ornithocephala, a new species of freshwater halfbeak (Pisces: Hemiramphidae) from the Vogelkop Peninsula of New Guinea Bruce B. Collette Morphometrics and distinctness of the hedgehog genera (Insectivora: Erinaceidae) C. Brian Robbins and Henry W. Setzer The correct identity of the pelagic amphipod Primno macropa, with a diagnosis of Primno abyssalis (Hyperiidea: Phrosinidae) Thomas E. Bowman Polychaete worms from a cave in the Bahamas and from experimental wood panels in deep water of the North Atlantic (Polynoidae: Macellicephalinae, Harmothoinae) Marian H. Pettibone An additional new scale worm (Polychaeta: Polynoidae) from the hydrothermal rift area off western Mexico at 21°N Marian H. Pettibone The Oligochaeta of Georges Bank (NW Atlantic), with descriptions of four new species Dale Davis Penaeoid shrimp fauna from tropical seagrass meadows: species composition, diurnal, and sea- sonal variation in abundance Raymond T. Bauer Two new species of two new gammaridan genera (Crustacea: Amphipoda) from the Florida Keys James Darwin Thomas and J. Laurens Barnard Three new species of thread snakes (Serpentes: Leptotyphlopidae) from Hispaniola Richard Thomas, Roy W. McDiarmid, and Fred G. Thompson A new skink (Reptilia: Sauria: Leiolopisma) from Fiji George R. Zug Halophila decipiens, an unreported seagrass from the Philippines Ernani G. Mefiez and Hilconida P. Calumpong Redescription of Heterocarpus laevis A. Milne Edwards (Crustacea: Decapoda: Pandalidae) Brian Kensley and William Tobias Lucayarina catacumba, new genus, new species, a Bahamian sea-cave amphipod (Crustacea: Amphipoda: Lysianassidae) Janice Clark and J. L. Barnard New species of Isopoda from the Florida Middlegrounds (Crustacea: Peracarida) Allan Hooker Cryptopenaeus crosnieri, a new species of shrimp, and a new record of C. sinensis (Penaeoidea: Solenoceridae) from Australian waters Isabel Pérez Farfante and Brian Kensley The fauna of Atlantic marine caves: evidence of dispersal by sea floor spreading while maintaining ties to deep waters C. W. Hart, Jr., R. B. Manning, and T. M. Iliffe International Commission on Zoological Nomenclature: Opinions and Notices 107 112 121 127 150 158 177 191 204 22k 232 237, 243 255 281 288 298 THE BIOLOGICAL SOCIETY OF WASHINGTON 1983-1984 Officers President: Donald R. Davis Secretary: Gordon L. Hendler Vice President: Austin B. Williams Treasurer: Leslie W. Knapp Elected Council J. Laurens Barnard Maureen E. Downey Frederick M. Bayer Louis S. Kornicker Isabel C. Canet Storrs L. Olson Custodian of Publications: David L. Pawson PROCEEDINGS Editor: Brian Kensley Associate Editors Classical Languages: George C. Steyskal Invertebrates: Thomas E. Bowman Plants: David B. Lellinger Vertebrates: Richard P. Vari Insects: Robert D. Gordon Membership in the Society is open to anyone who wishes to join. There are no prerequisites. Annual dues of $15.00 include subscription to the Proceedings of the Biological Society of Washington. Library subscriptions to the Proceedings are: $18.00 within the U.S.A., $23.00 elsewhere. 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 Treasurer, Biological Society of Washington, National Museum of Natural History, Smithsonian Instutution, Washington, D.C. 20560. 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. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 295-302 A NEW SPECIES OF THE COLUBRID SNAKE GENUS LIOPHIS FROM BRAZIL James R. Dixon Abstract.—A new species of the colubrid snake genus Liophis (L. maryellenae) is described from the Brazilian state of Minas Gerais. Its closest congener is L. jaegeri, differing only in the number of body scale rows. The new species shows some affinities to L. viridis by its color and body scale rows. A total of approximately 5600 specimens of 31 species of the colubrid snake genus Liophis were examined during the past eight years. Nearly all species were represented by 15 or more specimens. However, one taxon, represented by only ten individuals, was unique in having a combination of relatively small size at sexual maturity (474 mm total length), a scale row reduction of 19-19-17, no apical scale pits, and relatively high numbers of long, slender maxillary teeth (25— 28). This taxon is similar to Liophis jaegeri in size, somewhat in color and color pattern, number of ventrals, subcaudals, head scales, maxillary teeth; tail length/ total length ratio, and length of in situ hemipenes. However, it differs in the number of scales around the body. The species is also similar to Liophis viridis in color, number of scales around the body, but differs by fewer number of ventrals, higher tail length/total length ratio, and number of maxillary teeth. Other Liophis species that occur sympatricly or parapatricly with the new form are L. miliaris, L. poecilogyrus, L. jaegeri, L. almadensis, and L. dilepis. Liophis maryellenae, new species Fig. 1 Holotype.— American Museum of Natural History (AMNH) 62202, adult male, taken (with AMNH 62206) from Annapolis (16°20’S—48°58'W), Goias, Brazil, 7 Mar 1936, by R. M. Gilmore (found unlabelled with Gilmore collection when material was cataloged). Paratypes (All from Brazil). —AMNH 62206 female, Annapolis, Goias; Texas Cooperative Wildlife Collection (TCWC) 57701, female, 13°10’S, 46°00’W, ca. 150 km SW Barreiras, Bahia; Natural History Museum, Paris (MNHP) 3565, female, “Brazil”; Museum of Zoology, University of Sao Paulo (MZUSP) 6609 male, D. F., Brazilia (16°12'S, 44°26'W); MZUSP 8059 male, Minas Gerais, Itambe do Mato Dentro (19°24'S, 43°19’W); MZUSP 7989 female, Minas Gerais, Grao Mogol (16°34’S, 42°54'W); Instituto Butantan (IB) 1209 male, IB 5680 female, Minas Gerais, Araguari (18°39’S, 48°12’W); IB 12559 male, Minas Gerais, Ouro Branco. Description of holotype.—Measurements in mm; total length 325; tail length 81; tail/total length ratio 0.249; head length 11.5; head width 6.8; eye diameter 2.4; eye/nostril distance 2.2; eye/snout distance 3.4; frontal width 2.1; frontal length 4.0; head width/head length ratio 0.59; eye diameter/eye—nostril distance 0.936; frontal width/frontal length ratio 0.525. 296 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Se SS CLT Cn) | 7S 2S aaaw, 2~ SZ ZS Fig. 1. Dorsal, lateral and ventral views of the head of Liophis maryellenae (TCWC 57710). Black bar equals one cm. Scale rows number 19-19-17, smooth, no apical scale pits, reduction to 17 scale rows occurs through fusion of scale rows 3 + 4 over 79th ventral; tail scale row reduction 8-6-4-2 over 9th, 18th, and 41st subcaudals, respectively. Ventrals number 150, subcaudals 69; supralabials are 8-8, with 4th and 5th entering orbit; infralabials 10-10; preoculars 1-1, postoculars 2-2, temporals 1 + 2, loreal 1-1, 27 maxillary teeth, last two enlarged, ungrooved and separated from remainder by diastema equal to width of ectopterygoid process of maxillary; anal shield divided. Color (in alcohol): overall grayish brown; head uniform grayish brown; anterolateral portion of nasal scale and dorsal tip of rostral, whitish; supralabials 1—4 tipped with grayish brown, supralabials 7-8 dark grayish brown, remainder of supralabials, infralabials, chin, throat, and anterior 10-12 ventrals whitish. Hemipenes spinose, without calyces, but with apical, smooth disc; in situ hemi- penis 12 subcaudals in length, slightly bilobed, lobe about two subcaudals in VOLUME 98, NUMBER 2 297 length; sulcus spermaticus forks at sixth subcaudal; basal naked pocket present; shank of hemipenis with moderate sized spines that slowly decrease in size towards tip of apical disc. ‘Variation. — The five female paratypes differ from the holotype as follows: num- ber of ventrals vary from 146-153, x = 149.6; number of subcaudals vary from 62-73, x = 66.2; infralabials 10(4), 11(1); supralabials, preoculars, postoculars, temporals, supralabials entering orbit, loreal, and anal plate invariate, 8, 1,2, 1 + 2,4 + 5, 1, divided, respectively; tail/total length ratios vary from 0.221-0.254, X = 0.237; total length varies from 295-530, x = 420.8 mm; reduction site varies from 82nd to 85th, x = 84.2 ventral; diameter of eye/snout distance ratio varies from 0.605-0.697, x = 0.666; tail scale row reduction of 8-6-4-2 occurs over subcaudals 9, 23, 48, or 8, 23, 50, respectively (not recorded for MNHP specimen). The AMNH female paratype has a paravertebral dark stripe beginning at the 96 ventral on the seventh scale row, passing posteriorly onto the upper edge of the fourth scale row of the tail. Gilmore’s field notes for the latter specimen are “neck greenish, below and behind yellowish brown.” Color in life of the Bahia (TCWC) specimen “dark gray-green dorsum, yellow-orange venter.” Neither color descriptions mention the presence of lateral and/or paravertebral dark lines of the body, but they are present in preservative (Fig. 2a, b). Besides the faint “lined” patterns that appear in preservation, definite areas of darkened pigment appear as well as defined patterns in other specimens (Fig. 2c, d). The patterns may appear as paravertebral lines on the eighth scale row and a series of dashes or dots on scale row four and frequently parts of scale rows three and five. Occasionally, a series of black flecks (spots?) occur in scale rows one, two, three, four, seven and/or eight, or any combination of these. The venter is immaculate cream or yellowish in preservation, with some darkening along the outer edge of scale row one and the adjoining ventral. The dark color may extend completely across the ventral along its anterior edge. Five males vary as follows: number of ventrals vary from 144-159, x = 152.2; number of subcaudals vary from 64-82, x = 69.6; supralabials, infralabials, pre- oculars, postoculars, temporals, supralabials entering orbit, loreal, and anal plate invariate, 8, 10, 1, 2, 1 + 2,4 + 5, 1, divided, respectively. The tail/total length ratio varies from 0.234—-0.262, x = 0.243; total length varies from 325-435, x = 394.4 mm; hemipenial length in situ varies from 9.5 to 12, xX = 10.5 subcaudals; reduction site varies from the 78th to 95th, x = 83.4 ventral; diameter of eye/ snout distance ratio varies from 0.632-—0.844, O = (e) WIDTH (100 pm) 0.5 1.0 2.0 3.0 4.0 LENGTH (100 pm) Fig. 4. Dimensions of atria of Paranais litoralis and P. birsteini. nearly equal in length. Penial setae in V paired in each bundle, with short distal ends and short teeth of equal length. Dorsal setae 3-4, rarely 5 in V, 2-4 in rest of dorsal bundles. Spermathecal and atrial ampullae ovoid, atrial length to breadth ratio 1.2-1.9 (x = 1.5, n = 19, s = 0.19), atrial ampullae with thick muscular walls. Spermathecal and atrial ducts distinct, spermathecal pores vary in position from in front of and below or adjacent to ventral setae. Body wall distinctly papillate, with transverse ridges with foreign material in grooves. Kamchatka, USSR, and Pacific coast of Canada. Material examined.—19 mature specimens, from: Wainwright Basin, Tsimp- sean Peninsula below Prince Rupert; Tomahawk Island, Nowlan Passage; Ho- karson Point and Ram Bluff, Dean Channel, Rattenbury Point, Fisher Channel, near Ocean Falls; Mowitch Point and Ecstall River confluence with Skeena River, Skeena River estuary; Alberni Inlet, Tahsish Inlet, Moyeha Bay, Sydney Inlet, Vancouver Island, all British Columbia, coll. H. R. Baker and K. A. Coates, 1979- 1981. Skagit River Flats, mouth of the north fork, Skagit River, Washington coll. E. Gallagher, Dec 1979. Remarks.—The new material from British Columbia and Washington is re- markably similar to that described from Kamchatka, and it would be reasonable to assume that this is a Pacific Rim species. The atrial walls in the original material are said to be 17 um thick; here they vary from 9 to 24 wm with a mean of 17 um. The length/breadth ratio of the atria of the original is 1.7, that of the new material is 1.5 (see Fig. 4). The species combines the somatic setae and muscular 310 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON atrial walls of the form seen in P. /itoralis with penial setae like those of P. frici. The length and breadth of the atria and spermathecae are intermediate between those of P. litoralis and P. frici. Paranais grandis (Harman, 1977), new combination Figs. 2-3, Table 1 Wapsa grandis Harman, 1977:83, fig. 1. Diagnosis.—3.0—4.0 mm, s = 28 plus (all fragments). Ventral setae of II 6-7, longer than other setae (105-150 um versus 80-122 wm) with upper teeth much longer than lower. Ventral setae of IIJ-IV 3 per bundle, with similar teeth, ventral setae of V 4, strongly modified penial setae with short, thin, recurved distal ends (88-140 um long). Most other ventral bundles with 3 setae, all with long upper teeth. Dorsal setae 2-3 per bundle, similar to rest in form. Spermathecal and atrial ampullae ovoid, length to breadth ratios of 1.3 and 1.4 respectively, with walls thin when fully distended (5 wm). Body wall papillate with foreign matter attached. Louisiana, Texas, U.S.A. Material examined. — 13 whole-mounted specimens, Sea Rim State Park, Texas, 29 Jan, 1979, coll. J. Wern, det. M. Loden. Remarks.—Harman (1977) cited the setal lengths in this species as 128-148 pum in II, 98-106 wm in III, 109-121 um in other ventrals, and a total range of 74-123 wm for dorsals, the anterior setae being longer than the more posterior setae. He compared these with setal lengths for other Wapsa species (W. mobilis, W. evelinae) and showed that those of W. grandis were distinctly longer than those of the other two species. He did not compare the setae with those of Paranais species. The setal measurements based on mature specimens of the four species available to us confirm the fact that there is a much greater difference between the setae of II and the other bundles in P. grandis than in the other species (Table 1), and that the setae are generally longer overall. Otherwise this species differs from P. frici primarily by the possession of fully developed penial setae of the form observed in P. litoralis, and by the larger number of somatic and penial setae (Fig. 2). Other Species Only one of the following species has been examined, the type-specimens of P. botniensis having been studied for us by C. Erséus. Type-material is not available for eastern European species, and the South American species were described from immature specimens. Paranais simplex Hrabe, 1936 Paranais simplex Hrabe, Kasprzak, 1977:93, figs. 1-4, plates 1-5. This Ponto-Caspian species has 5—6 ventral setae in II with the teeth equally long, and 3—4 in other bundles with the upper teeth shorter than the lower. The penial setae are paired, sigmoid and bifid. The reproductive system, described by Kasprzak (1977), includes ovoid spermathecae and atria, the latter with thick muscular walls. No dimensions were given in this description, but the illustration VOLUME 98, NUMBER 2 311 shows the atrial muscles to be about 15 um thick. This species is very similar to P. birsteini apart from the form of the setal teeth. Because of the limited distri- bution and freshwater habitat, recognition of this as a separate species causes no practical problems, and so unless the setal form can be shown to be the result of the ambient salinity by experimentation, the question of synonymy of P. birsteini with this species will not be pursued. Paranais botniensis Sperber, 1948 This species shares the small setal number and thin muscle layer of the atrial ampulla (2—7 um) with P. frici, but the length/breadth ratio of the atria approaches that of P. birsteini (64 by 48 um, or 76 by 37 in another specimen) as does the form of the setae. The penial setae are paired and bifid, sigmoid as in both of the above. This species has not, to our knowledge, been found outside the brackish water type-locality in Sweden. The setal numbers together with other differences noted by Sperber, strongly suggest that this be regarded as a distinct species. In any case, it does not appear to be a geographic race of P. /itoralis, one possibility suggested by Sperber. Paranais macrochaeta Cernosvitov, 1939 and Paranais salina Cernosvitov, 1939 These two species were recorded from saline water in the Lake Titicaca basin, Peru, described from very few immature specimens. The principal difference between these and existing taxa was claimed to be the lateral position of the ventral setae of II relative to those of other bundles, a suggestion maintained by Sokolskaya (1971) in her key to the species in the genus. In all of the species examined by us the setae of II seem to be in this position. The number of setae in P. macrochaeta (8-9 in II, 4—5 in all others bar III-IV) is much higher than in P. frici, which shares the pattern of setal teeth, with the upper teeth always much longer than the lower. Paranais salina resembles P. litoralis, but mention of the anus Opening into a rectangular fossa.raises doubts. Until mature specimens are collected, these species must remain dubious taxa. Paranais mobilis Liang, 1958 As nephridia and foreign matter on the body wall have now been discovered in P. frici, and the question of the form of the penial setae has been resolved (see above), this species is here regarded as synonymous with P. frici (q.v.). Wapsa evelinae Marcus, 1965 The primary reason for the erection of this species as the type of a new genus was the presence of nephridia. Marcus (1965) did compare the species briefly with the existing Paranais species other than P. mobilis. The species clearly differs from P. litoralis in that the penial setae are sigmoid and bifid, and the atria and spermathecae are globular to ovoid in shape. The muscular layer of the atrial wall is said to be thick, but no measurements are presented and the illustration suggests that they are thinner than those of P. birsteini (for example). It is clear that there was considerable variation in setal number and form in the material described 312 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON by Marcus (ventrals of II 3-4 or 5-6, in III-IV 2 or 3-4, the upper teeth twice as long as the lower, or less than twice as long). The specimens with the smaller setal number distinctly resemble P. frici in both number and form of the setae. Marcus interpreted literally the statement by Hrabe (1941) regarding the penial setae resembling the other ventral setae, and supposed the slightly modified penial setae of his specimens to be a distinguishing characteristic. He also quotes the presence of intracellular stomach canals in the new species, a character utilised by Sperber (1948). However, in view of our experience with nephridia, and in view of the level of distinction applied to the separation of species throughout the family, we do not feel that this is a sound character, and would not use a single residual character to maintain a distinct species in the absence of other differences. Some part, at least, of the Marcus material would seem to be attributable to P. frici depending on the reliability of observations on the thickness of the muscle layer of the atrial walls. Otherwise some, or even all, of the material closely resembles P. botniensis although no mention of papillation of the body wall appears in the original description of that species. This species cannot be identified for certain. Paranais multisetosa Finogenova, 1972 Paranais multisetosa Finogenova, 1972:94, figs. 1-5. In this species there are 8—10 setae in II, with the teeth of equal length. In other ventral bundles there are 3-6, and 3-4 in dorsal bundles. The penial setae number 3—4 and are said to resemble the ventrals (see discussion above). The setal lengths are 62—73 um (anterior ventrals), 59-64 um (posterior ventrals), and 61-68 wm (dorsals). The atrial ampullae are 112-117 wm long by 50—75 um wide, the sper- mathecae are 70-129 wm long by 36 um wide, with a long duct. The species is restricted to the Dnieper Bug Firth, in the Black Sea, salinity 0.14—5.6 parts per thousand. The species is close to P. simplex, the Ponto-Caspian species, but has more setae of a shorter length. Paranais palustris Udaltsov, 1907 This species was tentatively attributed to Homochaeta setosa (Moszynski, 1933) by Marcus, 1965, but the available descriptions do not justify such an action, so that it is in the interest of stability to regard this species as a dubious taxon. Access to the type, even if one exists, is not feasible. Other Homochaeta species have been identified as Paranais species at various times (see Brinkhurst 1971). Acknowledgments We wish to thank C. Erséus, W. Harman, J. K. Hiltunen, and M. S. Loden in particular for their generous loan of material and help in reviewing this manuscript. C. Erséus examined type-material for us in Sweden. The Museum of Victoria, Melbourne, loaned material to the late H. R. Baker, and these as well as the personal collections made by Dr. Baker in Canada and Europe and other material VOLUME 98, NUMBER 2 313 loaned to him has been used extensively in this study. Sections were prepared by the Laboratory of Analytical Systematics of the Royal Ontario Museum. Literature Cited Brinkhurst, R. O. 1971. Part 2. Systematics. 7. Family Naididae. Jn Brinkhurst, R. O., and B. G. M. Jamieson, Aquatic Oligochaeta of the World.— Oliver and Boyd, Edinburgh, xi + 806 pp. pp. 304-443. 1978. Freshwater Oligochaeta in Canada.—Canadian Journal of Zoology 56:2166—-2175. , and R. D. Kathman. 1983. A contribution to the taxonomy of the Naididae (Oligochaeta) of North America.— Canadian Journal of Zoology 61:2307-2312. , and M. L. Simmons. 1968. The aquatic Oligochaeta of the San Francisco Bay system.— California Fish and Game 54:180-194. Cernosvitov, L. 1939. Volume 1, Part 1. VI. Oligochaeta. pp. 81-116. Jn H. C. Gilson. The Percy Sladen Trust Expedition to Lake Titicaca in 1937.—Transactions of the Linnean Society. Finogenova, N. P. 1972. New species of Oligochaeta from the Dneiper and Bug Firth and the Black Sea and revision of some species. — Transactions of the Zoological Institute of the USSR Acad- emy of Sciences 52:94—-116. Harman, W. J. 1977. Three new species of Oligochaeta (Naididae) from the Southeastern U.S.— Proceedings of the Biological Society of Washington 90:483-490. Hiltunen, J. K.,and D. J. Klemm. 1980. A guide to the Naididae (Annelida: Clitellata: Oligochaeta) of North America.— United States Environmental Protection Agency Research Reports, En- vironmental Monitoring Series 600/4-80-031. Hrabe, S. 1936. Zur Kenntnis der Oligochaeten des Aral-Sees.—Izvestia Akademia Nauk USSR 6:1265-1276. . 1941. Zur Kenntnis der Oligochaeten aus der Donau. Acta Societatis Scientarium Naturalium Moravicae 13:1-36. 1981. Vodni malostetinatci (Oligochaeta) Ceskoslovenska.—Acta Universitatis Carolinae- Biologica 1979:1-167. Kasprzak, K. 1977. Remarks on histological structure of some parts of reproductive organs of Paranais simplex Hrabe, 1936 (Oligochaeta, Naididae).— Zoologica Poloniae 26:93-102. Laakso, M. 1969. Oligochaeta from brackish water near Tvarminne, south-west Finland.— Annales Zoologici Fennici 6:98-111. Liang, Y.-L. 1958. On some new species of Naididae from Nanking including remarks on certain known species.— Acta Hydrobiologia Sinica 7:41-58. Marcus, E. 1965. Naidomorpha aus brasilianischem Brackwasser.— Beitrage zur Neotropischen Fau- na 4:61-82. Moore, J. P. 1905. Some marine Oligochaeta of New England.—Proceedings of the Academy of Natural Sciences, Philadelphia 57:373-399. Sokolskaya, N. L. 1964. The new species and subspecies in family Naididae (Oligochaeta) from brackish reservoir in Kamchatka and South Sakhalin.—Byul Moskov Obshchest Ispytatelei Prirody Otd Biol 69:57-64. 1971. A new species of the genus Paranais (Oligochaeta, Naididae) from Kamchatka. — Zoologicheskii Zhurnal 50:930-933. Sperber, C. 1948. A taxonomical study of the Naididae.— Zoologiska Bidrag fran Uppsala 28:1-296. (ROB) Ocean Ecology Laboratory, Institute of Ocean Sciences, P.O. Box 6000, Sidney, British Columbia V8L 4B2, Canada; (KAC) Department of Biology, Uni- versity of Victoria, P.O. Box 1700, Victoria, British Columbia, V8W 2Y2, Canada. Note added in proof: Since this MSS. was prepared, two reports published by N. P. Finogenova have become available and have been translated. Paranais tjupensis, described from Issyk-kul Lake, is similar to P. botniensis (1977, Hydrobiological studies on the Tyup River and of Tyup Bay on Lake Issyk-kul. Collection of Scientific Papers; U.S.S.R. Academy of Sciences. Leningrad, Zoological In- stitute). In 1982 the subspecies orientalis was promoted to specific rank, though it is very similar to P. litoralis and within the range of variation of that species as defined here (Marine Invertebrates of coastal biocenoses of the Arctic Ocean and the Pacific Ocean. Explorations of the fauna of the seas; Academy of Sciences of the U.S.S.R., Zoological Institute). PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 314-320 PANDION LOVENSIS, A NEW SPECIES OF OSPREY FROM THE LATE MIOCENE OF FLORIDA Jonathan J. Becker Abstract.—Pandion lovensis n. sp. (Aves: Pandionidae) is described from the late Miocene (latest Clarendonian) of Florida. This species, based on pelvic limb elements, appears to be the most primitive member of the genus. The modern osprey (Pandion haliaetus) is the sole living representative of an enigmatic family of diurnal raptors. A number of detailed studies have investigated the morphology of the osprey in order to clarify its taxonomic position (Compton 1938; Hudson 1948; Sibley and Ahlquist 1972; Jollie 1976-1977). These studies have placed the osprey in a separate family or suborder, usually allied with the hawks and eagles. Warter (1976) discussed the fossil record of the Pandionidae and described the first paleospecies, Pandion homalopteron, from Sharkstooth Hill, but did not find any convincing evidence to ally the modern osprey closely with any other fal- coniform group. Sharkstooth Hill near Bakersfield, Kern County, California, is early middle Miocene (about 14.5—13 million years B.P.) and is “closely tied into the late “Temblor’ megainvertebrate stage and the Luisian microinvertebrate stage” (Repenning and Tedford 1977:79). Pandion homalopteron, based on associated humeri and ulnae, represents an osprey slightly larger than the average modern osprey, that shows evidence of weaker wing musculature (Warter 1976). Brunet (1970) proposed the transfer of Palaeocircus cuvieri Milne-Edwards from the Accipitridae to the Pandionidae, but because of the incompleteness of the holotype, a fragmentary carpometacarpus, this has not been accepted by other workers (Warter 1976). Reported here, from the late Miocene of Florida is the second known paleo- species of osprey. Abbreviations. —Specimens cited below are housed in the following institutions: Florida State Museum (UF), collection of Pierce Brodkorb (PB), and Natural History Museum of Los Angeles County (LACM). Recent specimens examined.—Pandion haliaetus carolinensis four male speci- mens, PB 20312, PB 39212, PB 27958, UF 19406; four female specimens, PB 17061, UF 14546, UF 17082, UF 18215; four specimens of unknown sex, PB 34670, PB 39613, PB 37976, PB 34669. Fossil specimens examined. — Referred proximal end of left tibiotarsus of Pan- dion homalopteron (LACM 42815). Subsequent to Warter’s (1976) description of this species, this tibiotarsus was collected by Mr. William Hawes from the same location at Sharkstooth Hill (LACM locality 3205) that he collected the type- material of P. homalopteron (L. G. Barnes, in litt. 1982). It should be noted, that although this specimen bears the same catalog number as the holotype, it cannot be considered type-material (except as a referred hypotype), as it was not included in the original description. VOLUME 98, NUMBER 2 Sls) Descriptive statistics are based on all above recent specimens. Morphological comparisons are based on the seven specimens in the Brodkorb collection. Mea- surements (Table 1) were made with Kanon dial calipers, accurate to 0.05 mm and rounded to the nearest 0.1 mm. BMDP Statistical Software program BMDPI1D was used to calculate simple descriptive statistics (Dixon 1981). Computations were made at the Northeast Regional Data Center (NERDC) at the University of Florida, Gainesville. All fossil specimens are deposited in the Vertebrate Paleon- tology collections of the Florida State Museum, University of Florida (UF). An- atomical terminology follows Baumel et al. (1979) and Howard (1929). Order Accipitriformes (Falconiformes auct.) Family Pandionidae (Sclater and Salvin, 1893) Skeletal elements of pelvic limb distinguished from other accipitriform families by the following combination of characters: (1) femur with very deep popliteal fossa; (2) tibiotarsus with extensor canal very deep under tendinal bridge with single distal opening; (3) fibula fused far distad; (4) tarsometatarsus relatively short, with ossified retinaculi extensoris for M. extensor digitorum longus; (5) hypotarsus extremely large with a single circular opening for tendons of Mm. flexor digitorum longus and flexor hallucis longus; (6) calcaneal ridge grooved; (7) trochleae strongly arched; (8) claws rounded beneath. Genus Pandion Savigny, 1809 Pandion lovensis, new species Figs. 1, 2 Holotype.— Nearly complete left tarsometatarsus. Vertebrate Paleontology col- lections of the Florida State Museum, UF 25950 (Fig. 1b, c); collected in 1979 by personnel of the Florida State Museum. Type-Locality.—Love Bone Bed local fauna. Florida, Alachua County, along State Road 241, NW 4, SW 4, NW 4, Sec. 9, T. 11S., R. 18 E., Archer Quadrangle, U.S. Geological Survey 7.5 minute series topographical map, 1969. Webb et al. (1981) give an overview of this local fauna. Horizon. — Late Miocene, latest Clarendonian land mammal age (approximately 9 million years B.P.). The Love Bone Bed local fauna originates from fluvial sediments of the Alachua Formation (Williams et al. 1977). Etymology.—For the type locality, the Love Bone Bed. Paratypes.— Distal half of right femur, UF 25766; distal end of right tibiotarsus, UF 25884; complete left tibiotarsus, UF 25928; right tarsometatarsus lacking proximal end, UF 25863; ungual phalanges, UF 26055, UF 26056, UF 29660. Measurements.—Table 1. Diagnosis.— Distinguished from P. haliaetus by: longer and more slender tar- sometatarsus, lateral proximal vascular foramen opening within hypotarsal canal; femur with patellar sulcus broader and caudal intermuscular line more mediad; tibiotarsus with anterior and posterior intercondylar sulci wider and less deep, cranial opening of extensor canal larger and more transversely oriented, and distal end wider. Distinguished from P. homalopteron by a tibiotarsus with smaller transverse width of proximal end and deeper fossa retrocristalis. 316 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Measurements of Pandion spp. Data are number of specimens (n), mean + standard deviation (x + SD) and range. Tibiotarsus—A, Total length; B, Length fibular crest; C, Least depth shaft; D, Depth proximal end; E, Transverse width proximal end; F, Transverse width distal end, across anterior portion of condyles; G, Transverse width distal end, across posterior portion of condyles; H, Depth medial condyle; I, Depth lateral condyle; J, Least depth intercondylar area. Tarsometatar- sus—K, Total length, from eminentia intercondylaris through trochlea IIJ; L, Length metatarsal I facet; M, Transverse width trochlea III; N, Transverse width distal end; O, Depth trochlea III; P, Transverse width proximal end; Q, Depth proximal end, excluding hypotarsus. Femur—R, Transverse width of lateral condyle; S, Transverse width of medial and lateral condyles; T, Transverse width lateral condyle and trochlea fibularis; U, Transverse width distal end; V, Depth distal end; W, Depth femoral shaft cranial to condyles. P. h. carolinensis MedSUT C= 6 ee eee eee ee ee homalopteron P. lovensis ment x + SD (n) Range Xj Xj Tibiotarsus: A 123.59 + 4.62 (12) 119.2-130.8 — 124.8 B 34.82 + 2.09 (13) 31.2-38.2 — 32.7 Cc 5.59 + 0.29 (12) 5.2-6.1 — Boe Dat/ D 17.00 + 0.92 (13) 15.9-18.5 17.4 17.0 E 13.17 + 0.60 (13) 12.3-14.0 14.3 13.1 F 14.08 + 0.69 (12) 13.1-15.1 — 14.9; 15.0 G 10.36 + 0.50 (12) 9.6-11.1 — 12.0; 12.1 H 13.18 + 0.68 (12) 12.4-14.2 — 13.2; 13.3 I 12.77 + 0.57 (12) 12.0-13.6 — 12.0; 12.5 J 5.83 + 0.29 (12) 5.4—6.5 — 6.6; 6.6 Tarsometatarsus: K 51.86 + 1.64 (12) 49.7-54.9 — 59.5 L 9.38 + 0.80 (12) 8.0-10.6 — 8.0; 8.4 M 6.80 + 0.56 (12) 5.8-7.6 — 7.8; 7.8 N 15.00 + 0.54 (12) 14.4-15.9 — 16.4 O 4.98 + 0.26 (12) 4.6-5.5 — EAs sil Pp 14.46 + 0.76 (12) 13.4-15.7 — 14.7 Q 5.49 + 0.28 (12) 5.1-6.0 — 6.7 Femur: R 3.12 + 0.27 (13) 2.8-3.6 — 3.1 S 12.14 + 0.69 (13) 11.2-12.9 — 12.4 T 6.81 + 0.56 (13) 6.0-—7.9 = 6.6 U 15.29 + 0.89 (13) 14.0-16.6 — 15.2 V 13.51 + 0.47 (13) 12.7-14.2 — 12.8 W 7.70 + 0.35 (13) 7.2-8.3 — U2? Comparisons and description.—Unless otherwise stated, all comparisons are made in relation to 7 specimens (Pierce Brodkorb collection) of Pandion haliaetus carolinensis Gmelin, 1788. Femur.—Pandion lovensis n. sp. has caudal intermuscular line more mediad, merging smoothly with the crista supracondylaris medialis, forming a sharp caudo- medial border immediately above the medial epicondyle. Caudal aspect of the medial condyle broader. Popliteal fossa slightly broader. Caudal aspect of lateral condyle (i.e., tibial articular surface) extending less craniad and is not inclined laterad. Crista tibiofibularis and lateral epicondyle less pronounced. Patellar sulcus slightly broader. Tibiotarsus.—Fibular crest shorter. Both anterior and posterior intercondylar VOLUME 98, NUMBER 2 Sil7/ Fig. 1. Stereophotographs of tarsometatarsi of Pandion lovensis n. sp. A, Paratype UF 25863, distal view. B, C. Holotype, UF 25950. B, Caudal view; C, Cranial view. Scale equals 10 mm. (A); 20 mm. (B, C). sulci wider and less deep. Both lateral and medial epicondylar depressions deeper and more distinct. Cranial opening of extensor canal larger and more transversely oriented. Internal ligamental prominence more distinct. Distal end wider, espe- cially caudal portion. The referred proximal end of tibiotarsus of Pandion homalopteron (LACM 42815), when compared with P. haliaetus, is robust, with a greater transverse width of proximal end. Crista cnemalis lateralis slightly elongated, producing a more pronounced incisura tibialis and a broader sulcus intercristalis. Facies gas- 318 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Paratypes of Pandion lovensis. A, Caudal view femur UF 25766; B, Cranial view tibiotarsus UF 25928; C, Caudal view tibiotarsus UF 25928. Scales = 10 mm. trocnemalis flatter. Distinct notch present on caudo-lateral margin of proximal articulating surface. Slight ridge extends proximad from tuberositas poplitea. In comparison with Pandion homalopteron, P. lovensis has a smaller transverse width, a less distinct notch on the caudo-lateral margin of the proximal end, and a distinctly deeper fossa retrocristalis. Tarsometatarus. — Shaft longer, more slender, and less flattened dorso-plantarly. Fossa parahypotarsalis medialis slightly more excavated. Crista medialis hypotarsi extends a proportionally shorter distance down shaft. Lateral foramen vascularia proximalia opens within hypotarsal canal (outside in all specimens of Pandion haliaetus examined). Fovea ligamentum collateralis on trochlea IV larger and deeper. Trochlea IV less recurved, anterior surface flattened. Trochlea III larger. Medial foramen vascularia proximalia proximal to origin of inner strut of arcus extensoris. Fossa infracotylaris dorsalis deeper. Medial border of trochlea III projects laterad (dorsad in P. haliaetus). Distal end larger. In medial view, caudal process on trochlea II proportionally longer. VOLUME 98, NUMBER 2 319 Discussion. — The power-arm ratio of the tarsometatarsus has been the focus of many investigations (Miller 1911, 1912; Howard 1932; Jollie 1976-1977; among others). The major flexor of the tarsometatarsus on the shank is the M. tibialis anterior (=M. tibialis cranialis of Baumel et al. 1979) and, to a lesser degree, the M. extensor digitorum longus. In Pandion haliaetus, the tibialis anterior arises by two heads, a tibial head on the anterior side of the tibial crest, extending in a narrow line down the medial side of the tibial shaft; and a femoral head extending from the distal apex of the external condyle of the femur. This muscle inserts by a single tendon on the tibialis anterior tuberosity on the proximal end of the tarsometatarsus (Hudson 1937, 1948). The power-arm ratio (Miller 1912, 1925) is calculated by dividing the length from the proximal end to the midpoint of the tibialis anterior tubercle (=power- arm) multiplied by 100, by the total length of the tarsometatarsus (=resistance or weight-arm). Miller (cited in Jollie 1976-1977) noted that species with long tarsi have a short power-arm ratio while those species with a short broad tarsus have a relatively large ratio. Miller (1911), Howard (1932), and Jollie (1976-1977) provide tables of power-arm ratios for comparison. It is interesting to note that the modern osprey has the greatest power-arm ratio (32.2%) of any accipitriform species. Pandion lovensis has a much smaller power-arm ratio (17.0 mm/59.5 mm X 100 = 28.6%). This is approximately 11% less than the modern osprey. An increase in length of the tarsometatarsus, without a concomitant shift in the position of the tibialis anterior tubercle is responsible for the decrease in the power-arm ratio in P. /ovensis. The increase in length would also allow the distal end of the tarsometatarsus to be moved at a faster rate, all other things being equal. The interpretation of these differences is difficult. Fisher (1945:742) states ““The development of this great flexor of the tarsus may be correlated with ability to walk or run, ability to grasp with the foot as in perching or in predation, and with weight of the foot or of the entire body. In fact it is impossible to define and distinguish individual adaptations.” Because P. homalopteron, P. lovensis, and P. haliaetus have only one known skeletal element in common, any proposed phylogeny is tenuous. Pandion hom- alopteron is not very distinct in wing morphology from the modern osprey, even though a large interval of time separates them (Warter 1976). The only known hindlimb element of this species, a proximal end of a tibiotarsus, also appears close to that of the modern osprey. Pandion lovensis is less derived than either of these species and shares a number of characters with the Accipitridae, the proposed sister taxa of the Pandionidae (Jollie 1976-1977). These characters include a femur with a broader and less deep patellar sulcus, and the caudal intermuscular line medial; a tibiotarsus with broader intercondylar sulci; and a tarsometatarsus which is longer and less broad, with a reduced power-arm ratio. Pandion lovensis appears to be the least derived member of the genus and represents a lineage distinct from that of P. homalopteran and P. haliaetus. Acknowledgments I thank Pierce Brodkorb, Department of Zoology, University of Florida and the personnel of the Division of Ornithology, Florida State Museum for loan of skeletal specimens. S. David Webb and B. J. MacFadden, Division of Vertebrate 320 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Paleontology, Florida State Museum, University of Florida, and L. G. Barnes, Los Angeles County Museum of Natural History, made fossil specimens or casts available for study. I especially thank R. G. Wolff, Hildegarde Howard, Storrs L. Olson, Cécile Mourer-Chauviré and P. Brodkorb for their comments on the manu- script. Literature Cited Baumel, J. J., King, A. S., Lucas, A. M., Beazile, J. E., and Evans, H. E. (Eds.). 1979. Nomina Anatomica Avium. Academic Press, New York. 637 pp. Brunet, J. 1970. Oiseaux de l’éocéne Supérieur du Bassin de Paris.—Annales de Paleontologie (Vertébrés) 56:1—57, plates A-D. Compton, L. V. 1938. The pterylosis of the Falconiformes with special attention to the taxonomic position of the Osprey.— University of California Publications in Zoology 42(3):173-212. Dixon, W.J. 1981. BMDP Statistical Software.— University of California Press. Berkeley, California. 725 pp. Fisher, H. I. 1945. Locomotion in the fossil vulture Teratornis.—American Midland Naturalist 33(3):725-742. Howard, H. 1929. The Avifauna of Emeryville Shellmound.— University of California Publications in Zoology 32:301-394. 1932. Eagles and eagle-like vultures of the Pleistocene of Rancho La Brea.—Carnegie In- stitution of Washington, Publication 429:1—-82, 29 pls. Hudson, G. E. 1937. Studies on the muscles of the pelvic appendage in birds.—American Midland Naturalist 18(1):1-108. 1948. Studies on the muscles of the pelvic appendage in birds. II. The heterogenous order Falconiformes.— American Midland Naturalist 39(1):102—127. Jollie, M. 1976-1977. A contribution to the morphology and phylogeny of the Falconiformes. — Evolutionary Theory 1:285-—298, 2:115-300, 3:1-141. Miller, L. 1911. A series of eagle tarsi from the Pleistocene of Rancho La Brea.—University of California Publications, Bulletin of the Department of Geology 6(12):305-316. 1912. Contributions to avian paleontology from the Pacific Coast of North America.— University of California Publications, Bulletin of the Department of Geology 7(5):61—-115. . 1925. The birds of Rancho La Brea.—Carnegie Institution of Washington, Publication 349: 63-106, 6 pls. Repenning, C. A., and R. H. Tedford. 1977. Otarioid seals of the Neogene.—Geological Survey Professional Paper 992, 93 pp. Sibley, C. G., and J. E. Ahlquist. 1972. A comparative study of the egg white proteins of non- passerine birds.— Peabody Museum of Natural History, Yale University Bulletin 39, vi + 276 pp. Warter, S. 1976. A new Osprey from the Miocene of California (Falconiformes: Pandionidae).— Smithsonian Contributions to Paleobiology 27:133-139. Webb, S. D., B. J. MacFadden, and J. A. Baskin. 1981. Geology and paleontology of the Love Bone Bed from the late Miocene of Florida.— American Journal of Science 281:513-544. Williams, K. E., D. Nichol, and A. F. Randazzo. 1977. The geology of the western part of Alachua County, Florida.—Florida Department of Natural Resources, Bureau of Geology, Report of Investigations No. 85, 98 pp. Department of Zoology, University of Florida, Gainesville, Florida 32611. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 321-331 NEOTROPICAL MONOGENEA. 7. PARASITES OF THE PIRARUCU, ARAPAIMA GIGAS (CUVIER), WITH DESCRIPTIONS OF TWO NEW SPECIES AND REDESCRIPTION OF DAWESTREMA CYCLOANCISTRIUM PRICE AND NOWLIN, 1967 (DACTYLOGYRIDAE: ANCYROCEPHALINAE) Delane C. Kritsky, Walter A. Boeger, and Vernon E. Thatcher Abstract.— Dawestrema cycloancistrium Price and Nowlin, 1967, D. cycloan- cistrioides n. sp., and D. punctatum n. sp. are reported and described from the pirarucu, Arapaima gigas (Cuvier), collected from the Solimoes River near Man- aus, Amazonas, Brazil. An emended generic diagnosis of Dawestrema Price and Nowlin, 1967, is presented incorporating new information on internal anatomy and structure of the copulatory complex and haptor. A summary of the parasitic helminths reported from A. gigas is provided. The pirarucu, Arapaima gigas (Cuvier), Osteoglossidae, inhabits the Amazon River drainage, the western Orinoco and the Rupununi and Essequibo river sys- tems of the Guianas, and is thus limited in its distribution to parts of northern South America. This fish is of considerable economic importance and is regarded as a food fish of the highest quality. Studies on the parasites of the pirarucu were begun during the early nineteenth century with the description of larval Gnathostoma gracilis by Diesing (1838). Including that record, a total of 14 helminth species have been reported from this fish. Baylis (1927) listed the following (names appearing below are as given by Vicente and Pinto 1981, Rego et al. 1974, and Noronha 1981): Nematoda— Goezia spinulosa (Diesing, 1839), Terranova serrata (Drasche, 1884),* Camal- lanus tridentatus (Drasche, 1884), and Gnathostoma gracilis (Diesing, 1838); Acanthocephala— Polyacanthorhynchus macrorhynchus (Diesing, 1856); and Ces- todaria—Schizochoerus liguloideus (Diesing, 1850), and Nesolecithus janickii Poche, 1922. Baylis (1927) also described the nematode, Philometra senticosa (Nilonema senticosa in Vicente and Pinto 1981), from the swim-bladder of the fish. Machado Filho (1947) reported Polyacanthorhynchus rhopalorhynchus (Dies- ing, 1851) and P. macrorhynchus (Diesing, 1856) from A. gigas; and Travassos (1960) reported the nematode, Rumai rumai, as a new species from the host’s body cavity. Prudhoe (1960) described Caballerotrema brasiliense (Trematoda) * According to Baylis (1927), Porrocaecum draschei (Stossich, 1896) is a synonym of this species since Ascaris serrata “was very briefly described by v. Drasche (1884), who had before him only a single male specimen. Stossich (1896), who changed the name of the species to A[scaris]. draschei (apparently on the ground of the previous existence of Ancyracanthus serratus Wright, 1879, which he believed to be an Ascaris, but which is probably to be referred to Cystidicola), added nothing to the description.” 322 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON from the intestine of the host. This species was redescribed, and C. arapaimense proposed as a new species from pirarucu by Thatcher (1980). Himasthla piscicola, described by Stunkard (1960) from the intestine of the host, is likely a synonym of one of the Caballerotrema species. The only monogene previously reported from A. gigas is Dawestrema cycloancistrium Price and Nowlin, 1967. The present study adds two new species to the parasite fauna of this host and includes a redescription of D. cycloancistrium. Materials and Methods The host was collected from the Solimoes River near Manaus, Amazonas, Brazil, on 15 April 1983. Fish gills were placed in finger bowls and covered with a 1:4000 formalin solution. After one-half hour, the gills were agitated in this liquid and then removed from the bowl. The helminths were allowed to settle to the bottom and were subsequently removed with the aid of a small probe and dissecting microscope. They were immediately fixed and stored in AFA. Some were mounted unstained in Gray and Wess’ medium for study of sclerotized structures. Other specimens stained with Semichon’s carmalum or Gomor1’s tri- chrome were used to determine internal features. Measurements were taken only on specimens collected during the present study; all, in micrometers, were made according to the procedures of Mizelle and Klucka (1953) except that the cirrus measurement is the diameter of the largest ring of the coil. Numbering of hook pairs follows that proposed by Mizelle (1936). Illustrations were prepared with the aid of a camera lucida or microprojecter. Type-specimens are deposited in the collections of the Instituto Nacional de Pesquisas da Amazonia (INPA), the U.S. National Museum Helminthological Collection (USNM), and the University of Nebraska State Museum (UNSM) as indicated below. Species of many genera of Monogenea from the Neotropical Region are char- acterized by a cirrus comprising a variably developed base from which a coiled shaft arises. The coil of the shaft may consist of less than one complete ring to many rings. While the direction of the coil has not been specifically addressed in previous studies on Neotropical forms, a counterclockwise or clockwise coil may be shown to have diagnostic value at the specific and/or generic level as under- standing of the Monogenea from this region develops. Thus, the coil direction is determined by viewing the cirrus in ventral view. If the cirrus shaft is directed in a clockwise direction from the base to the ventral tip of the shaft, the rings are defined to have a clockwise direction, and conversely so, counterclockwise. In the present study, all species of Dawestrema were found to have counterclockwise rings, and this character is incorporated into the emended diagnosis of the genus. Dawestrema Price and Nowlin, 1967 Emended diagnosis.—Dactylogyridae, Ancyrocephalinae. Body elongate, di- visible into cephalic region, trunk, peduncle, and haptor. Tegument thin, smooth. Head organs, cephalic glands present. Four eyes. Mouth subterminal; pharynx muscular, glandular; esophagus elongate; intestinal crura 2, confluent in posterior trunk, lacking diverticulae. Gonads intercecal, slightly overlapping or tandem; testis postovarian. Vagina sinistroventral; seminal receptacle overlapping or lying immediately anterior to ovary along body midline; uterus delicate; genital pore VOLUME 98, NUMBER 2 323 midventral. Vitellaria well developed as 2 bilateral bands coextensive with gut; eggs with terminal filament(s). Vas deferens looping left intestinal crus; 2 seminal vesicles, simple dilations of sperm duct; prostatic reservior present; cirrus com- prising a coil of one to many counterclockwise rings; accessory piece articulated to cirral base, proximal part lying within cirrus coil, terminal portion serving as cirrus guide. Haptor with 2 pairs of anchors (dorsal and ventral), dorsal and ventral bars (nonarticulated), 7 pairs of hooks. Hook pairs 1, 2, 3, 4, 6, 7 arranged concentrically around haptor usually anterior to anchors. Ventral bar with medial anterior projection arising from posterior or dorsal margin of bar. Parasitic on gills of freshwater fishes of the Osteoglossidae. Type-species, host, and locality.— Dawestrema cycloancistrium Price and Now- lin, 1967, from the gills of Arapaima gigas (Cuvier), from Amazon River and tributaries, Brazil. Other species. — Dawestrema cycloancistrioides n. sp., D. punctatum n. sp., both from the gills of Arapaima gigas, Solimoes River near Manaus, Amazonas, Brazil. Dawestrema cycloancistrium Price and Nowlin, 1967 Figs. 1-11 Specimens studied. — Holotype, USNM 62985; voucher specimens, INPA 00000, USNM 78224, UNSM 22092. Description (based on 22 specimens and holotype).— Body graciliform; length 2180 (1618-2552), greatest width 112 (83-138) near level of vagina or cephalic region. Cephalic margin rounded, lobes poorly developed; head organs comprising loosely associated cephalic gland ducts; postpharyngeal bilateral groups of uni- cellular cephalic glands well developed. Each eye with lens; members of posterior pair of eyes larger, farther apart than those of anterior pair; eye granules small, variable in shape; accessory granules absent. Pharynx spherical, 62 (45-72) in diameter; gut confluent apparently at several locations posterior to testis. Peduncle elongate, with pair of conspicuous glands; haptor bulbous, with anchors situated on posterior lobe; haptor 156 wide, 129 (86-159) long. Anchors similar; ventral anchor robust, with elongate straight point, heavy base, ornate deep root, super- ficial root with conspicuous saddle-like fold; ventral anchor 35 (33-37) long, base 25 (23-28) wide. Dorsal anchor with curved point and shaft, fold of superficial root weakly developed; dorsal anchor 51 (47-54) long, base 35 (29-43) wide. Anchor filament simple, double, conspicuous. Ventral bar plate-like, with anterior medial projection arising near posterior margin; bar 25 (21—29) long. Dorsal bar with globose ends, heavy ridge along posterior margin; bar 38 (28—43) long. Hook distribution typical with ring of hooks well anterior to anchors; hook 16 (13-18) long; pairs 1, 2, 3, 4, 6, 7 similar; each with depressed thumb, shank with small terminal enlargement, point recurved, FH loop % shank length; pair 5 delicate, with small enlargement of proximal end of shank, FH loop 2 shank length. Testis Ovate, postovarian, 29 (21-37) wide, 57 (46-74) long; seminal vesicles fusiform; prostatic reservoir elongate, thin walled, frequently twisted. Cirrus with 5-7 rings, slightly enlarged base, largest ring diameter 42 (37—48); accessory piece terminally enclosing citrus shaft. Ovary elongate ovate, 28 (24-33) wide, 108 (94-121) long; oviduct short; ootype not observed; uterus (when empty) a delicate duct with terminal expansion, extending to right of midline; genital pore at level of cirrus; 324 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 00s VOLUME 98, NUMBER 2 325 vagina tubular, proximally coiled, with distal petal-shaped sclerotization pro- truding from aperture; seminal receptacle pyriform, lying anterior to vagina coil. Vitellaria appearing as hollow tubular longitudinal structures. Egg elongate ovate, with proximal filament; filament exceptionally long; egg 136 (126-147) by 34 (29- 37). Remarks. —Examination of the holotype of Dawestrema cycloancistrium and comparison of it with our specimens showed them to be conspecific. The holotype, mounted unstained in glycerine jelly, is apparently contracted as a result of fixation procedures, which explains the significant differences of body shape and size between our specimens and measurements and drawings provided by Price and Nowlin (1967). Further, we determined several erroneous interpretations in the original description concerning the internal anatomy and structure of the scler- otized haptoral armament and copulatory complex. These are evaluated in the discussion below. Dawestrema cycloancistrioides, new species Figs. 12-20 Type-specimens.— Holotype, INPA 00000; paratypes, INPA 00000, USNM 78222, UNSM 22093. Description (based on 24 specimens).— Body cylindrical, tapered at extremes; length 1281 (1015-1766), greatest width 92 (73-118) at or posterior to gonads. Two terminal, 2 subterminal cephalic lobes well developed; head organs distinct in each lobe; cephalic glands well developed, comprising prepharyngeal, pharyn- geal, and postpharyngeal bilateral paired groups of unicellular glands; bilateral pair also lying immediately anterior to level of vagina. Members of posterior pair of eyes with lenses, larger and farther apart than those of anterior pair; eyespot granules small, variable in size and shape; accessory granules usually restricted to immediate region of eyes. Pharynx spherical, 53 (40-60) in diameter. Peduncle moderately elongate; haptor bulbous, tapered posteriorly with ventral anchors situated on posterior lobe; haptor 83 (60-107) wide, 82 (65-122) long. Anchors dissimilar; ventral anchor robust, with evenly curved point and shaft, exaggerated and truncate superficial root, small deep root; anchor 49 (46-51) long, base width 28 (24-30). Dorsal anchor with delicate, evenly curved point and shaft, vestigial deep root, tapered superficial root; anchor 32 (29-34) long, base width 22 (17- 28). Anchor filament simple, double, conspicuous. Ventral bar plate-like, with anterior projection originating from posterior margin of bar; bar 21 (20-23) long. Dorsal bar broadly U-shaped with slightly enlarged ends; bar 28 (26-30) long. Hook pairs 1, 2, 3, 4, 6, 7 distributed around the widest part of the haptoral bulb at level of dorsal anchor bases; hooks 1, 2, 3, 4, 6, 7 similar, each with two-part shank, slightly depressed thumb, tapered shaft and point; hook pair 5 with slender — Figs. 1-11. Dawestrema cycloancistrium: 1, Composite drawing of whole mount (ventral); 2, Cop- ulatory complex; 3, Enlargement of worm at level of reproductive systems (ventral); 4, Vagina; 5, Hook (pair 1); 6, Hook (pair 5); 7, Ventral bar; 8, Dorsal bar; 9, Ventral anchor; 10, Dorsal anchor; 11, Egg. All figures are reproduced to the same scale (30 micrometers) except Figs. 1, 11 (S00 mi- crometers, 100 micrometers, respectively). 326 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 12 20 Figs. 12-20. Dawestrema cycloancistrioides: 12, Ventral view of holotype; 13, Vagina; 14, Cop- ulatory complex; 15, Ventral bar; 16, Hook (pair 1); 17, Hook (pair 5); 18, Dorsal bar; 19, Dorsal anchor; 20, Ventral anchor. All figures are drawn to the same scale (30 micrometers) except Fig. 12 (200 micrometers). VOLUME 98, NUMBER 2 327 shank and small proximal enlargement; hook length 19 (17-23). FH loop extending to union of 2 parts of shank. Cirrus a coil of 7-8 rings, with flared base, largest ring diameter 33 (28-37); accessory piece terminally enclosing cirral shaft. Gonads overlapping. Testis dorsoposterior to ovary, elongate ovate, 16 (14-19) wide, 64 (43-85) long; seminal vesicles stout, fusiform, anterior vesicle larger; prostatic reservoir a delicate pyriform sac lying near level of anterior seminal vesicle; prostrate not observed. Ovary spindle-shaped, elongate, 34 (20—47) wide, 77 (74— 79) long; oviduct short; ootype not observed; uterus delicate, extending along ventral midline opening into midventral genital atrium by darkstaining structure apparently functioning as sphincter; vagina comprising a distal funnel with elon- gate twisted tube, opening into inconspicuous seminal receptacle; vitellaria con- fluent posteriorly, vitelline commissure lying at level immediately anterior to vaginal coils; egg ovate with proximal short filament, 80 (61—98) by (45-46). Remarks. — Dawestrema cycloancistrioides most closely resembles D. cycloan- cistrium, as shown by the comparative morphology of the copulatory complex. The new species differs from D. cycloancistrium by having 1) a smaller body size, 2) a vaginal tube without a tight proximal coil, 3) a short anteromedial process of the ventral bar, 4) robust hooks (except pair 5), 5) anchors of a different morphology, and 6) a short proximal egg filament. The specific name, from Greek, indicates the relationship of these two species. Dawestrema punctatum, new species Figs. 21-29 Type-specimens.— Holotype, INPA 00000; paratypes, USNM 78223, UNSM 22094. Description (based on 5 specimens).— Body spindle shaped; length 852 (796— 923), greatest width 86 (65-114) near midlength or in anterior half. 2 terminal, 2 bilateral cephalic lobes well developed; some specimens with an incipient lobe between major pairs. Head organs well developed, one in each cephalic lobe; cephalic glands not observed. Members of posterior pair of eyes larger, closer together than those of anterior pair, lens usually visible in posterior pair; eye granules subspherical, small, variable in size; accessory granules generally absent (eye granules frequently disassociated in flattened specimens). Pharynx spherical, 40 (37-43) in diameter; esophagus moderate in length; gut apparently normal. Peduncle elongate to moderately long; haptor bulbous, tapered posteriorly, with ventral anchors situated on posterior lobe; haptor 108 (98-119) wide, 89 (75- 106) long. Anchors similar, each with well-developed base, roots moderately developed, elongate point with terminal recurved tip, ventral anchor 54 (53-55) long, base width 30 (28-32); dorsal anchor 41 (35-45) long, base width 25 (21- 30). Anchor filament simple, double, inconspicuous. Ventral bar subrectangular, with anterior projection arising from dorsal surface, bar 30 (26-33) long; dorsal bar rod-shaped with slightly enlarged ends, bar 36 (32-39) long. Hook pairs 1, 2, 3, 4, 6, 7 lying in ring at level of dorsal anchor bases, similar in shape, 21 (18- 24) long, each with slender distal shank, proximal shank enlarged, thumb erect, point recurved; hook pair 5 with slender shank and small proximal enlargement, 13 (12-14) long. FH loop extending to union of 2 parts of shank. Cirrus a coil of about 1% rings, base of cirrus lying ventral in body, with first ring of shaft directed dorsally, termination of shaft curved ventrally, ring diameter 39 (37—41); accessory 328 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 28 29 27 Figs. 21-29. Sclerotized parts of Dawestrema punctatum: 21, Ventral bar; 22, 23, Dorsal bars; 24, Vagina; 25, Copulatory complex; 26, Hook (pair 1); 27, Ventral anchor; 28, Hook (pair 5); 29, Dorsal anchor. All figures are at the 30 micrometer scale. piece variable, terminally enclosing cirrus shaft. Gonads overlapping. Testis elon- gate, dorsoposterior to ovary; loop of vas deferens immediately posterior to va- gina; seminal vesicles c-shaped; prostatic reservoir with outer circular muscles and enlarged duct. Limits of ovary not observed; oviduct short; ootype not ob- served; uterus delicate, frequently containing single egg; genital pore midventral at level of cirrus; vagina a short sclerotized tube flaring proximally and opening simply at left margin; vitellaria generally distributed throughout trunk, commis- sure anterior to seminal receptacle; egg elongate ovate, with terminal short filament at each pole; egg 132 long, 32—33 wide. Remarks.—This species differs significantly from both Dawestrema cycloan- cistrium and D. cycloancistrioides in the comparative morphology of the haptoral armament, copulatory complex, vagina, and egg. However, it is apparently closest to D. cycloancistrioides based on the similar nature of the superficial roots of the anchor bases. The specific name is from Latin (punctata = thorn) and refers to the recurved tips of the anchor points. Discussion Price and Nowlin (1967) proposed Dawestrema for monogenes characterized by having a circle of hooks located well anterior to the remainder of the haptoral armament, two prostatic reservoirs, an intercecal vas deferens, and apparently VOLUME 98, NUMBER 2 329 non-confluent intestinal crura. However, our examination of the holotype con- firmed that one of our forms was the type-species, D. cycloancistrium, and that errors in the original study had been made concerning the structure of the cop- ulatory complex, ventral bar, and internal anatomy which necessitated a redefi- nition of the genus. While the hook arrangement in Dawestrema cycloancistrium is as originally described, our discovery of how two new species show that the anterior position of the ring of hooks is only a specific trait. In D. cycloancistrioides and D. punc- tatum, the hook ring is at the level of the bases of the dorsal pair of anchors. Hook pairs | (submedial), 2, 3, and 4 are ventral, each pair situated more laterally, respectively. Pairs 6 and 7 continue the ring on the dorsal surface of the haptor with pair 7 being nearest the midline. In those species in which the hook ring is at the level of the dorsal anchor bases, pair 7 lies lateral to the anchors. Pair 5 (ventral) lies between the shafts and points of the more posterior ventral anchors. All known species of Dawestrema possess a medial, anteriorly directed projec- tion on the ventral bar. Price and Nowlin (1967) indicate in their fig. 8 that the origin .of the process is the anterior bar margin, while in all of our specimens, the process originates from the posterodorsal or posterior bar surfaces. Availability of three species of Dawestrema, some specimens of which were stained for study of internal structure, provided an opportunity to clarify mor- phologic features of the reproductive and digestive systems. Our findings differed from the original description as follows: (1) the presence of two tandemly arranged seminal vesicles (one described by Price and Nowlin 1967, which probably rep- resents the anterior expanded part of the uterus); (2) a vas deferens looping the left intestinal crus (intercecal in the original description); (3) one prostatic reservoir (two in Price and Nowlin 1967); (4) confluent intestinal crura (Price and Nowlin apparently mistook the hollow nature of the bilateral vitelline bands in D. cy- cloancistrium as intestine); and (5) a testis smaller than ovary. Lastly, our specimens clearly show that the proximal portion of the accessory piece (connecting piece of Price and Nowlin 1967) is located within the rings of the cirrus coil and not external to them as originally described. This configuration of the copulatory complex, along with observations on specimens in which the terminal portion of the cirral shaft was partially protruded from the genital pore, provides some insight to the functional morphology of this structure during cop- ulation. In specimens with partly protruded cirri, the cirrus base may have the bottom surface facing ventrally, while in others this surface faces dorsally. In addition, the diameter of the cirral rings does not change appreciably in specimens with extruded cirri. Since the terminal part of the accessory piece, serving as a cirrus guide, is apparently fixed in location at the genital pore or within the genital atrium, these observations suggest that protrusion of the cirrus during copulation is not a result of a tightening of the cirrus coil, but rather an unwinding of the cirral shaft with the cirrus base rotating as the shaft is extruded. The twisted nature of the proximal part of the accessory piece, its articulation with the base of the cirrus, and the supposed fixed location of the terminal part of the accessory piece in the genital atrium, suggest that the accessory piece serves as a means of retracting the cirrus shaft much the same as the spring of a carpenter’s tape measure. This functional aspect, based on observed structure of fixed specimens, however, is assumed as the mechanism employed, and confirmation will depend on ob- 330 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON servation of living specimens and/or analysis of the composition of the proximal part of the accessory piece. This is the first time, nonetheless, that insight into mechanical function of these structures has been available. Price and Nowlin (1967) suggested that the long cirrus shaft of D. cycloancis- trium provides a survival advantage for copulation over some distance. However, comparison of the structure of the vaginae and the length of the cirral shafts of respective species in the genus would not suggest that this occurs. In D. cycloan- cistrium and D. cycloancistrioides, the elongate cirral shaft would be necessary to reach the seminal receptacle through the long, tortuous, and/or coiled vaginal tube, while in D. punctatum, the comparatively short cirral tube could effectively reach the seminal receptacle since the vaginal tube is correspondingly short. Sim- ilar relationships in cirral and vaginal tube lengths occur in other monogeneans (e.g., Gonocleithrum aruanae and G. cursitans, see Kritsky and Thatcher 1983). Acknowledgments The authors wish to express thanks to Dr. Ralph Lichtenfels, USNM, for loan of the holotype of Dawestrema cycloancistrium. The assistance of the personnel of SUDEPE in Manaus during collection of hosts is gratefully acknowledged. This study was supported, in part, by a grant (#523) from the Faculty Research Com- mittee, Idaho State University, Pocatello, Idaho. Literature Cited Baylis, H. A. 1927. Some parasitic worms from Arapaima gigas (teleostean fish) with a description of Philometra senticosa n.sp. (Filarioidea).— Parasitology 19:35—47. Diesing, K. M. 1838. Abbildungen neuer Gattungen brasilianischer Binnenwiirmer (Entozoén).— Amtlicher Bericht tiber die Versammlung Deutscher Naturforscher und Aerzte:189. Drasche, R. von. 1884. Revision der in der Nematoden-Sammlung des k.k. zoologischen Hofca- binetes befindlichen Original-Examplare Diesing’s und Molin’s.— Verhandlungen der Kaiser- lich-K6niglichen Zoologische-Botanischen Gesellschaft in Wien 28:107-118. Kritsky, D. C., and V.E. Thatcher. 1983. Neotropical Monogenea. 5. Five new species from the aruana, Osteoglossum bicirrosum Vandelli, a freshwater teleost from Brazil, with the proposal of Gonocleithrum n. gen. (Dactylogyridae: Ancyrocephalinae).— Proceedings of the Biological Society of Washington 96:581-597. Machado Filho, D. A. 1947. Revisao do genero ““Polyacanthorhynchus” Travassos, 1920 (Acantho- cephala, Rhadinorhynchidae).— Revista Brasileira de Biologia 7(2):195—201. Mizelle, J.D. 1936. New species of trematodes from the gills of Illinois fishes. American Midland Naturalist 17:785-806. , and A. R. Klucka. 1953. Studies on monogenetic trematodes. XIV. Dactylogyridae from Wisconsin Fishes.— American Midland Naturalist 49:720-733. Noronha, D. 1981. Acanthocephala. Jn Hurlbert, Rodrigues and Santos, eds., Aquatic Biota of Tropical South America. Part 2. Anarthropoda.—San Diego State University, San Diego, Cal- ifornia: 146-150. Price, C. E., and W. J. Nowlin. 1967. Proposal of Dawestrema cycloancistrium n. gen. n. sp. (Trem- atoda: Monogenea) from an Amazon River host.—Rivista di Parassitologia 28:19. Prudhoe, S. 1960. On two new parasitic worms from Brazil.—Libro Homenaje Eduardo Caballero y Caballero:415—421. Rego, A. A., Santos, J. C., and P. P. Silva. 1974. Estudos dos cestoides de peixes do Brasil.— Memorias do Instituto Oswaldo Cruz 72:187-204. Stossich, M. 1896. Il Genere Ascaris Linné. Lavoro monografico.—Bolletino Societa Adriatica di Scienze Naturali in Trieste 17:9-120. VOLUME 98, NUMBER 2 331 Stunkard, H. W. 1960. Further studies on the trematode genus Himasthla with descriptions of H. mcintoshi n. sp., H. piscicola n. sp., and stages in the life-history of H. compacta n. sp.— Biological Bulletin 119:529-549. Thatcher, V. E. 1980. Duas novas espécies de Caballerotrema (Trematoda: Echinostomatidae) do pirarucu e do aruana (Osteoglossidae), com uma redefinigao do género e uma redescricao de C. brasiliense Prudhoe, 1960.—Acta Amazonica 10:419-423. Travassos, L.P. 1960. Sébre nematoides cavitarios de peixes do Rio Amazonas.—Atas da Sociedade de Biologia do Rio de Janeiro 4:15—20. Vicente, J. J., and R. M. Pinto. 1981. Nematoda. Zooparasitic forms. Jn Hurlbert, Rodrigues and Santos, eds., Aquatic Biota of Tropical South America. Part 2. Anarthropoda.—San Diego State University, San Diego, California: 136-145. (DCK) Department of Allied Health Professions and Idaho Museum of Natural History, Idaho State University, Box 8002, Pocatello, Idaho 83209; (WAB and VET) Instituto Nacional de Pesquisas da Amazonia, Manaus, Brazil. Note added in proof: Specimens for deposition in the helminthological collection at INPA were shipped on 31 May 1984 and to this date (11 March 1985) have not been received in Manaus. We assume that they have been lost, at least temporarily, and thus no INPA numbers have been assigned. Lost specimens include the holotypes of D. cycloancistrioides and D. punctatum, 6 paratypes of D. cycloancistrioides, and 6 vouchers of D. cycloancistrium. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 332-340 FOUR NEW SPECIES OF SCALIBREGMATIDAE (POLYCHAETA) FROM THE GULF OF MEXICO, WITH COMMENTS ON THE FAMILIAL PLACEMENT OF MUCIBREGMA FAUCHALD AND HANCOCK, 1981 Jerry D. Kudenov Abstract.—Four new species of Scalibregmatidae (Polychaeta) are described from the Gulf of Mexico: Asclerocheilus mexicanus, Hyboscolex quadricincta, Neolipobranchius blakei, and Sclerocheilus unoculus. Mucibregma Fauchald and Hancock, 1981, is probably referable to the Fauveliopsidae. Recent systematic studies on scalibregmatid polychaetes have resulted in a generic review and classification (Kudenov and Blake 1978), and subsequent clarification and refinement of this scheme (Blake 1981). In all, 15 genera and 46 species of scalibregmatids have been described (Blake 1981; Fauchald and Han- cock 1981), not including the undescribed genus represented by Asclerocheilus californicus Hartman (Blake 1981). Four new species are described here, including Asclerocheilus mexicanus, Hyboscolex quadricincta, Neolipobranchius blakei, and Sclerocheilus unoculus. The generic definition of Neolipobranchius is emended, and the recently described genus Mucibregma Fauchald and Hancock, 1981, is tentatively referred to the Fauveliopsidae Hartman and Fauchald 1971. This study is based on two small collections from the Gulf of Mexico. One collection derives from the Hourglass Cruises (Kudenov 1985), and was kindly made available by Thomas H. Perkins, Florida Department of Natural Resources, St. Petersburg, Florida. The other was taken as part of the Bureau of Land Man- agement’s Outer Continental Shelf Baseline Environmental Survey along the coasts of Mississippi, Alabama, and Florida (MAFLA) and southwest Florida (SOW- FLA). All MAFLA samples were collected by Dames and Moore, and SOWFLA samples by Woodward-Clyde Environmental Consultants. Details of the MAFLA survey are given by Dames and Moore (1979). These materials were generously made available by Paul G. Johnson and Joan Uebelacker, Barry Vittor and As- sociates, Mobile, Alabama (Kudenov 1984). All materials including most types are deposited in the National Museum of Natural History, Smithsonian Institu- tion, Washington D.C. (USNM); two paratypes are also deposited in the Inver- tebrate Reference Collection of the Florida Department of Natural Resources Marine Research Laboratory, St. Petersburg, Florida (FSBC I). Asclerocheilus mexicanus, new species Fig. 1 Material examined.—FLORIDA, GULF OF MEXICO: MAFLA sta 2211G, 27°56'29.5"N, 83°52'59.5”W, 43 m, Nov 1977; 1 paratype (USNM 97267). Sta 2211H, same, Jun 1976; 1 paratype (USNM 97268). Sta 2423J, 29°37'00.8’N, 84°29'58.4”W, 19 m, silty fine sand, Jul 1976; 1 paratype (USNM 97265). Sta VOLUME 98, NUMBER 2 333 Fig. 1: Asclerocheilus mexicanus (holotype, USNM 97264): A, Anterior end, dorsal view; B, Para- podium, right setiger; C, Furcate seta; D, Notoaciculum from first row of setiger 1; E, Same from second row of setiger 1; F, Detail of tip of aciculum from setiger 1; G, Pygidium, lateral view. Scale: A, B = 2 mm; D, E = 40 um; C, F = 10 um; G = 100 um. 2852E, 28°30'00.4’N, 83°29'58.4”W, 22 m, medium sand, Aug 1977; 1 paratype (USNM 97266). Sta 2528H, 29°54'58.6’N, 86°04'58.5”W, 37 m, coarse sand, no date; holotype (USNM 97264). Sta 2853C, 29°18'01.9"N, 84°19'59”W, 29 m, coarse sand, Aug 1977; 1 specimen (USNM 97269). Hourglass sta C, 27°37'N, 83°28’W, 38 m, sponge-algae-coral bottom, 5 Oct 1967; 1 paratype (FSBC I 31310). Description.— Body arenicoliform, widest anteriorly, measuring up to 10 mm long, 0.9 mm wide, for 46 segments; opaque white in alcohol. Prostomium T-shaped, wider than long, with lateral process stout, distally blunt (Fig. 1A). Eyes numbering 2 pairs, partly fused along inner margins of a given pair (Fig. 1A). Buccal segment apodous, achaetous, uni- to biannulate dorsally and laterally, fused ventrally with setiger 1 (Fig. 1A). Parapodia biramous, poorly developed, resembling small conical papillae anteriorly (Fig. 1A), and flattened lobes pos- teriorly (Fig. 1B). Setae include smooth capillaries in all rami, furcate setae having tines of unequal lengths in all rami except first notopodium (Fig. 1C), and acicular spines having recurved, distally pointed shafts with minutely hirsute distal sheaths in the notopodia of setiger 1 only (Fig. 1A, F). Acicular spines arranged in 2 transverse rows, with 5-6 spines in the first (Fig. 1A, D), and 4—5 in the second (Fig. 1A, E). Pygidium damaged, with at least 3 anal cirri (Fig. 1G). Body segments indistinctly annulate, with anterior and posterior ones normally uniannulate (sometimes biannulate); medial ones tri- or quadriannulate (Fig. 1A, B, G). Remarks. — Asclerocheilus mexicanus is most closely allied to A. acirratus (Hart- man, 1966) and A. tropicus (Blake, 1981), in having acicular spines confined to the notopodia of setiger 1. There are two rows of these spines in both A. mexicanus and A. tropicus, and 1 in A. acirratus. Asclerocheilus mexicanus differs from A. tropicus in having stout, conical prostomial processes instead of flattened lateral 334 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON wings; in having two pairs of eyes with dense ocelli, arranged obliquely and partially fused along their inner margins, rather than longitudinal lines of separate ocelli; in having acicular spines of two different sizes; in having furcate setae with parallel instead of divergent tines; in lacking body reticulae; and in having pygidial cirri. Tine ratios for the two species also differ, with 2.48 for A. mexicanus, and 2.66 for A. tropicus (Blake, 1981). Refer to Discussion for a further treatment of this character. Asclerocheilus mexicanus is also related to A. ashworthi Blake, 1981, A. ber- ingianus Uschakov, 1955, and A. heterochaetous Kudenov and Blake, 1978, in having only notoacicular spines. However, these spines are present in setigers 1— 2 in both A. ashworthi and A. beringianus, and setigers 1—4 in A. heterochaetus. Etymology.—The epithet derives from the Gulf of Mexico, and is an adjective formed from the noun Mexico. Distribution. — Northeastern Gulf of Mexico, 18-31 m. Hyboscolex quadricincta, new species Fig. 2 Material examined.—FLORIDA, GULF OF MEXICO: MAFLA sta 2852E, 28°30'00.4’N, 83°29'58.4”W, 22 m, medium sand, Aug 1977; 4 paratypes (USNM 97277). Hourglass sta B, 27°37'N, 83°07'W, 18 m, EJ-65-365, sponge-coral, 3 Dec 1965; 1 paratype (USNM 97271). Same, EJ-66-73, mud, coral, dead shell, 3 Mar 1966; 1 paratype (USNM 97272). Same, EJ-67-160, sand, rock, algae, 11 May 1967; 2 paratypes (USNM 97273). Same, EJ-67-179, sponge, coral and shell, 20 May 1967; holotype (USNM 97270), 1 paratype (USNM 97425). Sarne, EJ- 67-213, sponge, coral, algae and shell debris, 2 Jun 1967; 1 paratype (USNM 97274). Same, EJ-67-328, sponge, algae, coral, 11 Sep 1967; 1 paratype (USNM 97275). Same, EJ-67-389, shell, algae, sponge and coral bottom, 20 Nov 1967; 1 specimen (USNM 97278). Sta J, 26°24'’N, 82°28’W, 18 m, coral and sponge bottom, 6 Dec 1966; 1 paratype (USNM 97276). Same, 26°24’N, 82°25'W, EJ- 66-194, 18 m, coral and algae, 11 May 1966; 3 specimens (USNM 97279). Same, EJ-66-460, coral and sponge bottom, 6 Dec 1966; 1 paratype (USNM 97425). Same, EJ-66-469, 1 paratype (FSBC I 31388). Sta K, 26°24’N, 82°58'W, 31 m, EJ-67-383, sand, sponge bottom, 14 Nov 1967; 1 specimen (USNM 97280). Sta 12 I-BC-C, EJ-81-104, Citrus County, Crystal River, approximately 18 km at 214 from mainland, mouth of discharge canal, 0.6 m, primarily exposed limestone overlain by muddy sand, 14 Apr 1981; 1 specimen (USNM 97281). Description.— Body arenicoliform, widest anteriorly, measuring up to 13 mm long, 1.4 mm wide, for up to 69 setigers; opaque white in alcohol. Prostomium with stout lateral processes about as long as wide (Fig. 2A). Eyes numbering 2 pairs, arranged in 2 parallel to diverging rows of pigment (Fig. 2A, B). Buccal segment apodous, achaetous, uniannulate dorsally and laterally, fused ventrally with setiger 1 (Fig. 2A, B). Parapodia biramous, projecting slightly above body wall, resembling flat mounts anteriorly, becoming small conical lobes posteriorly (Fig. 2A). Spinulose capillary setae (Fig. 2C) and furcate setae with unequal tines (Fig. 2D) present in all rami. Pygidium variably developed, with 4-6 anal cirri (Fig. 2E-G). Body segments annulated with buccal segment and setigers 1, 44— 59 uniannulate; setigers 2-4 and 27-43 biannulate; setigers 5—6 triannulate; se- tigers 7-32 quadriannulate (Fig. 2A, E-G). VOLUME 98, NUMBER 2 335 Fig. 2. Hyboscolex quadricincta: A-E, holotype (USNM 97270); F, paratype (MAFLA 2852E: USNM 97277); G, paratype (EJ-66-73:USNM 97272). A, Anterior end, dorsal view; B, head region, dorsal view; C, Spinulose capillary seta, detail of shaft; D, Furcate seta; E, Pygidium, left frontolateral view; F, Same, right frontolateral view; G, Same, left frontolateral view. Scale: A-B, E-G = 0.5 mm; C, D= 10 um. Remarks. — Hyboscolex quadricincta is most closely allied to H. pacificus (Moore, 1909) in the shape of the prostomium, configuration and placement of the eyes (Berkeley and Berkeley 1952:59), and in having a variable number of anal cirri (Imajima and Hartman 1964:304). Hyboscolex quadricincta differs in having an- terior body segments quadriannulate, instead of triannulate or biannulate, in having spinulose capillaries, instead of smooth ones, and in having 4—6 anal cirri, rather than O or 6-7. It is noteworthy that furcate setae of H. quadricincta have a tine ratio of 2.4 compared to 1.7 determined from seven specimens of H. pacificus loaned by the Allan Hancock Foundation (AHF N2394). Etymology.—The epithet derives from the presence of four transverse annuli on most body segments. Distribution. —Northeastern Gulf of Mexico, 0.6—31 m. Neolipobranchius Hartman and Fauchald, 1971, emended Type-species.—Neolipobranchius glaber (as glabrus) Hartman and Fauchald, 1971, by original designation. Description.—Body maggot-like, widest anteriorly, tapering posteriorly. Pro- stomium entire, conical to bluntly conical, not incised; eyes and frontal processes absent. Buccal segment achaetous, apodous. Segments 2—3 reduced, apodous, with or without setae. Parapodia biramous from segment 4 to end of body. Parapodial lobes papillar to low-lying glandular mounds. Setae include limbate to smooth capillaries and furcate setae. Branchiae, parapodial cirri absent. Pygidium with or without anal cirri. Body segments either with transverse annuli or longitudinal striations; epidermis smooth. Remarks.—The occurrence of a new species, described below, necessitates 336 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON emending the generic definition of Neolipobranchius. The type-species, N. glaber, was recognized by Hartman and Fauchald (1971:124) as a highly modified scal- ibregmatid having an entire prostomium, and lacking segmental annuli, para- podial lobes, furcate setae, and pygidial cirri. The emended generic definition of Neolipobranchius recognizes that the latter four characters are present in the genus. However, it must be emphasized that Neolipobranchius is unique amongst scal- ibregmatids in having an entire prostomium. As such, this genus is not as strongly isolated in the family as previously sus- pected (Hartman and Fauchald 1971:124; Kudenov and Blake 1978; Blake 1981). For example, Neolipobranchius, Polyphysia Quatrefages, 1865, and Kebuita Chamberlin, 1919, are all assigned to Division II sensu Kudenov and Blake (1978: 440-441; see also Blake 1981:1157) for scalibregmatids having maggot-like bodies and lacking tentacular prostomial processes. In all, these three genera are repre- sented to date by eight described species. The latter two genera possess incised prostomia (Division IIA), while Neolipobranchius has an entire prostomium (Di- vision IIB). Mucibregma Fauchald and Hancock, 1981, may be incorrectly as- signed to the scalibregmatids and is here excluded from Division II. Refer to the Discussion where Mucibregma is treated separately. Division II species exhibit typical scalibregmatid features such as an achaetous buccal segment, branchiae, segmental annuli, epidermal tesselations, parapodia, furcate setae and pygidial cirri. However, these characters are not evenly distrib- uted amongst species of this group, are generally reduced, and may be absent in such highly modified species as Neolipobranchius glaber. Described species of Division II differ generally from those of Division I (arenicoliform scalibregmatids having prostomial tentacular processes) in lacking acicular setae, parapodial cirri, and normally also pygidial cirri. Neolipobranchius blakei, new species Fig. 3 Material examined. — FLORIDA, GULF OF MEXICO: Citrus County, Crystal River, FSBC sta I-29172, EJ-81-104, approximately 18 km at 214 from mainland, 0.6 m, mouth of discharge channel, mixed algae primarily Caulerpa species, exposed limestone overlain by muddy sand and oyster shells, 14 Apr 1981, Ap- plied Biology, Inc., colls; holotype (USNM 97283). Description.—Body maggot-like, widest anteriorly, tapering posteriorly, mea- suring 3 mm long, 0.7 mm wide for 38 setigers (Fig. 3A); opaque white in alcohol. Prostomium bluntly conical, simple, lacking both eyes and lateral processes (Fig. 3A, B). Buccal segment and segments 2-3 (=setigers 1, 2) reduced, apodous, forming complete uniannulate rings (Fig. 3B); buccal segment achaetous; segments 2-3 with notopodial fascicles reduced or absent, lacking neurosetal fascicles (Fig. 3B). Segment 4 (=setiger 3) also reduced, about 2 the size of segment 5 (Fig. 3B). Parapodia biramous from segment 4 to end of body, as low-lying mound- to papilla-shaped lobes (Fig. 3A). Setae include smooth capillaries in all fascicles, and furcate setae with unequal tines from setiger 3, the latter having thin shafts anteriorly (Fig. 3C), becoming thicker posteriorly (Fig. 3D, E). Pygidium with 5 anal cirri (Fig. 3F). Body segments annulated with buccal segment and setigers 1-3, 31-38 uniannulate; setigers 4-18 quadriannulate; setigers 19-24 distended, annuli not recognizable; setigers 25-30 biannulate. VOLUME 98, NUMBER 2 337 Fig. 3. Neolipobranchius blakei (holotype, USNM 97283): A, Entire worm, dorsal view; B, Anterior end, dorsal view; C, Furcate seta, setiger 5; D, Same, setiger 17; E, Same, setiger 29; F, Pygidium, right frontolateral view. Scale: A = 2 mm; B, F = 50 wm; C-E = 10 um. Remarks. — Neolipobranchius blakei differs from the only other described species, N. glaber (Hartman and Fauchald, 1971), in having capillary setae on segments 2-3, furcate setae, parapodia, pygidial cirri, and segmental annuli. This is the first record of the genus from a littoral habitat. Etymology.—It is an honor to name this species for James A. Blake, teacher, friend, and fellow polychaetologist. The epithet is a noun in the genitive case. Distribution. —Gulf of Mexico, 0.6 m. Sclerocheilus unoculus, new species Fig. 4 Material examined. —FLORIDA, GULF OF MEXICO: Sta 2852E, 28°30'00.4’N, 83°29'58.4”W, 22 m, medium sand, Aug 1977; 1 paratype (USNM 97287). Hourglass sta B, 27°37'N, 83°07'W, 18 m, EJ-67-111, coral, shell, sponge and algae bottom, 3 Apr 1967; holotype (USNM 97285). Same, EJ-67-370, same, 2 Nov 1967; 1 paratype (USNM 97284). Sta C, 27°37'N, 83°28’W, 37 m, EJ-67- 28, coral and sponge bottom, 20 Jan 1967; 1 paratype (USNM 97286). Description. — Body arenicoliform, widest anteriorly, tesselate, measuring up to 6 mm long, 1 mm wide without setae, for up to 46 setigers; opaque white in alcohol. Prostomium T-shaped, wider than long, with lateral processes stout, distally blunt (Fig. 4A). Eyes fused, connected medially by amorphous red pig- ment, forming a yoke across prostomium (Fig. 4A). Buccal segment apodous, achaetous, uniannulate dorsally and laterally, fused ventrally with setiger 1. Para- podia biramous, well developed, resembling flat triangular lobes anteriorly (Fig. 4A), becoming inflated, conical posteriorly (Fig. 4B, C). Dorsal cirri absent. Ven- tral cirri present in setigers 24—36, short, digitiform, inconspicuous (Fig. 4C, D). Setae include smooth capillaries in all rami; furcate setae having tines of unequal 338 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON E Fig. 4. Sclerocheilus unoculus: A-B, E-F, holotype (USNM 97285); C—D, paratype (EJ-67-28: USNM 97286). A, Anterior end, dorsal view; B, Right parapodium, setiger 16, posterior view; C, Right parapodium, setiger 29, posterior view; D, Furcate seta; E, Aciculum from setiger 1; F, Pygidium, right dorsolateral view. Scale: A = 2 mm; B—D = 50 um; E = 10 um; F = 20 um. lengths in all rami except the first notopodium (Fig. 4E); and short acicular spines having nearly straight shafts, blunt tips and clear, loose fitting hyaline sheaths in notopodia of setiger 1 (Fig. 4F). Acicula arranged in | transverse row of 4—5 spines (Fig. 4A). Pygidium with 4 long anal cirri (Fig. 4D). Body segments annulated with setigers 1-2 biannulate; setigers 3-5 triannulate; and setigers 6 to end of body quadriannulate. Remarks. —Sclerocheilus unoculus is most closely allied to Sclerocheilus mi- nutus Grube in having a single transverse row of acicular spines in the notopodia of setiger 1. Sclerocheilus unoculus differs from S. minutus in having eyes fused and connected medially across the prostomium by a yoke of pigment, triannulate anterior segments instead of biannulate ones, acicular spines having distal sheaths, and greatly reduced ventral cirri. There is also a difference in the relative tine lengths of furcate setae (tine ratio 2.1 in S. wnoculus; 1.3 in S. minutus). Etymology.—The epithet unoculus derives from Latin terms uwnus meaning one, and oculus meaning eye. It is a noun in apposition. Distribution. — Gulf of Mexico, 18-38 m. Discussion Preliminary data on tine ratios for the furcate setae of Asclerocheilus, Hybo- scolex, and Sclerocheilus (this study) corroborate findings presented by Blake (1981:1132) for five species of Asclerocheilus. For example, tine ratios for As- clerocheilus mexicanus do not coincide with values listed by Blake (1981). How- ever, some variability must be present, although it is probably small. For example, tine ratios for A. beringianus Uschakov, 1955, from the Gulf of Mexico are 1.8 (Kudenov 1985), compared to 1.7 reported by Blake (1981:1132). Although tine VOLUME 98, NUMBER 2 339 ratios are probably valid and useful taxonomic criteria, their ranges of variability must be determined before they can be used routinely. ' The fact that scalibregmatids were becoming a polyphyletic assemblage of sim- ply structured polychaetes (Fauchald 1977:43) was central to the review by Ku- denov and Blake (1978) and the recent revision by Blake (1981). Fauchald and Hancock (1981:20) restated this concern in their description of a new scalibreg- matid genus, Mucibregma. This genus is described as having a “T”’-shaped pro- stomium, an extremely muscular peristomial segment, a dorsoventrally flattened body encased in a tough mucous sheath attached to a glandular peristomial ridge, biramous parapodia having acicular spines accompanied by long capillary setae, and in lacking furcate setae. However, examination of the type-species, Muci- bregma spinosum Fauchald and Hancock, 1981, revealed significant anatomical features that are not associated with scalibregmatids. For example, the prosto- mium is not as conspicuously ““T’’-shaped in that the small prostomial processes project laterally and posteriorly (Fauchald and Hancock 1981, Pl. 3, fig. a). The extremely muscular and glandular peristomium is a character not encompassed by the family definition of scalibregmatids, and may include the presence of perhaps more than one achaetous segment in Mucibregma. Furthermore, the presence of heavy, spindle-shaped acicular spines accompanied by long, pliable capillary setae in all parapodia is simply not a scalibregmatid characteristic. Fi- nally, scalibregmatids occupy burrows, and apparently do not construct tubes (Fauchald and Jumars 1979:245) such as that made by Mucibregma. However, in light of such unusual features, Fauchald and Hancock (1981) assigned Muci- bregma to Division II scalibregmatids (sensu Kudenov and Blake 1978) for worms having maggot-like bodies and lacking distinct tentacular processes. However, Fauchald and Hancock do not define their new category “‘C,”’ and it is surmised that their subcategory ““1” refers to the absence of gills in Mucibregma. Overall, it appears that the inclusion of Mucibregma in the Scalibregmatidae results in an inappropriately polyphyletic taxon. By contrast, Mucibregma appears to exhibit a greater affinity to the Fauveliop- sidae Hartman and Fauchald 1971. For example, the primary criterion suggesting this relationship is the fascicles of thick spindle-shaped spines accompanied by long capillary setae present in all fascicles of both Mucibregma and Fauveliopsis. There is also a superficial similarity in the shape of the prostomia, which are bluntly rounded anteriorly; the posterolateral regions are not fused to the body in Mucibregma and are fused with the buccal segment in Fauveliopsis (Kudenov, unpublished data). Furthermore, both taxa lack furcate setae, which may simply be an example of convergence, and should probably not be weighted unduly. These two taxa differ from one another in that the anterior fragment on which Mucibregma is defined lacks the interramal papillae that are typically present in Fauveliopsis; it also has an extremely well developed peristomium. It is therefore suggested that Mucibregma be temporarily referred to the Fauveliopsidae. Al- though this association needs to be examined in greater detail, such a study may reveal that Mucibregma will need to be assigned to a separate family. Acknowledgments I wish to thank Paul Johnson and Joan Uebelacker, Barry Vittor and Associates, Mobile, Alabama, and Thomas H. Perkins, Department of Natural Resources, Tallahassee, Florida, for allowing me to work on scalibregmatids collected as part 340 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON of the Bureau of Land Management’s and Hourglass Cruise’s programs in the Gulf of Mexico. I am also indebted to Susan J. Williams, Allan Hancock Foun- dation, University of Southern California, Los Angeles, California, both for lend- ing specimens of Hyboscolex pacificus and for generously making space and fa- cilities available during a visit in 1982. Finally, I thank James A. Blake, Battelle, New England Marine Research Laboratory, Duxbury, Massachusetts, and Kristian Fauchald, National Museum of Natural History, for reviewing this manuscript. Literature Cited Berkeley, E., and C. Berkeley. 1952. Annelida Polychaeta Sedentaria. Canadian Pacific Fauna.— Fisheries Research Board of Canada, No. 9b(2):1-139. Blake, J. A. 1981. The Scalibregmatidae (Annelida: Polychaeta) from South America and Antarctica collected chiefly during the cruises of the R/V Anton Bruun, R/V Hero and USNS Eltanin. — Proceedings of the Biological Society of Washington 94(4):1131-1162. Chamberlin, R. V. 1919. The Annelida Polychaeta.—Memoirs of the Museum of Comparative Zoology at Harvard College 48:1-514. Dames, T.,and W. Moore. 1979. Final report—The Mississippi, Alabama, Florida Outer Continental Shelf Survey MAFLA 1977/1978, Vols. I-A, B, for the U.S. Bureau of Land Management, Washington, D.C. NTIS PB-294 288. New Orleans, Louisiana. Fauchald, K. 1977. The polychaete worms definitions and keys to orders, families and genera.—Los Angeles County Museum of Natural History, Science Series 28:1—190. , and P. Jumars. 1979. The diet of worms: a study of polychaete feeding guilds.—Annual Review of Oceanography and Marine Biology 17:193-284. , and D. R. Hancock. 1981. Deep water polychaetes from a transect off central Oregon. — Monographs of the Allan Hancock Foundation 1 1:1—73. Hartman, O. 1966. Quantitative survey of the benthos of San Pedro Basin, Southern California. Part 2. Final results and conclusions.— Allan Hancock Pacific Expeditions 19:187—456. , and K. Fauchald. 1971. Deep-water benthic polychaetous annelids of New England to Bermuda and other North Atlantic areas. Part 2.—Allan Hancock Monographs in Marine Biology 6:1-—327. Imajima, M., and O. Hartman. 1964. The polychaetous annelids of Japan. Part I1.—Allan Hancock Foundation Occasional Paper 26:239-452. Kudenovy, J. D. 1984. Chapter 18. Family Scalibregmatidae (Malmgren, 1867). In J. Uebelacker and P. Johnson, eds., Taxonomic Guide to the polychaetes of the northern Gulf of Mexico. — Final Report to The Minerals Management Service, contract 14-12-001-29091. Barry A. Vittor & Associates, Inc. Mobile, Alabama. Vol. 3, pp. 18-1 to 18-14. . 1985. Scalibregmatidae (Polychaeta).— Memoirs of the Hourglass Cruises. [In preparation]. , and J. A. Blake. 1978. A review of the genera and species of the Scalibregmadae (Polychaeta) with descriptions of one new genus and three new species from Australia.—Journal of Natural History 12:427-444. Moore, J. P. 1909. Polychaetous annelids from Monterey Bay and San Diego, California.—Pro- ceedings of the Academy of Natural Sciences of Philadelphia 61:235-—295. Quatrefages, A. de. 1865. Histoire naturelle des annéles marins et d’eau douce. Annélides et Gé- phyriens 1:1—588. Paris: Librairie Encyclopédique de R6ret. Uschakov, P. V. 1955. [Polychaeta of the Far Eastern Seas of the U.S.S.R.]. Opredeliteli Po Faune, S.S.S.R. 56:1—445 (In Russian). (English edition published in 1965 by the Israel Program for Scientific Translations, Washington: U.S. Department of Commerce, 419 pp). Department of Biological Sciences, University of Alaska, Anchorage, 3211 Providence Drive, Anchorage, Alaska 99508. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 341-346 A NEW SPECIES OF PAROTOCINCLUS (PISCES: LORICARIIDAE) FROM GUYANA Robert E. Schmidt and Carl J. Ferraris, Jr. Abstract.—A new species of loricariid fish, Parotocinclus collinsae, is described from the Essequibo drainage, Guyana. This species is most similar to P. macu- licauda from the Amazon River drainage and P. britskii from the Coppename River, Surinam. This is only the second species of the genus reported from north of the Amazon drainage. The most recent review of the loricariid genus Parotocinclus Eigenmann and Eigenmann (1889) listed thirteen species (Garavello 1977). In this paper we de- scribe a new species of the genus Parotocinclus from the Essequibo River drainage, taken during recent collections of freshwater fishes made by RES in Guyana. Methods. — Morphometric measurements were made with a Wild M-8 binocular microscope and an ocular micrometer. Morphometrics are those used by Gar- avello (1977) to differentiate species. Body depth was measured at origin of dorsal fin, head length was from tip of snout to rear margin of bony opercle, body width was measured at widest region of scapular bridge, and orbital diameter was mea- sured horizontally. Morphometrics of holotype are given first followed by range of paratypes in parentheses. Spinous fin elements are designated with a small “I”’ since ostariophysans do not have true spines and unbranched rays are designated with “i.” Meristics of the holotype are indicated by an asterisk (*). The following museum abbreviations are used: AMNH = American Museum of Natural His- tory, USNM = National Museum of Natural History, Smithsonian Institution. Parotocinclus collinsae, new species Figs. 1-3 Holotype.—AMNH 55433, 25.0 mm SL; Guyana, Essequibo Province, tribu- tary to Takutu River about 2 mi from Mazarahally Takutu lumber camp in Takutu Mountains, approximately 6°15’N, 59°5’W; R. E. Schmidt and A. Pappantoniou, 17 Aug 1983. Paratypes.—Same data as holotype; AMNH 55434, 4 specimens, 18.5—25.5 mm SL; USNM 265091, 2 specimens, 18.5—22.0 mm SL. Diagnosis.—A Parotocinclus with the following meristics and color pattern: 21— 22 lateral line plates, more than 25 premaxillary and dentary teeth on each side, abdomen covered with regular series of plates; one light saddle at dorsal fin and two dark saddles: one between dorsal and adipose fins, and one between adipose and caudal fins. Description.—Fin element counts typical for Parotocinclus (Garavello 1977): Dorsal—i, 7; Anal—i, 5; Pectoral—i, 6; Pelvic—i, 5; and Caudal—1i, 14, 1. Pre- maxillary teeth 29—33* on each side, mandibular teeth 23—32* on each side with the smallest specimens having the fewest teeth. Pored lateral line plates 21* or 22. 342 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Ventral view of head and abdomen of holotype of Parotocinclus collinsae, AMNH 55433 showing arrangement of abdominal plates. VOLUME 98, NUMBER 2 343 Fig. 2. Dorsal view of holotype of Parotocinclus collinsae, AMNH 55433 (25.0 mm SL) showing reticulated appearance of plate junctions. Standard length/body depth 5.2 (4.8-6.6), SL/body width 3.9 (3.6—4.1), head length/orbital diameter 7.4 (6.1—8.1) and interorbital width/orbital diameter 2.8 (2.4—3.0). Lips covered with large papillae, lower lip extends approximately half way to exposed coracoids. Barbel extends posteriorly from the corner of mouth no further than posterior margin of lower lip. Body covered with plates except for area between lower lip and exposed cor- acoids, triangular area around anal papilla, and opening of air bladder capsule ventral to posttemporal plate; plates uniformly covered with recurved denticles, those on snout somewhat enlarged. Plates without ridges except median carina on posterior two-thirds of snout to between nares, 2 shorter carinae extending from lateral to nares to above eye. One pair of plates posterior to occiput, followed by single median plate, second pair of plates, and second median plate, latter anterior to origin of dorsal fin in holotype; smaller specimens with fewer plates on nape. Six plates between dorsal and adipose fins, 4 between adipose and caudal fins. Abdomen with 3 rows of 6 plates each in holotype (Fig. 1), only lateral plates apparent in smaller specimens. Posttemporal plate imperforate. Base color on dorsolateral surface light orange-brown in ethanol. Junctions between plates on head, back, and sides of body darker than plates, producing reticulated appearance, especially prominent on holotype (Fig. 2). Very faint, wide dark lateral band from anterior to eye to insertion of dorsal fin, with equally faint dorsal saddles at supraoccipital and base of dorsal fin. Nape comparatively light. Dark saddle between posterior of dorsal fin and anterior of adipose fin extends ventrally to mid-lateral line where it divides into anterior and posterior ventral projections (Fig. 3); latter almost completely encircling peduncle. Second, nar- rower dark saddle at base of caudal fin with similar pattern on ventral portion of sides, completely encircling peduncle. Ventral surface of head, abdomen, breast, and ventral and ventrolateral portions of peduncle white. Some discrete black spots on breast, abdomen, and ventral surface of head in holotype; spots fewer or absent on paratypes. Spines and rays of all fins with black dashes arranged to appear as bars; pectoral with 5 bars, pelvic-1 or 2*, anal-1 or 2*, dorsal-4, adipose- 1, caudal-3. 344 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON lateral plate arrangement and color pattern. wing f holotype of Parotocinclus collinsae, AMNH 55433 (25.0 mm SL) sho Lateral view 0 Fig. 3. VOLUME 98, NUMBER 2 345 Etymology.—Named for Dr. Margaret Collins (Alfred Emerson Field Station, ‘Kartabo, Guyana) who made it possible for the senior author to collect fishes in Guyana. Biology.—The type-series was collected in a stream of moderate gradient in virgin rainforest in the Takutu Mountains, approximately 300 m elevation. The substratum was predominantly sand with many fallen trees. Collections were made with a 3 m seine in less than | m depth by kicking up and driving fishes downstream into the net. Discussion. —Parotocinclus collinsae is most similar to P. maculicauda and P. britskii in having the abdomen covered with regular rows of plates and possessing more than 25 premaxillary and dentary teeth on each side. It differs from P. maculicauda in having 21-22 plates in the lateral line rather than 24—26. Paro- tocinclus maculicauda has four dark dorsal saddles situated differently. Paroto- cinclus britskii has five rows of plates on the abdomen (rather than three) and three dark dorsal saddles: one under the anterior part of the dorsal fin, one under the adipose fin, and one just anterior to the caudal fin. Most Parotocinclus species are known from coastal Brazil (Garavello 1977). Parotocinclus amazonensis from Rio Solimoes and P. britskii from the Coppe- name River, Surinam are exceptions. Parotocinclus collinsae is the first record of the genus from the Essequibo River drainage. Inclusion of P. collinsae in the genus Parotocinclus is somewhat problematical. Boeseman (1974) distinguished Parotocinclus from other hypoptopomatine genera by the presence of an adipose fin, 3—5 series of plates on the abdomen and a slightly flattened head with eyes situated dorsolaterally, not visible from below. Garavello (1977) used a broader definition of the genus, to include species with naked abdomens or nearly lateral eyes. Further, he identified two species groups within Parotocinclus but did not designate them as genera because he was unable to place P. spilurus into either group. As with P. spilurus, P. collinsae shares characters with both groups and further blurs the distinction between them. If the genus were to be subdivided, P. collinsae would quite possibly remain in Parotocinclus due to its regular pattern of abdominal plates (also present on P. maculicauda, the type-species) which are, we believe, a derived feature within the group. Acknowledgments We thank the government of Guyana for making it possible for RES to collect fishes in its country. The Mazarahally Timber Company arranged transportation to the collection site and Mr. Roy van Long, Takutu Camp, was especially helpful in providing facilities and transportation. Dr. Antonios Pappantoniou helped in the field, Mary Rauchenberger helped in the darkroom and Kathleen A. Schmidt drew the figures. This is Bard College Field Station-Hudsonia Contribution #30. Literature Cited Boeseman, M. 1974. On two Surinam species of Hypoptopomatinae, both new to science (Lorica- riidae, Siluriformes, Ostariophysi).— Koninklijke Nederlandse Akademie van Wetenschappen- Amsterdam, Proceedings, Series C, 77(3):257-271. Eigenmann, C. H., and R. S. Eigenmann. 1889. Preliminary notes on South American Nematognathi II.— Proceedings of the California Academy of Science 2:28-56. 346 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Garavello, J. C. 1977. Systematics and geographical distribution of the genus Parotocinclus Eigen- mann & Eigenmann, 1889 (Ostariophysi, Loricariidae).—Arquivos de Zoologia, Sao Paulo 28(4):1-37. (RES) Hudsonia Limited, Bard College, Annandale, New York 12504; (CJF) Department of Ichthyology, American Museum of Natural History, Central Park West at 79th Street, New York, New York 10024. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 347-350 TWO NEW SPECIES OF STRATIODRILUS, S. AEGLAPHILUS AND S. PUGNAXI (ANNELIDA: HISTRIOBDELLIDAE) FROM CHILE Irma Vila P. and Nibaldo Bahamonde N. Abstract.—Two new species of Histriobdellidae (Annelida) from Chile of the genus Stratiodrilus are described: Stratiodrilus aeglaphilus from the gill chamber of Aegla laevis and Stratiodrilus pugnaxi from the gill chamber of Parastacus pugnaxi. A key for the identification of all known species of Stratiodrilus is given and their known biogeographic distribution is described. The family Histriobdellidae includes Histriobdella and Stratiodrilus, epizoic annelids living in the branchial chamber or on the eggs of marine and freshwater crustaceans, respectively. Histriobdellidae have the following characteristics: small, vermiform, head conspicuous, clearly separated from rest of body, coelom re- duced, sexes separate, pharyngeal sac present, adhesive glands at the posterior end of the body, and male with copulatory organ. We agree with Borradaile and Potts (1958) that these characteristics, together with the absence of parapodia and chaetae, indicate an evolutionary specialization typical of parasites. Haswell (1900) proposed the genus Stratiodrilus, which now includes epizoic freshwater animals of circumantarctic distribution with species found in Australia, Tasmania, the southern region of South America, and Madagascar. They possess a conspicuous head, a trunk of five body segments and a caudal region. The head carries an unpaired median and two pairs of lateral tentacles, followed by paired conic retractile locomotive appendages. The trunk has three paired lateral cirri and in addition in the males, retractile gonopodia. The caudal region carries cirri, which in the genus Stratiodrilus are of great taxonomic significance. The high degree of specialization of Stratiodrilus, together with its peculiar geographical distribution, supports the geological relationships supposed to have existed between Australia, Madagascar and South America. The life cycle of Stratiodrilus, entirely confined to the branchial chamber of the host, differs substantially from that of most known polychaetes. However, the significance of this feature remains to be studied in detail. Further knowledge of the life cycle, taxonomy and ecology of Stratiodrilus will certainly help in tracing phylogenetic relationships between host freshwater crustaceans of the Southern Hemisphere. To date, four species of Stratiodrilus have been described: S. tasmanicus Has- well, 1900; S. novahollandiae Haswell, 1913; S. haswelli Harrison, 1928, and S. platensis Cordero, 1927. We describe herein two new species of Stratiodrilus: S. aeglaphilus and S. pugnaxi and give a key for the identification of all known species of Stratiodrilus. 348 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. The species of Stratiodrilus, habitus: a, S. pugnaxi; b, S. novaehollandiae; c, S. platensis; d, S. haswelli; e, S. aeglaphilus; f, S. tasmanicus. c1—c3, lateral cirri; c4—c7, caudal cirri. Stratiodrilus aeglaphilus, new species Diagnosis.— Adult males reach an average total length of 750 wm (maximum 1200 um); females average 500 um (maximum 1000 um). The mandibular ap- paratus of both sexes has an average length of 90 um. As in other species, the trunk bears 3 pairs of lateral unsegmented simple cirri on the second, third, and fifth trunk segments. Each of the caudal appendages of the posterior region has a simple cirrus and a small tubercle, well separated, along the caudal border (Fig. le, Table 1). Differences from other species of the genus.—Stratiodrilus aeglaphilus is the smallest of the species so far known. It has a maximum total length of 1200 um VOLUME 98, NUMBER 2 349 Table 1.—Geographical distribution of Stratiodrilus. Species Host Geographic distribution SS. tasmanicus Astacopsis franklinii (Gray) As- Tasmania, Hobart. Haswell, 1900 tacopsis franklinii tasmanicus Erickson S. novaehollandiae Astacopsis serratus Shaw New South Wales. Streams 2000 to Haswell, 1913 3000 feet in blue mountains. Hawkes- bury River System. Cataract River and London River. Post Hacking Waterfall Creek. Middle Harbour, Port Jackson, Pitt water off Broken Bay. Murrum Bridgee River of Mur- ray River System. S. platensis Aegla laevis* (Latreille) Uruguay. Stream Solis Chico, Departa- Cordero, 1927 mento de Canelones, Arroyo Migue- lete. S. haswelli Astacoides madagascariensis, Madagascar. Harrison, 1928 Milne Edwards S. aeglaphilus, n. sp. _Aegla laevis (Latreille) Central Chile. Rio Maipo. S. pugnaxi, n. sp. Parastacus pugnax (Poeppig) Southern Chile, Reumén (Valdivia). * The available evidence (Schmitt 1942) indicates that two species of Aeg/a are present in Uruguay: A. platensis Schmitt and A. uruguayana Schmitt; whereas A. /aevis has only been reported present in Central Chile (Bahamonde and Lopez 1963). in the male and 1000 um in the female. Their simple unsegmented lateral cirri differ from the bifurcate ones of S. haswelli and the bisegmented ones of S. tasmanicus and SS. novaehollandiae. Stratiodrilus aeglaphilus differs from S. pla- tensis in having only one pair of caudal cirri. Material examined.—Rio Maipo, Chile, 33°45’S, 70°45’W, from branchial chamber of Aegla laevis laevis Latreille. Holotype: M.N.H.N. AN-2001 (Museo Nacional de Historia Natural) Santiago, Chile, Penaflor tributary of Rio Maipo, Chile. Paratypes: M.N.H.N. AN-2003, 2 males and 2 females. Stratiodrilus pugnaxi, new species Diagnosis.— The average length of the adult male is 1200 wm (maximum 1360 um) and that of the female is 1100 um (maximum 1180 um). It has the longest mandibular apparatus with an average length of 430 um. The 3 pairs of lateral cirri are simple, unsegmented and clearly ciliated at their distal ends. The caudal region has 4 paired retractile cirri: the first 2 pairs are highly retractile and located lateral and close to the anus. One pair is dorsal and the other is ventral. When the dorsal pair of caudal cirri is fully extended, it becomes bifurcate. The other 2 pairs of cirri are located distally on the caudal region (Fig. la, Table 1). Differences from other species of the genus.—The most distinctive feature of S. pugnaxi, is the number of caudal cirri: 4 pairs in S. pugnaxi and 1-2 pairs in the other species. Material examined. —Chile (Valdivia): Reumén, 39°45’S, 73°45'W. In the bran- 350 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON chial chamber of Parastacus pugnax. Holotype: Male, M.N.H.N. AN-2002, Reu- mén 50 km, Valdivia, Chile. Paratypes: M.N.H.N. AN-2004, 5 males and 5 females. Key to the Species of Stratiodrilus Haswell, 1900 1( 2) One or twoypairs ol caudal’ Cini) 9. ae ek ee ee 3 2(o)) Eourspairsyoficaudaleciar 4 0.405). S. pugnaxi, n. sp. (Fig. 1a) 30e4) slworpainsnof caudalueinms 2 52k cigs os eae a oo ok ee ee 5 4@3)'Oneipair oftcaudal Cit 227 oh Asis os os ns bee 2 ee 7 5( 6) Caudal cirri bisegmented ... S. novaehollandiae Haswell, 1913 (Fig. 1b) 6( 5) Caudal cirri unsegmented .......... S. platensis Cordero, 1927 (Fig. 1c) HC 8) eateralicumbibituccates... 05.0040. S. haswelli Harrison, 1928 (Fig. 1d) S@) peateralicismiomnon=bituncate | sss tae ieee 9 9(10) Lateral cirri unsegmented ............... S. aeglaphilus, n. sp. (Fig. le) 10( 9) Lateral cirri bisegmented ........ S. tasmanicus Haswell, 1900 (Fig. 1f) Acknowledgments We thank Dr. Juan Fernandez for a critical review of the manuscript, Mr. Victor Bocic for providing some of the specimens of S. pugnaxi, and Mrs. Luz Uribe and Marta Cariceo for the illustrations and typing of the manuscript. Literature Cited Bahamonde, N., and M. T. Lopez. 1961. Estudios biolégicos en la poblacion de Aegla laevis laevis (Latreille) en el Monte.—Investigaciones Zoolégicas Chilenas 7:19—58. , and 1963. Decapodos de las aguas continentales de Chile.—Boletin del Museo Nacional de Historia Natural Chile 10:123-149. Borradaile, L. A., and F. A. Potts. 1958. The Invertebrata. Chapter VIII. The phylum Annelida. — University Press, Cambridge, pp. 266-316. Cordero, B. H. 1927. Un nuevo arquianélido, Stratiodrilus platensis n. sp. que habita sobre Aegla laevis laevis (Latreille).—Physis 7:574—-578. Harrison, L. 1928. On the genus Stratiodrilus (Archiannelida: Histriobdellidae), with a description of a new species from Madagascar.— Records of the Australian Museum 16:116-121. Haswell, W. A. 1900. On a new Histriobdellidae.— Quarterly Journal of Microscopical Science N.S. 43:299-335. . 1913. Notes on Histriobdellidae.— Quarterly Journal of Microscopical Science N.S. 59:91- 99. Lang, K. 1949. Morphology of Stratiodrilus plantensis.— Arkiv fiir Zoology 42a:43-67. Schmitt, W. 1942. The species of Aegla, endemic South American freshwater crustaceans. —Pro- ceedings of the United States Natural Museum 91:431—520. Departamento de Ciencias Ecolégicas, Facultad de Ciencias Basicas y Farma- céuticas, Universidad de Chile, Casilla 653, Santiago, Chile. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 351-365 SAXIPENDIUM CORONATUM, NEW GENUS, NEW SPECIES (HEMICHORDATA: ENTEROPNEUSTA): THE UNUSUAL SPAGHETTI WORMS OF THE GALAPAGOS RIFT HYDROTHERMAL VENTS Keith H. Woodwick and Terry Sensenbaugh Abstract.— Saxipendium coronatum, new genus, new species, is described from a rocky habitat at the edge of a geothermal vent of the Galapagos Rift. The spaghetti worms represent a new family, Saxipendiidae, in class Enteropneusta, phylum Hemichordata. Important differentiating characteristics at the familial level which are found in the spaghetti worms include weakly developed muscles of the proboscis, coronate proboscis skeleton, long recurved skeletal crura, simple digestive tract, and antra associated with testis. Structures lacking in spaghetti worms but present in other enteropneusts include vermiform process, cauliflower organ, synapticulae, pygochord, hepatic caeca, and large yolky eggs. Most enteropneusts have been taken from the intertidal or shallow areas offshore (Hyman 1959); however a few forms from deeper waters have been described including Spengelia sibogae (Spengel, 1907) at 275 m, Glandiceps talaboti (Mar- ion, 1886) at 30-350 m, and G. abyssicola (Spengel, 1893) at 4500 m. An addi- tional deep-sea form was taken recently from near the hydrothermal vents of the Galapagos Rift area. These spaghetti worms were first seen at the original ““Clam- bake” site in 1977 and were then collected in 1979 along with clams, mussels, crabs, and vestimentiferans (Jones 1981). Dive 878 by the deep-diving submersible A/vin on January 19, 1979 carried it to a maximum depth of 2478 m at 00°47.9'N; 86°13.5'W where the spaghetti worms were found loosely attached to rocks located at the periphery of the vent area. These enteropneusts, as viewed in videotapes, were draped over the rocks with the posterior end on the rock and with the anterior part of the body alternately drifting freely in the water and then back over the surface of the rock. Many incomplete specimens were collected using a suction apparatus which pulled in the spaghetti worms and a certain amount of associated mucous debris and rock slivers. The holotype (USNM 97395) and paratypes (USNM 97396-8) have been de- posited in the collections of the Division of Worms, National Museum of Natural History (USNM). Methods and Materials Some specimens were prepared for histological studies and others for investi- gation using an electron microscope. Specimens for histological studies were de- hydrated through an ethanol-toluene series, embedded in Paraplast (56—57°C), sectioned at 10, 13 or 15 w using an American Optical Rotary Microtome. Serial cross and sagittal sections were stained with Harris’ Hematoxylin and Eosin Y. 352 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Specimens were studied using an American Optical Stereozoom and Zeiss GFL compound with optivar and photographic attachments. Both Tri-X and Technical Pan black and white film were used. Other specimens (glutaraldehyde fixed) were processed for scanning electron microscopy (SEM) using either histological sections or freeze fractured material. For sections the glutaraldehyde fixed material was dehydrated in a graded series of ethanol and paraffin embedded. Sections varying in thickness (10—25 yu) were obtained using a Leitz Rotary Microtome. Sections were attached to round, 13 mm cover slips and deparafhinated with xylene; cover slips holding the sections were transferred to filtered Freon TF and critical point dried in liquid CO,. The cover slips were then mounted using double-sided tape. Following glutaraldehyde fixation the material for freeze fracturing was rinsed in cold (5°C) 0.1 M sodium cacodylate buffer solution and post-fixed for one hour in cold osmium tetroxide in the same buffer. The material was then rinsed in cold buffer and dehydrated through a graded series of ethanol up to 100%. Material was then, piece by piece, encapsulated in a thin parafilm tube (containing 100% ethanol), immersed in liquid N,, and fractured with a scalpel. Fractured material was then returned to 100% ethanol, transferred to filtered Freon TF and critical point dried in liquid CO,. The fractured material was mounted using graphite glue. All SEM materials were metal coated with gold-palladium using a Jeol JFC- 1100 ion sputter and examined at 15 or 25 kV using a Jeol JSM-35 scanning electron microscope. Saxipendiidae, new family Type-genus.—Saxipendium, new genus Diagnosis.—Enteropneusta with poorly developed proboscis muscles, coronate (in cross section) body of the proboscis skeleton, skeleton also with long recurved crura, simple post branchial gut (no regional specialization), testicular antra, small eggs; lack-synapticulae, vermiform process of buccal diverticulum, hepatic caeca, lateral septa (in trunk). Saxipendium, new genus Type-species. —Saxipendium coronatum, new species. Gender. — Feminine. Diagnosis.—Characters of the family (above). Etymology.— From Latin, saxi (rock, boulder) + pend (hanging) + ium, in ref- erence to the habitat of the organism. Saxipendium coronatum, new species Figs. 1-12 Unidentified organisms (draped like spaghetti).—Corliss and Ballard 1977:450 (with fig.). “Spaghetti” (acorn worm, enteropneust) Sedgwick 1980:50 and 51 (figs.) 52, 53. Acorn worms.— Hessler 1981:741, 744, 746. Diagnosis.—Characters of the family (above). VOLUME 98, NUMBER 2 358 Etymology.— From Latin, corona (crown) in reference to the appearance of the proboscis skeleton. Type-locality.—Near “‘Rose Garden”? geothermal vent, Galapagos Rift (00°47.9'N; 86°13.5’W), 2478 m depth. Material examined. — Holotype (USNM 97395), and 8 male and 3 female spec- imens (paratypes) serially sectioned, cross or sagittal. General characteristics. — The preserved specimens are in tortuous coils, wrapped upon themselves and welded together with mucous. Some detrital material as well as a few slivers of black rock are attached to their surface. The specimens fixed in glutaraldehyde are in better condition than those in alcohol. None of the specimens was complete. The holotype (glutaraldehyde fixed) has the following measurements in mm: overall length—215, proboscis—11.0, col- lar— 3.0, trunk—201. Maximum width which occurs at the base of the proboscis, at the collar, and at the anterior trunk region, was 8.0 mm. Two other long specimens (alcohol preserved) had the following measurements: overall— 220 and 154, proboscis— 10.0 and 8.0, collar—3.0 and 4.0, trunk—207 and 142, respec- tively. The proboscis is softly pointed anteriorly and enlarges posteriorly to produce an arrow shape, overall (Fig. 1). It is indented dorsally in preserved specimens but this characteristic is not as strong on specimens viewed in the videotape. A narrow peduncle joins the proboscis to the collar. The cuff-like collar is as wide as the proboscis but is very short. It has a definite raised ring just anterior to the juncture with the trunk. Anteriorly the trunk is the same width as the collar but is reduced slightly in overall dimensions in succeeding regions. The branchial region is more rounded in cross section than the posterior regions which are slightly flattened and have a median longitudinal depression with raised lateral ridges. These ridges have undulations produced by the varied size of the individual gonads in the two longitudinal rows. The gill pores are not readily visible but the raised gonopores are. Spaghetti worms are yellow-white in color. The proboscis is the lightest shade and the anterior part of the collar is the darkest. The posterior raised ring of the collar is, however, very light. The iodine-like odor characteristic of many enteropneusts was not present in the preserved material or in the preservatives. Internal anatomy-proboscis.—In cross section, external to internal, the bodywall of the proboscis includes a simple epidermis, nerve fiber layer, a basement mem- brane, and connective tissue and muscle (Fig. 2). The simple epidermis includes several cell types but is dominated by gland cells. In general the epidermis is thicker ventrally than dorsally. The muscles and connective tissues include fibers which are circular, longitudinal, or oblique in pattern but all are very weakly developed. Dorsally the body wall of the proboscis is about 130 wm thick. The epidermis (100 yu) includes a basal grouping of nuclei, an open vacuolated area in the middle, an overlying grouping of nuclei and a few scattered distal goblet glands. Pseu- dostratification is evident. Laterally and ventrally the body wall is generally 130 u, but in some areas is thicker. The epidermis basally is open with a few dispersed nuclei and distally is packed with goblet cells. Centrally it is dominated by a layer of dark staining nuclei. 354 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON The nerve layer is fibrous and is about 20 u thick except near the base of the proboscis close to the peduncle where it may be as thick as 50 u dorsomedially. Epidermis overlying this cord is made up of 50 uw of heavy basal nuclei and a 50 uw distal area of packed glands. The muscle-connective tissue component of the body wall is poorly developed, only 10 uw, consisting mainly of short fibers of the circular muscle component and attachment fibers. The internal muscular and connective tissue of the proboscis, in general, is weakly developed, also, and consists of circular, longitudinal, and oblique fibers traversing the entire interior area anteriorly and in the area surrounding the central proboscis coelom posteriorly. This sparsely muscular portion (Fig. 2) makes up the anterior two-thirds of the proboscis; the posterior one-third includes the be- ginning of the proboscis coelom and the contained proboscis complex. The proboscis coelom makes up about 10% of the total proboscis area near the coelom’s anterior end, 40% midway, and only a small percentage posteriorly. There it is reduced and divided into right and left lateral and dorsal and ventral cavities by the presence and increase in size of the proboscis complex. The right dorsal coelomic area is blind ending but the left dorsal area opens to the outside through the proboscis canal and pore. The proboscis complex has a double attachment dorsally and a single ventral mesentery which slants posteriorly, first attaching to the coelomic lining and then penetrating to and attaching to the subepidermal basement membrane. The com- plex includes the glomerulus, central sinus, pericardial vesicle, and buccal diver- ticulum (Fig. 3). The glomerulus is a tortuous mass of tubules filled with vascular fluid. This vascular plexus covers the anterior end of the developing complex and then posteriorly covers the dorsal and lateral portions but not the ventral. The lateral glomerulus is reduced in size and the dorsal portion disappears as the double dorsal mesentery of the complex joins the dorsum of the pericardial vesicle. The vesicle overlies the central sinus and the buccal diverticulum is ventrally placed in the complex with the central sinus and pericardial vesicle dorsal to it. The buccal diverticulum continues posteriorly through the peduncle and opens into the buccal region (the anterior gut area) of the collar. Anteriorly the proboscis coelom which more or less surrounds the proboscis complex is nearly round and measures 500 u in diameter. It has a distinct lining and there is open space all around the contained complex which measures about 400 uw in diameter. At this level the glomerulus shows a few dorsal and lateral folds and the anterior tip of the buccal diverticulum appears along with the pericardial vesicle just dorsal to it. A little posterior the coelom elongates dorsally and enlarges to measure 750 u and 650 uw wide. The pericardial vesicle also elongates dorsally and the paired dorsal mesenteries which enclose it join the coelomic wall at the nerve fiber layer on each side; mesenteries are initially about 300 wu apart. The vesicle contains a few loose connective tissue and muscle fibers. The buccal diverticulum has en- larged to a diameter of about 475 uw and has a large central cavity, 170 u high and 100 uw wide. The epithelium making up the buccal diverticulum is about 200 u thick and is highly vacuolated with scattered elongate distal nuclei. The central cavity is lined with cilia. Just dorsal to the buccal diverticulum the blood-filled central sinus appears. A ventral mesentery is present but it is oriented antero- VOLUME 98, NUMBER 2 355 dorsal to postero-ventral and is not attached ventrally at this level. The connective tissue and muscle of the proboscis external to the coelom continues to be weakly developed but shows a slight concentration midventrally. To the posterior, the central sinus has a direct relationship with the open spaces of the glomerular plexus. Dorsally on the proboscis an external medial infolding occurs and the underlying internal structures also show changes. The epidermis is composed of a heavy nerve fiber layer (50 yu), a layer (50 uw) of many nuclei round in shape basally and elongate distally, and a layer (50 uw) of glands. To the posterior the nerve fiber layer expands to a width of 60 uw and the coelom broadens to 2000 u (wide) < 800 u (high). Farther posteriorly, the coelom is reduced to a breadth of 600 uw and a greatly reduced height of 70 uw as the dorsal mesenteries shorten. The buccal diverticulum is 600 uw across assuming the shape of a square with rounded corners; the central cavity is 100 uw in height and 80 w across. The glomerulus is well-developed laterally next to the buccal diverticulum and on the lower external surface of the dorsal mesenteries. The ventral mesentery is com- plete. At this point in one specimen the central sinus had an X-shaped cross section. The two ventral arms represented the original sinus and the upper arms, anterior lateral extensions. To the posterior, they merged ventro-medially and flowed laterally into the glomerulus and up between the two layers of each dorso- lateral mesentery where each formed a subepidermal vessel. Posterior to this level as the proboscis decreases in overall width the glomerulus begins to disappear and the lateral horns of the buccal diverticulum fill the space which was originally part of the coelom and divide the remaining coelom into dorsal and ventral spaces. The internal cavities of the horns and the central area of the buccal diverticulum interconnect. The dorso-lateral vessels merge medially to form a major longitudinal vessel. There is no racemose organ. The proboscis includes the preoral ciliary organ (Fig. 1) first described by Bram- bell and Cole (1939) which in this species is located on the posterior external face of the proboscis. The U-shaped organ has two parallel grooves with associated short epidermal cells and heavy ciliation. The medial groove of each arm has longer cells and shorter ciliation. The raised area between the grooves is not glandular but the medial groove area has granular acidophilic and reticular ba- sophilic gland cells. Internal anatomy-peduncle.—The peduncle (Fig. 1) connects the proboscis and collar and contains important nerve fibers, muscles, blood vessels, buccal diver- ticulum, skeleton and related materials, coelom remnants of both the proboscis and collar which, however, do not interconnect, and a surrounding epidermis which is different on the dorsal and ventral surfaces. The dorsal epidermis which is underlain by a heavy nerve layer (60-70 yp) is about 100 uw. The basal 75 uw includes a layer of rounded and then centrally more elongate nuclei. The distal surface is vacuolated and contains only a few glands. Ventrally the nerve is only 20 uw and the epidermis about 70 y; the basal half is nuclei and the distal is mixed glands (goblet, reticular, granular). Near the posterior end of the peduncle the reticular glands dominate ventrally. The buccal diverticulum has a large central cavity and through its expansion as lateral horns it crowds out the ventro-lateral portions of the proboscis coelom and reduces in size the dorso-lateral portions. On both sides the lining of the dorsal coelomic cavity shows a distinctly nucleated simple columnar epithelium 356 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON “Zs ai : > SW ASS =, Ge os ae Noite Biatacr rls . se) PNY ae ype CUA te) Wi ae ee) ra ‘ak OEE! Abed) NG Pena 7 VOLUME 98, NUMBER 2 Si) which on the left side forms the proboscis canal and lines the proboscis pore which opens externally. The pore is not round but is an elongate slit running antero-posteriorly. Farther posterior in the peduncle the proboscis canal and pore disappear and the proboscis skeleton begins to replace the buccal diverticulum positionally. The main part of the skeleton, which is in the peduncle, is beautifully crown-shaped in cross section (Fig. 4), and the spikes of the crown divide into subsections the central part of the buccal diverticulum; the large lateral horns and a small median-ventral caecum remain intact, each with an open central area. Closer to the collar the skeleton crowds out the ventral caecum and divides the lateral horns into subsections (Fig. 5). Then the spikes of the skeleton are reduced in length and the central body develops medio-lateral concentrations evidenced by a two-whorl pattern. Those two areas become the separate lateral skeletal crurae of the collar. The dorsal median subepidermal vessel is present throughout the peduncle and posteriorly it joins medio-ventrally with a vessel which lies between the two perihaemal cavities (Fig. 6). The free anterior cuff of the collar overlaps the posterior part of the peduncle but as the dorsal half of the cuffturns ventrally and posteriorly it joins the peduncle, first laterally and then dorso-medially. An anterior pocket is produced just above the heavy nerve fiber layer of the peduncle (Fig. 5). The pocket partly closes off and as a collar epidermis shifts dorsally the nerve fiber layer comes to have a more internal position. Throughout the collar this neurocord (Fig. 6) is the dom- inant part of the nervous system. In the posterior peduncle this nerve fiber layer which is at first flat, develops lateral dorsal curlings, and then a strong dorsal and medial recurving of the two lateral areas so that the fibers fold over the central cellular area. As the dorsal half of the collar cuff joins the peduncle the coelom of the collar directly interconnects with small lateral coelomic and muscular areas which first appeared lateral to the proboscis skeleton in the posterior region of the peduncle. The two lateral muscle groups attach anteriorly to the proboscis skeleton and posteriorly in the collar. Internal anatomy-collar.—The cuff-shaped collar (Fig. 1) has a complex of epidermal zones, the neurocord, the anterior portion of the digestive tract, and a coelom with connective tissue poorly developed peripherally but with well-de- veloped muscles associated with the crura of the proboscis skeleton (Fig. 6). The epidermis is characterized by five main zones arranged generally as suc- cessive rings from anterior to posterior. A five-zone pattern is not uncommon (e.g., Protoglossus koehleri, Burdon-Jones 1956) but a few species have been — Figs. 1-6. Saxipendium coronatum: 1, Overall view of holotype, <1.2; 2, Proboscis, transverse section, x64; 3, Proboscis complex, transverse section (SEM), x90; 4, Peduncle, transverse section of skeleton (SEM), x 162; 5, Peduncle-collar interphase, transverse section, x58 (paratype, USNM 97396); 6, Collar, transverse section, x81 (USNM 97396). BD: buccal diverticulum, C: coelom, CR: skeletal crura, CS: central sinus, E: epidermis, EM: epithelium, G: gut cavity, GL: glomerulus, M: muscle (and connective tissue), MED: dorsal mesentery, MEV: ventral mesentery, N: neurocord, NF: nerve fibers, P: peduncle, PC: preoral ciliary organ, PH: perihaemal cavity, PV: pericardial vesicle, SP: proboscis skeleton, V: blood vessel, X: proboscis, Y: collar, Z: trunk. 358 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON described as having three or four zones. The number of zones may be a matter of varied specificity in histological interpretation. In Saxipedium coronatum two of the zones show variations designated as subzones. The epidermis of Zone I which comprises the inner portion and turned edge of the anterior part of the cuff is 100 uw in thickness and has a heavy basal band of nuclei and the distal portion consists of packed acidophilic goblet cells. There are two subzones with the more posterior having the nuclei in a wider and more unevenly dispersed band, a slightly greater concentration of goblet cells and an occasional basophilic reticulated gland cell. Zone II has an epidermis of 125 uw. The nuclei are still basal but form a less distinct band followed distally by an open vacuolated region and then a superficial region lightly stained with goblet cells and a few granular gland cells. Zone III is the widest region and has the thickest epidermis. It includes the area preceding the raised posterior ring as a subzone and the ring itself as another. Overall, the epidermis is 175 w thick and has a basal and a distal grouping of nuclei. In the anterior subzone the basal nuclei dominate and a few goblet cells are present distally; in the posterior subzone the distal elongate nuclei form a distinct band. This extremely thick zone of the epidermis is highly vacuolated throughout. Zone IV comprises the posterior edge and infolding of the ring. The epidermis is 100 w and includes a dispersed basal grouping of nuclei, a vacuolated central region, and a distal region which includes a heavy concentration of acidophilic cells and sparse granular gland cells. Zone V comprises the juncture of the collar and trunk. The epidermis is 100 u and includes a concentrated basal layer of nuclei similar to Zone I but with only sparse goblet cells distally. Underlying the epidermis and included in all measurements is a nerve fiber layer 12-15 w thick. The fibrous portion of the neurocord is continuous anteriorly with the epidermal nerve fiber layer of the peduncle and posterior face of the proboscis as well as the dorso-median concentration and other nerve fibers in the proboscis. Posteriorly the neurocord is continuous with the solid dorso-median nerve cord of the trunk and the collar-trunk lateral connectives which join to form the solid ventral nerve cord of the trunk. The neurocord is 650 uw wide and 150 yp thick and the lateral and ventral fibrous portion makes up about 65% of the cross sectional surface. There are anterior and posterior openings into the cellular portion of the neurocord which is also irregularly open throughout (Figs. 5 and 6). Just ventral to the neurocord are located the dorsal vessel and the perihaemal cavities which are thin paired anterior extensions of the trunk coelom (Fig. 6). These cavities which contain longitudinal muscles reach into the peduncle. In some specimens at their anterior extremity they become an undivided cavity. Posteriorly they open into the general trunk coelom and the muscles become part of the weak dorsal musculature of the trunk. The anterior part of the perihaemal cavities is at the level where the proboscis skeleton separates into left and right crura. The crura (Fig. 6) bend down and around the gut as they extend posteriorly to near the end of the collar. The main anterior skeleton and the crura serve as the attachment points for the two muscle bundles noted above which connect VOLUME 98, NUMBER 2 359 posteriorly to the interface of the collar and trunk at a ventro-lateral position. The crura are unusual in having a thin recurved portion which bends back an- teriorly for a short distance in the posterior collar. The position of the crura in relation to the gut epithelium causes the epithelium to push into the gut cavity on each side to produce an upper and lower chamber. Contraction of muscles and movement of the skeleton caused by fixation produced in some specimens dorso-lateral blind ending (posteriorly) gut pockets above the crura in the posterior part of the collar. The muslces and connective tissue of the collar coelom are poorly developed. The anterior collar gut epithelium dorso-medially and lateral to each of the crura is similar to the tissue of the buccal diverticulum (Fig. 6). It is about 100 pw thick with a distal dispersed band of elongate nuclei and the majority of the thickness made up of vacuolated epithelium. The lateral components have a few distal basophilic reticular gland cells and the ciliation is heavier than it is medially. The dorsal central band has continuity with the central portion of the buccal diverticulum which is made possible by the reduction in size of the proboscis skeleton and the separation of the crura. The epithelium covering the crura is very thin and the nuclei dominate in a heavy basal band (Fig. 6). The ventro- lateral and ventral gut epithelium is about half as thick as the dorsal. In the central collar thicker epithelium appears ventro-laterally and it extends into the ventro- medial region in the posterior collar. The gut cavity decreases in overall size posteriorly, but the chamber above the crura increases in size relative to the lower chamber as the crura shift their position. In the posterior collar as it interfaces with the trunk the neurocord begins to show a reduction in the dorso-medial recurving of the fiber portion and the central cellular area begins to open to the outside. The fibrous portion of the neurocord centrally forms the dorsal solid nerve cord of the trunk and bilaterally forms the connectives which shift ventrally and rejoin to form the ventral subepidermal solid nerve cord of the trunk. The perihaemal cavities widen laterally and increase in size as they approach a direct relationship with the dorsal trunk coelom. The last important structures of the collar to appear are the collar pores which via short canals connect the collar coelom to the outside. In this species the canal connects the collar coelom to the outside through pores located in the dorso- medial area but the canals also connect to the most anterior gill pouch. The specialized epithelium lining the canals and pores is 200-250 wu in thickness. It has a distinct basal layer of heavy elongate nuclei, a narrow open distal zone, and heavy surface ciliation. Internal anatomy-trunk.—The anterior trunk matches the collar in shape and width but the trunk is slightly flattened dorso-ventrally throughout most of its length. The dorso-median area is depressed and deeper sub-lateral grooves extend antero-posteriorly. The extreme dorso-lateral surface which is raised contains the gonads (Figs. 1 and 11). The trunk is not highly specialized in structure or variable in appearance but it does contain one of the more important enteropneust char- acters. A series of paired gill slits and pores and associated skeletal elements occurs in the anterior branchial region of the trunk. Following the unspecialized post branchial anterior esophagus, a series of paired canals and pores connect the gut with the surface dorsolaterally in the posterior esophagus. The intestine proper follows this esophageal pore region. The gonads occur with the gills anteriorly, 360 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON and posteriorly to the gills the multiple paired gonads dominate the trunk. They are present in the esophageal region and continue throughout the entire trunk. The ventral epidermis of the trunk is generally thicker, contains more glands, and is more heavily ciliated than the dorsal (Fig. 9). Just beneath the epidermis occur the nerve fibers which are concentrated medially to form dorsal and ventral solid longitudinal nerve cords. The dorsal cord anteriorly has short lateral dorsal wings which are a modified continuation of the posterior collar neurocord. Throughout the remainder of the trunk the dorsal cord (50 u thick) has a triangular shape with the point dorsally placed pushing into the epidermis. The body does not bulge here because the epidermal cells are shorter where the nerve cord thickens. The ventral nerve cord (75 u) is more strongly developed than the dorsal; it also has a triangular shape with the point dorsally placed but here it pushes into the basement membrane and towards the coelom (Fig. 9). The gut of Saxipendium coronatum is not highly specialized. It contrasts greatly with that of many ptychoderids (e.g., Balanoglossus stephensoni Horst, 1940) and some spengeliids which have well developed hepatic caecea. It also contrasts with harrimaniids, especially those of Saccoglossus (e.g., S. otagoensis Thomas, 1972) which have many specialized regions of the gut with a variety of epithelial linings and strong ciliated grooves. The gut epithelium of the spaghetti worm is relatively unchanged from region to region and the gut cavity, although it changes in overall shape, is simple in pattern (Figs. 8, 9, 10, and 11). The ventral longitudinal muscles of the trunk are slightly heavier than the dorsal series but both are weakly developed. Dorsal and ventral blood vessels (longi- tudinal) are present (Fig. 9). In the branchial region the dorsal epidermis (60-70 yu) is only one half the thickness of the ventral. It has a strong band (15 wy) of slightly elongate basal nuclei. The remaining distal portion is open and vacuolated with a few superficial goblet glands. Laterally the ventral epidermis grades into the dorsal as it becomes thinner and the number of goblets decreases. The dorsal and ventral median nerve areas are 80 wu in thickness and the dorsal cord has small finger-like extensions into the epidermis. There are dorsal and ventral mesenteries and longitudinal bands of muscles. The weaker dorsal muscles reach laterally only to the gill pore region but the ventrals (SO—70 yu) extend about 90° to each side gradually decreasing in thickness laterally. The lateral body regions have a thin epidermis and very little muscle. With slight variation the above characteristics apply to the entire trunk. In the branchial gut the epithelium of the dorsal and ventral areas is similar. It is about 150 wu in thickness with an indistinct grouping of elongate distal nuclei. Central and basal regions are openly vacuolated. The lateral gill tongues and bars are covered mainly by a low epithelium with dark staining basal nuclei. The lateralmost wall of the gill pouch area resembles the dorsal gut epithelium but changes to the epidermal pattern at the gill pore (Fig. 7). The branchial skeleton embraces only the upper half of the gut cavity and although the skeletal elements are not strongly curved their arrangement produces an upper and lower chamber in this part of the trunk. The number of gill slits in enteropneusts varies from species to species from only a few to one hundred or more. They also vary during the life of an individual; there are more in a mature adult than a young developing specimen. In one VOLUME 98, NUMBER 2 361 representative specimen of Saxipendium coronatum which was cleared and stud- ied as a whole mount there was a total of 54 pairs of gill slits in a space of 27 mm. The gill slits are typically U-shaped and skeletal elements extend into the septa and tongues but are not strongly developed in either. There are no synap- ticulae. Anterior gill slits are larger. The branchial region includes about 4 of the length of the trunk. The next region, the anterior esophagus, has no opening from the gut to the outside. The posterior esophagus is as long as the anterior region. It has about 40 pairs of esophageal pores which have narrow canals connecting them to the dorsal lateral areas of the gut (Fig. 8). Small skeletal structures were associated with the openings of many of the pores. This is not unusual in enter- opneusts and has led to the suggestion that esophageal and gill pores may be related in origin. In the trunk posterior to the branchiae, the gonads fill most of the rounded lateral extensions of the body. Cross sectional views show three lobes to the body; two are lateral and include the gonads and one is central and includes the digestive tract (Figs. 1 and 2). The wall of the postbranchial gut is slightly folded and the cavity makes up about 4 of the cross section. The epithelium averages about 50 uw in thickness; it has a basal and a distal grouping of nuclei and a central open vacuolated region. The epithelium mid-dorsally and mid-ventrally is a little thick- er and each forms a medial groove containing heavier ciliation than the thinner lateral epithelium. In the posterior part of the esophagus, paired pores open dorsally just lateral to the nerve cord and run ventrally 300 wu to the gut proper. The ventral half of the canal has a total diameter of 60 w and a central cavity of 20 uw. The canal narrows dorsally as it approaches the pore. The epithelium of the lower canal is similar to but thinner than that of the gut proper (50 wu) which is open and vacuolated with a slight basal grouping of nuclei and a heavier distal grouping of elongate nuclei. The gut has a square to rectangular shape in cross section and the gut cavity makes up about *% of the overall cross section (Fig. 8). The intestine proper located posterior to the esophageal pores makes up about ¥4 of the length of the trunk. Its gut and epithelium are even simpler than those of other trunk regions. The gut cavity is nearly round in cross section and comprises more than *% of the gut area. The gut epithelium is not folded and is 25-75 u in thickness. It has a heavy basal grouping of nuclei, very little central or distal vacuolation, a few granular acidophilic gland cells and scattered basophilic cells superficially. The gut has strong ciliation with heavier areas mid-dorsally and especially mid-ventrally (Fig. 9). Dorsal muscle bands are more concentrated near the gonopores. The gonads occur as repeated individual lobes organized into lateral rows ex- tending, as presently known, throughout the trunk. Sexes are separate. The testes, which are larger than the ovaries, average about 1000 wu in diameter. Lying one next to the other from anterior to posterior there would be about 200 testes on each side of one of the specimens described above which had a trunk of 205 mm. The main body of the testis is usually a little longer on the axis which leads to the gonopore (Figs. 11 and 12). The gonad is extra-peritoneal and as it extends into the coelom is covered by the peritoneum and underlain by a thin layer of muscle and heavy vascular concentrations. The testes have a thin superficial grouping of germinal cells and a concentrated central mass of spermatozoa. The 362 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON EM ' 4, ‘ Fea « ANS VR aR Ni lee ae) » Yet gis ZAI wet pf +, Tiions Ke | ase eae SRS Sees Yay? cise ales : == Re See —— Figs. 7-12. Saxipendium coronatum: 7, Trunk (branchial region), transverse section, x94 (USNM 97396); 8, Trunk (esophagus), transverse section, x52 (USNM 97397); 9, Trunk (intestine), transverse section, x39 (USNM 97397); 10, Trunk (intestine), transverse section, x39 (USNM 97397); 11, Trunk (intestine), freeze fracture (SEM), x13; 12, Trunk, transverse section, x62. A: antrum, C: coelom, E: epidermis, EC: esophageal canal, EM: epithelium, G: gut cavity, GIP: gill pore, GP: gonopore, M: muscles, MED: dorsal mesentery, MEV: ventral mesentery, ND: dorsal nerve, NV: ventral nerve, O: ovary, T: testis, V: blood vessel. VOLUME 98, NUMBER 2 363 sperm are about 29 wu in overall length including a head of about 4.0 uw. The head is not round; it has a full basal nuclear portion but the acrosomal region attenuates to a bluntly rounded tip. Between the main body of the testis and the gonopore there is a small specialized antrum (Fig. 12). The antrum has a heavy muscle covering which encloses a concentration of highly vacuolated cells surrounding the central canal leading to the gonopore. The epidermis near the gonopore in- cludes concentrations of granular acidophilic gland cells. The ovaries which are smaller than the testes measure about 750 wu in greatest width. The muscle covering is thin as in the testis but the vascular concentrations are greater. The gonopore canal does not have a distinct antrum but the cells surrounding the canal are vacuolated as in the testes. The internal tissues are organized as cords which attach peripherally and extend into the center of the ovary (Fig. 7). Each cord includes what appear to be covering follicle cells, dis- persed internal vitelline material, and occasional internal oocytes. In some gonads the most mature oocytes occurred on the dorsal wall. The ovaries have internal open spaces and do not appear to be mature. The largest oocytes found were in the 40-50 wu range. All specimens studied were incomplete; the terminal region of the trunk has not been found. Discussion The spaghetti worms are readily assigned to phylum Hemichordata and class Enteropneusta but they are not readily assigned to any one of the three presently accepted families (Hyman 1959). The arrangement including Ptychoderidae, Spengeliidae, and Harrimaniidae was established by Spengel (1901). Caullery and Mesnil (1904) did establish a new family for Protoglossus koehleri but Burdon- Jones (1956) placed the genus in Harrimaniidae and returned to the three family pattern. Members of Ptychoderidae are distinguished in part by the presence of lateral septa in the trunk, dorso-lateral ciliary grooves in the intestinal wall, hepatic caeca, synapticulae, pygochord, short skeletal crura, and a cauliflower organ (Horst 1939; Dawydoff 1948). Spaghetti worms lack all of the above. Ptychoderids lack esophageal pores which are present in spaghetti worms. Members of Spengeliidae are distinguished in part by the presence of well- developed muscles and a vermiform process in the proboscis as well as by circular muscles internal to the longitudinal muscles of the trunk. These characters are not found in the spaghetti worms. Separation of the branchial gut into dorsal and ventral areas is characteristic of spaghetti worms but not generally in spengeliids. Both have long skeletal crura but they are recurved in the spaghetti worms. Members of Harrimaniidae are distinguished in part by the presence of well- developed muscles in the proboscis (often arranged in concentric rings), and large yolky eggs and presumably direct development. The proboscis muscles of the spaghetti worms are very poorly developed, and the gonads of the females from the one collection contain small ova and have an internal organization which suggests a form with indirect development. Unlike the harrimaniids the spaghetti worms have widely spaced double dorsal mesenteries in the proboscis, a proboscis skeleton with a unique coronate pattern in cross section, a very simple epithelium and arrangement of the post branchial gut, and testicular antra. 364 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON When compared with each of the three families for overall similarity, Saxi- pendium coronatum is found to be closest to the family Harrimaniidae but rec- ognition of its many special characteristics complemented by a suite of other characteristics emphasizes the need to place it in a new family. Although many valuable contributions to our understanding of enteropneusts have been made during the past 40 years, it is important to note that nearly one- half of the species and three-fourths of the genera were described nearly a century ago between 1891 and 1910. In addition, since that time there has been limited consideration of the existing systematic pattern and evolutionary relationships of the species and genera of the class. It is hoped that the presence of the spaghetti worms in association with the interesting hydrothermal vent community along with the unusual characteristics of this form will lead to a general increase of interest in members of this group from all habitats and throughout the world. Acknowledgments I thank J. F. Grassle (WHOI) for assigning the material to me and J. A. Blake (Battelle New England) for initiating the arrangements for that assignment. Re- search was supported in part by a grant from NFR of Sweden through the Institute of Zoology, Uppsala University, with administrative and academic assistance of Carl-Olof Jacobson and Ake Franzen. Collaboration began at the institute and continued at California State University, Fresno, which has supported the project through internal funding. R. R. Hessler (SIO) provided the videotape, information, and the opportunity to discuss the spaghetti worms with other scientists involved in deep-sea research. This paper is Contribution no. 55 of the Galapagos Rift Biology Expedition, supported by the National Science Foundation. Literature Cited Brambell, F. W. R., and H. A. Cole. 1939. Saccoglossus cambrensis sp. n., an enteropneust occurring in Wales. — Proceedings of the Zoological Society, London Series B 109:211-—236. Burdon-Jones, C. 1956. Observations on the enteropneust, Protoglossus koehleri (Caullery and Mes- nil).— Proceedings of the Zoological Society, London 127:35-58. Corliss, J. B., and R. D. Ballard. 1977. Oases of life in the cold abyss.— National Geographic 152(4): 440-453. Caullery, M., and F. Mesnil. 1904. Contribution a l’étude des Entéropneustes. Protobalanus (n.g.) koehleri Caull. et Mesn.—Zoologische Jahrbucher, Abteilung fiir Anatomie und Ontegeni der Tiere 20:227—256. Dawydoff, C. 1948. Embranchement des Stomocordés. Pp. 365-532 in Traité de Zoologie, Tome XI. Masson et Cie, Paris. Hessler, R. 1981. Oasis under the sea.—New Scientist 92(1283):741-746. Horst, C. J. van der. 1939. Hemichordaten. Jn Bronn’s Klassen und Ordnungen des Tier-Reichs Band 4, Abteilung 4, Buch 2, Teil 2, 737 pp. 1940. Ona new South African species of Balanoglossus and a comparison between it and Balanoglossus capensis (Gilchrist).— Annals of the South African Museum 33(2):69-93. Hyman, L. H. 1959. Hemichordata. In “The Invertebrates. Vol. V, Smaller Coelomate Groups.” McGraw-Hill, New York, pp. 72-207. Jones, M. L. 1981. Riftia pachyptila, a new genus, new species. The vestimentiferan worm from the Galapagos Rift geothermal vents (Pogonophora).— Proceedings of the Biological Society of Washington 93(4):1295-1313. Marion, A. 1886. Etudes zoologiques sur deux espéces d’entéropneusts (Balanoglossus hacksi et Balanoglossus talaboti).— Archives Zoologique Expérimentale, Series 2, 4:305-326. VOLUME 98, NUMBER 2 365 Sedgwick, J. 1980. Fred Grassle dives to a strange new world.—International Wildlife 10(6):49-53. Spengel, J. W. 1893. Die Enteropneusten des Golfes von Neapel.— Fauna und Flora des Golfes von Neapel, Monograph 18. 1901. Die Benennung der Enteropneusten-Gattungen.— Zoological Jahrbucher, Abteilung fiir Systematik 15(2):209-218. . 1907. Studien iiber die Enteropneusten der Siboga-Expedition. —Siboga Expedition, mono- graph 26:1-126, 17 tables. Thomas, I.M. 1972. Action of the gut in Saccoglossus otagoensis (Hemichordata: Enteropneusta). — New Zealand Journal of Marine and Freshwater Research 6(4):560-569. (KHW) Department of Biology, California State University, Fresno, California 93740-0001; (TS) Department of Zoology, Uppsala University, Uppsala, Sweden. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 366-390 NOMENCLATURAL AND TAXONOMIC NOTES ON THE PTERIDOPHYTES OF COSTA RICA, PANAMA, AND COLOMBIA, II David B. Lellinger Abstract.—The purpose of this paper and the one which preceded it (Proc. Biol. Soc. Washington 89:703-—732. 1977) is to publish lectotypes, new combinations, and new species of pteridophytes that will be included in my forthcoming “‘Ferns and Fern-allies of Costa Rica, Panama, and the Choco.” Throughout this paper in the citation of type specimens, the word “‘photo”’ followed by a number refers to the numbered series of photographs taken by C. V. Morton and distributed from the U.S. National Museum. Photographs by others are identified by the name (and number, if any) of the person or institution who took them. The principal entries are in alphabetical order for easy reference. In those cases where the principal entries are taxonomic or nomenclatural synonyms, the dis- position of these names is given as I understand it at present. Acrostichum aureum L. Sp. Pl. 2:1069. 1753 Lectotype.—To fix the application of this name, I choose: plate 7 of Plumier’s “Description des Plantes de l’Amérique,”’ which is based on a specimen collected by Plumier at Ft. Royal, Martinique. The specimen bearing “‘Acrostich. 10 au- reum” in the hand of Linnaeus (LINN 1245.5 not seen microfiche Smithsonian Institution Library), according to Jackson (Ind. Linn. Herb. 28. 1912), was not present in Linnaeus’ herbarium until after 1755, and so cannot be a type. Alsophila ichthyolepis Christ, Bull. Herb. Boiss. II, 6:186. 1906 Lectotype.—To fix the application of this name, I choose: La Palma, Pcia. S. José, Costa Rica, 1459 m, Tonduz 12527 (P not seen photo 3458; isolectotypes NY not seen, US). The other syntypes are: Costa Rica, Wercklé 57 (P not seen); Navarro, Pcia. Cartago, Costa Rica, Wercklé (P not seen); and Tsaki, Talamanca, Pcia. Limon, Costa Rica, 200 m, Pittier 9469 (P not seen). Alsophila ichthyolepis is a synonym of Trichopteris stipularis (Christ) Tryon. Anemia hirsuta var. humboldtiana Hieron. Bot. Jahrb. Engler 34:566. 1905 Lectotype.—To fix the application of this name, I choose: Between Caripe and Santa Cruz, Edo. Monagas—Sucre, Venezuela, Humboldt 459 (B-Hb. Willd. 19495-2 not seen Tryon photo US). The other syntypes are: Panamanian Isthmus, Colombia, Lehmann 1866 (B not seen); near Chaparral, Depto. Tolima, Colom- bia, 700 m, Lindig 213 (B not seen); Edo. Cumana, Venezuela, Moritz 158 (B not seen); Venezuela, van Lansberge [probably Landsberg] (B not seen); and Peru, VOLUME 98, NUMBER 2 367 Mathews 3299 (B not seen; isotype US). Anemia hirsuta var. humboldtiana is a synonym of A. hirsuta (L.) Swartz. Anemia pastinacaria Moritz in Prantl Untersuch. Morph. Gefaesskrypt. 2:110. 1881 Lectotype.—To fix the application of this name, I choose: “In convalli del Tigre,” Venezuela, Moritz 26 (B). The other syntypes are: Venezuela, Landsberg (B not seen); Caracas, Distr. Fed., Venezuela, Galeotti 196 (B not seen): Chaparral, Dep- to. Tolima, Colombia, 700 m, Lindig 213 [not Karsten, as Prantl has it; see Bot. Jahrb. Engler 34:566. 1905] (W not seen); Panama, Duchassaing (GOET not seen); Trapiche de la Concepcion, Edo. Oaxaca, Mexico, Liebmann (C not seen); and Trinidad, Fendler 62 (M not seen). Asplenium cuspidatum var. triculum Lellinger, sp. nov. Fig. 1 Plantae epipetricae vel epiphyticae, per radices repentes et gemmiferas multi- plicatas. Rhizoma erectum parvum ad apicem paleaceum; paleis ovato-lanceo- latis, ca. 2 mm longis 1 mm latis griseo-brunneis clathratis subbullatis persisten- tibus. Stipites 1-6 cm longi, 0.25—0.5 mm lati tereti exalati viridi vel adaxialiter brunnei, ad basin squamati, paleis eis rhizomatis similibus, supra basin paleis linearibus praediti. Rachides virides vel adaxialiter ad basin brunnei teretes vel ad apicem alatae sparse squamatae, paleis linearibus contortis fuscis. Laminae 2— 20 cm longae 0.8—4.5 cm latae papyraceae lanceolatae vel anguste lanceolatae, ad basin obtusae, ad apicem acuminatae vel caudatae, (1)2-pinnato-pinnatifidae in- terdum tripinnatae; pinnis petiolatis alternis inaequilateralibus, ad basin infe- riorem excavatis (parvissimis exceptis); pinnulis acutis vel acuminatis, basalibus acroscopicis saepe petiolulatis, acroscopicis dilatatis; segmentis angustis curvatis apiculatis; soris rectis usque ad 2 mm longis albidis integris. Type. — Basin of El General, Pcia. S. José, Costa Rica, 675-900 m, Skutch 4776 (US; isotypes F, GH). Paratypes.—HONDURAS: Depto. Sta. Barbara: San Pedro Sula, 1200 ft, Thieme (US). COSTA RICA: Pcia. Cartago: Rio Reventazon, Turrialba, Godfrey 66328 (GH); Valley of the Rio Reventazon 3 km SE of Turrialba, 500-600 m, Holm & IIltis 37 (US), ca. 600 m, Mickel 3355 (UC, US), ca. 1600 ft, Rossbach 3557 (GH); Vicinity of Pejivalle, ca. 900 m, Standley & Valerio 46892 (GH, US). Pcia. S. José: Cerro Turubales, 600 m, Brade & Brade 427 (UC, US); Vicinity of El General, 880 m, Skutch 2154, 2328 (both GH, US); Along the Rio Sonador near the Panamerican Hwy., 600 m, L. O. Williams et al. 28775 (US). Pcia. Puntarenas: Boruca, Pittier 4823 (US), Las Cruces Tropical Botanical Garden 6 km S of S. Vito de Java, 4000 ft, McAlpin [Las Cruces 74-445, 74-447] (both US); Osa Peninsula ca. 20 km S of Rincon de Osa, ca. 150 m, ca. 150 m, Mickel 23814 (US); Bahia Yglesias, Cocos Island, 10 m, G6mez 6555 (US). PANAMA: Pcia. Panama: Rio Piedras ca. 25 km NE of Cerro Azul, 550 m, Mori & Kallunki 3468 (US). Pcia. Darién: Cana, R. S. Williams 902 (US). VENEZUELA: Edo. Monagas: Montana de Aguacate between Caripe and Caripito, 600-900 m, Steyermark 62209 (US). Edo. Yaracuy: Fila La Enjalma S of Chivacoa, 500 m Vareschi & Pannier 2629 (US). PERU: Depto. Huanuco: Near the confluence of the Rio 368 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Holotype of Asplenium cuspidatum var. triculum Lellinger, Skutch 4776 (US). Fig. 2. Holotype of Asplenium excelsum Lellinger, Maxon 5205 (US). VOLUME 98, NUMBER 2 369 Cayumba with the Rio Huallaga, 860 m, Mexia 8290 (US). Depto. S. Martin: ' Near Tarapoto, Spruce 4035 (US). Discussion. — This is an entirely lowland variety that can be confused with small (pinnate-pinnatifid or 2-pinnate) forms of var. tripinnatum (Fourn.) Morton & Lellinger. Asplenium excelsum Lellinger, sp. nov. Fig. 2 Plantae terrestres. Rhizoma erectum (0.5)1—1.5 cm diam. paleaceum; paleis lanceolatis ca. 10 mm longis 2 mm latis brunneis clathratis marginibus subre- pandis. Stipites 12—30(40) cm longi ca. 3 mm lati profunde canaliculati exalati virides vel brunneoli, ad basin paleacei, paleis eis rhizomatis similibus. Rachides virides vel brunneolae canaliculatae uniformiter anguste alatae glabrae vel sparse squamatae, paleis linearibus ca. 1 mm longis 2 cellulis latis clathratis curvatis. Laminae ovatooblongae (32)35—65(70) cm longae (10)15-—30(42) cm latae papy- raceae, ad basin obtusae vel truncatae, ad apicem obtusae et abrupte acuminatae in modo Struthiopteridis, subtripinnatifidae vel tripinnato-pinnatifidae; pinnis manifeste petiolatis alternatis vel suboppositis subaequilateralibus, ad basin in- feriorem leviter excavatis, ad apicem acuminatis, costulis alatis; pinnulis oblongo- Ovatis; segmentis et venulis simplicibus vel ad apicem furcatis; soris brevibus rectis usque ad 3(4) mm longis albidis integris. Type.—Slopes of Cerro de Lina, above El Boquete, Pcia. Chiriqui, Panama, 1300-1560 m, Maxon 5205 (US; isotypes GH, NY, US). Paratypes.—NICARAGUA: Depto. Jinotega: Region of Las Mercedes, sierra E of Jinotega, 1200-1500 m, Standley 10725 (F, US). Depto. Managua: Sierra de Managua, 600-900 m, Bro. Antonio Garnier A744 (US). COSTA RICA: Pcia. Alajuela: Zarcero, A. Smith P.C.311 (F), 4500 ft, A. Smith F71 (F), 1550 m, A. Smith 48/140 (US), 1575 m, A. Smith 48/128 (US), 2000 m, A. Smith H.479 (US); San Antonio de Zarcero, 1475 m, A. Smith 48/310 (US). Pcia. Heredia: Vara Blanca de Sarapiqui, 1310 m, Skutch 3634 (NY, US). Pcia. S. José: La Palma, April 1910, Wercklé (US); Rancho Redondo, slopes of Volcan Irazu, ca. 6000 ft, Scamman 7744 (GH, US); Cerro de Piedra Blanca above Escazt, Standley 32519 (US); Tablazo, 1700 m, Brade & Brade [Ros. Fil. Costar. Exs. 43] (GH, NY); S. Pablo de Tarraza, O. Jiménez 1100 (US); El Copey, 1800 m, Tonduz 11704 (US), 7500 ft, Stork 1544 (US), ca. 2100 m, Lellinger 1788 (US); 5 mi S of Sta. Maria de Dota, 6600 ft, Stork 1779 (NY, US), 6800 ft, Stork 1756 (US); Vicinity of Sta. Maria de Dota, 1500-1800 m, Standley 41762 (US); Vicinity of El General, 1525 m, Skutch 4185 (GH, MO, NY, US); 16.5 mi S of Villa Mills on Interamerican Highway, 4700 ft, McAlpin 271 (DUKE); 25 km S of Villa Mills on Interamerican Highway, ca. 1450 m, Mickel 3277 (NY, US); Pacific slopes of Chirrip6 Massif at Abra, 2500 m, Davidse & Pohl 1531 (MO). Pcia. Cartago: Cerro Carpintera, 5000 ft, Stork 2124 (US), 1500-1800 m, Standley 34269 (US), 1500-1850 m, Standley 35504 (GH, US), 1800 m, Torres R. 111 (US); Mountains above Tres Rios, 5000 ft, Scamman & Holdridge 80521 (GH, US); Navarro, 1500 m, Torres R. 85 (US); Finca Navarro, 1350 m, Maxon 663 (NY, US); El Muneco, 1400-1500 m, Standley & Torres 51035 (US); La Estrella, Standley 39183 (US). PANAMA: Pcia. Chiriqui: Vicinity of El Boquete, 1000-1300 m, Maxon 5079 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 370 (SA) 109s uoxDypy ‘I9BUTTIIT MUuoxDU UniIUajdsy JO adi ojoy “p “3Iq (SA) £0ZE YoINYS “IsBUTIsT Wnuvizawos Wniuadsy Jo adAjojoH, +“¢ “314 VOLUME 98, NUMBER 2 371 (US), 1000-1500 m, Cornman 944 (US), 987 (F, US); Between Holcomb’s Trail and Monniche Plantation above El Boquete, 1500-1725 m, Killip 5082 (GH, NY, US); Vicinity of Casita Alta, Volcan de Chiriqui, ca. 1500—2000 m, Woodson, Allen & Seibert 837 (US); Rio Piedras Candela on logging road ca. 4 mi with junction of road to Las Mellisas, 1800-2000 m, McA/lpin 2208 (DUKE). Discussion. — This species is fairly common in dense forests and along streams in ravines. Commonly it has been identified as Asplenium solmsii Baker, which is a much smaller plant from Guatemala. It is related to A. commutatum Mett. ex Kuhn and to A. achilleifolium (Liebm.) Mart. & Gal. Asplenium gomezianum Lellinger, sp. nov. Fig. 3 Plantae epiphyticae. Rhizoma erectum vel ascendens ca. 3 mm diam. ad apicem paleaceum; paleis lanceatis ca. 5 mm longis 0.75 mm latis aeneis clathratis mar- ginibus integris. Stipites approximati 6-22 cm longi 1-2 mm lati canaliculati anguste alati virides vel brunneoli glabri. Rachides virides vel brunneolae an- gustissime alatae glabrae ad apicem leviter flexuosae. Laminae oblongae (10)18-— 42 cm longae (4.5)8—12(20) cm latae papyraceae, ad basin obtusae vel truncatae, ad apicem subobtusae abrupte acuminatae vel caudatae, pinnatae; pinnis oblongo- lanceolatis petiolatis alternatis inaequilateralibus, ad basin inferiorem excavatis ad basin superiorem truncatis non auriculatis, ad apicem acuminatis interdum acutis marginibus crenulatis; venis furcatis leviter curvatis; indusiis usque ad 13 mm longibus viridibus integris. Type.— Vara Blanca de Sarapiqui, Pcia. Herdia, Costa Rica, 1500-1750 m, Skutch 3203 (US; isotypes GH, NY). Paratypes.—COSTA RICA: Pcia. Puntarenas: Monteverde, 4500 ft, Palmer 100 (NY). Pcia. Alajuela: La Palma de S. Ramon, 1310 m, Brenes 3956 (NY); Rio La Balsa ca. 23 km NE of S. Ramon, 850 m, Taylor 17817 (NY); S of S. Ramon ca. 3 km above S. Rafael, ca: 1200 m, Lellinger 1346, 1347 (both CR, F, MO, US). Pcia. Heredia: Zarcero, 1500 m, A. Smith 48/198 (US). Pcia. S. José: La Palma, 1500 m, Valerio A94, 198 (both US). Pcia. Cartago: Cartago, 4250 ft, Cooper 6037 (US); 5 mi S of Cartago, ca. 1800 m, Maxon 507 (NY, US); Estrella, 4400 ft, Cooper 6038 (US), 5200 ft, Stork 3281 (GH, US); Navarro, 1500 m, Torres R. 76, 79 (both US); El Muneco, ca. 1400 m, Standley 33502, 33631 (both US); Vicinity of Orosi, Standley 39840 (US); 2.2 km SSE of Purisil above Finca La Concordia, ca. 1800 m, Lellinger 1479 (US), 1499 (CR, F, US), 1518 (F, US); Ca. 6 km SE of Tapanti, Wilbur & Stone 8901 (DUKE). PANAMA: Pcia. Chiriqui: Vicinity of El Boquete, 1000-1500 m, Cornman 876, 1012 (both US); Slopes of la Sierra del Boquete, 4400 ft, Bro. Maurice 828 (GH), 830 (US); Valley of the Rio Piarnasta above El Boquete, 1525-1575 m, Killip 5165 (US); Roballo Trail above El Boquete, 1600-1700 m, Killip 5426 (US). Pcia. Veraguas: Near the Escuela Agricola Alto Piedra near Sta. Fé, 730 m, Croat 34071 (MO). Discussion. — This species resembles A. auritum Swartz, but has broader, thinner pinnae, and A. auriculatum Swartz, but has pinnae lacking a basal acroscopic lobe. 372 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Asplenium harpeodes var. major Hieron. Hedwigia 60:237. 1919 Lectotype.—To fix the application of this name, I choose: Caracas, Distr. Fed., Venezuela, Funck & Schlim [probably actually Moritz] 246 (B not seen; isolec- totypes FI-Hb. Webb not seen, GH not seen, W not seen). This is a synonym of A. harpeodes Kunze. Asplenium maxonii Lellinger, sp. nov. Fig. 4 Plantae epiphyticae terrestres raro epipetricae. Rhizoma ascendens ca. (5)10 mm diam, usque ad 6(8) cm longum squamatum; paleis anguste lanceolatis usque ad 4 mm longis 0.5 mm latis subbicoloribus brunneis clathratis marginibus pal- lidioribus subclathratis integris. Stipites congesti 1-6 cm longi exalati. Stipites rachidesque canaliculati atropurpurei vel brunnei glabri. Rachides uniformiter anguste alati, alis viridibus, prolongati ad apicem bulbiliferi. Laminae rhombicae (9)25—45 cm longae (2.5)4—13 cm latae papyraceae, ad basin acuminatae raro obtusae, ad apicem obtusae vel subobtusae (rachidi prolongati excepto), tripin- natae; pinnis sessilibus lanceatis alternatis aequilateralibus; pinnulis oblongis in- aequilateralibus segmentis obovatis vel elongatis saepe bilobatis ad apicem ro- tundatis non apiculatis; indusiis ca. 2 mm longis subviridibus integris. Type.— Upper Caldera River, near “Camp I,’’ Holcomb’s trail above El Bo- quete, Pcia. Chiriqui, Panama, 1450-1650 m, Maxon 5601 (US; isotypes GH, NY, US). Paratypes.—COSTA RICA: Pcia. Puntarenas: Vicinity of the biological field station at Wilson finca, 1100-1400 m, Mickel 2013 (NY), 1100-1200 m, Mickel 3051a (NY), 3051b (NY, US); Finca las Cruces, 4000 ft, Gillis 10154 (A); Las Cruces, ridge road, Dressler 3589 (US); 3-5 km NW of the biological field station at Finca Wilson, 1300-1400 m, Lellinger 813 (US); Ca. 5 km NE of Finca las Cruces, ca. 1400 m, Burch 4514 (NY). Pcia. Alajuela: Upper drainage of the Rio Penas Blancas below the Monteverde Cloud Forest Nature Reserve, 1250-1350 m, Burger, Visconti & Gentry 10728 (NY); 11 km N of S. Ramon, 1000 m, Lellinger 747 (US), Mickel 2942 (NY, US); Viento Fresco, 1600-1900 m, Standley & Torres 47865 (US); San Antonio de Zarcero, 1500 m, A. Smith 48/286 (US). Pcia. Limon: Suerre, Llanuras de Sta. Clara, 300 m, Donnell-Smith 6890 (GH, NY, US); La Concepcion, Llanuras de Sta. Clara, 250 m, Donnell-Smith 6932 (NY). Pcia. Heredia: Vara Blanca, 1600-1700 m, Maxon 8316, 8341 (both US), 1500-1750 m, Skutch 3133 (GH, US); Yerba Buena NE of S. Isidro, ca. 2000 m, Standley & Valerio 49668 (US); Cerro de las Caricias N of S. Isidro, 2000—2400 m, Standley & Valerio 52056 (US). Pcia. S. José: La Palma, 1400 m, Brade & Brade [Ros. Fil. Costar. Exs. 27| (NY, US), ca. 1600 m, Standley 33110, 33124 (both US); vicinity of La Palma, 1450-1550 m, Maxon 368, 387 (both NY, US); Las Nubes, ca. 5000 ft, Scamman & Holdridge 8046 (GH, US); Quebrada Bajo Maquina, 3 km NE of Cascajal, 1640 m, Lent 2495 (GH); SW of Santiago de Puriscal, between Cerbatana and Mercedes Sur, ca. 1100 m, Lellinger 1574 (US); Quebrada de los Yases, between La Guaria and Palmilera, 1700 m, Jiménez M. 1459 (CR, F, NY, US); Vicinity of Sta. Maria de Dota, 1500-1800 m, Standley & Valerio 43318 (GH); El Copey, O. Jiménez 1103 (US); Lower SE slopes of Cerro Chirrip6, 5500-7000 ft, Evans & Lellinger 11 (US). Pcia. Cartago: Vicinity VOLUME 98, NUMBER 2 373 _ of Cartago, 1500 m, 19 April 1906, Maxon (NY); Finca Navarro, 1350 m, Maxon 627 (NY); El Muneco, 5000 ft, Stork 2661 (US); Vicinity of Orosi, Standley 39754 (US); Ca. 2.2 km SSE of Purisil, above Finca La Concordia, ca. 1800-2300 m, Lellinger 1550 (US); Ridge above Platanillo, 1200-1450 m, Mickel 3411 (NY, US); Platanillo, 650-900 m, de la Sota 5247 (US); New road from Tapanti ca. 7 km S of bridge, 1500 m, Hauke 416 (NY); E slope above the Rio Grande de Orosi ca. 10 km S of Tapanti, 1400-1600 m, Burger & Stolze 5696 (F, GH, US), ca. 15 km S of Tapanti, 1500 m, Burger & Liesner 6751 (F, GH); Road above the Rio Grande de Orosi, 12—16 km S of Tapanti, 1500-1600 m, Stolze 1482 (F, GH, NY, US); Flood plain of the Rio Reventazon 3 km SE of Turrialba, 500- 600 m, Holm & IItis 56 (US). PANAMA: Pcia. Chiriqui: Ca. 5 km NE of Boquete, 1700-1800 m, Skog 4048 (US); Cerro Horqueta, 6500 ft, von Hagen & von Hagen 2109 (NY); Valley of the Rio Caldera from El Boquete to the Cordillera, 1400- 1600 m, Killip 5062 (GH), 1650 m, 5238 (GH, US); Vicinity of El Boquete, 1000- 1500 m, Cornman 1010, 1187 (both US). Pcia. Coclé: 5 mi N of El Valle, Armond 375 (DUKE); El Valle de Anton, Vergara 8 (PMA, US). COLOMBIA: Depto. Choc6: Rio Nuqui, ca. 300 m, Haught 5497 (US); Rio Mutata ca. 3 km above its junction with the Rio El Valle, NW of Alto del Buey, ca. 800 m, Lellinger & de la Sota 145 (LP, US); NW side of Alto del Buey, ca. 1300 m, Lellinger & de la Sota 329 (COL, HUA, LP, US). ECUADOR: Pcia. Napo: Talag, 15 km SSW of Tena, 2000 ft, Grubb et al. 126 (US); Between Tena and Archidona, Asplund 9274 (US). PERU: Depto. Huanuco, 1600 m, Woytkowski 34518 (US). BOLIVIA: Depto. La Paz: Polo-Polo, 1100 m, Buchtien 3479 (US). Discussion. — This species is similar to A. rutaceum (Willd.) Mett., from which it differs in pinnule outline and in having round, rather than apiculate, segment apices (a character especially useful in distinguishing juvenile plants of the two species). This species was formerly called A. conquisitum Underw. & Maxon ex Christ, but that name has proved to be a synonym of A. rutaceum. Blechnum occidentale f. pubirhachis (Rosenst.) Lellinger, comb. nov. Blechnum occidentale var. pubirachis Rosenst. Hedwigia 46:94. 1906. Type.— Passo Mansa, Munic. Blumenau, Est. S. Paulo, Brazil, Haerchen 18 (S not seen; isotype NY not seen). Discussion.— This form appears throughout the range of the species, although it tends to occur at higher elevations than does f. occidentale, and often has smaller, more congested fronds. Ctenitis bidecorata Lellinger, sp. nov. Fig. 5 Plantae terrestres. Rhizoma erectum 5 mm diam. (basibus stipitum exclusis) paleaceum; paleis linearibus usque ad 10 mm longis 1 mm latis concoloribus aeneis pilosulis integris. Stipites approximati 7-20 cm longi squamati. Stipites rachides costaeque paleaceae dense aut sparse pilosae, paleis anguste lanceolatis 1.5-8 mm longis 0.1-—0.5 mm latis concoloribus aeneis aut brunneis epilosulis integris, pilis usque ad 1 mm longis multicellularibus hyalinis. Laminae lanceo- latae (12)15-37 cm longae (4)6—18(20) cm latae papyraceae, ad basin obtusae 374 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 5. Holotype of Ctenitis bidecorata Lellinger, Stork 1518 (US). Fig. 6. Basal pinna of holotype of Ctenitis skutchii Lellinger, Skutch 2337 (US). VOLUME 98, NUMBER 2 375 bipinnato-pinnatifidae, ad apicem acuminatae pinnatifidae, pro parte maxima ' pinnato-pinnatifidae; pinnis sessilibus lanceatis inaequilateralibus (pinnis basa- libus triangularibus basiscopice productis); segmentis oblongis obtusis, venis us- que ad 6-jugis; soris rotundis ca. 1 mm diam. submedialibus exindusiatis. Type.—Estrella—Sta. Maria road, Pcia. San José, Costa Rica, 7000 ft, Stork 1518 (US). Paratypes.—COSTA RICA: Pcia. Heredia: Aromal del Volcan Barba, 2800 m, Valerio 28 (CR, US); Porrosati, 1900 m, Gémez 3553, 3564 (both CR). Pcia. S. José: Zurqui, 1900 m, Gdémez 3525 (CR); between Estrella and Sta. Maria, 7000 ft, Stork 1523 (UC). Pcia. Cartago: Road above the Rio Grande de Orosi, 12-16 km S of Tapanti, 1500-1600 m, Stolze 1495 (F, US). Discussion. — Christensen (in herb. US) considered this species to be a variety of what is now called Ctenitis macrotheca (Fée) Ching, a member of Ctenitis subg. Subincisae. However, the relationship does not seem close to me. Ctenitis skutchii Lellinger, sp. nov. Fig. 6 Plantae terrestres. Rhizoma erectum ca. 1 cm diam. (basibus stipitum exclusis) paleaceum; paleis linearibus usque ad 15 mm longis 1 mm latis concoloribus aeneis denticulatis raro integris. Stipites approximati 15-55 cm longi paleacei et sparse pilosuli, paleis eis rhizomatis similibus, pilis appressis usque ad 0.5 mm longis hyalinis. Rachides costaeque paleaceae et dense pilosulae, paleis lanceolatis 1-3 mm longis 0.1-0.75 mm latis concoloribus subclathratis brunneis integris, pilis eis stipitum similibus. Laminae oblongae (30)50—100(150?) cm longae (25)30-— 60(80) cm latae papyraceae, ad basin obtusae manifeste tripinnato-pinnatifidae, ad apicem acuminatae pinnato-pinnatifidae, pro parte maxima tripinnatae vel obscure tripinnato-pinnatifidae; pinnis pinnulisque sessilibus oblongis aequila- teralibus (pinnis basalibus triangularibus basiscopice productis); segmentis ob- longis obtusis vel truncatis sparse pilosulis marginibus pilosulis, pilis curvatis; venis ca. 4-jugis; soris rotundis medialibus exindusiatis. Type.— Vicinity of El General, Pcia. S. José, Costa Rica, 1190 m, Skutch 2337 (US). Paratypes.—COSTA RICA: Pcia. Puntarenas: Atlantic-facing ridge beyond the Monteverde Forest Reserve near road across the ridge, ca. 1600 m, Lellinger 1748 (US); Canas Gordas, 1100 m, Pittier 10993 (US); Vicinity of the biological field station at Finca Wilson, 1100-1200 m, Lellinger 814 (US). Pcia. Alajuela: Upper drainage of the Rio Penas Blancas below the Monteverde Cloud Forest Nature Reserve, 1250-1350 m, Burger, Visconti & Gentry 10774 (CR, F). Pcia. S. José: 8-12 km SE of Desemparados on Calle Tablazo, 1800-1900 m, Utley & Utley 3046 (F). PANAMA: Pcia. Chiriqui: Rio Quebrada, 1650 m, Killip 5502 (US, misnumbered 5/36 at GH); Ca. 5 km NW of the town of Cerro Punta, 2000- 2300 m, Wilbur et al. 15230 (DUKE). Discussion.—This species belongs to Ctenitis sect. Subincisae and has com- monly been called C. subincisa (Willd.) Ching, but differs from that species in having pilosulous axes bearing usually flat, slightly toothed scales. It is closer to C. adenopteris (C. Chr.) Ching than to C. subincisa. 376 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Cyathea pseudonanna (Gomez) Lellinger, comb. nov. Trichopteris pseudonanna Gomez, Phytologia 50:69. 1981. Type.—Cerro Tuti, Edo. Veraguas, Panama, Folson & Edwards 3370 (MO). Discussion.—A second gathering of this species from the type locality (Knapp & Sytsma 2581, MO, US) has minute, hemitelioid indusia. The specimen has alate rachises reminiscent of C. parva (Maxon) Tryon, but probably is most closely related to C. decorata (Maxon) Tryon, both members of the C. speciosa group as defined by R. M. Tryon (Contr. Gray Herb. 206:46—47. 1976). Diplazium ferulaceum (Moore ex Hook.) Lellinger, comb. nov. Asplenium ferulaceum Moore ex Hook. Sp. Fil. 3:216. 1860. Syntypes.—Co- lombia, Hartweg 1519 (K not seen); and Quito, Pcia. Pichincha, Ecuador, Jameson (K not seen). Gymnogramme sciatrophis Donn.-Sm. Bot. Gaz. 19:266, t. 26. 1894, as “‘scia- traphis.”’ Type.—Rio Jiménez, Llanuras de Sta. Clara, Pcia. Limon, Costa Rica, J. D. Smith 5084 (US; isotype NY). Discussion. — This species apparently is the endpoint in lamina reduction leading from D. ordinatum (Christ) Lellinger and D. solutum (Christ) Lellinger through D. myriomerum (Christ) Lellinger. See D. solutum for a further comment. Diplazium multigemmatum Lellinger, sp. nov. Riga, Plantae terrestres. Rhizoma erectum(?) paleaceum(?). Stipites ca. 70 cm longi sparse paleacei, paleis ca. 6-15 mm longis 1.25-—3 mm latis concoloribus atro- brunneis sparse denticulatis, ad basin atrati, distaliter viriduli. Rachides glabrius- culae viridulae. Laminae oblongae ca. 75 cm longae 25—40 cm latae papyraceae, ad basin obtusae vel truncatae bipinnatae, ad apicem anguste acutae pinnato- pinnatifidae; pinnis petiolulatis (usque ad 1.5 cm) lanceatis (paribus infimis lan- ceolatis) usque ad 27 cm longis 7 cm latis aequilateralibus subfalcatis pinnatifidis raro ad basin pinnatis, ad apicem acutis aut acuminatis; segmentis oblongis sub- obtusis aut rotundis marginibus crenatis; venis 5—8(10)-jugis bifurcatis; indusiis 2-5 mm longis marginibus sparse aut dense ciliatis; gemmulis minutis nigris saepe in ramis acroscopicis venis basalibus acroscopicis. Type.—North slopes of Cerro Chompipe ca. 10 km NNE of Heredia, Pcia. Heredia, Costa Rica, 2200 m, Lellinger 1006 (US). Paratypes.—COSTA RICA: Along Interamerican Highway, Scamman 7143 (GH). Pcia. Alajuela: La Palma de S. Ramon, Brenes 10560 (NY). Pcia. Heredia: Porrosati, 1900 m, Gémez 3561 (CR). Diplazium myriomerum (Christ) Lellinger, comb. nov. Athyrium myriomerum Christ, Bull. Herb. Boiss. II, 6:169. 1906. Syntypes.— Navarro, Pcia. Cartago, Costa Rica, Wercklé in 1905 (P not seen; presumable isosyntype US); and Costa Rica, Wercklé 100 p. p. in 1903 (P not seen). Diplazium bradeorum Rosenst. Repert. Sp. Nov. Fedde 9:69. 1910. Type.— Carrillo, Pcia. S. José, Costa Rica, 400 m, Brade & Brade 383 (S not seen photo 6265; isotypes P not seen photo 4077, US). Discussion.—See D. solutum (Christ) Lellinger. VOLUME 98, NUMBER 2 377 Diplazium navarrense Lellinger, sp. nov. Fig. 8 Plantae terrestres. Rhizoma erectum dendroideum usque ad 25 cm longum 1 cm latum, ad apicem sparse paleaceum, paleis lanceatis ca. 4 mm longis 1 mm latis concoloribus brunneis integris. Stipites congesti 30-40 cm longi. Stipites rachidesque glabri viriduli sulcati exalati. Laminae oblongo-lanceatae 35-45 cm longae 20-30 cm latae papyraceae, ad basin truncatae, ad apicem acuminatae pinnatae, distaliter bipinnatae proximaliter pinnato-pinnatifidae; pinnis sessilibus alternatis oblongis usque ad 15 cm longis 3 cm latis aequilateralibus aut subae- quilateralibus, ad basin basiscopicis interdum excavatis, pinnatis pinnatisectis vel pinnatifidis, ad apicem lobatis caudatis; pinnulis segmentisque ovatis vel oblongis, venis usque ad 6-jugis usque ad 3(4)-furcatis; soris usque ad 4 mm longis curvatis saepe singularibus; indusiis brunneis subscariosis integris. Type.— Ravine of the Rio Naranjo, Finca Navarro, Pcia. Cartago, Costa Rica, 1350 m, Maxon 658 (US; isotype NY). Paratype.—Same as the type, Maxon 662 (NY, US). Diplazium ordinatum (Christ) Lellinger, comb. nov. Athyrium ordinatum Christ, Bull. Herb. Boiss. II, 4:967. 1904. Type.—Costa Rica, Wercklé (P not seen). Athyrium reductum Christ, Bull. Herb. Boiss. I], 4:966. 1904. Lectotype.— Alajelita, Pcia. S. José, Costa Rica, 1300 m, Alfaro 16472 (US; isolectotypes NY, P not seen), chosen by Lellinger (Proc. Biol. Soc. Washington 89:708. 1977). Athyrium costaricense Christ, Bull. Herb. Boiss. II, 5:252. 1905. Type.—Costa Rica, Wercklé (P not seen). Discussion.—See Diplazium solutum (Christ) Lellinger. Diplazium skutchii Lellinger, sp. nov. Fig. 9 Plantae terrestres. Rhizoma erectum dendroideum aetate usque ad 20 cm lon- gum 1 cm latum (stipitibus excluso), ad apicem dense paleaceum, paleis usque ad 10 mm longis 1.5 mm latis concoloribus atrobrunneis integris. Stipites ap- proximati 30-60 cm longi ad 5 mm lati straminei villosuli, villis contortis ca. 0.5 mm longis multicellularibus subcatenatis. Laminae oblongo-lanceolatae 50- 75(100?) cm longae 25—40(50) cm latae papyraceae, ad basin obtusae bipinnato- pinnatifidae, ad apicem acuminatae pinnato-pinnatifidae aut pinnatifidae; pinnis sessilibus alternis suboppositis (infimis petiolulatis usuge ad 1 cm) aequilateralibus oblongo-lanceolatis usque ad 25 cm longis 5 cm latis (infimis usque ad 30 cm longis 12 cm latis), ad basin truncatis, ad apicem anguste acutis; pinnulis vel segmentis oblongis, ad apicem rotundis, abaxialiter pilosulis, pilis hyalinis, venulis usque ad 12-jugis 2- vel 3-furcatis; soris 2—5 mm longis, plus minusve rectis saepe singularibus; indusiis brunneis ciliato-erosis. Type.— Vicinity of El General, Pcia. S. José, Costa Rica, 1160 m, Skutch 2959 (US; isotype NY). Paratypes.—NICARAGUA: Depto. Matagalpa: Sta. Maria de Ostuma, ca. 1500 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 378 (SA) 859 uoxnypy “19d3UTT]9T asuauivanu wnizpjdig Jo adho0joH °8 “314 (S A) 9001 4asuyjaT ‘1e80T9T wnipuumasiyjnu wnizpjdiq jo sdAyojoy Jo puoy Jo uoriod jeseg “1 “314 ry ermams wamen 4 mbar Oe 1 Any i ua .: w rad 2 eat A WNIHVENSH “WNOLLYN - ae > a or PINE “LE te 7 5 sa8urq 7} svuorxenwu anyzeidrd JO AdALOIOK ey pe Fy sacplae O9TtSbe 10 1 g2>) jo0qs--Ad AUTIOH : Sucppr] sawmwaszapnm anyzerdrd OLYN 63LV1S O3.LIND wld 40 SINV14 S3i¥iS O31INN VOLUME 98, NUMBER 2 379 _ m, Gomez, Cittar & Villa 6375 (CR). COSTA RICA: Pcia. S. José: Vicinity of El General, 1190 m, Skutch 2329 (NY, US). Pcia. Cartago: Carpintera, 1700 m, Brade 805 (UC). PANAMA: Pcia. Chiriqui: Vicinity of Camp Aguacatal, E slope of Volcan Chiriqui, Maxon 5288 (US); Vicinity of El Boquete, 1000-1300 m, Maxon 4939 (US), 1000-1500 m, Cornman 1098, 1125 (both US), 1362 (UC, US). Diplazium solutum (Christ) Lellinger, comb. nov. Fig. 10 Athyrium solutum Christ, Bull. Herb. Boiss. II, 4:967. 1904. Type. —Costa Rica, Wercklé in 1903 (P not seen photo 4038; isotype US). Discussion.—This species, along with D. ferulaceum (Moore ex Hook.) Lellin- ger, D. myriomerum (Christ) Lellinger, and D. ordinatum (Christ) Lellinger, were formerly placed in Athyrium; Tryon and Tryon (Ferns Allied Pls. 544-545. 1982) pointed out their affinity with Diplazium. Although most have single, rather than the typical double sori found in Diplazium, double sori are found occasionally in D. solutum, and the segments of these species are not spinulose nor are the veins readily visible on the abaxial lamina surface, as is usual in most Athyrium species. Dryopteris aspidioides var. subhastata C. Chr. K. Danske Vidensk. Selsk. Skr. VII, 4:287. 1907 Lectotype. —To fix the application of this name, I choose: Near Tarapoto, Depto. S. Martin, Peru, Spruce 3964 (C not seen; isolectotypes B not seen, presumably K not seen, US). The other syntypes are: Loreto, Depto. Loreto, Peru, Ule 6518 (P not seen); S. Gavan, Depto. Puno, Peru, Lechler 2311 (B not seen); Colombia, Lindig 53 (B not seen); and Caracas, Distr. Fed., Venezuela, Otto 596 (B not seen). Dryopteris aspidioides var. subhastata is a synonym of Thelypteris aspi- dioides (Willd.) Tryon. Dryopteris tablaziensis Christ Bull. Herb. Boiss. II, 7:262. 1907 Lectotype.—To fix the application of this name, I choose: La Palma, Pcia. S. José, Costa Rica, 1500 m, Wercklé 17010 (P not seen), examined by A. R. Smith (in litt. 22 Mar 1977). The other syntype is: Tablazo, Pcia. S. José, Costa Rica, 1900 m, Biolley 67 (P not seen); isosyntype US). Drypteris tablaziensis is a syn- onym of Thelypteris pilosula (Mett.) Tryon. Grammitis barbensis Lellinger, sp. nov. Fig. 11 Plantae epiphyticae. Rhizoma ascendens minutum manifeste paleaceum; paleis lanceatis ca. 3 mm longis 0.5 mm latis cinereis clathratis marginibus pilosis, pilis simplicibus furcatis vel stellatis ca. 0.15 mm longis. Stipites 3—9 mm longi 0.4 mm lati. Stipites rachidesque teretes atropurpureae striatae glabrae. Laminae anguste oblongae (3.5)5—20 cm longae (0.7)1—2 cm latae membranaceae, ad basin acutae vel acuminatae, ad apicem acutae, pinnatisectae; segmentis oblongis vel aliquando triangularibus leviter ascendentibus 4—10 mm longis 1—3 mm latis, ad basin basiscopicis decurrentibus, ad basin acroscopicis truncatis raro subgibbosis, PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 380 ‘JosuTT[eT (SYD) “Ol (4) $696 2Z/01§ = 4a8ing wiNnInjos unizDjdiq Jo puosy “OL “StI (SN) 6562 YIMS soBUTa] uyomys wnizyjdiq Jo edAojoy Jo puoy jo uonsod jeseg 6 “SIA hie tS > a as LL ay boa wet 4 ak WE 12 tofurp [oT tt4yoan4s enyaerdra yo AAALOTON we sf, a 4 aay8) VOLUME 98, NUMBER 2 381 ad apicem acutis vel raro rotundis, glabris marginibus integris vel leviter repandis; ' soris submedialibus rotundis leviter impressis. Type.— Volcan Barba, Pcia. Heredia, Costa Rica, 3000 m, Valerio 209 (US; isotype CR). Paratypes.—Osa Peninsula, Pcia. Puntarenas, Costa Rica, 50-600 m, Mickel 1942 (NY). Discussion.—This species is a member of subg. Cryptosorus and seems to be related to G. suprasculpta (Christ) Seymour and to G. micula Lellinger. Grammitis cornuta Lellinger, sp. nov. Fig. 12 Plantae epiphyticae. Rhizoma ascendens ca. 4 mm diam. paleaceum; paleis anguste lanceatis usque ad 10 mm longis | mm latis aeneis vel rufobrunneis nitidis repandis, ad apicem setaceis. Stipites approximati 7-30 cm longi usque ad 2 mm lati tereti brunnei setiferi et sparse paleacei, setis rectis usque ad 2 mm longis brunneis, paleis eis rhizomatis similibus. Stipites rachidesque glanduliferae, glan- dulis submoniliformibus saepe irregulariter furcatis, cellulis globosis leviter elon- gatis. Laminae oblongae vel lanceolatae (5)8—25 cm longae 4—9(12) cm latae subcoriaceae, ad basin truncatae vel obtusae, ad apicem acutae, profunde pin- natifidae; segmentis oblongis saepe subfalcatis 2-5(9) cm longis 5—8 mm latis, ad basin basiscopicis decurrentibus vel subtruncatis, ad basin acroscopicis truncatis vel leviter surcurrentibus, ad apicem acuminatis, paginis abaxialibus glanduliferis, costis adaxialibus leviter setiferis, marginibus involutis maturitate; venulis ob- scuris anastamosantibus; soris leviter submedialibus rotundis 2—3 mm diam. non impressis. Type. —Las Nubes, Pcia. S. José, Costa Rica, ca. 1500-1900 m, Standley 38843 (US; isotype GH not seen). Paratypes.—COSTA RICA: Pcia. Heredia: Alto del Roble, ca. 11 km NNE of Heredia, 2100 m, Lellinger 1062 (US). Pcia. Cartago: Sta. Clara de Cartago, 1950 m, Maxon 8201 (US). PANAMA: Pcia. Chiriqui: Between Alto de las Palmas and the top of Cerro de la Horqueta, 2100-2268 m, Maxon 5509 (US). Discussion.— This species is a member of subg. Cryptosorus and is distinct in having areolate venation, a most unusual condition in Grammitis and one prob- ably resulting from the wide segments. Grammitis micula Lellinger, sp. nov. Fig. 13 Plantae epiphyticae. Rhizoma erectum vel ascendens paleaceum; paleis lan- ceatis usque ad 2 mm longis 0.25 mm latis cinereis clathratis marginibus sparse pilosis, pilis simplicibus ca. 0.2 mm longis hyalinis. Stipites 0.5—1 mm longi 0.15- 0.2 mm lati exalati. Stipites rachidesque teretae atrae striatae sparse pilosuli juventute, pilis simplicibus aut furcatis hyalinis. Laminae oblongae (3)5—15(22) cm longae (0.8)1—1.5(3) cm latae membranaceae, ad basin et apicem acutae, pinnatae vel pinnato-pinnatifidae; pinnis oblongis leviter ascendentibus 4-17 mm longis 2—3.5 mm latis, ad basin basiscopicis decurrentibus, ad basin acroscopicis lobatis, ad apicem acutis vel acuminatis, crenatis aut lobatis glabris; soris sub- medialibus rotundis leviter impressis. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 382 GA) EF RSE AaIpUDIg ‘xeBUTIOT VINULOD sHMULUDI Jo aAMOIOH “71 “BA (SA) 60Z MAIDA “xOBUNIPT SISUIqING SUIUUDAD J o odkojoH “TT “31d VOLUME 98, NUMBER 2 383 Type.—End of access road just above face of mine, Cerro Colorado, on the border of Pcias. Chiriqui and Bocas del Toro, Panama, 1600-1700 m, Folsom, Small & Robbins 4744 (US; isotype MO not seen). Paratypes.—COSTA RICA: Pcia. S. José: Vicinity of El General, 1190 m, Skutch 2831 (US), 1000 m, 3015 (GH, NY, US). Discussion.—This species is a member of subg. Cryptosorus and seems to be related to G. suprasculpta (Christ) Seymour and to G. barbensis Lellinger. Grammitis pseudomitchellae Lellinger, sp. nov. Fig. 14 Plantae epiphyticae. Rhizoma erectum minutum epaleaceum. Stipites obsoleti aut usque ad 7 mm longi atri pilosi, pilis ca. 1 mm longis tenuibus laxis. Laminae oblongae, 1.5—6 cm longae 2—4 mm latae membranaceae, ad basin attenuatae, ad apicem leviter anguste rotundae, profunde crenato-lobatae, ad apicem fertilem crenatae; lobis ascendentibus rotundis sparse pilosis, pilis 1 mm longis rectis hyalinis vel brunneolis; venulis non furcatis; soris submedialibus rotundis vel leviter elongatis non impressis. Type.—10km N ofS. Rafael de Heredia on Volcan Barba, Pcia. Heredia, Costa Rica, 1950 m, Mickel 3033 (NY). Paratypes.—PANAMA: Pcia. Chiriqui: Cordillera above ““Camp I,’’ Holcomb’s Trail 10 mi above El Boquete, 1800-2000 m, Killip 5289a (US); Cordillera, E of the Rio Caldera, 2150 m, Killip 5331 (US). Discussion. — This species is a member of subg. Xiphopteris. It closely resembles G. mitchellae (Baker) Stolze from Guatemala, from which it differs in having simple veins, non-setose sporangia, fewer and shorter marginal setae, and more and shorter setae on the abaxial surface of the laminae. Grammitis rigescens (Bory ex Willd.) Lellinger, comb. nov. Polypodium rigescens Bory ex Willd. Sp. Pl. ed. 4, 5:183. 1810. Type. — Bourbon [Réunion], Bory (B-Hb. Willd. 19668 not seen microfiche Smithsonian Institution Library; isotypes FI not seen photo 16023, P-Hb. Juss. not seen photo 2951). Discussion. — According to Pichi Sermolli (Webbia 37:118-125. 1983), Morton (Contr. U.S. Natl. Herb. 38:57-59. 1967) mislectotypified Polypodium flabelli- formis Poir. in Lam., and so considered it to be an earlier name for P. rigescens Bory ex Willd. Pichi Sermolli has lectotypified the name on plate 87 (left) of Plumier’s ““Tractatus de Filicibus Americanis.” His arguments for doing so are persuasive, and so Grammiitis flabelliformis (Poir, in Lam.) Morton is the correct name for the small Antillean species which has been passing as G. taenifolia (Jenm.) Proctor. Grammitis zeledoniana Lellinger, nom. nov. Polypodium taxifolium var. fragillimum Christ, Bull. Herb. Boiss. II, 4:1103. 1904, non Grammitis fragillima (Copel.) Morton. Type.—Costa Rica, Wercklé in 1903 (P). Selected specimens examined.—COSTA RICA: Pcia. Alajuela: 11 km N of S. Ramon, 1000 m, Lellinger 754 (US). Pcia. Heredia: Cerros de Zurqui, NE of S. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 384 (AN) €£0€ 12191 croBuTaT] anjjayomuopnasd smuUuDsy Jo eAMO[OH “pl “BI (SA) HELP SUIQION P [1DMS ‘mospo.T “SOBUT[ST] VINDIUA SHIMADA) JO adMojoH “El “Sty whidWauaH IWNOLLWN kour[ [91 aetrausarsopaowd uw a® TINS * ro8urT [91 edd Sebel Ss my arnt co [nore (ansosoadka9) #¢ayeeeAD FO FARLOTON Gp qdoydtx) sfayemesy 3° AALLOTOH ¥9Id V1SO9 4O SINV Td o/s VOLUME 98, NUMBER 2 385 Isidro, 2000-2400 m, Standley 52000 (US), 50453 (US). Pcia. S. José: La Palma area NE of S. Jeronimo above the La Hondura valley, 1500 m, Burger & Stolze 5381 (F). Pcia. Cartago: SE of Orosi, ca. 2.2 km SSE of Purisil, above Finca la Concordia, ca. 1800-2300 m, Lellinger 1525 (US). PANAMA: Pcia. Chiriqui: Cerro Colorado, along mining road 31.6 km beyond the bridge over the Rio S. Felix, 1690 m, Croat 37133 (MO); Between Alto de las Palmas and the top of Cerro de la Horqueta, 2100-2268 m, Maxon 5449 (US), 5481 (US). Pcia. Darién: Serrania de Pirre, ca. 8 km W of Cana Gold Mine, 1430-1480 m, Croat 37833 (MO). Discussion. — This species has usually been confused with G. taxifolia (L.) Proc- tor, from which it differs in having glabrous, rather than sparsely setose, segment margins and in being densely setose at the base of the stipes and more sparsely setose distally, rather than in having evenly setose stipes. Hymenophyllum mortonianum Lellinger, sp. nov. Fig. 15 Plantae epiphyticae. Rhizoma repens, ca. 0.4 mm diam. sparse pilosum, pilis ca. 1 mm longis hyalinis vel-brunneolis manifeste pluricellularibus. Stipites 3-13 mm longi, basin versus alati. Stipites rachidesque teretes atrobrunneae striatae alatae pilosae, pilis stellatis sessilibus vel subsessilibus 4—6-brachiatis, brachiis ca. 0.4 mm longis hyalinis vel rufobrunneolis. Laminae anguste rhombicae in- determinatae(?) usque ad 23 cm longae 2.5 cm latae pinnato-pinnatifidae, ad basin et apicem acuminatae, apice plusminusve elongatae, pinnis oblongis vel trian- gularibus usque ad 10-lobatis, lobis 1 mm latis, marginibus minute stellato-pilosis; venulis simplicibus (venulis ad basin acroscopicis furcatis exceptis); involucris rotundis 0.75 mm diam., ad basin truncatis vel obtusis, marginibus dense stellato- pilosus. Type. —Moyjarras de Tado, 8.5 km E of Istmina, Depto. Choc6, Colombia, 150- 250 m, Lellinger & de la Sota 425 (US; isotypes COL, CR, HUA, LP). Discussion. — This species is a member of subg. Leptocionium. It is known only from the type. Hypolepis grandis Lellinger, sp. nov. Fig. 16 Plantae terrestres. Rhizoma ascendens ca. | cm diam. pilosum, pilis rufobrun- neis catenatis. Frondes usque and ca. 3 m longae repertis. Stipites ad basin atro- purpurei, supra basin rufobrunnei, nitidi pilosi juventute non spinosi. Rachides laminaeque pilosae et pilosulae, pilis 1-3 mm longis, multicellularibus catenatis hyalinis vel brunneolis, pilosulis 0.25—1 mm longis, 1—3-cellularibus plus minusve acicularibus hyalinis. Laminae probabiliter deltatae usque ad 2 m longae 1 m latae, tripinnato-pinnatifidae papyraceae; pinnis basalibus usque ad 70 cm longis 35 cm latis basiscopice productis, pinnulis acroscopicis 20 cm longis 7 cm latis, pinnulis basiscopicis 15 cm longis, 4 cm latis; pinnis medianis oppositis lanceolatis vel oblongis usque ad 50 cm longis 30 cm latis, pinnulis ca. 20-jugis subsessilibus oblongis usque ad 15 cm longis 4 cm latis; segmentis oblongis vel lanceatis pro- funde pinnatifidis, lobulis 1.75—2.5 mm latis ad apicem rotundis; indusiis scariosis manifeste pilosis deciduis. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 386 (SA) [EFI ‘SXA 4DISOD [LT SOY] OPE apoAG *P apdsg “JoBUTTIF] Sipusd sidajodap] Jo (euurd [eseq e A[qeqoid jo uotjiod) adAjojoH “91 ‘BL (SA) STP DIOS 0] ap Pp sasuyjaT “1osuta] Wwnuviuojiou wnydydouamAy jo adAojoyH “SI “34 i ¥ x an OES »prag VOLUME 98, NUMBER 2 387 Type.—La Palma, Pcia. S. José, Costa Rica, 1400 m, Brade & Brade 348 [Ros. Fil. Costar. Exs. 143| (US; isotypes NY, UC). Paratypes.—COSTA RICA: Pcia. S. José: 5 mi S of Sta. Maria de Dota, 6800 ft., Stork 1754 (GH, UC, US). Pcia. Cartago: El Muneco, 4800 ft, Stork 2732 (NY, UC, US); La Estrella, Standley 39214 (US). PANAMA: Pcia. Chiriqui: Valley of the Rio Quebrada above El Boquete, 1650 m, Killip 5137 (US). Discussion.— This species resembles the South American H. stuebelii Hieron. in indument, but has larger laminae that are tomentose on the adaxial surface. Osmunda hirsuta L. Sp. Pl. 2:1064. 1753 Lectotype.—To fix the application of this name, I choose: plate 162 of Plumier’s ‘““Tractatus de Filicibus Americanis,”’ which is based on a specimen collected by Plumier near the Leogane quarter, Sto. Domingo, Hispaniola. According to Jack- son (Index Linn. Herb. 111. 1912), there was no specimen of this species in the Linnaean herbarium in the hand of Linnaeus prior to 1767. Osmunda hirsuta is a synonym of Anemia hirsuta (L.) Swartz. Pecluma ptilodon var. caespitosa (Jenm.) Lellinger, comb. nov. Polypodium pectinatum var. caespitosum Jenm. Bull. Bot. Dept. Jamaica, n.s., 4:125. 1897. Type.—Old England, St. Andrew Parish, Jamaica, 4000 ft, Jenman (NY? not seen). Pityrogramma ebenea var. aurata (Moore) Lellinger, comb. nov. Gymnogramma tartarea var. aurata Moore, Gard. Chron. 1870:493. 1870. Type. — Cultivated by Veitch, originally from Peru (not seen); according to Tryon (Contr. Gray Herb. 189:66. 1962), Peru, Pearce 182 (K not seen photo GH not seen) is authentic or perhaps even the type. Polypodium ambiguum Mett. ex Kuhn Linnaea 36:134. 1869, non Desv., 1827, nom. illeg. Lectotype.—To fix the application of this name, I choose: Near Colonia Tovar, Edo. Aragua, Venezuela, Fendler 254 (US). The other syntype is: Colonia Tovar, Edo, Aragua, Venezuela, Moritz (not seen). No specimens, but only a drawing of part of one frond of the Fendler specimen, exist at B. Polypodium ambiguum is a synonym of P. ursipes Moritz ex C. Chr. Polypodium beyerianum Rosenst. Repert. Sp. Nov. Fedde 22:17. 1925 Lectotype.—To fix the application of this name, I choose: Turrialba, Pcia. Car- tago, Costa Rica, 650 m, A. C. Brade 21 (S; isoelectotype UC). The other syntype is: Puntarenas, Pcia. Puntarenas, Costa Rica, Beyer 44 (S not seen). Polypodium beyerianum is a synonym of P. loriceum L. Polypodium fraxinifolium var. articulatum Christ Bull. Herb. Boiss. II, 6:49. 1906 Lectotype.—To fix the application of this name, I choose: Turrialba, Pcia. Car- tago, Costa Rica, 550 m, Pittier 9061 (US; presumable isolectotype P not seen). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 388 (SN) SO8rr OMavA ~P Aajpuvnig ‘IaBUT[[I] S1SUII11D]SOI VIIDIZ1A JO adAjofoHJ “81 “3 (SN) ZITI 4pauuay “1o8uTeT] apliaqns unipoddjog Jo adAojoy, “LI “3I4 Voly VLSOD AO SLNV 1d Sl Nan teem AdALOION ant pads 10 4 Tol b692 VOLUME 98, NUMBER 2 389 The other syntypes are: Carillo, Pcia. S. José, Costa Rica, Pittier 1162 (P not seen); Tsaki, Talamanca, Pcia. Lim6én, Costa Rica, Tonduz 9451 (P not seen; isosyntype US); and Costa Rica, Wercklé (P not seen). Polypodium fraxinifolium var. articulatum is a synonym of P. giganteum Desv. Polypodium subviride Lellinger, sp. nov. Fig. 17 Plantae epiphyticae. Rhizoma late repens 2—3.5(4) mm diam., phyllopodiis 2- 4 mm longis, irregulariter striatum atrum brunneum vel rufobrunneum pale- aceum, paleis lanceolatis peltatis ca. 2 mm longis 0.5 mm latis, ad basin fuscatis clathratis, alibi hyalinis, deciduis (aliquot basibus circularibus exceptis) margi- nibus erosis. Stipites 2.5-8 cm distantes 2-14 cm longi 1—2 mm lati anguste alati brunneoli aut rufobrunneoli. Stipites rachides laminaeque pilosae, pilis 2 mm longis laxis multicellularibus hyalinis. Laminae oblongae vel raro anguste ellip- ticae (15)20—33(40) cm longae (3.5)4—5.5 cm latae papyraceae subviridae, ad basin obtusae vel truncatae, lobis abortivis infra basin, ad apicem acuminatae vel acutae, subpinnatisectae; segmentis oblongis saepe leviter falcatis, ad basin basiscopicis truncatis, ad basin acroscopicis surcurrentibus, ad apicem acutis vel rotundis, marginibus integris leviter revolutis; soris 0.75-1.25 mm diam. submedialibus 1-seriebus. Type.—La Eneida region near Cerro Jefe, Pcia. Panama, Kennedy 1117 (US; isotype DUKE). Paratypes.—PANAMA: Pcia. Veraguas: 6-7 km W of Sta. Fé on road past agriculture school, 2900 ft, Nee 9727 (US). Pcia. Coclé: 6 mi N of El Valle de Anton, 582 m, Armond 343 (DUKE); La Mesa, 850-900 m, Luteyn 4070 (DUKE), ca. 800 m, Skog 4112 (US); Above Penonomé, Williams 515 (NY, US). Pcia. Col6n: Road from Portobello to Maria Chiquita, Armond 490 (DUKE); Along Sta. Rita Ridge lumber road, Correa & Dressler 751 (US); Near bridge over the Rio Buenaventura, near Portobello, Foster 1705 (DUKE). Pcia. Panama: Along road from Cerro Jefe to La Eneida, 1007 m, Armond 325 (DUKE); Cerro Jefe, Correa & Dressler 160 (US). COLOMBIA: Depto. El Valle: Rio Cajambre, S. Isidro, 5-100 m, Cuatrecasas 17346 (US). Depto. Narino: Cordillera de Pasto, 1000-1300 m, Lehmann 20 (US). Discussion.—This rather common species (especially in central Panama) is a member of subg. Goniophlebium. Often it has been identified as P. dasypleuron Kunze, a species of South America that is less pilose on the lower surface and that lacks abortive segments on the stipe. It is related to P. loriciforme Rosenst. in the latter character, but that species has glabrous laminae. Vittaria costaricensis Lellinger, sp. nov. Fig. 18 Plantae epiphyticae. Rhizoma repens ca. 1.5 cm diam. (paleis inclusis) dense paleaceum, paleis anguste lanceatis usque ad 10 mm longis 0.75 mm latis obscure rufobrunneis laxis tenuiter clathratis, ad apicem elongatum filiformibus, margi- nibus repandis vel distante et grosse dentatis. Stipites ca. 3 cm longi straminei canaliculati. Laminae lineares usque ad 60(75?) cm longae 4.5 mm latae subcor- iaceae dilute virides, ad basin et apicem attenuatae, costis obscuris, sulcis fertilibus 390 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ca. 1 mm ex marginibus distantis, marginibus saepe subinvolutis stramineis; paraphysibus ca. 0.2 mm longis rufobrunneis multifurcatis, furcis subclavatis; sporis bilateralibus. Type.—El Silencio near Tilaran, Pcia. Guanacaste, Costa Rica, ca. 750 m, Standley & Valerio 44805 (US; isotype GH). Discussion.— This species is known only from the type. It differs from V. gra- minifolia Kaulf. in having monolete spores and from V. lineata (L.) J. E. Smith, which has monolete spores, in having wider laminae and larger, less clathrate, and greatly toothed rhizome scales. The rhizome scales are very finely clathrate and have short cells only 2-3 times longer than wide, quite unlike most other New World Vittaria species. Department of Botany, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 391-402 A REVIEW OF CUMMINGSIA FERRIS (MALLOPHAGA: TRIMENOPONIDAE), WITH A DESCRIPTION OF TWO NEW SPECIES Robert M. Timm and Roger D. Price Abstract.— Descriptions and illustrations are given for the four previously rec- ognized species of Cummingsia: C. inopinata Méndez, C. intermedia Werneck, C. maculata Ferris, and C. peramydis Ferris; and for two new species: C. albujai from Caenolestes fuliginosus (Tomes) and C. perezi from Caenolestes convelatus Anthony, both from Ecuador. A key is provided for the identification of these six species. Acanthomenopon Harrison is recognized as a subgenus to contain a single species, C. peramydis, parasitic on the marsupial genus Monodelphis. The sub- genus Cummingsia contains the other five species which are parasitic on the marsupial genera Caenolestes, Lestoros, and Marmosa, and the rodent genus Thomasomys. For nearly half a century, the mallophagan genus Cummingsia Ferris (Trimen- oponidae) was known by only three sepcies distributed on South American mar- supials: C. intermedia Werneck, 1937, from Marmosa dryas Thomas and Mar- mosa incana (Lund); C. maculata Ferris, 1922, from Lestoros inca (Thomas); and C. peramydis Ferris, 1922, from Monodelphis brevicaudata (Erxleben) and Mono- delphis domestica (Wagner). In 1971 Méndez described a fourth species, C. ino- pinata, from a rodent, Thomasomys cinereiventer J. A. Allen. We recently collected two new species of Cummingsia from two other South American marsupials, Caenolestes convelatus Anthony and Caenolestes fuligi- nosus (Tomes). The collections are of interest in that these two species of Cum- mingsia represent the first records of Mallophaga on hosts in the genus Caenolestes. Description of these two new species has prompted us to review the entire genus Cummingsia and to provide a key for the identification of all six species. The genus Cummingsia appears to be restricted to the New World marsupials and the rodent genus Thomasomys. We now recognize two subgenera, the nominate sub- genus containing five species and Acanthomenopon Harrison containing a single species. Genus Cummingsia Ferris Cummingsia Ferris, 25 Apr 1922:83. Type-species.—C. maculata Ferris. Acanthomenopon Harrison, 17 May 1922:156. Type-species.—A. horridum Harrison. Remarks. —Ferris (1922:83) characterized the genus Cummingsia as follows: “lateral margin of the head at the most but slightly notched; with two pairs of strong, flattened, backward-pointing spines on the ventral side of the head, these arising from just before the bases of the antennae; with the clypeal region not 392 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON delimited dorsally by a transverse suture; with the antennal fossae not covered beneath by a flap; with the mesonotum quite distinct; without pulvilli on the first segment of the tarsi; with the genitalia of the males of a simple type.” To this characterization we would add the following: esophageal sclerites re- duced; antenna 4-segmented, with pedunculate third segment and relatively small fourth segment bearing large sensillum and stout blunt seta and small setae (Fig. 2); each leg with 2 pairs of claws; with 5 pairs of abdominal spiracles, these distributed on III—VII; male genitalia of type in Fig. 3, parameres outwardly curved with short subapical seta (Fig. 4), medioposteriorly with broad bifurcate plate, and spinous sac with small sclerite of various shapes. Subgenus Cummingsia Ferris Remarks.—The five species in this subgenus may be characterized by the fol- lowing combination of features: head shaped much as in Fig. 1; temple with each side having 3 very long setae, placed submarginally (Figs. 1, 9, 12, 15, or 17); with shorter postmental setae (Fig. 1); gula with chaetotaxy of 1 very long + 4 much shorter setae on each side and without spinous sculpturing (Figs. 1 or 17); metanotal chaetotaxy as in Fig. 1, with 2 spiniform setae placed mediad to outer very long marginal seta; marginal tergal and sternal setae mostly subequal in length (Fig. 1); and female subgenital plate as in Figs. 8 or 14, with 4 short to minute medioposterior setae. Cummingsia albujai, new species Figs. 1-8 Type-host. —Caenolestes fuliginosus (Tomes) [Marsupialia: Caenolestidae]. Male.—As in Fig. 1. Dorsal head chaetotaxy with seta mediad to large spiniform seta long, over twice length of spiniform; other setae with lengths as shown. Both ventral spinous processes on each side blunt, close together. With well-developed heavily pigmented carina across posterior head margin and lacking medioposterior protrusion such as in Fig. 17 or 21. Thoracic sternal plates as in Fig. 5; prosternal plate with 7 long, 7 short stout setae; mesosternal plate with 5 long, 5 (less often 4 or 6) short stout setae; metasternal plate with 15 (less often 16 or 17) short to long setae. Tergal setae: I, 6 (with outermost very short, innermost twice length of adjacent seta (Fig. 6)); II-VII, 6; VIII, 4 (less often 3); [X, 1 very long, 1 medium on each side. Sternal setae: I, 9-10; II, 20-22; III, 12-14; IV, 12-13; V, 10-12; VI, 7-9; VII, 6 (less often 5 or 7); those on I-II including shorter anterior setae. Subgenital plate (fused VIII-IX) with 10 setae, as in Fig. 7. Dimensions: preocular width (POW), 0.26—0.28; temple width (TW), 0.35—0.38; head length (HL), 0.23- 0.25; prothorax width (PW), 0.32—0.35; metathorax width (MW), 0.37-0.40; ab- domen width at segment V (AW), 0.54—0.59; total length (TL), 1.15-—1.25. Gen- italia as in Fig. 3, with genital plate broadly tapered, with sac sclerite thin, “U”’- shaped, and with tip of paramere with subapical seta well removed from end (Fig. 4); genitalia width (GW), 0.08-0.10; genital plate width (GPW), 0.06-0.07; genital plate length (GPL), 0.05-0.06. Female.—Much as for male, except as follows. Sternal setae: I, 9-11; II, 21- 25; IH, 13-15; IV, 12-14; V, 10-13; VI, 9-12; VII (fused with VIII), 8-12. VOLUME 98, NUMBER 2 393 Subgenital plate (fused VIII-IX) with 12 (less often 11) setae, including 4 minute medioposterior setae associated with break in plate (Fig. 8). Anus circular, with 25-28 minute to medium setae (Fig. 8). Dimensions: HL, 0.23-0.26; PW, 0.35— 0.36; MW, 0.43-0.45; AW, 0.68-0.73; TL, 1.21-1.40. Remarks. — This is the only known species of Cummingsia for which both sexes have no abdominal tergites with more than six setae; this separation is further supported by differences in dimensions, dorsal head chaetotaxy, details of male genitalic structure, and sternal chaetotaxy. Etymology.—This new species is named in honor of Dr. Luis Albuja V. of the Escuela Politécnica Nacional, Quito, in recognition for his studies on Ecuadorian mammals, and for generous assistance with this project. Holotype.— Adult 6, ex Caenolestes fuliginosus, Ecuador, Napo Province, 6.9 km W of Papallacta, 24 Jun 1978, collected by R. S. Voss (UMMZ 127107); in collection of University of Michigan. Paratypes.—Ex C. fuliginosus: 5 6, same as holotype; 2 6, 2 9°, Ecuador, Napo Province, vic. Papallacta, 4 May 1980, R. S. Voss #739 (UMMZ 155578); 3 88, 3 99, same except 6 May 1980, #749 (UMMZ 155693); 1 6, same except 23 Mar 1980, #630; 1 2, same except 7 May 1980, #763 (UMMZ 155580); 1 2, Pichincha Province, Volcan Pichincha, 10 Apr 1939, C. Ollalla (FMNH 53296). Paratypes will be distributed among the University of Michigan, University of Minnesota, Oklahoma State University, U.S. National Museum of Natural History, and Field Museum of Natural History. Cummingsia maculata Ferris Figs. 9-11 Cummingsia maculata Ferris, 1922:83. Type-host. — Caenolestes sp. = Lestoros inca (Thomas) [Marsupialia: Caenoles- tidae]. Male.—Close to C. albujai, differing as follows. Head with seta mediad to large spiniform seta short, of approximately same length; with 2 series of short + long + short setae posterior to spiniforms (Fig. 9). Tergite I with medial 4 setae subequal in length (Fig. 10). More tergal setae: III, 7-8; IV—VI, 8; VII, 7-8; VIII, 6. Sternal setae: II, 21-26; III, 14-16; IV, 12-14. Dimensions: HL, 0.22-0.24; PW, 0.29- 0.30; AW, 0.50-0.56; TL, 1.05—1.16. Genitalia (Fig. 11) with small “butterfly’’- shaped sclerite associated with genital sac. Female.—A\lso close to C. albujai, differing in having chaetotaxy of head and tergite I same as for male. Metasternal plate with 18 setae. Tergal setae: III, 8; IV, 9; V, 10; VI, 9. Sternal setae: I, 12; II, 31; III, 20; IV, 17. Anus with 24 setae. Remarks.—This species differs from the preceding species by consistently hav- ing tergites IIJI-VII with more than six setae, by the lengths of certain setae on the dorsal head and tergite I, by the male with a differently-shaped genital sac sclerite, and by the female with more sternal setae on I-IV. Ferris (1922:83) reported the type-host and type-locality for Cummingsia mac- ulata as ““Caenolestes sp., Cedrobamba Ruins, Peru,’ and mentioned that the specimens are from a skin in the United States National Museum. Emerson and Price (1981) subsequently listed Lestoros inca as the type-host for C. maculata. 394 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON \ WW We Figs. 1-11. 1-8. Cummingsia albujai: 1, Dorsal-ventral male; 2, Male terminal antennal segments; 3, Male genitalia; 4, Tip of male genitalic paramere; 5, Male thoracic sternal plates; 6, Male abdominal tergite I; 7, Ventral male terminalia; 8, Ventral female terminalia. 9-11. C. maculata: 9, Dorsal male head; 10, Male abdominal tergite I; 11, Apical portion of male genitalia. VOLUME 98, NUMBER 2 395 As Caenolestes fuliginosus recently has been reported in northern Peru (Barkley and Whitaker 1984), we thought it advisable to double-check the identity of the host. The mammal collection of the U.S. National Museum does contain a series of 17 specimens of Lestoros inca collected by Heller in May and June of 1915. The locality data for the series are listed on the skin tags as follows: Peru: Machu Picchu, Cedrobamba, Timber Line, 12,000 ft. On the back of the tag on two of these (USNM 194403 and 194427) it reads Cedrobamba Ruins rather than just Cedrobamba. Hence, it is likely that one of these two specimens represents the host animal from which Ferris obtained his series of one female and four males of Cummingsia that he described as C. maculata. Lestoros inca is the only cae- nolestid known from southern Peru and there is little doubt about the fact that it is the type-host for C. maculata. Material examined.— Holotype 2, allotype 3, 3 6¢ paratypes of C. maculata, ex Lestoros inca, Peru, Cedrobamba Ruins. Cummingsia perezi, new species Figs. 12-14 Type-host.— Caenolestes convelatus Anthony [Marsupialia: Caenolestidae]. Male.—Close to C. albujai, but differing as follows. Dorsal head setae generally longer (Fig. 12). Metasternal plate with 16 setae. More tergal setae: IJ-IV, 8; V— VI, 9; VII, 8; VIII, 6. Lengths of setae on tergite I near to those of C. maculata (Fig. 10). More sternal setae: I, 15; II, 27; HII, 22; IV, 20; V, 19; VI, 15; VII, 11. Subgenital plate with 11 setae. Much larger dimensions: POW, 0.31; TW, 0.40; PW, 0.39; MW, 0.44; AW, 0.69; TL, 1.29. Genitalia (Fig. 13) with broader terminal portion of genital plate and with different shape of sac sclerite; GW, 0.10. Female.— Much as for male, except as follows. Metasternal plate with 17 setae. Tergal setae: I, 7; II, 11; IV, 9; VI-VII, 10. Sternal setae: I, 13; II, 31; II, 24; IV, 21; V, 18; VI, 16; VII, 13. Subgenital plate not medially divided, with longer medioposterior 4 setae and with 2 long lateroposterior setae on each side (Fig. 14). Anus with 30 setae, medial ones in both ventral and dorsal rows longer than for other species (Fig. 14). Also with large dimensions: POW, 0.32; TW, 0.42; PW, 0.42; MW, 0.50; AW, 0.84; TL, 1.50. Remarks.—The combination of very large dimensions, large number of tergal and sternal setae, and details of dorsal head chaetotaxy for both sexes, the structure and chaetotaxy of the female ventral terminalia (it is the only Cummingsia with two long lateroposterior setae on each side of subgenital plate instead of only one), and the details of the male genitalia separate this species from all others of the genus. It is interesting that the two adults representing the type-series of this new species occurred on the same individual host also yielding three adults of another Cummingsia species; of five fairly advanced nymphs found on the same host, two were consistent with this new species in size and chaetotaxy, three with the other species. Etymology.—This new species is named in honor of Sr. Carlos Perez Q., owner 396 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON of Hacienda La Vega, for graciously allowing us to work on his property where the new species of Cummingsia was obtained. Holotype.— Adult 2, ex Caenolestes convelatus, Ecuador, Imbabura Province, Hacienda La Vega, 5 km ESE of San Pablo del Lago, elevation ca. 2600 m, 11 Sep 1983, collected by R. M. Timm (FMNH 124620); in collection of Field Museum of Natural History. Paratype.—1 6, same as for holotype. Cummingsia inopinata Méndez Figs. 15-16 Cummingsia inopinata Méndez, 1971:23. Type-host.— Thomasomys cinereiventer J. A. Allen [Rodentia: Cricetidae]. Male.—Close to C. albujai (Fig. 1), but differing as follows. Unique dorsal head chaetotaxy (Fig. 15), without any mediolateral heavy spiniform setae. Prosternal plate with 7 long, 5 (less often 6) short setae. Tergite I with lengths of setae intermediate between those in Figs. 6 and 10. Tergal setae: II, 4 (less often 5); IV, 8-10; V—VII, 10; VIII, 8; larger number of setae due to 1—2 shorter setae at end of each row (Fig. 16). Sternal setae: III-V, 14-16; VI, 10-13; VII, 6-8; larger number of setae as for tergites (Fig. 16). Dimensions (generally wider head and longer, slender body): POW, 0.30-0.31; TW, 0.39-0.40; PW, 0.34-0.36; MW, 0.39-0.42; AW, 0.46-0.51; TL, 1.27-1.38. Genitalia close to Fig. 3, but sac sclerite smaller and variably ““U’’- to “V”-shaped. Female.—Essentially as for C. albujai, except for differences noted for male above and as follows. Sternal setae: III-IV, 14-17; V, 12-15. Dimensions: POW, 0.31-0.33; TW, 0.40-0.44; HL, 0.24—0.26; PW, 0.35-0.42; MW, 0.41-0.46; AW, 0.54-0.57; TL, 1.39-1.59. Remarks.—This species can be separated from all others by the unique dorsal head chaetotaxy, the small number of short prosternal setae, and the large number of tergal and sternal setae due to additional shorter lateral setae in each row. Cummingsia inopinata, the louse parasitizing Thomasomys, is a ““marsupial” louse phylogenetically. The original description by Méndez (1971) of the new Cummingsia from a rodent host, 7. cinereiventer, left some suspicion as to the possibility of contamination having occurred, and that the true host of C. ino- pinata might actually be a marsupial. Subsequently, we have obtained additional specimens of Cummingsia from two specimens of Thomasomys erro Anthony in Ecuador and two of T. laniger (Thomas) in Colombia. Thus, there is little doubt about the fact that rodents of the genus Thomasomys are parasitized by Cum- mingsia. As most Mallophaga found on small mammals are extremely host spe- cific, we made a detailed examination of our specimens from Thomasomys, and were able to detect no significant variation in external morphology or in genitalia among the populations of Cummingsia found on the three distinct species of Thomasomys over a considerable geographic area. Our conclusion is that, at this time, there is no justification for recognizing more than the single species, C. inopinata, parasitizing Thomasomys. How and when could a “‘marsupial’’ louse have transferred to a rodent host? Méndez (1971) had a series of five males and 18 females from a single 7. cine- reiventer at the type-locality, and a single female from a second T. cinereiventer VOLUME 98, NUMBER 2 397 from another locality in Colombia. He suggested that ““7. cinereiventer may have acquired the original stock from which the new species of Cummingsia ... was derived, many years ago from opossum nests” (Méndez 1971:27). Thomasomys is part of the “complex penis” group of South American cricetines (subfamily: Sigmodontinae) that originated from North American stock, migrating to South America prior to the connection of the two continents by the Panamanian land bridge in the early Pleistocene. The ancestoral cricetine may have arrived in South America by the early Pliocene. Thomasomys has undergone a tremendous radia- tion in South America, with some 25 extant species now recognized. With three different species of Thomasomys, each being parasitized by the same species of Cummingsia, we suggest that an ancestral population of Thomasomys acquired Cummingsia from a marsupial host prior to the later major radiation in Tho- masomys. This may have occurred in the Pliocene. Cummingsia has undergone little differentiation, even though considerable time has elapsed and speciation within Thomasomys has occurred. It will be most interesting to see if additional species of Thomasomys are found to harbor Cummingsia, and, if so, whether any differentiation has occurred. Material examined.—1 6, 1 2 paratypes of C. inopinata, ex Thomasomys ci- nereiventer, Colombia, Depto. de Narino, Laguna de la Cocha, elevation 2700 m. 3 66, 3 22, ex Thomasomys laniger, Colombia, Antioquia, Paramo. 2 44, 3 °°, ex Thomasomys erro, Ecuador, Napo Province, Papallacta. Cummingsia intermedia Werneck Figs. 17-20 Cummingsia intermedia Werneck, 1937:70. Type-host.—Marmosa incana paulensis Tate [Marsupialia: Didelphidae]. Male. — Differing from C. albujai as follows. Dorsal head chaetotaxy with only minute setae or alveoli medially; mediolateral setae close to C. maculata (Fig. 17). Slender pointed inner head spinous process well separated from outer (Fig. 17). With reduced carina across posterior head margin and with medioposterior protrusion (Fig. 17). Prosternal plate with 11 short stout setae; mesosternal plate with 9 such setae; metasternal plate with total of 22 setae. Tergal setae: II, 7; III- IV, 8; V—-VI, 9; VII, 8. Sternal setae: I, 18; II, 34; III, 26; IV, 25; V, 21; VI, 20; VII, 11; included in these 12, 9, 7, 6, and 1 short anterior setae on IIIJ-—VII, respectively. Dimensions: POW, 0.29; HL, 0.26; MW, 0.43; AW, 0.61; TL, 1.35. Genital plate shaped as in Fig. 19, much like that of foregoing species; genital sac sclerite semicircular (Fig. 18); tip of paramere (Fig. 20) with seta near end in close proximity to dark button-like projection; GW, 0.11. Female.—Much as for male, except as follows. Prosternal plate with 10 short stout setae; metasternal plate with 21 setae. Tergal setae: II, 6; IV, 9; V, 8; VII, o-Vililiieese Sternalesetacs ly ls ly 36211 232IMin 222 Ve 203 Vill 6. Vile 10: included in these 8, 8, 6, 4, and 2 short anterior setae on III—VII, respectively. Anus with 30 setae; ventral terminalia essentially as in Fig. 8. Dimensions: POW, 0.31; TW, 0.39; HL, 0.26; PW, 0.36; MW, 0.44; AW, 0.69; TL, 1.50. Remarks.—This species differs from all of the preceding species by its medial dorsal head chaetotaxy, the shape and proximity of the ventral spinous head processes, the large number of short stout spiniform setae on the prosternal and 398 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON mesosternal plates, the shape of the male genital sac sclerite and the details of the tip of the paramere, and the large number of abdominal setae. In the absence of specimens from the type-host, M. incana, we tentatively assign our material from M. dryas Thomas to C. intermedia in accordance with Emerson and Price (1975). There are slight discrepancies between the illustrations provided by Werneck (1937) and the material from M. dryas that we have before us. As Werneck’s illustrations and descriptions are of insufficient quality to evaluate these discrepancies, we chose not to treat them at this time, hoping additional material will be available in the future. Illustrations of the adult male and female and details of the genitalia of C. intermedia from M. dryas may be found in Emerson and Price (1975:figs. 19- 2D), Material examined.—1 6, 1 2, ex Marmosa dryas, Venezuela, Merida, Tabay. Subgenus Acanthomenopon Harrison Acanthomenopon Harrison, 1922:156. Type-species.—A. horridum Harrison. Remarks. —In his description of the genus Cummingsia, Ferris (1922:83) stated ‘It is with some hesitation that I refer the two species above named to the same genus, for in the structure of the head there is a rather notable difference. I would especially call attention to the apparent absence of the maxillary lobes in C. peramydis. However, they are not so different that their association in the same genus is entirely incongruous and in some respects they are very similar.” We concur with the belief of Ferris that C. peramydis presents an array of differences from the other species of Cummingsia, but, rather than place it in a genus of its own, we prefer to place it in a subgenus of Cummingsia. The following features characterize the subgenus Acanthomenopon: head shaped much as in Fig. 21; temple with each side having 3 very long, 1 long, and 1 minute setae placed at margin (Fig. 21); postmentum with 1 long seta (Fig. 21); gula with chaetotaxy of 2 very long + 3 much shorter setae on each side and with spinous sculpturing on posterior portion (Fig. 21); metanotal chaetotaxy as in Fig. 22, with 3 very long marginal setae on each side and shorter setae placed as shown; marginal tergal and sternal setae with short among long (Fig. 23); and female subgenital plate as in Fig. 23, with 4 prominent medioposterior setae. Cummingsia peramydis Ferris Figs. 21-24 Cummingsia peramydis Ferris, 1922:85. Type-host.—Peramys domesticus = Monodelphis domestica (Wagner) [Marsu- pialia: Didelphidae]. Acanthomenopon horridum Harrison, 1922:156. Type-host.—Peramys domesticus = Monodelphis domestica (Wagner). Male.—Grossly as in Fig. 1. Dorsal head chaetotaxy as in Fig. 21, with at most only minute seta medioanteriorly. Ventral spinous head processes (Fig. 21) much as for C. intermedia. With reduced carina across posterior head margin and with VOLUME 98, NUMBER 2 399 22 Figs. 12-24. 12-14. Cummingsia perezi: 12, Dorsal male head; 13, Apical portion of male genitalia; 14, Ventral female terminalia. 15-16. C. inopinata: 15, Dorsal male head; 16, Dorsal-ventral male abdominal segments V—VI. 17-20. C. intermedia: 17, Dorsal-ventral male head; 18, Male genital sac sclerite; 19, Male genital plate; 20, Tip of male genitalic paramere. 21-24. C. peramydis: 21, Dorsal- ventral male head; 22, Dorsal male metathorax margin and abdominal tergite I; 23, Dorsal-ventral female terminalia; 24, Apical portion of male genitalia. medioposterior protrusion (Fig. 21). Prosternal plate with 5—6 long, 11-13 short stout setae; mesosternal plate with 4 long, 12—15 short stout setae; metasternal plate with 18-21 setae. Tergal setae: I, 5—6; II, 10-12; III, 13-16; IV, 15-18; V, 16-19; VI, 15-19; VII, 14-18; VIII, 12-17; IX, 8. Sternal setae: I, 11-12; II, 24—- 29; III, 21-22; IV-V, 22-25; VI, 21-22; VII, 15-19. Subgenital plate with 13-14 setae. Dimensions: POW, 0.29-0.32; TW, 0.35-0.40; HL, 0.23-0.26; PW, 0.31- 0.38; MW, 0.34-0.43; AW, 0.51-0.56; TL, 1.18-1.39. Genitalia as in Fig. 24, 400 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON with semicircular sac sclerite much as for C. intermedia, greatly elongated genital plate, and pronouncedly outwardly curved paramere tips; GW, 0.09-0.10; GPW, 0.06-0.07; GPL, 0.10. Female.—Much as for male, differing as follows. Prosternal plate with 12-15 short stout setae. Tergal setae: I, 6—7; II, 12-13; IV—VII, 17-20; IX, 4. Sternal setae: II, 27-32; III, 22—23; IV—VI, 22-26; VII, 17—21. Subgenital plate with 20-— 22 setae (Fig. 23). Anus with 30-33 setae. Dimensions: POW, 0.31-0.33; TW, 0.38-0.43; PW, 0.35-0.40; AW, 0.64—-0.66 TL, 1.46—-1.60. Remarks.—As the sole member of the subgenus Acanthomenopon, C. peramydis is easily separated from the other five species of Cummingsia on the basis of the subgeneric characters. Ferris (1922:85) listed the type-host and type-locality of C. peramydis as “‘Per- amys domesticus, Quixada, Brazil, from skins in the Field Columbian Museum.” The mammal collections of the Field Museum of Natural History contain a series of nine Monodelphis domestica from: Brazil, Ceara Province, Quixada, collected between 18 and 29 June 1913 by R. H. Becker (FMNH 20245-20251, 20254— 20255). We were able to obtain five additional specimens of C. peramydis from these skins. Illustrations of the adult male and female of C. peramydis and details of the genitalia were provided by Emerson and Price (1975:figs. 15-18). Material examined.—5 66, ex Monodelphis domestica, Brazil, Ceara, Quixada. 3 36, 2 22, ex M. brevicaudata (Erxleben), Venezuela, Trujillo, Valera (Isnotu), and Falcon, 5 km N, 13 km E of Mirimire near La Pastora. 1 6, 2 99, ex M. brevicaudata touan (Shaw) Brazil, Amapa, Serro do Navio. Discussion The genus Caenolestes as now defined contains five parapatric species: C. can- iventer Anthony known from southwestern Ecuador, C. convelatus Anthony known from north-central Ecuador, C. fuliginosus known from north-central Ecuador and extreme northern Peru, C. obscurus Thomas, known from northern and western Colombia and extreme western Venezuela, and C. tatei Anthony known from south-central Ecuador. All five species are distributed throughout the Andean mountains from northern and western Colombia and extreme western Venezuela south through Ecuador. The populations of Caenolestes now recognized as species probably represent only two or three valid species (Kirsch 1977). Three of the currently named taxa are known only from or very near the type-localities. Les- toros and Rhyncholestes are both monotypic. Simpson (1970) also suggested that the three genera of the family Caenolestidae, Caenolestes, Lestoros, and Rhyn- cholestes, would perhaps best be regarded as three species of a single genus. Marshall (1980), in a review of both the fossil and Recent forms of the family, considered the genera Caenolestes and Lestoros as marginally separable. This close relationship between Caenolestes and Lestoros is strongly supported by Hayman et al. (1971) who reported that the karyotypes of Caenolestes obscurus and Lestoros inca are nearly identical. The genus Marmosa contains some 44 species distributed throughout much of Central and South America. Mallophaga have been collected on only two of those species, M. dryas which is found in VOLUME 98, NUMBER 2 401 western Venezuela and M. incana which is found in eastern Brazil. The genus ' Monodelphis contains some 17 species distributed throughout much of South America. Again, Mallophaga have been collected on only two species, M. brevi- caudata which is found in Venezuela, the Guianas, Brazil, Paraguay, and northern Argentina, and M. domestica which is found in eastern and central Brazil, Bolivia, and Paraguay. In a cladistic analysis of the carnivorous marsupials, Kirsch and Archer (1982:616) suggested that Marmosa and Monodelphis (along with Les- todelphys) formed a monophyletic unit and that the specialized caenolestids were derived from “forms like Monodelphis brevicaudata.”’ Herein, we recognize six species of Cummingsia from seven species of South American caenolestid and didelphoid marsupials and from three species of the rodent genus Thomasomys. Except for the caenolestids, all of the mammalian genera involved as hosts for Cummingsia are widespread in South America and all are in need of revision. However, combined they provisionally contain over 80 species and we have lice from only nine of these. Thus, we suspect that numerous new species of Cummingsia have yet to be collected, and much remains to be learned about the systematics and host relationships in this genus. Key to the Species of Cummingisa 1. Marginal setae of abdominal tergites and sternites III—-VII with obviously shorter among longer setae (Fig. 23); gula with 2 very long setae on each Siden(siga21) ee ace Subgenus Acanthomenopon ....... peramydis Ferris — Marginal setae of abdominal tergites and sternites III-VII essentially of subequal lengths, except for occasional 1—2 setae at end of row (Fig. 1 or 16); gula with only 1 very long seta on each side (Fig. 1 or 17) ........ Pe PRA Sob BIC Ab mated SUbDSEMUSECUITITIINES Gm eLearn) a> 2. Abdominal tergites II-VII each with only 6 setae (Fig. 1) ..... albujai n. sp. — At least several of abdominal tergites II-VII with more than 6 setae ... 3 3. Median dorsal head with only minute setae or alveoli (Fig. 17); prosternal plate with at least 10 short stout setae; male genitalia with paramere tip as in Fig. 20 and sac sclerite as in Fig. 18 ............ intermedia Werneck — Median dorsal head with short to medium setae (Figs. 9, 12, or 15); prosternal plate with only up to 7 short stout setae; male genitalia with paramere tip near to Fig. 4 and sac sclerite nearer to Fig. 11 or 13..... 4 4. Laterodorsal head without stout spiniform setae (Fig. 15); prosternal plate Wit he@Onlye4—ORshOnteStOUt SLAC a. kc.wisgslensysmiolen caewla chores: inopinata Méndez — Laterodorsal head with stout spiniform setae (Fig. 9 or 12); prosternal platcawithwreshontestoutsetae, Byles cr ON Oe RM | es hs, ste eues 5 5. Head with long seta immediately mediad to inner dorsal spiniform seta (Fig. 12); posterior portion of female subgenital plate as in Fig. 14, with pair of very long lateroposterior setae on each side and lacking medio- posterior division; male genital sac sclerite as in Fig. 13 ....... perezi n. sp. — Head with short seta immediately mediad to inner dorsal spiniform seta (Fig. 9); posterior portion of female subgenital plate as in Fig. 8, with only single very long lateroposterior seta on each side and with medioposterior division; male genital sac sclerite as in Fig. 11 ............ maculata Ferris 402 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Acknowledgments We are grateful to Barbara L. Clauson, to Angelitos and Richard Garrett for assistance in the field, and to Luis Albuja, Ramiro Barriga, and Gustavo Orcés of the Escuela Politécnica Nacional, Quito, and Donald J. Stewart and Myriam Ibarra for logistic support while in Ecuador. Sr. Carlos Perez Q., owner of Hacienda La Vega, graciously allowed us to work on his property where the two new species of Cummingsia were obtained. We thank Barry M. OConnor, University of Mich- igan, K. C. Emerson, Sanibel, Florida, William A. Drew, Oklahoma State Uni- versity, and Bernard C. Nelson and John A. Chemsak, University of California, Berkeley, for making critical specimens of Cummingsia available to us. This project was supported in part by a grant from the National Science Foundation, INT-8303194. Support from the Escuela Politécnica Nacional and Field Museum of Natural History is gratefully acknowledged. This paper is published as Paper No. 14,078 of the Scientific Journal Series of the Minnesota Agricultural Exper- imental Station on research conducted under Minnesota Agricultural Experiment Station Projects No. Min-17-015 and Min-17-016. Literature Cited Barkley, L. J., and J. O. Whitaker, Jr. 1984. Confirmation of Caenolestes in Peru with information on diet.—Journal of Mammalogy 65:328-330. Emerson, K. C., and R. D. Price. 1975. Mallophaga of Venezuelan mammals.—Brigham Young University Science Bulletin, Biological Series 20:1-77. , and 1981. A host-parasite list of the Mallophaga on mammals.— Miscellaneous Publications of the Entomological Society of America 12:1-72. Ferris, G. F. 1922. The mallophagan family Trimenoponidae.— Parasitology 14:75-86. Harrison, L. 1922. On the mallophagan family Trimenoponidae; with a description of a new genus and species from an American marsupial.—The Australian Zoologist 2:154—-158. Hayman, D. L., J. A. W. Kirsch, P. G. Martin, and P. F. Waller. 1971. Chromosomal and serological studies on the Caenolestidae and their implications for marsupial evolution. — Nature 231:194— 195. Kirsch, J. A.W. 1977. The comparative serology of Marsupialia, and a classification of marsupials. — Australian Zoology, Supplement, series 52:1—152. , and M. Archer. 1982. Polythetic cladistics, or, when parsimony’s not enough: The relation- ships of carnivorous marsupials.—Jn M. Archer, ed., Carnivorous marsupials. Volume 2.— Royal Zoological Society of New South Wales, pp. 595-619. Marshall, L. G. 1980. Systematics of the South American marsupial family Caenolestidae. — Field- iana: Geology, New Series 5:vii + 1-145. Méndez, E. 1971. A new species of the genus Cummingsia Ferris from the Republic of Colombia (Mallophaga: Trimenoponidae).— Proceedings of the Entomological Society of Washington 73: 23-27. Simpson, G. G. 1970. The Argyrolagidae, extinct South American marsupials.—Bulletin of the Museum of Comparative Zoology 139:1—86. Werneck, F. L. 1937. Nova especie do genero Cummingsia (Mallophaga: Trimenoponidae).— Me- morias do Instituto Oswaldo Cruz 32:69-73. (RMT) Division of Mammals, Field Museum of Natural History, Chicago, Illinois 60605; (RDP) Department of Entomology, University of Minnesota, St. Paul, Minnesota 55108. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 403-410 CRYSTALLODYTES PAUCIRADIATUS (PERCIFORMES), A NEW CREEDIID FISH SPECIES FROM EASTER ISLAND Joseph S. Nelson and John E. Randall Abstract.—A new species of creediid fish, Crystallodytes pauciradiatus, is de- scribed from Easter Island. It is distinguished from C. cookei, the other member of the family with scales only along the lateral line, in having only 30—32 dorsal- fin rays and 48-52 lateral-line scales. The new species represents an eastern ex- tension in the range of the family. Crystallodytes cookei, previously reported from Hawaii and Enderbury and Tau islands, is reported here also from the Pitcairn Group (Pitcairn, Henderson, and Ducie islands), Swains Island in northern Amer- ican Samoa, and perhaps Tahiti. The creediid Chalixodytes tauensis is recorded from Pitcairn and Mangareva. A key to the species of the family is presented. Some 14 species are recognized in the Indo-West Pacific family Creediidae (Nelson 1983). One of the seven genera, Crystallodytes, is currently considered to be monotypic, although the nominal species is recognized with two subspecies, C. cookei cookei Fowler from Hawaii, and C. cookei enderburyensis Schultz from Enderbury Island (Phoenix Islands) and Tau Island (American Samoa) in the South Pacific. Crystallodytes cookei is distinguished from all other creediids in having scales only along the lateral line. Relatively little is known of the biology of Crystallodytes; Leis (1982) described its spawning time, eggs, larvae, and oc- currence in the plankton. We describe here a second species of Crystallodytes, from Easter Island, distinguished from the other species primarily in having fewer meristic parts. Measurements were made to the nearest 0.1 mm with needle-point dial calipers and are expressed as thousandths (0/00) of standard length (SL). The pectoral ray count excludes the short dorsal splint. Vertebral counts were made from radio- graphs. Material of the new species is deposited in the following museums: AMNH, American Museum of Natural History, New York; AMS, Australian Museum, Sydney; BPBM, Bernice P. Bishop Museum, Honolulu; CAS, California Academy of Sciences, San Francisco; USNM, National Museum of Natural History, Smith- sonian Institution, Washington, D.C. Material utilized of C. cookeiis from BPBM. Collections of Crystallodytes pauciradiatus and C. cookei enderburyensis consist primarily, if not exclusively, of females. Crystallodytes pauciradiatus, new species Figs. 1-2 Holotype.—BPBM 6734, female, 31.3 mm SL, Easter Island, off Ahu Akapu, depth 10.7 m, sand, 7 Feb 1969. Paratypes (all females unless otherwise noted). —AMNH 49554, 2 specimens, 34.1 and 30.4 mm SL, taken with BPBM 6733. AMS I.24606-001, 2 specimens, 404 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON )) EES Ss SS 5 mm Fig. 1. Crystallodytes pauciradiatus, holotype BPBM 6734, 31.3 mm SL. A, Lateral view of head region; B, Ventral view of pelvic fin region; C, Lateral view. 30.4 and 30.2 mm SL, taken with BPBM 6733. BPBM 6733, 3 specimens, 32.5, 32.3, and 31.7 mm SL (disarticulated while clearing and staining following taking of measurements and counts), Easter Island, west coast off south end of Hanga Roa, depth 12.2 m, sand near rocks, 10 Feb 1969. BPBM 29656, 28.7 mm SL, taken with the holotype. CAS 54946, 2 specimens, 27.5 and 27.3 mm SL, taken with BPBM 6733. USNM 266462, 5 specimens (3 of uncertain sex), Easter Island, Mataveri 0 Tai, depth 6.1 m, black and white sand, 2 Feb 1969. Non-type material. — The following material was too small to obtain comparable measurements and counts. BPBM 6735, 2 specimens, taken with USNM 266462. BPBM 29657, 6 specimens, taken with BPBM 6733. BPBM 29658, 2 specimens, disarticulated while clearing and staining, taken with BPBM 6733. All specimens were collected with rotenone by one of us (JER), accompanied by G. R. Allen or G. R. Allen and B. A. Baker. Diagnosis.—A creediid with scales only along the lateral line and having 30- 32 dorsal-fin rays, 48-52 lateral-line scales, and 50-53 vertebrae. The only other species with scales confined to the lateral line is the congeneric Crystallodytes cookei, which has a larger number of meristic parts (see key). The only other creediids with a comparable number of dorsal-fin rays are Limnichthys polyactis and L. rendahli of New Zealand, and the only species with a comparable number of lateral-line scales and vertebrae is Tewara cranwellae, also of New Zealand. Description.— Morphometric and meristic data are given in Tables 1—4. Snout relatively elongate, sloping gradually in dorsal profile. Upper jaw with fleshy extension, projecting beyond lower jaw; maxilla extending posteriorly to behind eye; posterior tip of maxilla rounded. Bony dorsal projection on symphysis of lower jaw. Lower jaw bordered by 1 row of short cirri, about 10-18 per side in specimens 27-34 mm in length. Eyes dorsolateral; total interorbital width about 16-22% SL, bony width much less. Ventral opercular flap overlapping branchios- tegal membrane and extending far forward; posterior portion of gill cover over- lapping base of pectoral fin. Branchiostegal rays 7. Subopercle heavily splintered, interopercle moderately splintered. Scales present only along lateral line. Lateral line arising at upper edge of gill cover, running above pectoral fin, descending gradually and reaching ventral profile only posteriorly; scales behind pectoral fin VOLUME 98, NUMBER 2 405 a at Ee eS oe eee a Canis — : Fig. 2. Crystallodytes pauciradiatus, showing pigmentation pattern in fresh specimens. A, Holotype, BPBM 6734; B, Paratype (30 mm SL when fresh), USNM 266462. elongate, with rounded posterior margin, not trilobate. Base of pelvic fins slightly anterior to base of pectoral fins; inter-pelvic distance small, less than length of base of pelvic fin; pelvic fins each with 1 spine and 5 soft rays. Dorsal, anal, pectoral, and pelvic rays unbranched. Branched caudal rays 8, countable in only one specimen. Teeth along upper and lower jaws ending well before anterior tip; small isolated patch of vomerine teeth on each side. Iliac spurs (Nelson 1979) elongate and curved and with short posteriorly-directed processes at midline. Last haemal and neural spines relatively narrow. The color pattern is not apparent in the preserved material. However, in a color transparency and a black and white photograph of the holotype taken at capture by one of us (JER) the body is whitish with 12 short dorsal saddles (described in field notes as consisting of yellow spots edged in dark red-orange) with a faint trace of a lateral band. In a black and white photograph of a 30 mm specimen of USNM 266462, originally part of BPBM 6735, there are 13 dorsal saddles (each with a central light area) with some extending ventrally to join the conspicuous lateral band. Color notes of fresh material of BPBM 6733 describe the yellowish spots in the dorsal saddles as being edged with dull orange to dark brown. Etymology.—The specific name pauciradiatus (Latin) is for the low number of dorsal, anal, and pectoral rays relative to Crystallodytes cookei. Discussion.—The new Easter Island form is more similar to Crystallodytes PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 406 CCT + OBE €v-9€ 61 + LYE 9€-CE 8c + 00P 8b—-9E Gy WQIO Jo YIsUST T€ +6LP €S-tv 6€ +S 0P Sb-9E 9C + 7 8h CS-EP Os yisus] nous 8° + ECL c8-19 97 + £89 1L-99 Tr + 069 LL-19 €L YipIM peoH LL + COC CCC—-C6I Oe + C6l L61-061 6L + LIC €£C-SOC LIZ yisus] peoH Iv + COl ITI-€6 CTE + 06 66-C6 CL + 9°98 66-SL cL yisug] uy [e10}09q CST + 6LE Elp—-vSe €1 + OLE TLE-89€ COIL + 9IP ver—-loOv CCV yisua] jeuesig Lyt + LS€ 6LE-STCE yCI + 89€ LLE-OSE OCI + EbP SSb-LCP Oty yisus] [esiopelg 8° + CLL 88-S$9 65 +S EL 8L-S9 Ly + 8CL p8-S9 €L yidop Apog 6€ + 16E CEP-C VE 1S ar tS ths SIp-E ve cc + TOE Tpe-€ 9¢ Calls (urur) yIUS] PrepueIg as # x osuel as #x osuerl as = x osuel adAjo0[oH, 19}081eYyD LT =u p=u IIV sl=u LL sodAjeieg sisuadinqgiapua *D “D 194OOI 1aY00 ~D snjpippsionvd Dy “Y3us] PrepuLis Jo syypuRsnoY} Ul Saj{poyjyjsdiD Jo satsadsqns pue so1deds 9y} JO s}uUsWoINseoW [euOTLIOdOIg —"] 31qe 1 VOLUME 98, NUMBER 2 407 Table 2.—Frequency distribution of number of pectoral-fin rays in the species and subspecies of . Crystallodytes. Species 9 10 11 12 13 x n C. pauciradiatus 7 7 2 9.7 16 C. cookei cookei 2 yD, 10.5 4 C. c. enderburyensis 11 6 12.4 17 cookei cookei from the Hawaiian Islands than it is to its geographically nearest neighbor C. c. enderburyensis in the meristic characters examined. The difference between C. c. cookei and the new form is greater than that between the two recognized subspecies of C. cookei in three of the five variable meristic characters. The occurrence on Easter Island of a new creediid represents an eastern exten- sion for the family (to approximately 109'4°W longitude) and the first recorded occurrence of a member of the family on the Nasca Plate. The nearest record of other creediids to Easter Island is that of C. c. enderburyensis, here recorded and well represented from Ducie, Henderson, and Pitcairn islands (based on BPBM collections, as are the following except as noted). Poorly fixed material from Tahiti may belong to this subspecies; it occurs on Swains Island, American Samoa (Scripps Institution of Oceanography 67-215). Chalixodytes tauensis occurs on Pitcairn and also on Mangareva of the Tuamotu Archipelago (one specimen of three from BPBM 13543 is virtually fully scaled). Oeno Island, of the Pitcairn Group, has a species of Limnichthys, perhaps L. donaldsoni (BPBM 16536). Key to the Species of Creediidae 1. Pelvic-fin rays I,4 or I,3 or fins absent; lateral-line scales behind pectoral fin not trilobate or only weakly so (but with posterior lobe) .......... my, — Pelvic-fin rays I,5; lateral-line scales behind pectoral fin distinctly tri- lobate except in Crystallodytes spp. where the posterior margin 1s rounded and without indentations ..... Ae ie A POSED at BENE leh WE 2 7 Deebelvichinsvabsent anal-tin trays, 32—36 5. sae ese ee eee ac ae sh gd hh ER eR AS 8 8 ee aie OF Apodocreedia vanderhorsti de Beaufort — Pelvic fins present; anal-fin rays fewer than 29 or more than 35 ...... 3 Table 3.— Frequency distribution of numbers of dorsal and anal-fin rays in the species and subspecies of Crystallodytes. Species 30 31 32 33 34 35 36 37 38 39 40 41 42 43 x n Dorsal-fin rays C. pauciradiatus 1 6 9 31.5 16 C. cookei cookei Papell | pos 2 37.7 4 C. c. enderburyensis 6d 1a 2 eidlele 17 Anal-fin rays C. pauciradiatus | ee © Gee 35.2 16 C. cookei cookei Lise pil geil 1 37.5 4 C. c. enderburyensis 4 10 2 1 39.0 17 408 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 4.—Frequency distribution of numbers of lateral-line scales and vertebrae in the species and subspecies of Crystallodytes (the non-type material of C. pauciradiatus is included in the vertebral counts). Species 48 49 50 51 a 38 Sb S55 50 57 58 59 = 60 x n Lateral-line scales C. pauciradiatus 2 1 4 YT 2 50.4 16 C. cookei cookei 2D. 55.5 4 C. c. enderburyensis 3 8 4 2 58.3 17 Vertebrae C. pauciradiatus Py Same ite meas 51.3 26 C. cookei cookei Mest Sree) Spo il C. c. enderburyensis 2, W304 2 983s 3. Dorsal-fin rays 12-16; anal-fin rays 24-28; lateral-line scales 40-47; lateral line descending abruptly behind pectoral fin to near ventral profile; bodystullyascaled:Gr notsnt:'t oa oe elds Roel cna bere 4 — Dorsal-fin rays 35-40; anal-fin rays 36—40; lateral-line scales 55-59; lateral line descending gradually; body scaleless except for lateral line, predorsal row, and caudal peduncle region [the two involved nominal species are poorly differentiated from one another and are only provi- sionally recognized here; one specimen of Chalixodytes tauensis from thesMiangareva Islandsns tullysscaledi]) P24. s7142 1-6. Sale eee 6 4. Pelvic-fin rays I,3; dorsal-fin rays 12 or 13; lateral-line scales and ver- tebrac: usually: 40=40,8 nee Se eee? Sits at re Eb Creedia alleni Nelson — Pelvic-fin rays I,4; dorsal-fin rays 13-16; lateral-line scales and vertebrae USUAL 24 SAP ran NG OC Sees hte cube bas Manu eee 5 5. Body fully scaled; maxilla extending to about center of eye, tip forked; profile of snout sloping gradually, not strongly convex .............. Nes ms abel as eh Stat a RN te alt Creedia haswelli (Ramsay) — Body scaleless in anterior half except for lateral line and paired predorsal row; maxilla extending well behind eye and tip blunt, not forked; profile OSNOULCONVEX 04) eee ee ee Creedia partimsquamigera Nelson 6. Dorsal-fin rays about 35-37; anal-fin rays about 36-38 .............. Ne area ene Ge REN eek (os dc Rh Ae Se Bs tl Chalixodytes tauensis Schultz — Dorsal-fin rays about 37—40; anal-fin rays about 37-40 ............. BERR Breet ARO ea Mas Url aes Bly Oe Chalixodytes chameleontoculis Smith 7. Anal, pectoral, and most of pelvic-fin rays branched; usually 16 or 17 pectoral-fin rays, the lowermost thickened; 9 branched caudal rays; usu- allyls—2 Oldorsal-tinsray sy. nee ee Schizochirus insolens Waite — Anal, pectoral, and pelvic-fin rays unbranched; usually 10-14 pectoral- fin rays, the lowermost not thickened; 8 branched caudal rays; usually 20=42 dorsal "rays: ocd, 2: s-ceb oe ee ee ee Oe 8 8. Lateral-line scales 48-60; anal-fin rays 35—41; usually 11-18 dorsal sad- GES ig a see ay ole ae 0 As Oo ee 9 — Lateral-line scales 36-47; anal-fin rays 25-34; 5-12 dorsal saddles ... 11 9. Body fully scaled; body depth more than 9% of standard length ...... pe eed SGN aieae ait cath UAE ger Oe eae eens Tewara cranwellae Griffin VOLUME 98, NUMBER 2 409 — Body scaleless except for lateral line; body depth less than 9% of standard KSTAYEAE AY 2 ono: 5 oh Gio "eo Gotoh BAS ako Bla DRNICRS Geek Eiote (eGR aL nC Ree Pape ant ae 10 10. Dorsal-fin rays 30-32; lateral-line scales 48-52 ..................... 5 bo 68k ee SRE RR reer pear Crystallodytes pauciradiatus n.sp. — Dorsal-fin rays 36—43; lateral-line scales 55-60 ..................... Ree Set, Mee, even Y thy aiivorle xaitiee tte, alte Crystallodytes cookei Fowler 11. Dorsal-fin rays 28 or more; anal-fin rays 30 or more ................ 12 — Dorsal-fin rays 27 or fewer; anal-fin rays 30 (rarely) or fewer ........ 13 12. Origin of anal fin in front of vertical through origin of dorsal fin; snout length usually less than half maximum body depth; cirri along either side of lower jaw usually fewer than 20, of near uniform length .......... 5618 FRR a rce sO 6 EO Be RNs elem aee GRY Deze seer Limnichthys polyactis Nelson — Origin of anal fin beneath or behind vertical through origin of dorsal fin; snout length usually greater than half maximum body depth; cirri along either side of lower jaw usually more than 20, of alternating size pos- (STRONG es Gis bos Coe eC cee ee ee eee Limnichthys rendahli Parrott 13. Combined number of dorsal and anal-fin rays 50-55 (usually 52 or more); at least some of the 5—9 dorsal saddles reaching lateral band ......... it acs A PR He ilca oie Locusts Sopa chai aw Limnichthys fasciatus Waite — Combined number of dorsal and anal-fin rays 45-53 (usually 51 or fewer); dorsal saddles, usually 9-11, not reaching lateral band (band may be PLO SSTO De teahla.'s CSG. 50s Cae Mier ae ee Oe a ate ree nn er Pea eee 14 14. Lateral-line scales 39-41; combined number of dorsal and anal-fin rays 48-53; lateral band usually present ............ Limnichthys nitidus Smith — Lateral-line scales 36-38; combined number of dorsal and anal-fin rays 46-50; lateral band usually absent ....... Limnichthys donaldsoni Schultz The above key excludes the following two forms: a new species of Creedia from Japan being described by Mr. Kazuhiko Shimada; a form similar to Limnichthys fasciatus but with fewer meristic parts. Material of the latter collected in Fiji was sent to one of us (JSN) by R. Winterbottom of the Royal Ontario Museum (ROM). It may be deserving of subspecific status or may represent a separate but unde- scribed species. It has the following diagnostic characters: dorsal-fin rays 22-24; anal-fin rays 25-27; vertebrae 37—40 (most with 39); two epurals; 5-8 dorsal saddles (including a faint one on nape) and a lateral band (most or all saddles join the band in specimens over 19 mm SL, but four small specimens have short saddles not reaching the band); in addition, there is a dark interorbital bar. Spec- imens similar to this form have been collected from New Britain and Solomon Islands, based on BPBM and ROM material, respectively, and from Lizard Island, Great Barrier Reef (taken with L. donaldsoni), based on AMS material. Further study and more specimens are required to determine if this material is the same form as represented by the above-mentioned Fiji specimens. Acknowledgments The figures were prepared, under supervision, by Diane Hollingdale, and ra- diographs were made by Wayne Roberts. The study was supported by grant No. A5457 of the Natural Sciences and Engineering Research Council of Canada (to JSN) and a grant (to JER) from the National Geographic Society for field work at Easter Island. 410 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Literature Cited Leis, J. M. 1982. Hawaiian creediid fishes (Crystallodytes cookei and Limnichthys donaldsoni): development of eggs and larvae and use of pelagic eggs to trace coastal water movement.— Bulletin of Marine Science 32(1):166-180. Nelson, J.S. 1979. Some osteological differences between the blennioid fishes Limnichthys polyactis and L. rendahli, with comments on other species of Creediidae.—New Zealand Journal of Zoology 6:273—277. 1983. Creedia alleni and Creedia partimsquamigera (Perciformes: Creediidae), two new marine fish species from Australia, with notes on other Australian creediids.— Proceedings of the Biological Society of Washington 96(1):29-37. (JSN) Department of Zoology, The University of Alberta, Edmonton, Alberta T6G 2E9, Canada; (JER) Division of Ichthyology, Bernice P. Bishop Museum, P.O. Box 19000-A, Honolulu, Hawaii 96819. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 411-415 THE VOICES AND RELATIONSHIPS OF THE CHILEAN FROGS EUPSOPHUS MIGUELI AND E. CALCARATUS (AMPHIBIA: ANURA: LEPTODACTYLIDAE) J. R. Formas Abstract.—The voices of Eupsophus migueli and E. calcaratus are described. The calls of both species are short in duration and consist of a single note. The call data together with genetic information indicates that E. migueli, E. cal- caratus, and E. roseus comprise a related species group that is quite different from E. vittatus. Frogs of the genus Eupsophus (E. vittatus, E. roseus, E. migueli, E. calcaratus, and E. insularis) are endemic to the Nothofagus temperate forest of southern Chile and Argentina. Eupsophus migueli is restricted to the Coastal Range in Valdivia Province (Formas 1978), whereas E. calcaratus occurs widely in southern Chile (Formas and M. I. Vera 1982). Eupsophus migueli is sympatric but not syntopic with E. roseus (Iturra and Veloso, 1981) and this latter species is allopatric with E. calcaratus (Formas and M. I. Vera, 1982). Calls of frogs can be useful in revealing systematic and evolutionary relation- ships at the species level. In general, closely related species have some acoustic characteristics in common (Blair 1958; Kuramoto 1974, 1977; Mecham 1971; Schigtz 1973). In order to establish the evolutionary trends of the species of the genus Eupsophus we analyzed the voices of E. migueli and E. calcaratus. These data are compared with the calls of E. roseus and E. vittatus which were previously described (Formas and M. A. Vera 1980). The voice of E. insularis remains unknown. The call data presented in this paper demonstrate that E. migueli and E. calcaratus show strong similarities with E. roseus but that this species group is different from LE. vittatus. Methods and Materials Field recordings were made at 19 cm/sec on an Uher 4000 Report-IC portable tape-recorder and an Uher m 517 microphone. Audio-Spectrograms were made with a Kay Electric audio spectrograph model 675 employing 85-8000 Hz fre- quency scale and narrow (45 Hz) and wide (300 Hz) band filters. Temperature, location and behavior of all individuals used in the analysis were taken at the time of recording. Call repetition rates were measured in the field by counting consecutive calls over one minute. Specimens and tapes were deposited in the collection of Amphibians of the Institute of Zoology at the Universidad Austral de Chile (IZUA), Valdivia. Eupsophus migueli Formas The call characteristics here described are based on the call of 25 individuals recorded at 9—12°C at the locality of Mehuin (Valdivia Province) during September 412 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 0) 25) 1 sec. Fig. 1. Sound spectrograms of calls of Eupsophus migueli (a, b, c) and E. calcaratus (d). a, Type A call (45 Hz); b, Type B call (45 Hz); c, Type B call (300 Hz); d, Call of EZ. calcaratus (300 Hz). 1981. Eight calls, from five specimens were analyzed in detail. Males of this species were observed and collected while they were calling from cavities in the ground near a cool stream. No aggregation of individuals was observed when they were calling but calls of two or more males were organized into duets or trios. Two types of calls were observed. The call type A consists of a single note (Fig. 1a, Table 1) lasting 0.20-0.35 seconds. The repetition rate is 3-8 calls/minute. The call shows poor modulation and a well defined harmonic is present between 3200-— 3900 Hz. Maximum sound energy is spread over the frequency range 1500-2500 Hz. Type B calis (Fig. 1b, Table 1) of E. migueli are trills consisting of 19-33 notes per call. The call duration is 2.7—4.4 seconds and each note last about 0.08— 0.09 seconds with poorly defined silent intervals between notes within a call. Each note is composed of 4—7 pulses per note. Maximum sound energy of the poorly modulated notes is spread over the frequency range 900-1500 Hz. The repetition rate is 5—8 calls/minute. Harmonics are absent; the fundamental frequency equals the dominant. This type of call was only observed in two animals. Eupsophus calcaratus (Giinther) The description of the call is based on 15 calls from three individuals. The calls were recorded at 11°C at the locality of Puntra (Chiloé Province) during December 1982. Males were collected while they were calling at the bank of a stream. In this area males are isolated; however, in the locality of Pucatrihue (Osorno Prov- ince, October 1983) a moderate aggregation was observed. The call of this species (Fig. 1c, Table 1) consists of a single note lasting 0.15- 0.21 seconds. A well defined harmonic (2200-3800 Hz) is present and the call is poor in modulation. Maximum sound energy is distributed between 1100-2700 Hz. The repetition rate is 16—25 calls/minute. Eupsophus roseus (Duméril and Bibron) and E. vittatus (Philippi) The calls of both species were previously described by Formas and M. A. Vera (1980). The calls characteristic of E. roseus and E. vittatus are included in the Table 1. VOLUME 98, NUMBER 2 413 Table 1.—Call characteristic (mean and ranges) of Eupsophus species. Call length Dominant frequency Species Call type Notes per call (sec) Repetition rate (Hz) E. migueli A 1 0.24 (0.20-0.35) 6 (3-8) (1500-2500) E. migueli B 24 (19-33) 3.4 (3.3-4.4) 6 (5-8) (900-1500) E. calcaratus 1 0.19 (0.15-0.21) 19 (16-25) (1100-2700) E. roseust 1 0.20 (0.19-0.21) 64 (60-72) (1600-2900) E. vittatust 5 (4-6) 0.6 (0.4-0.8) 4 (2-10) (1100-2500) + From Formas and M. A. Vera (1980). Discussion When two types of calls are produced as in E. migueli and some other species (Heyer 1971; Straughan and Heyer 1976; Narins and Capranica 1976; Pyburn 1978), the signals may have a different meaning for the two sexes. Type A call observed in E. migueli shows a frequency range between 1500-2500 Hz, a rep- etition rate of 5—8 calls/minute and the single note lasts 0.20—0.35 seconds. The characteristics here described suggest that the type A call could code information in order to attract gravid females because the frequency is relatively narrow and not difficult to locate, and especially because the call is repeated at predictable intervals. The unusual call type B is a trill with a dominant narrow energy band (900-1500 Hz). This voice was observed at regular intervals (5-8 call/minute) when two males were at the same burrow. Mostly on the basis of this observation we suggest that this signal codes information in order to maintain the territory of a calling male in a particular burrow. When we compare the voices of Eupsophus migueli (call type A), E. calcaratus, and E. roseus (Fig. 1, Table 1) it is noteworthy that all three species show strong similarities. As suggested above this type of call could be useful in attracting females and, since the calls are similar, they could attract females of any species of this species group. However, this appears unlikely because these species do not occur in the same place. Eupsophus roseus and E. calcaratus are allopatric in distribution (Formas and M. I. Vera 1982), and E. migueli and E. roseus are sympatric but not syntopic at the locality of Mehuin (39°26'S, 73°10’W) (Iturra and Veloso 1981). If we compare the voice of E. vittatus with other Eupsophus species, two dis- tinctive groups can be established. The first group which includes E. roseus, E. migueli, and E. calcaratus is characterized by producing a short single note (Table 1). The calls have a distinct harmonic in the frequency range 2200-3900 Hz and a maximum sound energy range between 1500-2500 Hz. All species calls have poor modulation and pulsation. The second group, contains only one species, E. vittatus. In this frog the call duration is 0.4—0.8 second and is composed of four to six well pulsed notes. The dominant frequency range is between 1900 and 2500 Hz (Table 1). Some authors (Blair 1958; Kuramoto 1974, 1977; Mecham 1971; Schigtz 1973) have suggested that related species have acoustic characteristics in common. If this hypothesis is true, the strong similarities of the calls of E. roseus, E. migueli, and E. calcaratus suggest that these species are closely related to each other. 414 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Formas (1980) established two species groups in the genus Eupsophus based on chromosomal information. The first contains only E. vittatus, which has 28 biarmed chromosomes and a fundamental number (FN) of 56. The second group contains the following species: E. roseus and E. calcaratus (2n = 30, 8 biarmed pairs and 7 monoarmed, and FN 46), and E. migueli (2n = 30, 7 biarmed pairs and 8 monoarmed pairs, and FN 44). In addition, Formas et al. (1983) analyzed the allozymic differentiation in these species and found that Nei’s genetic distance of E. vittatus with respect to the other species indicates a higher level of genetic differentiation, whereas the remaining species (E. roseus, E. migueli, and E. cal- caratus) have a lower differentiation when compared with E. vittatus. If acoustical and genetic data are compared it is interesting to observe the agreement between both sets of information. This could be useful in establishing systematic and evolutionary trends in the genus Eupsophus. On the basis of the evidence pre- viously cited we conclude that the E. roseus, E. migueli, and E. calcaratus group is comprised of very closely related species while EF. vittatus remains clearly separated. Acknowledgments I am grateful to Carlos Varela for his field assistance. Sonia Lacrampe typed the manuscript. This work was supported by Proyecto de Investigacién RS-82- 9, Direccion de Investigacion y Desarrollo, Universidad Austral de Chile. Literature Cited Blair, W.F. 1958. Call structure and species group in U.S. tree frogs (Hy/a).— Southwestern Naturalist 3:77-89. Formas, J. R. 1978. A new species of leptodactylid frogs (Eupsophus) from the Coastal Range in Southern Chile.—Studies on Neotropical Fauna and Environment 13:1-9. . 1980. The chromosomes of E. calcaratus and the karyological evolution of the genus Eup- sophus (Anura: Leptodactylidae).—Experientia 36:1163-1164. ,and M.A. Vera. 1980. Reproductive patterns of Eupsophus roseus and E. vittatus.—Journal of Herpetology 14:11-14. ——, and M. I. Vera. 1982. The status of two Chilean frogs of the genus Eupsophus (Anura: Leptodactylidae).— Proceedings of the Biological Society of Washington 95:594—601. —, M. I. Vera, and Sonia Lacrampe. 1983. Allozymic and morphological differentiation in the South American frogs genus Eupsophus.—Comparative Biochemistry and Physiology 75B:475-— 478. Heyer, W. R. 1971. Mating calls of some frogs from Thailand.—Fieldiana 58:61-82. Iturra, P., and A. Veloso. 1981. Evidence for heteromorphic set chromosomes in male amphibians (Anura: Leptodactylidae).— Cytogenetics and Cell Genetics 31:108-—110. Kuramoto, M. 1974. Mating calls of Japanese tree frogs (Rhacophoridae).— Bulletin of the Fukuoka University, Faculty of Education 24:67-77. —. 1977. Mating call structures of the Japanese pond frog Rana nigromaculata and Rana brevipoda (Amphibia, Anura, Ranidae).—Journal of Herpetology 11:249-254. Mecham, T. S. 1971. Vocalization of the leopard frog, Rana pipiens, and the related Mexican species. —Copeia 71:505-—516. Narins, P. M., and R. R. Capranica. 1976. Sexual differences in the auditory system of the tree frog Eleutherodactylus coqui.—Science 192:378-380. Pyburn, W. F. 1978. The voice and relationships of the tree frog Hyla hobbsi (Anura: Hylidae).— Proceedings of the Biological Society of Washington 91:123-131. Schigtz, A. 1973. Evolution of anuran mating calls, ecological aspects. Jn J. L. Vial, ed., Evolutionary Biology of the Anurans.— University of Missouri Press, Columbia, Missouri, pp. 311-319. VOLUME 98, NUMBER 2 415 Straughan, I. R., and W. R. Heyer. 1976. A functional analysis of the mating calls of the neotropical frog genera of the Leptodactylus complex (Amphibia, Leptodactylidae).—Papéis Avulsos de Zoologia 29:221-245. Instituto de Zoologia, Universidad Austral de Chile, Casilla 567, Valdivia, Chile. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 416-421 A NEW SPECIES OF THE AQUATIC BEETLE GENUS DRYOPOMORPHUS FROM BORNEO (COLEOPTERA: ELMIDAE: LARINAE) Paul J. Spangler Abstract.— A new species of an asiatic elmid genus, Dryopomorphus satoi, from Sabah, Borneo, is described, illustrated with line drawings and scanning electron micrographs, and distinguished in a key from the other three described species. The genus Dryopomorphus was described by Hinton (1936) with D. extraneus from Japan as the only included species. Since that time, one additional species, D. nakanei Nomura (1958), was described from Japan, and a third species, D. bishopi Hinton (1971), was described from Malaysia. The new species described below is the second known species of Dryopomorphus from Malaysia. Dryopomorphus satoi, new species Figs. 1-9 Holotype male.—Body form and size. Oblong; moderately convex dorsally. Length, 4.0 mm; width, 1.7 mm (Figs. 1-3). Color.—Black dorsally except antennae yellowish-brown basally and slightly darker brown apically. Dorsum covered with long, sparse, black setae and short, dense, light yellowish-brown hydrofuge setae. Venter dark reddish-brown except genae, sides of prosternum, anterior half of hypomeron, all palpi, labium, coxae, trochanters, and tarsi lighter reddish brown. Head. —Surface with dual punctation; coarse punctures sparse, widely separated; fine punctures dense, separated by distance equal to their diameters. Patch of dense pubescence posteromediad of each eye. Eyes moderately strongly convex. Clypeus with anterior margin nearly truncate; surface punctate similarly to head; without conspicuous transverse row of long setae bent over labrum. Labrum with anterior margin bordered with dense golden setae; lateral angles with moderately dense tuft of long golden setae curved toward midline; surface punctate similarly to head. Labium with prementum bearing 2 transverse tufts of long, stout, black, bristle-like setae; submentum with numerous, long, yellowish setae on middle. Thorax.—Pronotum 0.8 mm long, 0.9 mm wide; widest across base; sides moderately arcuate; disc without median longitudinal impression, evenly convex; surface of disc punctate similarly to head; sublateral basal sulcus deep, straight, and subequal to length of scutellum; sides not steeply sloping; anterolateral angles obtuse, rounded; posterolateral angles obtuse, not rounded; lateral margins each distinctly rimmed. Hypomeron with anterior cavity for reception of antenna moderately deep and separated from posterior portion by distinct, transverse ridge; surface behind cavity coarsely, moderately densely punctate. Prosternum in front of coxae about one and one-half times longer than length of procoxae. Prosternal process broad, margined laterally, and tapering broadly to protuberant apex. Mesosternum deeply foveate for reception of protuberance of prosternal process. VOLUME 98, NUMBER 2 417 Figs. 1-3. Dryopomorphus satoi, new species, habitus: 1, Lateral view; 2, Dorsal view; 3, Ventral view. Metasternum with shallow median longitudinal impression; sides with punctures about 4 times as coarse as facets of eyes and separated by 1 to 3 times their diameters; disc densely, moderately coarsely punctate and with 4 or 5 very coarse punctures laterally adjacent to declivous sides. Foreleg uniformly pubescent. Mid- dle and hind legs pubescent except posteromedial surfaces of tibiae glabrous; 418 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON DRYDPOMORPHUS # DS S pe rd EF’. Figs. 4-6. Dryopomorphus satoi, new species: 4, Mesotibial fringe, 200; 5, Mesotibial fringe, 1200 x; 6, Metatibial fringe, 200 x. mesotibia with rather slender cleaning fringe of longer setae on inner, apical three- fourths (Fig. 6); metatibia with moderately dense cleaning fringe on inner apical three-fourths (Figs. 4, 5). Scutellum flat, ovate; base and apex about equally rounded; surface finely, densely punctate. Elytron with 10 rows of coarse punc- tures; surface finely densely punctate; with fine erect setae as long as intervals are wide; with fine, much denser, shorter, suberect setae between erect setae; base adjacent to scutellum with short, indistinct, transverse plica in distinct impression. Abdomen. — First sternum with sublateral carinae prominent and complete; sur- VOLUME 98, NUMBER 2 419 Figs. 7-9. Dryopomorphus satoi, new species, genitalia: 7, Male, dorsal view; 8, Male, lateral view; 9, Female, dorsal view. face between carinae densely, moderately coarsely punctate; with large, deep, ovoid cavity laterally on each side of each carina; cavities surrounded with long setae that extend across openings. Surface with coarse, sparse punctures and fine, dense punctures. Male genitalia.—As illustrated (Figs. 7, 8). Female.—Similar to male externally. Genitalia as illustrated (Fig. 9). Type data.— Holotype male: BORNEO: SABAH: Poring, 18 Sep 1983, Warren E. Steiner, Jr., and Gary F. Hevel; USNM Type-No. 100122; deposited in the National Museum of Natural History, Smithsonian Institution. Allotype and para- types (29), same data as holotype. Paratypes: Specimens will be deposited in the British Museum (Natural History) London; California Academy of Sciences, San Francisco; Canadian National Col- lection, Ottawa; Muséum National d’Histoire Naturelle, Paris; Institut royal de Sciences Naturelles de Belgique; Stovall Museum of Science and History, Norman, Oklahoma; Zoologische Sammlung Bayerischen Staates, Miinchen; and Dr. Sat0’s personal collection. Etymology.—This species is named for my friend Dr. Masataka Sat6 in honor of his numerous contributions to our knowledge of Asiatic water beetles. 420 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Habitat.—The type-series was collected from leaf packs among rocks in a par- tially shaded stream. Comparative notes.—Dryopomorphus satoi, n. sp., resembles D. bishopi and keys to that species in Hinton’s (1971) key to the species. However, D. satoi may be distinguished from D. bishopi as tabulated below. Dryopomorphus satoi . Pronotum with fine and coarse punctures. . Hypomeron with anterior cavity for reception of antenna very deep; plane of floor of cavity at an angle of about 80° to plane of posterior surface of hypomeron. . Intervals between rows of coarse elytral punc- tures distinctly, finely, and densely punctate. . Surface of first abdominal sternum between carinae densely, moderately coarsely punctate. . Scutellum about equally rounded basally and apically; surface finely, densely punctate. Dryopomorphus bishopi Pronotum uniformly finely, densely punctate. Hypomeron with anterior cavity for reception of antenna moderately deep; plane of floor of cavity at an angle of about 20° to plane of posterior surface of hypomeron. Intervals between rows of moderately coarse elytral punctures microalutaceous and with very few, indistinct, fine punctures. Surface of first abdominal sternum between carinae very sparsely, moderately coarsely punctate. Scutellum wider basally, angulate apically; surface not punctate. The following key, modified from Hinton (1971), will serve to distinguish the four species presently known in the genus Dryopomorphus. Key to the Species of Dryopomorphus 1. Eyes flattened, scarcely convex. Hypomeron with anterior cavity for re- ception of antenna only moderately concave and without transverse carina separating cavity from posterior part of hypomeron. Sublateral pronotal sulci distinctly longer than scutellum. Length, 4.9 mm. Japan ......... ip e810 ESTP © Ok Bah RY Ge TEN UO OES ke ROC ge extraneus Hinton — Eyes strongly convex. Hypomeron with anterior cavity for reception of antenna deep and separated from posterior part of hypomeron by distinct LIAN SVEESC bl Se wr eer eet ve crate nick doko she aha CR Li ace ape 2 2. Clypeus, near anterior margin, with row of long, conspicuous setae that are bent over labrum. Sublateral pronotal sulci present on basal two-fifths, nearly twice as long as scutellum. Length 2.6 mm. Japan ............. Ota pea Ie UE RM PLM Rie des MS as choo el earalt Fade es. SOO al nakanei Nomura — Clypeus, near anterior margin, without conspicuous transverse row of long setae. Sublateral pronotal sulci present on basal fourth, about as long as SCUPC MMII. Pree is Abate a hak Ue oa aie Sas lel ce 3 3. Pronotum with fine and coarse punctures. Hypomeron with anterior cavity for reception of antenna very deep; plane of floor of cavity at angle of about 80° to plane of posterior surface of hypomeron. Intervals between rows of coarse elytral punctures distinctly finely, densely punctate. Length, AOMMMMALAVSI At ent nwt eG COM tr: ennai a hres satol, new species — Pronotum uniformly finely, densely punctate. Hypomeron with anterior cavity for reception of antenna moderately deep; plane of floor of cavity VOLUME 98, NUMBER 2 421 at angle of about 20° to plane of posterior surface of hypomeron. Intervals between rows of moderately coarse punctures microalutaceous and with very few, indistinct, fine punctures. Length, 3.4 mm. Malaysia ........ 2 Bn hbase ehcves 8 eu led LASTS 6) SAS CRE A CTL Un A aR ane OLE bishopi Hinton Acknowledgments I thank the following people who contributed to this article: Warren E. Steiner and Gary F. Hevel, for collecting this new species and other interesting aquatic beetles for me in Borneo; Christine von Hayek of the British Museum (Natural History), for making the type-series of Dryopomorphus bishopi Hinton available, which allowed me to recognize D. satoi as new and distinct; Young T. Sohn, entomological illustrator, for the line drawings; Robin A. Faitoute, museum tech- nician, and Susann Braden, scanning electron microscopist, for the micrographs; and my wife, Phyllis M. Spangler, for typing the manuscript. Literature Cited Hinton, H. E. 1936. New Dryopidae from the Japan Empire (Coleoptera).—The Entomologist 69: 164-168. . 1971. The species of Dryopomorphus (Coleoptera, Elmidae).—The Entomologist 104:292- 297. Nomura, S. 1958. Notes on the Japanese Dryopoidea (Coleoptera).—Toho-Gakulo 8:45-—59. Department of Entomology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 422-438 MIOCENE AND PLIOCENE MOLIDAE (RANZANIA, MOLA) FROM MARYLAND, VIRGINIA, AND NORTH CAROLINA (PISCES: TETRAODONTIFORMES) Robert E. Weems Abstract. — Three species of fossil molid fish (Tetraodontiformes) are represented in the Miocene and Pliocene deposits of the central Atlantic Coastal Plain. Two species belong to the genus Ranzania (R. grahami and R. tenneyorum) and one belongs to the genus Mola. Premaxillary beaks were found to be best suited for osteologically defining species and genera in this family. Osteological definitions for the living and fossil genera and species in this family are proposed based principally on the character states present in the premaxillary elements. Fossil molid remains were first recognized in 1883, when Van Beneden de- scribed the upper and lower “beaks” of ‘““Orthogoriscus chelonopsis” from Bel- gium. Later, Leriche (1907) pointed out that Van Beneden already had described two dermal bones of molids in 1881, but that he had incorrectly assigned them to the genus Pagrus, as P. pileatus and P. torus. Because Leriche considered all of these materials to belong to a single species, he therefore called all of these specimens Orthogoriscus pileatus on the basis of page priority. In 1926, Leriche figured more material recovered from the Brussels basin and compared it to the living species Mola mola (then usually called Orthogoriscis mola). Leriche decided that all of his fossil material constituted a single species which was collectively distinct from the modern species. More recently, Van Deinse (1953) described some dermal bones of a molid from the upper Miocene strata of the Netherlands which he also chose to refer to Mola pileata. These Miocene dermal bones are quite comparable to the material Van Beneden and Leriche described, especially since dermal bones are highly variable in shape in the living Mola mola (Harting 1864, cited in Van Deinse 1953). Romer (1966:361) listed the genus Mola as occurring in the Tertiary of South America, but no other reference to such an occurrence was located. Apart from this one reference, all fossil specimens of molids reported in the literature have been from Miocene strata of France, Neth- erlands, and Belgium, and all of this material has been referred to a single fossil species of the extant genus Mola. Berry (1941) unknowingly described a molid premaxillary beak from the upper part of the Calvert Formation of Virginia; he considered it to be the fused dentaries of the sea turtle Sy/lomus. D. H. Dunkle brought this error to my attention and it was footnoted in my paper on Calvert sea turtles (Weems 1974). Because only a footnote was published, the true nature of this premaxillary beak is still not widely recognized by ichthyologists. Since Berry’s paper, much more molid ma- terial has been collected from the Calvert and from the overlying Choptank For- mation. Much of this material has good stratigraphic control. A large selection of premaxillary beaks are known from the upper Calvert which can be used to determine the degree of variability in the premaxillary beaks of what was probably VOLUME 98, NUMBER 2 423 a single genetic population. Three other specimens, one from Calvert-age beds in central Virginia, one from the Choptank Formation of Maryland, and the third from the Yorktown Formation of North Carolina, clearly fall outside of the range of variability seen in the upper Calvert population of specimens. This indicates that a major revision of the taxonomy of the fossil members of this family is warranted and that two new species of molids can be recognized. Material The following specimens were used for comparisons made in this paper. ““Zones”’ are from Shattuck (1904): 1) AMNH 1679. Premaxillary beak. Calvert Formation, upper “‘zone” 11, ‘“‘zone’’ 12, or lower “‘zone”’ 13, Horsehead Cliffs, Westmoreland State Park, Westmoreland County, Virginia. Charles T. Berry, 9 Jun 1940 (see Berry 1941). 2) USNM 16363. Premaxillary beak. Calvert Formation, Chesapeake Beach, Calvert County, Maryland. William Palmer. 3) USNM 16364. No data. 4) USNM 16615. Premaxillary beak. Calvert Formation, Scientists Cliffs, Calvert County, Maryland. W. E. Salter, 1941. 5) USNM 16668. Premaxillary beak. Calvert Formation, Scientists Cliffs, Calvert County, Maryland. Mr. Foshag and Remington Kellogg, 1941. 6) USNM 16743. Premaxillary beak. Calvert Formation, “‘zone” 11, Scientists Cliffs, Calvert County, Maryland. L. P. Schultz, 1943. 7) USNM 186982. Dentary beak. Calvert Formation, “‘zone” 11, about 2 feet above shell layer, 1.7 miles south of Plum Point, Calvert County, Maryland. R. Lee Collins, 6 Jul 1936. 8) USNM 186983. Premaxillary beak. Choptank Formation, “‘zone” 19, just north of Bay Haven Camp, Calvert County, Maryland. R. Lee Collins, 8 Sep 1933. 9) USNM 186984. Dentary beak. Found on beach at Randle Beach. Lauck W. Ward, Aug 1969. 10) USNM 186986. Premaxillary beak, dentary beak, dorsal armor shield, nasal plate, jugular plate, and branchial arch or vertebral spine fragments. Calvert Formation, upper “zone” 11, Horsehead Cliffs, Westmoreland State Park, Westmoreland County, Virginia. R. E. Weems, Mar 1967. 11) USNM 265391. Fragmentary premaxillary beak. Calvert Formation, upper “zone” 11, Stratford Cliffs, Westmoreland County, Virginia. James Kal- tenbach. 12) USNM 265392. Premaxillary beak. Calvert Formation, in the local basal phosphate horizon of Calvert Formation, Gravett’s Mill Pond, King Wil- liam County, Virginia. R. E. Weems, Dec 1969. 13) USNM 265393. Fused dentary beak. Calvert Formation, upper “‘zone”’ 11, 4 feet above beach, Stratford Cliffs, Westmoreland County, Virginia. Rob- ert E. Weems, around 1970. 14) USNM 265394. Premaxillary beak. Calvert Formation, upper “‘zone”’ 11, 4 feet above beach, Stratford Cliffs, Westmoreland County, Virginia. R. E. Weems, Mar 1970. 15) USNM 265395. Fragmentary premaxillary beak. Calvert Formation, “zone” 424 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 12, 6 feet above beach, south corner of second bluff north of Parker Creek, Maryland. James Kaltenbach. 16) USNM 265650. Fragmentary premaxillary beak. Yorktown Formation, Lee Creek Mines, Aurora, North Carolina. 17) USNM 265651. Isolated dermal plates. Calvert Formation, “‘zone”’ 12, Parkers Creek, Calvert County, Maryland. Collector unknown, 20 Apr 1935. 18) USNM 265653. Molid jugal plate. Probably Calvert Formation, Scientists Cliffs, Calvert County, Maryland. Collector Walter Simonson. 19) USNM 291211. Fragmentary premaxillary beak. Yorktown Formation, Lee Creek Mines, Aurora, North Carolina. 20) USNM 336431. Small premaxillary beak. Calvert Formation, “zone” 10, Plum Point, Calvert County, Maryland. David O. Bohaska, 15 Oct 1983. Description Of the above-cited specimens, five include much or all of the premaxillary beak and also come from either upper “‘zone”’ 11 of the Calvert or possibly from “‘zone”’ 12 or lower “‘zone”’ 13. Three others almost certainly come from this same interval. These specimens occur within a narrow stratigraphic range, so it is reasonable to compare them to see if they represent more than a single population. The angle of these beaks are all within the rather narrow range of 77° to 88°, with a mean value of 84.5° and a standard deviation of +2°. The beaks vary considerably in total size, but this is to be expected due to wide differences in the age of individuals at death. The most variable character is the ratio between the beak length and the length of the bony palate that lies behind the beak. Even here, however, the changes are readily correlated with size and can be ascribed to age variation within a single species (Fig. 1). Except for USNM 336431, each specimen is toothless in the region of the anterior biting edge of the beak, but farther back on the roof of the mouth the bony palate region contains three poorly defined rows and clusters of teeth sharply demarcated from the toothless area (Figs. 1, 3D, 6D, 7B). The teeth are poorly developed and show no clear sign of enamel covering. The bony palate above and behind the tooth-bearing region consists of a large mass of bone which acts as a brace for the teeth. This is variably developed, but generally is more massive in larger specimens. One specimen within this population (USNM 186986) is a partial skeleton of a single large individual. This specimen includes the premaxillary beak, the den- tary beak, the nasal plate, the jugular plate, a large sheet of dermal armor, and poorly ossified bones possibly representing branchial arch supports or vertebral spines (Figs. 3—5). This material constitutes most of what could reasonably be expected to be fossilized in a molid. This specimen and the previously discussed isolated premaxillary beaks from the upper Calvert Formation can be referred to a single species that is variable within well-defined limits. Collectively, these beaks all differ from the described European specimens in the massive development of the tooth-bearing bony palate behind the beak (Fig. 7). Such a palate is absent in all but one of the specimens described by Leriche and in the living Mola mola. Therefore, the upper Calvert population represents a single species that is distinct from the one described European molid species. VOLUME 98, NUMBER 2 425 % / (mm) xX 100 / Total width / / / De) 4 ras : oe Je USNM 186986 / B lat USN / ony palate 16364 / e XZ 60 /,USNM 16743 Q / cad USMN ail oy 265391 / = AMNH / ee omInG79 Ne LUSNM 186982 USNM 2USNM 16668 265394 Beak width 20 0 ae) fe maa. TO MOR BAL 0 20 40 60 80 100 (mm) BEAK LENGTH Fig. 1. Graph showing the distribution in size of specimens of Ranzania grahami n. sp. from the upper Calvert Formation. Although total length and beak length vary greatly (see inset), the ratio of these two values are related by the simple algebraic relationship y = 0.35X + 5 (dashed line). Thus, these specimens probably represent different growth stages within a single population. Three other premaxillary beaks from Maryland, Virginia, and North Carolina fall well outside the range of variation seen in the upper Calvert specimens. One specimen comes from beds in central Virginia that appear to be equivalent to the upper Plum Point Marl Member of the Calvert Formation (L. W. Ward, written communication, 1983). This beak differs from the previously described specimens in the development of three pairs of well-defined tooth rows on the oral surface of the palate (Fig. 7D). These are quite distinct, unlike the poorly developed tooth rows of the specimens already considered. The other two beaks, a premaxillary beak from “zone” 19 of the Choptank Formation of Maryland (Fig. 7C), and a premaxillary beak from the Yorktown Formation of North Carolina (Figs. 6A, B), have lost all trace of the tooth rows and most of the mass of bone above the tooth rows. Their conformation is most like that of the European molids that have been described in the literature. In these two kinds of beaks we can see (1) an earlier morphological stage where teeth are better developed than in the pre- viously considered upper Calvert population and (2) a later morphological stage in which the teeth are essentially gone, closely approaching the condition seen in the all but toothless living Mola mola. Since the toothless beaks are still better ossified than those of the living M/. mola, they are best referred to M. chelonopsis and not to either of the living species. The beak with well developed tooth rows appears to represent a second new species of molid. 426 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Osteology of the Living Genera of Molids Although all fossil molid materials previously have been referred to the genus Mola, there are three extant genera in the family (Tyler 1980). Fraser-Brunner (1951) considered Mola to include two species (M. mola and M. ramsayi), but Tyler (1980) seems unconvinced that the two species are distinct. Both Tyler and Fraser-Brunner consider Ranzania and Masturus to be monotypic. Several papers have appeared on the anatomy of these fishes (for Mola: Gregory and Raven 1934; for Ranzania: Raven 1939a; for Masturus: Gudger 1937a, b; Raven 1939b), but the only paper that has dealt with the dental apparatus of any of these fishes is Tyler (1980). Although Tyler notes a crowded tooth battery in young Mola, study by the present author of adult skeletal remains has shown no instance in which distinct teeth are rooted in the bony portion of either the upper or lower beaks. Apparently the teeth are entirely embedded in cartilage in the living adult fish. Study of whole specimens of Ranzania by the author has shown that this genus possesses well developed rows of teeth which move past each other as the fish chews, apparently producing a shredding and chopping action (e.g., USNM 75155; ANSP 106723) (Fig. 2). Ranzania has been reported to eat only littoral seaweed (Barnard 1927), so the utility of this dental apparatus seems obvious. Only two specimens of Masturus could be located; neither was available for dissection (USNM 5704; AMNH 15962). Observation of the mouth, however, revealed a battery of rounded pavement teeth (Fig. 2) reminiscent of those de- veloped in pycnodont fishes and the living drumfish (Pogonias). The conformation of the teeth in the premaxillary beak differs somewhat from that shown in Tyler (1980:387), possibly due to age or size variation, but the dentary tooth battery in the specimens I observed agree well with the one figured by Tyler (p. 389). Al- though I could not observe if a bony mouth roof is present in Masturus, the figure of the skull of Masturus in Tyler (1980) suggests that one is possibly present. No such mass is present in the very cartilaginous Mola. Of the three living genera, only Ranzania has an extensive dermal skin external covering of regularly po- lygonal, bony armor plates; in both Masturus and Mola there are only scattered dermal patches bound in a cartilaginous sheet (Fraser-Brunner 1951, Tyler 1980). Taxonomy of the Miocene and Pliocene Molids From the foregoing descriptions, it is clear that only the Choptank and Yorktown premaxillary beaks can be referred properly to the genus Mola. The presence of regular rows of shredding teeth and a large mass of bone reinforcing the dental battery readily place the central Virginia beak in the genus Ranzania. The upper Calvert specimens afford a somewhat intermediate state between Ranzania and Mola, but since the bone reinforcing the dental battery is retained and an extensive bony dermal armor (albeit composed of irregular rather than polygonal plates) is retained, this form as well is best considered still to belong within the genus Ranzania. It is obvious, however, that this species had evolved far in a direction toward the character states present in the living genus Mola. Since the jugular and nasal bones are quite variable in shape and show no major change in form and shape from the upper Calvert population of Ranzania through the modern Mola mola, they are of little more than familial taxonomic value. VOLUME 98, NUMBER 2 427 ZET LSS {; ‘ Za MEGA— YEARS Mola Mola 9 —/HOLOCENE ners) aAeay Ranzania laevis Masturus lanceolatus PLEISTOCENE| _| = PLIOCENE ee a chelon- a = 2 | opsis : 10 > p s g 15 _| MIOCENE I a AS fa s 20 ie = Ranzania grahami : Ranzania tenneyorum 25 ~ Lu o a >) 30 ~ OLIGOCENE ~ > 35 36 Fig. 2. Morphology and oral characteristics represented in the various fossil and living species of the family Molidae. Age ranges (solid lines) and probable phyletic pathways (dotted lines) are indicated. Therefore, the species described by Van Beneden as Pagrus pileatus and Pagrus torus are considered to be each a nomen dubium; they could belong to any of at least three species of fish. Only “Orthagoriscus chelonopsis” shows enough char- acters to merit designation as a type, and the proper name for this species should be Mola chelonopsis. This specimen is quite comparable to the nearly toothless species present in the Choptank and Yorktown (cf. Leriche 1926). The upper Calvert species is undescribed, so the name Ranzania grahami is proposed after Robert A. Graham, who has accompanied the author on numerous fossil molid collecting trips. The central Virginia species also is undescribed, and the name Ranzania tenneyorum is proposed after Eleanor and Wilton Tenney, who also 428 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 10 CM fo) (ee St SENN Na) Fig. 3. Dentary and premaxillary beaks of Ranzania grahami n.sp. (type), USNM 186986. A, Dentary beak, ventral (aboral) view; B, Dentary beak, dorsal (oral) view. Note the notch cut in the right dentary by a predator’s tooth (dark shadow) and the piece of bone displaced outward by this injury; C, Premaxillary beak, dorsal (aboral) view; D, Premaxillary beak, ventral (oral) view. spent considerable time collecting with the author on the trips when molid remains were found. Because Masturus is so poorly known, it cannot be adequately defined yet in osteological terms. However, the presence of rounded crushing teeth is obviously a key distinguishing character. Osteologically the three recognizable living and fossil species of Ranzania can be defined as follows: Ranzania Nardo, 1840 Ranzania Nardo, 1840:10, 105 (type-species: Ranzania typus Nardo (=Ostracion laevis Pennant) by Whitley 1933). Generic diagnosis. —External dermal coating of polygonal plates, internal skel- eton largely cartilaginous except for premaxillaries and dentaries fused respectively into upper and lower beak. Upper beak posteriorly containing rows or patches of teeth fused against thick mass of bracing bone overlying roof of mouth. Range.— Middle Miocene to Holocene. Ranzania laevis (Pennant, 1776) Species diagnosis.—In addition to generic characters, this species shows a reg- ularly polygonal array of dermal plates. Teeth occur in four discrete intermeshing VOLUME 98, NUMBER 2 429 Fig. 4. Ventral view of the dorsally located dermal armor shield of Ranzania grahami n.sp. (type), USNM 186986. rows on both the dorsal and ventral surfaces of the mouth, ventral row not attached to a bony shelf. Range.— Holocene. Ranzania grahami, new species Figs. 3-5, 6C—D, 7A-B, 8 Holotype.—USNM 186986, Premaxillary beak, dentary beak, dorsal armor shield, nasal plate, jugular plate, and branchial arch or vertebral spine fragments; Calvert Formation, upper “‘zone” 11, Horsehead Cliffs, Westmoreland State Park, Westmoreland County, Virginia; R. E. Weems, Mar 1967. Species diagnosis.—In addition to generic characters, dermal coating made of very irregularly-shaped polygonal plates. Teeth present but in poorly defined rows and patches that are rooted directly to the palatal bracing bone. In the one known very young premaxillary beak specimen (less than 20 mm length), the anterior beak region may contain up to six pairs of tooth ridges crowded against the front of the beak, followed posteriorly by the irregular tooth rows on the bony palate typical of this species. Apparently the anterior six rows wear out with age, leaving this region toothless in larger individuals. Nasal plate and jugular plate similar to those of living Mola mola. Range.— Lower middle Miocene. 430 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ae | z f s O 3 10 CM Fig. 5. Isolated dermal bones of Ranzania grahami. A, Jugular plate (USNM 265651); B, Nasal plate (USNM 186986, type); C, Dermal element (USNM 186986, type); D, Fragmentary jugular plate (USNM 186986, type); E, Jugular plate (USNM 265653). VOLUME 98, NUMBER 2 431 Ranzania tenneyorum, new species Fig. 7D Holotype.—USNM 265392, premaxillary beak; Calvert Formation, in the local basal phosphate horizon of Calvert Formation, Gravett’s Mill Pond, King William County, Virginia; R. E. Weems, Dec 1969. Species diagnosis.— Based solely on the fused premaxillaries. Beak small, with three well developed pairs of tooth rows present behind it which are located on a bony shelf well below the level of the palatal bracing bone, so that there is a well developed notch behind the tooth rows. Other premaxillary beak characters as in genus. Range.— Lower middle Miocene. Mola Koelreuter, 1770 Mola Koelreuter, 1770:337 (type-species: Mola aculeata Koelreuter (=Tetraodon mola Linnaeus, juvenile) by Jordan, 1885). Generic diagnosis.—Dermal armor reduced to a nasal plate, a jugular plate, scattered small dermal plates along the body and along the clavus. Premaxillary and dentary beaks reduced nearly to totally toothless condition with only a few, isolated, scattered teeth present along the internal beak margin. Palatal bony brace for the teeth totally reduced to cartilage. Range.— Upper Middle Miocene to Holocene. Osteologically, the three living and fossil species of Mola can be defined as follows: Mola mola (Linnaeus, 1758) Species diagnosis.—In addition to the generic characters, 8 to 9 ossicles along the clavus, paraxial pair of ossicles fused. Premaxillary beak strongly reduced. Range.— Holocene. Mola ramsayi (Giglioli, 1883) Species diagnosis. —In addition to the generic characters, 12 ossicles along the clavus, paraxial pair not fused. Dentary and premaxillary beaks not described, presumably as for M. mola. Range. — Holocene. Mola chelonopsis (Van Beneden, 1883) Species diagnosis. —Premaxillary beak toothless, and lacking palatal tooth brace, toothless shelf anterior to location of the former tooth position much longer than in M. mola, such that the antero-posterior beak length is greater than the lateral beak width at the level of the back of the shelf. Dentary beak comparable to M. mola. Range.— Upper middle Miocene to lower Pliocene. Remarks.— All but one of the premaxillary beaks which Leriche (1926) de- scribed totally lack both the bony mass above the mouth roof and the rows of 432 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON O 3) 10 CM Tee) LEE Tee) Lae (YT (A aU VR) LDR ee (ST (es | Fig. 6. Dorsal and ventral views of the premaxillary jaw of Ranzania grahami and Mola chelon- opsis. A, Mola chelonopsis (USNM 265650), dorsal view; B, Mola chelonopsis (USNM 265650), ventral view; C, Ranzania grahami (USNM 265395), dorsal view; D, Ranzania grahami (USNM 265395), ventral view. teeth. These specimens therefore should be referred to the above species. All of these specimens presumably came from the upper middle Miocene. One small beak that Leriche described has two (but not three) tooth rows imbedded in a bony palate. In the number of tooth rows present it is somewhat different from VOLUME 98, NUMBER 2 433 Fig. 7. The premaxillary jaws of Ranzania tenneyorum, Ranzania grahami and Mola chelonopsis. A, Midline section through a beak of Ranzania grahami (USNM 265391); B, Ventral view of same beak of Ranzania grahami showing poorly developed and irregularly placed teeth on the bony palate; C, Ventral view of beak of Mola chelonopsis (USNM 186983) lacking development of a bony palate. Total length 23 mm; D, Ranzania tenneyorum (USNM 265392) showing three well developed pairs of tooth rows on the bony palate. Total length 25 mm. 434 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A B Fig. 8. Lateral view of the right dentary of Ranzania grahami, showing a partially healed fracture running down from a wedge shaped groove (A) produced by the tooth ofan attacking predator. Although the wound partially healed, jaw occlusion was permanently altered. As a result, the premaxillary beak margin cut a broad U-shaped notch into the dentary element (B). The dark, linear shadow to the right and below “‘C” is a second fracture line which healed. Both the healed and unhealed fracture lines are visible in Plate 1B, because the triangular bone fragment lying between them was displaced outward at the time of injury. R. tenneyorum, but otherwise it is similar to that species and thus probably should be assigned to it. Phylogeny of the Family Molidae The Calvert and Choptank specimens suggest that the genus Mola was derived from a form like Ranzania tenneyorum by way of R. grahami and Mola chelon- opsis (Fig. 2). Ranzania tenneyorum is both morphologically and temporally a suitable ancestral form for the living Ranzania laevis, even though no specimens of intermediate age are known. Thus, the genera Mola and Ranzania can be closely linked by intermediate fossil forms. The crushing dental apparatus of Masturus, on the other hand, is quite different and suggests that Masturus has a more remote affinity with both Ranzania and Mola. The family Molidae is con- sidered to be derived from a distant common stock with the mollusc-eating family Diodontidae and to be most closely related to the family Triodontidae (Tyler VOLUME 98, NUMBER 2 435 1980). While different in detail, the dental apparatus of Masturus more closely approaches the dental apparati of these related families than the dental apparati of either Ranzania or Mola within its own family. Moreover, only Masturus retains the primitive characteristic of a vestige of the true tail, which has been eliminated totally in the other two genera and been replaced functionally by the dorsal and anal fins, and the clavus (Fraser-Brunner 1951). Therefore, it seems most likely that Masturus separated from the Ranzania-Mola stock before the appearance of the Miocene species R. tenneyorum and R. grahami, perhaps as early as Oligocene time. This indicates relationships within the family as shown in Fig. 2. Working only with modern material, Raven (1939b), Fraser-Brunner (1951), and Tyler (1980) suggested that Mola and Masturus were the more closely related living genera. Fraser-Brunner considered Ranzania to be the most gen- eralized genus, while Raven and Tyler considered it to be the most derived. While I agree with Fraser-Brunner and Tyler that Ranzania is the most derived member of this family, the fossil sequence used for this study suggests that the gene pools of Ranzania and Mola became isolated from each other at a time later than when both’ became isolated from Masturus. This leads me to argue: 1) that the highly cartilaginous state of Mola and Masturus is the result of parallel evolution; 2) the tail remnant and dental apparatus of Masturus represent uniquely retained prim- itive features within the family; and 3) the numerous similarities between Mola and Masturus are commonly retained primitive familial characters that cannot be used to prove an especially close relationship between these two genera. This implies that the elongate and dorso-ventrally flattened body form of Ranzania (Fig. 2) is a derived character within this family (as Fraser-Brunner 1951, also pointed out), in contrast to the body form of Mola and Masturus which represent the primitive family character. The fact that the mouth of Ranzania closes along a vertical line, rather than along a horizontal line as in most vertebrates (Fraser- Brunner 1951), clearly shows that it has undergone unique specializations relative to the other genera in its family and is not simply a primitive and conservative member of this group. Therefore, it is reasonable to argue that the body shape of Ranzania also is a specialized rather than a primitive trait. On the basis of morphology alone, Mola and Masturus are similar, but this is believed to have resulted from persistent conservatism in these two genera rather than from an especially intimate phylogeny. If the above phylogeny is accepted, then a graded series of evolutionary steps can be postulated to derive the living pelagic Mola mola (and M. ramsayi?) from a bottom-dwelling, mollsucivorous ancestor. While the dental apparatus of Mas- turus 1s similar to that of the mollusc-eating Diodontidae, its body is not so rounded as that of most persistent bottom dwellers and catch records also suggest a world-wide pelagic distribution. Its known diet includes seaweed (Palmer 1936) as well as sponges, mollusks, and annelid worms (Yabe 1950). Yabe considered this diet to indicate that these fish normally feed on the sea beds. This implies that the first step in the derivation of this family was a change of feeding habit from gathering sea bed bivalves to foraging for food along the sea floor. This resulted in a major modification of the body from relatively wide and flattened to relatively narrow and high. In the line leading to modern Ranzania, the fish became specialized and adopted a diet that no longer included meat and consisted entirely of seaweed (Barnard 1927). This resulted in a change in the dental ap- 436 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON paratus from a crushing dentition to a shredding dentition like that found in R. tenneyorumi. In the other, less specialized branch of Ranzania(R. grahami) leading toward the modern Mola, the fish became adept at eating both plants and animals. The modern Mola has been reported to eat algae (Bigelow and Welsh 1925; Breder 1932; Fries et al. 1895), seaweed (Breder 1932; Reuvens 1897; Townsend 1918), eelgrass (Zostera) (Bigelow and Welsh 1925; Fries et al. 1895; Reuvens 1897), sponges (Breder 1932), hydroids (Bigelow and Welsh 1925), jellyfish (Bellomy 1961; Bigelow and Welsh 1925; Binney 1842; Breder 1932; Dieuzeide et al. 1955; Fries et al. 1895; Hargitt 1905; Hubbs and Schultz 1929; Linton 1901; MacGinitie and MacGinitie 1949; Nichols and Breder 1927; Zimbelman 1967), ctenophores (Bigelow and Welsh 1925; Hargitt 1905; Linton 1901), mollusks (Bigelow and Welsh 1925; Dieuzeide et al. 1955; Fries et al. 1895), crustaceans (Bigelow and Welsh 1925; Dieuzeide et al. 1955; Fries et al. 1895; Linton 1901; Nichols and Breder 1927), echinoderms (Bigelow and Welsh 1925), salps (Bigelow and Welsh 1925; Dieuzeide et al. 1955; Fries et al. 1895; Linton 1901; Nichols and Breder 1927), and fish (Boulenger 1936; Dieuzeide et al. 1955; Fraser-Brunner 1951; Fries et al. 1895; Grassi 1897; Norman 1931; Norman and Fraser 1949; Reuvens 1897; Schmidt 1921). This kind of a diet only requires a dental apparatus for plucking and swallowing. This set the stage for the nearly total loss of palatal teeth and their supporting bony shelf that has occurred in Mola. Thus the morphological changes that can be seen in the premaxillary beak, from Ranzania tenneyorum through R. grahami and Mola chelonopsis to M. mola and M. ramsayi, are interpreted here to reflect a progressive change in the habitat and diet of these forms. Pathology The type-specimen of Ranzania grahami (USNM 186986) has a broad notch in the dentary beak. Below this notch two radiating fractures are present. One fracture is completely knit but the other is not, even though a great deal of bone growth has occurred in the vicinity of the fracture (Figs. 3A, B, 8). This is obviously a pathological condition perhaps resulting from an attack by a large predator. The partial healing of the fracture indicates that the victim survived for an extended period of time, as does the strongly asymmetrical wear on the dentary beak tip and the presence of a deep wear-notch cut into it by the premaxillary beak (Fig. 8). The notch facet and wear pattern clearly show that jaw occlusion was severely distorted by the attack, and the fact that the animal survived at all suggests that it could swallow food whole without complex mastication. This suggests that R. grahami was far along in its feeding habits towards handling a Mola-like rather than a Ranzania-like diet. It is interesting to note that the osteological response to this wound was similar to that seen in untreated human fractures; 1.e. formation of bony masses around the persistently unknit fracture. Acknowledgments The author is indebted to Lauck W. Ward and James C. Tyler for critically reviewing this manuscript and suggesting improvements. Tyler made available to me his extensive annotated bibliography on the diet of molid fishes, and I am VOLUME 98, NUMBER 2 437 indebted to him for most of the references comprising the data base for this subject. Literature Cited Barnard, K. H. 1927. A monograph of the marine fishes of South Africa, Part 2.—Annals of the South Africa Museum 21(2):419-1065. Bellomy, M. D. 1961. The fish that is all head.—Sea Frontiers 7(1):12-18. Berry, C. T. 1941. The dentary of Syllomus crispatus Cope.—American Museum Novitates 1132: 1-2. Bigelow, Henry B., and William W. Welsh. 1925. Fishes of the Gulf of Maine. —Bulletin of the U.S. Bureau of Fisheries 40(1):1—567. Binney, Amos. 1842. Observations made during two successive summers at Nahant, on the habits of the Orthogariscus (sic) mola, or short sunfish. — Proceedings of the Boston Society of Natural History 1 (for 1841-1844):93. Boulenger, E.G. 1936. A Natural History of the Sea.—Appleton-Century Co., New York, 215 pp. Breder, Charles M. 1932. Fish notes for 1931 and 1932 from Sandy Hook Bay.—Copeia 1932 (4): 180. Deinse, A. B. van. 1953. Fishes in Upper Miocene and Lower Pleistocene deposits in the Nether- lands.— Netherlands, Geologische Stichting, Mededelinger, Nieuwe Serie 7:5-12. Dieuzeide, R., M. Novella, and J. Roland. 1955. Catalogue des Poissons des Cotes Algeriennes (III Osteopterygii, suite et fin).— Bulletin du Station Aquiculture Péche de Castiglione (Algerie), n. s. 6:11-384. Fraser-Brunner, A. 1951. The Ocean Sunfishes (family Molidae).—Bulletin, British Museum of Natural History (Zoology) 1(6):89-121. Fries, B., C. V. Ekstrom, and C. Sundevall. 1895. A History of Scandinavian Fishes, Second Edition, Revised and Completed by F. A. Smitt, v. II. Stockholm, pp. 567-1240. Giglioli, H. H. 1883. Zoology at the fisheries exhibition, II. Notes on the vertebrata.— Nature (London) 28(718):313-316. Grassi, G. B. 1897. The reproduction and metamorphosis of the common eel (Anguilla vulgaris). — Proceedings of the Royal Society of London 60:262-271. Gregory, W. K., and H. C. Raven. 1934. Notes on the anatomy and relationships of the Ocean Sunfish (Mola mola).—Copeia (4):145-151. Gudger, E. W. 1937a. The structure and development of the pointed tail of the Ocean Sunfish Masturus lanceolatus.— Annals and Magazine of the Natural History (10)19:1—46. . 1937b. The natural history and geographical distribution of the Pointed-tailed Sunfish (Mas- turus lanceolatus) with notes on the shape of the tail.—Proceedings of the Zoological Society of London, A 3:353-396. Hargitt, Charles W. 1905. The medusae of the Woods Hole region.— Bulletin of the U.S. Bureau of Fisheries 24:23-79. Harting, P. 1864. Leerboek van de Grondbeginselen der Dierkunde. 2° deel, 1° afd., Tiel. Hubbs, Carl L., and L. P. Schultz. 1929. The northward occurrence of southern forms of marine life along the Pacific coast in 1926.—California Fish and Game 15(3):235-241. Jordan, D. S. 1885. A list of the fishes known from the Pacific coast of tropical America from the tropic of Cancer to Panama.— Proceedings of the United States National Museum 8:361-394. Koelreuter, J. T. 1770. Piscium rariorum e Museo Petropolitano exceptorum descriptiones: 1) Gas- teropelecus Gronovii seve Clupea Linn.; 2) Trutta dentata; 3) Gobio; 4) Cyprinus; 5) Gobio pinna ventrali subrotunda; 6) Gobio pinna dorsualiunica; 7) Sparus; 8) Labrus valde oblongus; 9) Scomber,; 10) Lophius; 11) Mola aculeata, 12, 13) Perca.—Novi Commentarii Academiae Scientiarum Imperialis Petropolitanae 8:63—404. Leriche, M. 1907. Révision de la Faune ichthyologique des Terrains neogenes du Bassin du Rhone. — Association Francaise pour l’Avancement des Sciences, Compte Rendu de la 35¢ Session (Lyon, 1906), Notes et Memoires, 351 pp. 1926. Les Poissons Neogenes de la Belgique.—Memoires du Musée Royal d’Histoire Na- turelle de Belgique, 32:367—-472. 438 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Linnaeus, Carolus. 1758. Systema Naturae. Volume 1. Regnum Animale (10th edition). Upsala, Sweden, 824 pp. Linton, Edwin. 1901. Parasites of fishes of the Woods Hole region.—Bulletin of the U.S. Fish Commission 19:405—492. MacGinitie, G. E., and Nettie MacGinitie. 1949. Natural History of marine animals. McGraw-Hill, New York, 473 pp. Nardo, G. D. 1840. Considerazioni sulla famiglia dei pesci Mola, e sui caratteri che li distinguono. — Annali della Scienze Naturali del Regno Lombardo-Veneto 10:105-112. Nichols, John T., and C. M. Breder, Jr. 1927. The marine fishes of New York and southern New England. — Zoologica (New York) 9(1):1-192. Norman, J. R. 1931. A History of fishes. Ernest Benn, Ltd., London, 463 pp. , and Fraser, F.C. 1949. Field Book of Giant Fishes. G. P. Putnam’s Sons, New York, 376 pp. Palmer, R. H. 1936. Ocean sunfish in Habana Waters.—Science 83(2164):597. Pennant, Thomas. 1776. British Zoology. Volume 3—Reptiles. Fish (4th edition). W. Eyres, War- rington. Raven, H.C. 1939a. Notes on the anatomy of Ranzania truncata.—American Museum Novitates 1038:1-7. 1939b. On the anatomy and evolution of the locomotor apparatus of the Nipple-tailed Sunfish (Masturus lanceolatus).—Bulletin of the American Museum of Natural History 76(4): 143-150. Reuvens, C.L. 1897. Orthagoriscus nasus Ranz. on the Dutch Coast.— Notes of the Leyden Museum 18(26):209-21. Romer, A. S. 1966. Vertebrate Paleontology (3rd ed.). University of Chicago Press, Chicago, 465 pp. Schmidt, Johannes. 1921. New studies of sun-fishes made during the “Dana” expedition, 1920.— Nature (London) 107:76-79. Shattuck, G. B. 1904. Geological and paleontological relations, with a review of earlier investigations in Miocene.— Maryland Geological Survey, pp. 33-137. Townsend, C.H. 1918. The great ocean sunfish (Mola mola).— Bulletin of the New York Zoological Society 21(6):1677-1679. Tyler, James C. 1980. Osteology, phylogeny, and higher classification of the fishes of the order Plectognathi (Tetraodontiformes).— National Oceanic and Atmospheric Administration, Tech- nical Report, National Marine Fisheries Service, Circular 434:1—422. Van Beneden, P. J. 1881. Sur un poisson fossile nouveau des environs de Bruxelles et sur certains corps enigmatiques du crag d’Anvers.— Bulletin de l’ Academie Royale des Sciences, des Lettres et des Beaux-Arts de Belgique, 3° Ser. 1:124—-125. . 1883. Sur quelques formes nouvelles des terrains tertiares du pays.— Bulletin de l’ Academie Royale des Sciences, des Lettres et des Beaux-Arts de Belgique, 3° Ser. 6:132-133. Weems, R. E. 1974. Middle Miocene Sea Turtles (Sy/lomus, Procolpochelys, Psephophorus) from the Calvert Formation.—Journal of Paleontology 48(2):279-303. Whitley, G. P. 1933. Sunfishes.— Victorian Naturalist 49(9):207-—213. Yabe, H. 1950. Juvenile of the pointed-tailed ocean sunfish, Masturus lanceolatus.—Bulletin of the Japanese Society of Scientific Fisheries 16(1):40—42. Zimbelman, Edward. 1967. Mola mola go to the cleaners.—Sea Frontiers 13(2):120-122. United States Geological Survey, Mail Stop 928, Reston, Virginia 22092. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 439-446 SPEONEBALIA CANNONTI, N. GEN., N. SP., FROM THE CAICOS ISLANDS, THE FIRST HY POGEAN LEPTOSTRACAN (NEBALIACEA: NEBALIIDAE) Thomas E. Bowman, Jill Yager, and Thomas M. Iliffe Abstract. — Speonebalia cannoni is described from two marine caves on Prov- idenciales Island. It is characterized by the small rostrum, eyestalks without visual elements, mandible without incisor, maxilla 2 with very small exopod and uni- divided endopod, and broad caudal ramus with densely setose margins. A rich marine fauna has recently been found to inhabit marine caves in the Bahamas and associated islands. Cohen and Robins (1970) described a cavernic- olous brotulid fish from a limestone sink near Nassau on New Providence Island. Yager (1981) discovered Remipedia, a new class of Crustacea, from a marine cave on Grand Bahama Island. Carpenter (1981) identified a new genus of troglobitic cirolanid isopod from a seawater-filled cave on San Salvador Island. From the same cave, Lighthouse Cave, Carpenter and Magniez (1982) reported a new isopod genus in the Gnathostenetroidoidea; Van Soest and Sass (1981) found three new species of sponges, and Barr (1984) described a new demersal calanoid copepod. From a cave in the Turks and Caicos Islands, Stock and Vermuelen (1982) iden- tified a new genus of amphipod in the primarily abyssal family Pardaliscidae which also has a second species in the same genus inhabiting a lava tube cave on the Canary Islands. Buden and Felder (1977) have reported on the presence of cavernicolous shrimps in the Turks and Caicos. We describe herein a new genus and species of Nebaliacea (Phyllocarida: Leptostraca) collected during an expe- dition to marine caves in the Turks and Caicos Islands. This is the first known troglobite to be reported within the Leptostraca. Nebaliidae Baird 1850, emend. Hessler 1984 Speonebalia, new genus Diagnosis. — Blind, unpigmented. Carapace strongly compressed laterally, cov- ering thoracopods and pleopods 1—5. Rostrum small, shorter than eyestalks, with- out carina. Eyestalks rather narrow, tapering distally; margins smooth. Antenna 1 slightly shorter than antenna 2, about *4 length of carapace; last segment of peduncle without spines or teeth; scale narrow. Antenna 2 with 4-segmented peduncle. Mandible without incisor; palp long, distal segment tapering, with several lon- gitudinal rows of setae. Maxilla 2, distal (4th) endite of protopod well developed; endopod 1-segmented, with series of oval marginal organelles; exopod very small. Thoracopods all similar, like those of Nebalia, with broad exopods and epipods having plumose marginal setae. Pleopod 1 exopod without the usual dense row of short complex spines on lateral margin. 440 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Caudal rami short and broad, margins densely setose, setae on medial margin very long. Type-species.— Speonebalia cannoni, new species. Etymology. — From the Greek “‘speos,”’ cave, plus Nebalia. Speonebalia cannoni, new species Figs. 1-2 Material.—Turks and Caicos Islands, Caicos Islands, Providenciales Island: The Hole, 30 Oct 1982, leg. Thomas M. Iliffe, 11 paratypes, USNM 213480, from 5-7 m depths with suction bottle, using scuba. Airport Cave, 31 Oct 1982, leg. Thomas M. Iliffe, holotype, USNM 213478, and 14 paratypes, USNM 213479, from 10 m depths with suction bottle, free-diving. Description.— Length from anterior end of rostrum to posterior end of caudal ramus 8-11 mm (holotype 10.4 mm). Carapace oval, about 1.7 x as long as high, reaching posteriorly to pleonite 6, with angular posterodorsal corner below which posterior margin is armed with a series of close-set obtuse spines. Rostrum short, barely reaching beyond anterior margin of carapace, without keel, about ¥, as long as wide, narrowly rounded apically. Eyestalks narrowly pyriform, reaching beyond rostrum by about % of their lengths, completely lacking visual elements. Antenna | much shorter than carapace, flagellum 13—19-segmented; scale about 6x longer than broad, with cluster of apical setae, longest 1.5 x as long as scale, and row of setae on distal half of ventral surface. Antenna 2 slightly longer than antenna 1, segments 3 and 4 of peduncle not fused, armed with rows of long setae; flagellum 16—18-segmented, each segment with distomedial cluster containing long and short naked setae, seta with serrate section near midlength, and 1 or 2 esthetes (Fig. 1i, Table 1). Mandible with incisor absent or possibly represented by low triangular process distal to molar; molar well developed, with granular oval chewing surface; 1st segment of palp short, unarmed; 2nd segment long, with long seta on anterior margin distal to midlength; 3rd segment slender, slightly shorter than 2nd, with row of long setae rising on medial surface and extending anteromedially and 2 rows of short setae on posterior margin, medial row perpendicular to axis of segment, lateral row directed obliquely distad. Maxilla 1 proximal endite with 12 setae having shape resembling hypodermic syringe, 1 slender distal seta and 1 seta on surface; distal endite with 9 stout spines, some bifid apically, and 2 longer setae distally; palp with cluster of 10 setae at level of endites and about 13 well separated setae along rest of shaft. Maxilla 2 with 4 well-developed densely setose endites, proximal endite about 2 x as broad as subequal endites 2—4; endopod 1- segmented, about 3 x as long as wide, with 6-7 setae on medial margin, 2 at apex, and 1 on lateral margin distal to midlength; anterior margin with 8-9 papillae having central canal leading to pyriform glands within endopod; exopod very small, about 0.2 x as long as endopod, with 5-8 setae on anterior margin and apex. — Fig. 1. Speonebalia cannoni: a, Habitus, lateral; b, Anterior body and carapace, dorsal; c, Rostrum and left eyestalk, lateral; d, Eyestalks, dorsal; e, Right antenna 1, dorsal; f, Antenna 1, segment 9 of VOLUME 98, NUMBER 2 441 flagellum; g, Antenna | scale, ventral; h, Antenna 2 proximal segments, dorsal; i, Antenna 2 flagellum, setal cluster on segment 8; j, Labrum; k, Mandible; 1, Maxilla 1; m, Maxilla 1, distal endite; n, Maxilla 2; o, Maxilla 2 endopod, lateral, showing glands; p, Pleonite 4, left lateral. 442 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Composition of setal clusters on flagellum of antenna 2. Segment Short, naked Long, naked Serrate at midlength Esthete 1 Dy) 1 — 1 2 2 2 — 1 3 2 1 1 2 4 2 1 1 2) 5 yD) 1 1 1 6 D 1 1 2 7 2 1 1 1 8 D) 1 1 yD; 9 2 1 1 2 10 2 1 1 D9 11 2 1 1 2 12 2 1 1 2 13 D 1 1 1 14 2 1 1 1 15 Dy) 1 1 1 16—5 long terminal setae, 2 subterminal lateral setae Thoracopods all similar; endopod longer than exopod, curving laterad, with 4 segments distally, apical segment without marginal notches, medial margin dense- ly setose; exopod lamellate, lateral and apical margins sparsely setose; epipod well developed, lamellate, with proximal and distal lobes, lateral and proximal margins sparsely setose. Pleopods 1—4 with robust protopods having serrate posterior margins of lateral surface. Endopod longer and narrower than exopod, both margins setose, apex with long terminal spine lateral to which is shorter spiniform process; appendix interna with 2 retinacula. Exopod with setose medial margin; lateral margin armed with long spines and in pleopods 3 and 4 with plumose setae between each pair of spines except distal pair and sometimes proximal pair; apex with long spine. Pleopods 5 and 6 with setae on medial margin, apex, and distal part of lateral margin, those on medial margin shorter and more numerous in pleopod 5; apex with 2 long spines in pleopod 5, 3 in pleopod 6. Pleonites 4-7 with serrate posterior margins; pleonite 4 with posteroventral corner produced into point; tergum of pleonite 8 (anal segment) incised posteriorly for about '4 length; sternum incised more broadly, with rounded shoulders lat- erally. Caudal ramus slightly longer than pleonites 7 and 8 combined, slightly more than 3x as long as wide; lateral margin armed with progressively longer spines; medial margin densely armed with very long plumose setae; apical spine about half as long as ramus. We have been unable to determine the sex. Etymology.— Named for H. Graham Cannon (1897-1963) in recognition of his outstanding contributions to our knowledge of the Nebaliacea (1927, 1931, 1960). Comparisons. — Four genera of Nebaliidae are now recognized: Nebalia Leach, 1814; Paranebalia Claus, 1880; Nebaliella Thiele, 1904; Dahlella Hessler, 1984. The carapace of these genera is shorter than in Speonebalia, not reaching beyond pleonite 5. The rostrum is much larger than in Speonebalia, extending well beyond VOLUME 98, NUMBER 2 443 AZ Z, Y Z Za iM SN NS . A Fig. 2. Speonebalia cannoni: a, Thoracopod 1; b, Thoracopod 8; c, Left pleopod 1, anterior; d, Same, posterior; e, Left pleopod 2, anterior; f, Right pleopod 4, anterior; g, Same, lateral; h, Pleopod 5; i, Pleopod 6; j, Anal segment and uropod, dorsal; k, Posterior part of anal segment, ventral. 444 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. Entrance to The Hole, Providenciales Island, aerial view. Photograph by Dennis Williams, November 1983. the eyestalks in Nebalia, Paranebalia, and Dahlella, and the rostral carina present in Paranebalia and Nebaliella, is lacking in Speonebalia. Antenna 2 is much longer than antenna | in Nebalia, Paranebalia, and Ne- baliella, but only slightly longer in Dahlella and Speonebalia. The distal pedun- cular segment of antenna | has neither the armature of spines found in Nebalia nor the processes present in Paranebalia and Nebaliella. The mandibular incisor, well developed in Nebaliella, is weakly developed in Nebalia, Paranebalia, and Dahlella, and absent in Speonebalia. The mandibular palp is unusually large in Speonebalia, reaching the distal segment of the peduncle of antenna 2; the 3rd segment is unusual in its slender tapering shape and its complex setal armament. Maxilla 2 is the most distinctive appendage of Speonebalia. The 4th endite is well developed in Nebalia and Paranebalia as in Speonebalia, but reduced in Nebaliella and Dahlella. The 1-segmented endopod of Speonebalia is found in Nebaliella and Nebaliopsis; in Nebalia, Paranebalia, and Dahlella the endopod is 2-segmented. The glands in the endopod of Speonebalia have not been reported from any other Nebaliacean. Only in Dahlella is the exopod reduced as in Speo- nebalia. In Nebalia, Speonebalia, and Dahlella the thoracopods have well developed endopods, exopods, and epipods. The epipods are absent in Nebaliella and much reduced in Paranebalia. In all other genera of Nebaliidae the caudal rami are relatively slender. In Speonebalia the dense row of long setae along the medial margin presumably resists sinking and suggests a pelagic rather than a benthic life. VOLUME 98, NUMBER 2 445 Habitat.—The Turks and Caicos Islands are a group of islands lying southeast of the Bahamas and north of Haiti. The basic geological and geomorphological setting is generally similar to that of the Bahama Islands (Gregor 1981). Provi- denciales Island is located on the northern edge of the Caicos Bank. The main topographical feature of the island is a line of hills 20 to 40 m above sea level running parallel to the coastline. These hills are formed from eolian carbonates, probably of Pleistocene age. Both The Hole and Airport Cave are situated within these hills. The Hole is a shear-walled cenote-like pit about 15 m deep with a 15 m long by 10 m wide lake at the bottom, open to daylight (Fig. 3). This cave is located at the western end of Providenciales, 1.1 km from the nearest open water, the south coast. The 6 to 8 m deep lake is floored with breakdown and surface debris such that upon exploration of the lake with scuba, no human-sized cave passages were found extending off from it. Speonebalia specimens were collected from under a ledge in a dimly illuminated section of the lake. All specimens observed were swimming slowly about in the water column. Other species collected or observed in this lake include a new genus of ostracode in the family Thaumato- cyprididae now being described by Louis Kornicker, a new genus of amphipod being studied by John Holsinger, a crab identified as Sesarma (H.) miersii Rathbun by C. W. Hart, Jr. and an uncollected copepod. Airport Cave is located about 2.2 km inland from the north coast and 100 m north of the new airport terminal on the southern slope of Old Blue Hill. This cave was described by Buden and Felder (1977) as their collection site for the shrimp Barbouria cubensis. Airport Cave or one of several nearby caves may also be the nameless “‘waterhole”’ located northwest of the airfield which Stock and Vermeulen (1982) listed as the type-locality for the amphipod Spelaeonicippe provo. The cave consists of a nearly vertical collapse fissure in a large circular sink. A colony of bats, Erophylla sezekorni sezekorni (Gundlach), inhabits the cave, and their guano, dropping into the cave pool, has darkly stained the cave walls and sediments. The upper layer of the pool is anoxic as evidenced by a distinct hydrogen sulphide odor that was noted when the surface waters were disturbed. The visibility in the upper layers of the pool was somewhat reduced, but below about 2 to 3 m the water was very clear. Speonebalia specimens were observed to stay in loosely grouped clusters of up to 50 individuals. They were only found in the deeper waters of the pool in total darkness. Other animals observed in the pool include a new family of shrimps now being described by C. W. Hart, Jr. and Raymond Manning, and an amphipod of the same new genus as was found in The Hole. Several individuals of a new species of Remipedia were collected by Jill Yager from Airport Cave during a later visit in April 1983. Contrary to Buden and Felder’s (1977) observations, no Barbouria cubensis were observed, nor did we find any Spelaeonicippe provo that Stock and Vermeulen (1982) had collected here. During the April 1983 expedition, water salinity in Airport Cave was measured, with the following results: surface, 19.1%o0; 3 m, 19.1%0; 5 m, 26.5%o; 7 m, 28.7%o. Temperature was 23.1°C at the surface and 25.6°C at 5 m. Acknowledgments Collection of specimens from caves in the Turks and Caicos Islands was sup- ported in part by a National Science Foundation Grant (BSR 8215672) to Thomas 446 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON M. Iliffe. We especially thank Paul and Shirley Hobbs for arranging accommo- dations and helping us with cave location and collections, and Dennis Williams for flying us to various islands as well as assisting with field collections. Field expenses for the April 1983 expedition were supported by NSF Grant BSR 8212335. Howard Cosgrove assisted in field collecting and water analysis. We are grateful to Erik Dahl and Robert R. Hessler for helpful comments on the manuscript. All research divers were cave divers certified by the National Speleological Cave Diving Section. This paper is contribution No. 990 from the Bermuda Biological Station for Research, Inc. Literature Cited Barr, D. J. 1984. Enantiosis cavernicola, a new genus and species of demersal copepod (Calanoida: Epacteriscidae) from San Salvador Island, Bahamas.— Proceedings of the Biological Society of Washington 97(1):160-166. Buden, D. W., and D. L. Felder. 1977. Cave shrimps in the Caicos Islands.—Proceedings of the Biological Society of Wasington 90(1):108-115. Cannon, H.G. 1927. On the feeding mechanism of Nebalia bipes.— Transactions of the Royal Society of Edinburgh 55:355-369. . 1931. Nebaliacea.—Discovery Reports 3:199-—222. —. 1960. Leptostraca.—Bronn’s Klassen and Ordungen des Tierreichs, Bd. 5, Abt. 1, Buch 4, Teil 1:1-81. Carpenter, J. H. 1981. Bahalana geracei n. gen., n. sp., a troglobitic marine cirolanid isopod from Lighthouse Cave, San Salvadore Island, Bahamas. — Bijdragen tot de Dierkunde 5 1(2):259-267. , and G. J. Magniez. 1982. Deux asellotes stygobies des Indies Occidentales: Neostenetroides stocki n. gen., n. sp., et Stenetrium sp.—Bijdragen tot de Dierkunde 52(2):200-206. Claus, C. 1880. Grundziige der Zoologie, 4th edition, volume 1. VII + 822 pp, Marburg. Cohen, D. M., and C. R. Robins. 1970. A new ophidioid fish (genus Lucifuga) from a limestone sink, New Providence Island, Bahamas.— Proceedings of the Biological Society of Wahsington 83(11):133-144. Gregor, V. A. 1981. Karst and caves in the Turks and Caicos Islands, B.W.I.— Proceedings of the Eighth International Congress of Speleology, Bowling Green, KY, U.S.A., pp. 805-807. Hessler, R. R. 1984. Dahlella caldariensis n. gen., n. sp.: Leptostracan (Crustacea, Malacostraca) from deep-sea hydrothermal vents.—Journal of Crustacean Biology 4(4):655-664. lliffe, T. M., C. W. Hart, Jr., and R. B. Manning. 1983. Biogeography and the caves of Bermuda.— Nature 302(5904):141-142. Leach, W. L. 1814. Nebalia.—Zoological Miscellany 1:99. Stock, J. H., and J. J. Vermeulen. 1982. A representative of the mainly abyssal family Pardaliscidae (Crustacea, Amphipoda) in cave waters of the Caicos Islands.—Bijdragen tot de Dierkunde 52(1):3-12. Thiele, J. 1904. Die Leptostraken.— Wissenschaftliche Ergebnisse der deutschen Tiefsee-Expedition auf dem Dampfer “Valdivia” 1898-1899 8:1—26, pls. 1-4. Van Soest, R. W. M., and D. B. Sass. 1981. Amsterdam Expedition to the West Indian Islands, Report 13: Marine sponges from an island cave on San Salvadore Island, Bahamas. — Bijdragen tot de Dierkunde 52(2):332-344. Yager, J. 1981. Remipedia, a new class of Crustacea from a marine cave in the Bahamas. — Journal of Crustacean Biology 1(3):328-333. (TEB) Department of Invertebrate Zoology, National Museum of Natural His- tory, Smithsonian Institution, Washington, D.C. 20560; (SY) Department of Bi- ological Sciences, Old Dominion University, Norfolk, Virginia 23508; (TMI) Bermuda Biological Station for Research, Ferry Reach 1-15, Bermuda. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 447-469 ADDITIONAL BRANCHIATE SCALE-WORMS (POLYCHAETA: POLYNOIDAE) FROM GALAPAGOS HYDROTHERMAL VENT AND RIFT-AREA OFF WESTERN MEXICO AT 21°N Marian H. Pettibone Abstract.— A new subfamily of Polynoidae, Branchinotogluminae, is established for Branchinotogluma, new genus, for three new species: B. hessleri, B. sandersi, and B. grasslei, and Opisthotrochopodus alvinus, new genus, new species, all from hydrothermal vents in the eastern Pacific. They are characterized by the presence of well-developed arborescent brachiae and notopodial bracts on some or all of the elytrigerous segments. The posterior segments of O. alvinus are greatly mod- ified, having most unusual wheel organs. The present paper deals with the third subfamily of Polynoidae from the hy- drothermal rift-area off Western Mexico at 21°N and the Galapagos hydrothermal vent that is characterized by the presence of the well-developed branchiae, an unusual feature in the family. The two previously described branchiate groups include Branchipolynoe symmytilida in Branchipolynoinae, commensal with the deep-sea vent mussels in the Galapagos Rift (Pettibone 1984a) and Branchiplicatus cupreus in Branchiplicatinae from the hydrothermal rift-area at 21°N (Pettibone, 1985). Two new genera and four new species, collected in the same two areas, are referred herein to a third new subfamily. The available material was collected by the Galapagos Rift Biology Expedition in 1979, and by the OASIS group of Scripps Institution of Oceanography in 1982. The specimens were received from J. F. Grassle of the Woods Hole Oceanographic Institute (WHOJ), with preliminary sorting by I. Williams. Numerous specimens, including many young, were collected in the Mussel Bed, Garden of Eden, and Rose Garden areas of the Galapagos Rift, in 2451 to 2493 meters, from washings, slurp samples, and residues of mussels and vestimentiferans (Riftia pachyptila Jones). From 21°N, specimens were collected in 2612 to 2633 meters, associated with washings and slurp samples of vestimentiferans (Riftia pachytila Jones), giant clams (Calyptogena magnifica Boss and Turner) and ampharetid polychaetes (A/- vinella pompejana Desbruyéres and Laubier). The types and additional specimens are deposited in the National Museum of Natural History, Smithsonian Institution (USNM). This is OASIS Expedition Contribution number 26 and Galapagos Rift Biology Expedition number 54, supported by the National Science Foundation. Family Polynoidae Branchinotogluminae, new subfamily General Characteristics The body is short, flattened, slightly tapering anteriorly and posteriorly, with 21 segments, the first achaetous. The parapodia are longer than the body width. 448 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON The posterior few segments may be unmodified (B. hessleri, B. grasslei; Figs. 1C, D, 5E, F), the posterior 3 segments may be somewhat modified (B. sandersi; Fig. 4B-—H) or the posterior 4 segments may be greatly modified, including wheel organs (O. alvinus; Fig. 9A—-N). There are 10 pairs of elytra attached on prominent bulbous elytrophores on segments 2, 4, 5, 7, 9, 11, 13, 15, 17, and 19. The elytra are large, overlapping and cover the dorsum. They are oval, delicate, usually with branched “veins” arising from their place of attachment to the elytrophores; tubercles and papillae are lacking (Fig. 5G). Dorsal cirri are borne on segments 3, 6, 8, 10, 12, 14, 16, 18, 20, and 21. The cirrophores are short, cylindrical, attached posteriorly on the notopodia; the styles are long, filiform, smooth, extending to the tips of the neurosetae or far beyond (Fig. 2D). Dorsal tubercles on the cirrigerous segments, in line with the elytrophores, are inflated and indistinct (B. grasslei, B. sandersi) or elongate and tapered (B. hessleri, O. alvinus; Figs. 1B, C, 7C, E). The dorsum has transverse ciliated bands, 2 per segment, extending onto the elytrophores and dorsal tubercles Fig. 1A—C). Compact arborescent branchiae, with short (B. sandersi, B. grasslei; Figs. 3E, 6C, D) or long (B. hessleri, O. alvinus; Figs. 2D; 7E) terminal filaments, are present as 2 main groups attached to the lateral sides of the elytrophores or dorsal tubercles and on the dorsal sides of the notopodia beginning on segment 3 (Fig. 2B—D). The branchiae continue to the posterior end (B. hessleri; B. grasslei; Figs. 1C, 5D), except for the species with modified posterior ends where the branchiae are absent on the posterior 3 (B. sandersi; Fig. 4B) or 4 (O. alvinus; Fig. 9A) modified segments. The prostomium is oval, bilobed, with triangular or cylindrical anterior lobes with delicate frontal filaments (Figs. 1A, 3A). A median antenna with a bulbous ceratophore is inserted in the anterior notch. Paired palps are stout, smooth and tapered. Lateral antennae and eyes are absent. The first or tentacular segment is not visible dorsally. Tentaculophores, lateral to the prostomium, are achaetous, each with a pair of long dorsal and ventral tentacular cirri (Fig. 1A). Ventrally the tentacular segments contributes to the anterior and lateral lips of the mouth (Fig. SC). The second or buccal segment bears the first pair of elytrophores, elytra, biramous parapodia, and paired ventral or buccal cirri with short cirrophores inserted basally on the neuropodia; their styles are longer than the following ventral cirri, similar to the tentacular cirri (Figs. 5B, C, 7A, B). The buccal segment contributes to the posterior lip of the ventral mouth. The muscular pharynx is eversible, with 5 papillae around the opening: 3 dorsal and 2 ventral (Figs. 1F, 8H). The two pairs of hooked jaws are minutely denticled along the border (Figs. 1F, 5H, 8H, I). The parapodia are biramous, with the notopodia shorter than neuropodia, both rami having projecting acicular processes (Fig. 2B—D). Prominent notopodial bracts enclosing the acicular lobes and notosetae are present only on elytrigerous segment 2 (B. sandersi; B. grasslei, Figs. 3A, B, 5B, 6A) or on all the elytrigerous segments (B. hessleri, O. alvinus; Figs. 1A—C, 7A, B, D). The notosetae are stouter than the neurosetae, straight, acicular, smooth (Fig. 6E) or with scattered spines (Fig. 1E). The neuropodia have longer flattened conical presetal lobes with pro- jecting acicular processes and shorter, rounded postsetal lobes. The neurosetae are slender, finely spinous, with slightly hooked tips (Fig. 3G, H). The ventral VOLUME 98, NUMBER 2 449 cirri are short, tapered, and attached on the middle of the neuropodia (Fig. 2B— D). The pygidium is enclosed in the parapodia of the posterior few segments, with or without a pair of anal cirri (Fig. 1C, D, 4C). Ventral segmental papillae may be absent or variously developed, such as a pair of elongated papillae on segment 12, followed by 5 pairs of short rounded lamellae on segments 13-17 (O. alvinus; Fig. 9B); 4 pairs of long papillae on segments 12-15 and 3 pairs of rounded lamellae on segments 16-18 (H. sandersi; Fig. 41); 6 pairs of small rounded lamellae on segments 11-16 (B. hessleri; Fig. 2H) or 5 pairs of small squarish papillae on segments 11-15 (B. grasslei; Fig. 6H). Key to the Genera and Species of Branchinotogluminae 1. Posterior 4 segments (18-21) compressed, with parapodia greatly modi- fied, including wheel organs on segment 20 (Fig. 9N). Opisthotrochopodus, new genus. Notopodia with prominent rounded bracts on elytrigerous segments 2-17 (Figs. 7A, D, 8A, C). Arborescent branchiae on segments 3-17, with rather long terminal filaments (Figs. 7A, C-E, 8A—C). Dorsal tubercles elongate, tapered (Figs. 7C, 8A). Notosetae stout, acicular, smooth or with 1—2 rows of spines (Fig. 7F). With pair of elongated ventral papillae on segment 12 and 5 pairs of short rounded lamellae on segments 13-17 (TEINS 2) B) i ORR NARI rt te) Sem alan a Moa O. alvinus, new species — Posterior 4 segments not compressed or only 3 posterior segments (19- 21) slightly modified, without wheel organs on segment 20 ........... PE a Branchinotogluma, new genus ................. 2 2. Prominent notopodial bracts on all elytrigerous segments (Figs. 1 A—C, 2A, C). Arborescent branchiae with rather long terminal filaments (Fig. 2D). Posterior segments not modified, with branchiae on all segments from segment 3 on (Fig. 1A, C). Dorsal tubercles projecting posteriorly, with branchiae on lateral side, ciliated on posterior side (Fig. 1B). Notosetae stout, acicular, smooth or with 2 rows of spines (Fig. 2E). Ventral segmental lamellae 6 pairs, on segments 11-16, small rounded (Fig. 2H) ......... hte Be INE IAL ae GR eA) BLA O LES ieee oth, B. hessleri, new species — Notopodial bracts on segment 2 only (Figs. 3A, B, SA, B, 6A). Arborescent branchiae with short terminal filaments (Figs. 3E, 6D). Dorsal tubercles inflated, indistinct. Notosetae stout, acicular, smooth (Figs. 3F, 6E).... 3 3. Posterior 3 segments slightly compressed and modified, with branchiae absent from segments 19—21 (Fig. 4B—H). With 4 pairs of elongated ventral segmental papillae on segments 12-15 (Fig. 4I) and 3 pairs of rounded lamellacionysesmentsslG=l'S 4. 5.. 2546s soe eee B. sandersi, new species — Posterior segments not compressed or modified, with branchiae on all segments from segment 3 on (Fig. 5B, E, F). Without elongated ventral segmental papillae; with or without 5 pairs of small, squarish segmental papillae on segments 11-15 (Fig. 6H) .............. B. grasslei, new species Branchinotogluma, new genus Type-species.— Branchinotogluma hessleri, new species. Gender: feminine. Diagnosis.— Body short, with 21 segments, first achaetous. Elytra and elytro- 450 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON phores 10 pairs, on segments, 2, 4, 5, 7, 9, 11, 13, 15, 17, and 19. Dorsal cirri with short cylindrical cirrophores and long distal styles, and dorsal tubercles, in line with elytrophores, on segments 3, 6, 8, 10, 12, 14, 16, 18, 20, and 21. Arborescent branchiae attached to lateral sides of elytrophores, dorsal tubercles, and dorsal sides of notopodia beginning on segment 3 and continuing to near or end of body. Prostomium bilobed, with triangular anterior lobes bearing minute frontal filaments, with median antenna in anterior notch and paired palps; without lateral antennae or eyes. First or tentacular segment not visible dorsally; tenta- culophores lateral to prostomium, each with pair of dorsal and ventral tentacular cirri, without setae. Second or buccal segment with first pair of elytra, biramous parapodia and ventral or buccal cirri attached to basal parts of parapodia lateral to ventral mouth; styles longer than following ventral cirri. Parapodia biramous, with notopodia shorter than neuropodia. Notopodia of elytrigerous segments with prominent rounded bracts enclosing flattened conical acicular lobes and notosetae (B. hessleri) or present only on elytrigerous segment 2 (B. sandersi, B. grasslei). Neuropodia with longer flattened conical presetal and shorter rounded postsetal lobes. Notosetae stouter than neurosetae, straight, acicular, smooth or with few spines. Neurosetae long, slender, finely spinous, with slightly hooked tips. Ventral cirri short, tapered. Posterior 4 segments not greatly modified, without wheel organs or posterior 3 segments slightly modified (B. sandersi). Pharynx with 5 papillae around opening, 3 dorsal and 2 ventral; 2 pairs of dorsal and ventral hooked jaws each with row of numerous minute teeth along edge. Etymology.—Branchi from the Greek branchia for gills, plus noto from notos, Greek for back, and g/uma from the Latin g/uma, a bract, referring to the presence of branchiae and notopodial bracts. Branchinotogluma hessleri, new species Figs. 1, 2 Material examined.—Pacific Ocean off Western Mexico, 20°50'N, 109°06’W, OASIS Alvin dives in 1982: Dive 1214, 20 Apr, 2633 m, vestimentiferan wash, 19 paratypes including 6 young (USNM 97323). Dive 1218-15, 24 Apr, 2618 m, clam and crab trap wash, paratype (USNM 97324). Dive 1219-10A & B, 25 Apr, 2612 m, Riftia and clam wash, paratype (USNM 97325). Dive 1221-15, 4 May, 2618 m, Riftia and Clayptogena wash, coarse fraction, holotype (USNM 97321) and 7 paratypes (USNM 97322). Dive 1223-11, 7 May, 2616 m, Riftia and Calyptogena wash, coarse fraction, 4 paratype (USNM 97326). East Central Pacific, from dive of the A/vin along the Galapagos Rift in 1979: ROSE GARDEN, 00°48'15”N, 86°13'28”W: Dive 984-32, 1 Dec, 2451 m, mussel washings, 3 paratypes (USNM 97327). Description.— Length of holotype 14 mm, width 8 mm with setae, segments 21. Body flattened, tapering slightly anteriorly and posteriorly, with parapodia longer than body width. No color except for yellow amber-colored setae. Dorsum with transverse ciliated bands, 2 per segment, extending onto elytrophores and dorsal tubercles (Fig. 1 A—C). Elytra and prominent bulbous elytrophores 10 pairs (Figs. 1 A—C; 2A, C). Elytra large, covering dorsum, round to oval, delicate, with branched “‘veins”’ (Fig. 1E). Dorsal cirri with short cylindrical cirrophores and long tapered smooth styles extending beyond tips of setae (Figs. 1A—D, 2B, D). Dorsal tubercles elongate. VOLUME 98, NUMBER 2 451 D Fig. 1. Branchinotogluma hessleri, A-E, Holotype, USNM 97321; F, Paratype, USNM 97323: A, Dorsal view of anterior end, style of left dorsal tentacular cirrus missing; B, Dorsal view of left side of segments 8 (cirrigerous) and 9 (elytrigerous); C, Dorsal view of posterior end, including segments 19-21; D, Ventral view of posterior end, including segments 20 and 21; E, Right elytron from near posterior end; F, End view of opened pharynx, tipped slightly to left. b, notopodial bract; dT, dorsal tubercle. Scales = 1.0 mm for A; 1.0 mm for B—D, F; 1.0 mm for E. Both elytrophores and dorsal tubercles directed posterolaterally, with group of delicate arborescent branchiae attached on their lateral sides and additional small group of branchiae attached on bases of notopodia; branchiae compact, with rather long terminal filaments (Figs. 1A—C, 2B—D). Branchiae beginning on segment 3 as 2 small groups (Fig. 2B), becoming larger on following segments (Fig. 2C, D) and continuing to posterior end (Fig. 1C, D). Prostomium bilobed, oval, anterior lobes triangular with delicate frontal fila- ments; median antenna with bulbous ceratophore in anterior notch, with style about as long as prostomium; palps stout, tapered, smooth; without eyes (pair of shaded areas sometimes appearing as “eyes”; Fig. 1A). Tentaculophores of first segment lateral to prostomium, with 2 pairs of tentacular cirri similar in length and slightly shorter than palps; ventrally tentacular segment forming anterior and lateral lips of mouth. Second or buccal segment forming posterior lip of mouth 452 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON —_ LACM pang) ete ds ZZ ZS J, ‘ E tl d Fig. 2. Branchinotogluma hessleri, holotype, USNM 97321: A, Right elytrigerous parapodium of segment 2, anterior view, with detail of neuropodial border papillae; B, Right cirrigerous parapodium of segment 3, posterior view; C, Right elytrigerous parapodium of segment 9, anterior view, acicula dotted; D, Right cirrigerous parapodium of segment 10, posterior view with detail of branchial tips; E, Three notosetae; F, Upper neuroseta, with detail of tip; G, Middle neuroseta, with detail of tips; H, Ventral view of right side of segments 13-16, showing ventral segmental lamellae. b, notopodial bract. Scales = 0.5 mm for A—D; 0.1 mm for E-—G; 1.0 mm for H. and bearing biramous parapodia and first pair of elytrophores; notopodium with hood or bract encircling small bundle of notosetae; conical neuropodium with globular micropapillae on distal border; ventral buccal cirri similar to tentacular cirri, longer than following ventral cirri (Figs. 1A, 2A). Everted muscular pharynx showing 5 papillae around opening: 3 dorsal and 2 ventral; 2 pairs of prominent jaws, minutely denticled on inner border (Fig. 1F). VOLUME 98, NUMBER 2 453 Biramous parapodia with notopodia shorter than neuropodia, both rami with light yellow amber-colored setae (Fig. 2A—D). Notopodia with short conical acic- ular lobes and, on elytrigerous segments, with prominent bracts enclosing noto- setae dorsally and posteriorly (Figs. 1A—C, 2A, C). Notosetae forming radiating bundles, short to long, some almost as long as neurosetae, acicular, much stouter than neurosetae, smooth or with 1-2 rows of short spines (Fig. 2E). Larger neu- ropodia with subconical presetal lobes tapering to sharp tips; postsetal lobes short- er, rounded; some globular micropapillae on distal borders. Neurosetae slender, very numerous, forming fan-shaped bundles. Few upper neurosetae with more prominent spines, with tips flattened and finely spinous (Fig. 2F). Remaining neurosetae with borders finely spinous; tips slightly hooked, with finely spinous hood (Fig. 2G). Lower neurosetae shorter and more slender. Ventral cirri short, tapered, attached on middle of neuropodia (Fig. 2C, D). Pygidium consisting of small squarish to rounded lobe wedged between bases of posteriorly-directed parapodia of posterior 2 segments, with pair of anal cirri (mostly broken off; Fig. 1C, D). Six pairs of small rounded segmental lamellae near ventral bases of neuropodia of segments 11-16 (Fig. 2H); only 2-3 pairs on segments 11-13 of smaller specimens. Young specimens with 17 or fewer segments may be confused with Opistho- trochopodus alvinus. See below under Branchinotogluma sp. A. Etymology.—The species is named for Robert R. Hessler, one of the observers on the OASIS Alvin dives. Branchinotogluma sandersi, new species Figs. 3, 4 Material examined.—East Central Pacific, from dives of the Alvin along the Galapagos Rift in 1979: MUSSEL BED, 00°47'53’”N, 86°09'12”W: Dive 880, 21 Jan, 2493 m, mussel washings, paratype (USNM 97330). Dive 989, 6 Dec, 2482 m, bottle rack wash no. 1, paratype (USNM 97331). GARDEN OF EDEN, 00°47'41°N, 86°07'44"W: Dive 883, 25 Jan, 2493 m, slurp sample in mussel area, paratype (USNM 97332). Dive 884, 25 Jan, 2482 m, residue from vestimentiferan tubes, young paratype (USNM 97333). ROSE GARDEN, 00°48'15’N, 86°13'28”W: Dive 983-112, 30 Nov, 2457 m, washings, paratype (USNM 97329). Dive 990- 41, 7 Dec, 2451 m, vestimentiferan wash, holotype (USNM 97328). Pacific Ocean off Western Mexico, 20°50'N, 109°06’W, OASIS Alvin dives in 1982: Dive 1214, 20 Apr, 2633 m, vestimentiferan wash, 6 paratypes including 3 young (USNM 97334). Dive 1219-1B, 25 Apr, 2612 m, slurp sample in Riftia habitat, 3 paratypes (USNM 97335). Dive 1221-15, 4 May, 2618 m, Riftia and Calyptogena wash, coarse fraction, 6 paratypes (USNM 97336). Description. — Length of holotype from Galapagos Rift (USNM 97328) 26 mm, width 13 mm with setae, segments 21. Length of largest paratype from 21°N (Alvin dive 1214; USNM 97334) 16 mm, width 6 mm, segments 21. Body rectangular, flattened, tapering slightly anteriorly and posteriorly, with parapodia longer than body width. Posterior 3 segments (19-21) compressed, with parapodia modified, differing from more anterior parapodia as well as from one another (Fig. 4B—H). No color except for golden-colored setae. Dorsum with transverse ciliated bands, 2 per segment, extending onto elytrophores and dorsal tubercles (Fig. 4A). PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 454 , y | | | | HK | Fig. 3. Branchinotogluma sandersi, holotype, USNM 97328: A, Anterior end, dorsal view; B, Right elytrigerous parapodium of segment 2, anterior view, acicula dotted; C, Right cirrigerous parapodium of segment 3, posterior view; D, Right cirrigerous parapodium of segment 8, posterior view; E, Right elytrigerous parapodium of segment 9, anterior view, acicula dotted, with detail of branchial tips; F, Four notosetae; G, Supraacicular neuroseta, with detail of tip; H, Subacicular notoseta, with detail of tips; I, Two right elytra. b, notopodial bract. Scales = 1.0 mm for A; 1.0 mm for B-E; 0.1 mm for F— H; 2.0 mm for I. VOLUME 98, NUMBER 2 455 Elytra and prominent bulbous elytrophores number 10 pairs (Figs. 3A, B, E, 4A, B, D). Elytra large, covering dorsum, round to oval, stiff, opaque (Fig. 31). Dorsal cirri with rather long cylindrical cirrophores and long tapered smooth styles extending beyond tips of setae (Figs. 3C, D, 4A, B). Dorsal tubercles inflated, indistinct. Both elytrophores and dorsal tubercles with delicate arborescent bran- chiae attached on their lateral sides, with additional smaller groups of branchiae attached on dorsal bases of notopodia. Branchiae short, compact, with numerous short, curled branches and short terminal filaments (Figs. 3C—E, 4A, B). Branchiae beginning on segment 3 as single small group (Fig. 3C), becoming larger more posteriorly and continuing to segment 18 (Fig. 4B). Prostomium bilobed, anterior lobes prominent, cylindrical, with terminal fil- aments; median antenna with cylindrical ceratophore in anterior notch, with style short, subulate, with long terminal filament; palps stout, tapered, smooth; without eyes (pair of shaded areas sometimes appearing as eyes; Fig. 3A). Tentaculophores with 2 pairs of tentacular cirri, dorsal ones about length of palps, ventral ones slightly shorter; ventrally tentacular segment forming anterior and lateral lips of mouth. Second or buccal segment forming posterior lip of mouth and bearing biramous parapodia and first pair of elytrophores; notopodia with rounded thick- ened bract on anterodorsal side of large bundie of notosetae; ventral buccal cirri similar to tentacular cirri, longer than following ventral cirri (Fig. 3A, B). When everted, muscular pharynx showing 5 papillae around opening: 3 dorsal and 2 ventral; 2 pairs of prominent jaws minutely denticled on inner border. Parapodia of segments 2 to 18 biramous, with notopodia shorter than neuro- podia, both rami with golden-colored setae (Fig. 3A—E). Notopodia conical with projecting acicular process, without bracts except on segment 2. Notosetae nu- merous, short to long, forming radiating bundle, much stouter than neurosetae, tapered, straight, smooth, acicular (Fig. 3F). Neuropodia with long conical presetal acicular lobes; postsetal lobes shorter, rounded. Neurosetae very numerous, slen- der, forming fan-shaped bundles. Supraacicular neurosetae with 2 rows of more prominent spines; distal part finely spinous with tip slightly hooked (Fig. 3G). Subacicular neurosetae decreasing in length ventrally, finely spinous up to hooked tip, with flattened finely spinous hood (Fig. 3H). Ventral cirri short, tapered, attached on middle of neuropodia (Fig. 3D, E). Parapodia of posterior 3 segments modified, directed posteriorly and enclosed in parapodia of cirrigerous segment 18, all with ventral cirri but with branchiae lacking (Fig. 4B, C). Parapodia of elytrigerous segment 19 biramous,with rami similar in length; notopodia with thickened dorsal bract (Fig. 4B, D). Parapodia of segment 20 much smaller and enclosed in parapodia of segment 19 (Fig. 4B, C, E). Notopodial acicular lobe fused to cirrophore of dorsal cirrus, with short distal style; notosetae few (about 9), short, stout, curved, with 2 rows of spines on distal part (Fig. 4F). Neuropodia subconcial, rounded distally, with small bundle of neurosetae: upper ones stouter with scattered long spines and rounded tips; few lower ones slender, capillary, with single basal spine (Fig. 4G). Parapodia of segment 21 (Fig. 4B, C, H) with notopodia similar to that of 20, formed of fused acicular lobe and dorsal cirrophore, with very short distal style and with notosetae lacking. Neuropodia formed of short conical acicular lobe with small group of delicate neurosetae. Pygidium squarish, wedged between parapodia of segments 19 to 21, with pair 456 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 4. Branchinotogluma sandersi, holotype, USNM 97328: A, Dorsal view of left parapodia of segments 8 (cirrigerous) and 9 (elytrigerous); B, Dorsal view of posterior end including segments 17— 21; C, Same, ventral view; D, Right elytrigerous parapodium of segment 19, posterior view; E, Right cirrigerous parapodium of segment 20, anterior view, acicula dotted; F, Notoseta from same, with detail of part; G, Upper and lower neurosetae from same; H, Right cirrigerous parapodium of segment 21, anterior view, acicula dotted; I, Ventral view of left side of segments 1 1-14, showing long segmental papillae; aC, anal cirrus; b, notopodial bract. Scales = 1.0 mm for A-C, I; 0.5 mm for D, E, H; 0.1 mm for F, G. VOLUME 98, NUMBER 2 457 of long anal cirri (Fig. 4C). Four pairs of long segmental ventral papillae on segments 12 to 15, becoming shorter posteriorly, some showing “secretion”? em- anating from papillar opening (Fig. 41). Three pairs of rounded ventral lamellae on following segments 16 to 18. On juveniles, small developing papillae on seg- ment 12 to 13 or 14. Young specimens with 17 or fewer segments may be confused with Branchi- notogluma grasslei. See below under Branchinotogluma sp. B. Etymology.— The species is named for Howard L. Sanders, one of the observers on the OASIS Alvin dives. Branchinotogluma grasslei, new species Figs. 5, 6 Material examined.— Pacific Ocean off Western Mexico, 20°50’N, 109°06'W, OASIS Alvin dives in 1982: Dive 1214, 20 Apr, 2633 m, vestimentiferan wash, 7 paratypes (USNM 97317). Dive 1215-6A, 21 Apr, 2616 m, slurp sample, para- type (USNM 97318). Dive 1219-1A, 10A & B, 25 Apr, 2612 m, slurp sample in Riftia habitat, Riftia and clam wash, 4 paratypes (USNM 97319). Dive 1221, 4 May, 2618 m Riftia and Calyptogena wash, coarse fraction, holotype (USNM 97314) and 11 paratypes (USNM 97315, 97316). East Central Pacific, from dive of the A/vin along the Galapagos Rift in 1979: MUSSEL BED, 00°47'53”N, 86°09'12”W: Dive 989, 6 Dec, 2482 m, bottle rack wash no. 1, paratype (USNM 97320). Description. — Length of holotype from 21°N (USNM 97314) 19 mm, width 11 mm with setae, segments 21. Length of large paratype from Galapagos Rift (USNM 97320) 31 mm, width 20 mm, segments 21. Body flattened, tapering slightly anteriorly and posteriorly, with parapodia longer than body width. No color except for golden- or bronze-colored setae. Dorsum with transverse ciliated bands, 2 per segment, extending onto elytrophores and dorsal tubercles (Fig. 5B, D). Elytra and prominent bulbous elytrophores number 10 pairs (Figs. 5A, B, D, E, 6C). Elytra large, covering dorsum, round to oval, delicate, with branched “veins” (Fig. 5G). Dorsal cirri with cylindrical cirrophores and long tapering smooth styles extending to about tips of setae or beyond (Figs. 5B, D, E, 6B). Dorsal tubercles inflated, indistinct. Both elytrophores and dorsal tubercles with group of delicate arborescent branchiae attached on their lateral sides and addi- tional smaller group of branchiae on bases of notopodia; branchiae short, compact, with numerous short, curled branches (Figs. 5D, E, 6B—D). Branchiae beginning on segment 3 as 2 small groups (Fig. 5B), becoming larger on following segments (Fig. 6B—D) and continuing to posterior end (Fig. SE). Bilobed prostomium with prominent cylindrical anterior lobes with terminal filaments; median antenna with bulbous ceratophore in anterior notch, with short tapered style; palps stout, tapered, smooth; without eyes (shaded areas may appear as pair of eyes; Fig. 5A, B). Tentaculophores with 2 pairs of tentacular cirri, dorsal pairs longer than palps, ventral pairs slightly shorter; ventrally tentacular segment forming anterior and lateral lips of mouth (Fig. SA—C). Second or buccal segment forming posterior lip of mouth and bearing biramous parapodia and first pair of elytrophores; notopodia with kood or bract encircling bundle of notosetae; ventral buccal cirri similar to tentacular cirri, longer than following ventral cirri (Figs. 5A-C, 6A). 458 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 5. Branchinotogluma grasslei, A, paratype, USNM 97320; B-G, holotype, USNM 97314; H, paratype, USNM 97315: A, Anterior end, dorsal view, pharynx partially extended; left palp missing; B, Anterior end, dorsal view, right ventral, left dorsal and ventral tentacular cirri missing; C, Same, ventral view; D, Dorsal view left parapodium of segments 10 (cirrigerous) and 11 (elytrigerous); E, Dorsal view posterior end, including segments 19-21, left dorsal cirrus of segment 20 short, regen- erating; F, Ventral view posterior end, including segments 20 and 21; G, Four elytra; H, Dorsal jaws spread apart, with detail of part. b, notopodial bract. Scales = 2.0 mm for A; 1.0 mm for B—F; 2.0 mm for G; 0.5 mm for H. VOLUME 98, NUMBER 2 459 Everted muscular pharynx showing 5 papillae around opening: 3 dorsal and 2 ventral; 2 pairs of prominent jaws, minutely denticled on inner border (Fig. 5H). Biramous parapodia with notopodia shorter than neuropodia, both rami with golden- or bronze-colored setae (Fig. 6A—C). Notopodia conical, with prominent projecting acicular process; notopodia with notopodial bract on segment 2 only (Fig. 5A). Notosetae numerous, short to long, not as long as neurosetae, tapered, straight, acicular, smooth, much stouter than neurosetae (Fig. 6E). Larger neu- ropodia with conical presetal lobes tapering to sharp tips, diagonally truncate dorsally, rounded on ventral half; postsetal lobes shorter, rounded. Neurosetae slender, very numerous, forming fan-shaped bundle. Supraacicular neurosetae with 2 rows of prominent spines, with tips flattened and finely spinous (Fig. 6F). Subacicular neurosetae finely spinous along border, with tips slightly hooked and with finely spinous hood (Fig. 6G). Few lower neurosetae with capillary tips. Ventral cirri short, tapering, attached on middle of neuropodia (Fig. 6B, C). Pygidium consisting of small squarish lobe wedged between posteriorly-directed parapodia of posterior 2 segments (20, 21), with pair of anal cirri (all missing; Fig. SE, F). With or without 5 pairs of small squarish segmental ventral papillae near ventral bases of neuropodia on segments 11-15 (Fig. 6H). Young specimens with 17 or fewer segments may be confused with Branchi- notogluma sandersi. See below under Branchinotogluma sp. B. Etymology.—The species is named for J. Frederick Grassle, one of the observers on the OASIS Alvin dives. Opisthotrochopodus, new genus Type-species.— Opisthotrochopodus alvinus, new species. Gender: masculine. Diagnosis.—As in Branchinotogluma with following additions: Notopodia of elytrigerous segments 2-17 with prominent rounded bracts enclosing subconical acicular lobes and notosetae. Posterior 4 segments (18-21) without branchiae, compressed, with parapodia directed posteriorly and greatly modified, including expanded delicate lamellae, elongated cylindrical notopodia fused with dorsal cirrophores and distal styles on segments 20 and 21; achaetous notopodia on segments 19-21; achaetous neuropodia on segments 18 and 19; unique neuro- podial wheel organs on segment 20, including stout acicular and hooked neuro- setae. Etymology.—The name is derived from Greek: opistho, behind, plus trocho from trochos, a wheel, and podus from podos, foot, referring to the modified posterior parapodium forming a wheel organ. Opisthotrochopodus alvinus, new species Figs. 7-9 Material examined.—East Central Pacific, from dives of the Alvin on 3 vent areas along the Galapagos Rift in 1979: MUSSEL BED, 00°47'53”N, 86°09'12”W: Dive 880, 21 Jan, 2493 m, mussel washings, paratype (USNM 97253). GARDEN OF EDEN, 00°47'41’N, 86°07'44”W: Dive 884, 25 Jan, 2482 m, clam bucket with mussels, paratype (USNM 97254). ROSE GARDEN, 00°48'15’N, 86°13'28”W: Dive 983-112, 30 Nov, 2457 m, mussel washings, holotype (USNM 460 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON = SSS = SS F G Fig. 6. Branchinotogluma grasslei, holotype, USNM 97314: A, Right elytrigerous parapodium of segment 2, anterior view, acicula dotted; elytrophore not shown; B, Right cirrigerous parapodium of segment 6, posterior view; C, Right elytrigerous parapodium of segment 7, anterior view, acicula dotted; D, Upper and lower branchiae from same; E, Three notosetae; F, Supraacicular neuroseta, VOLUME 98, NUMBER 2 461 80639) and paratype (USNM 97251). Dive 984-32, 1 Dec, 2451 m, mussel wash- ings, 2 young paratypes (USNM 97252). Pacific Ocean off Western Mexico, 20°50’N, 109°06’W, OASIS Alvin dives in 1982: Dive 1214, 20 Apr, 2633 m, vestimentiferan wash, 21 paratypes including 8 young (USNM 97255). Dive 1219-2A, 25 Apr, 2612 m, slurp sample from Alvinella habitat, paratype (USNM 97256). Dive 1221-15, 4 May, 2618 m, Riftia and Calyptogena wash, 8 paratypes (USNM 97257). Dive 1223-11, 7 May, 2616 m, Riftia and Calyptogena wash, coarse fraction, 10 paratypes including 6 young (USNM 97258; 97259). Dive 1225-7, 9 May, 2618 m, Riftia and Calyptogena wash, fine fraction, paratype (USNM 97260). Dive 1226-7, 10 May, 2616 m, Riftia, Calyptogena and Alvinella wash, coarse and fine fractions, 2 paratypes (USNM 97261). Description. —Length of holotype from Galapagos Rift (USNM 80639) 10 mm, width 5 mm with setae, segments 21. Length of largest paratype from 21°N (USNM 97258) 14mm, width 6 mm, segments 21. Body subrectangular, flattened, tapering posteriorly, with parapodia longer than body width. Posterior 4 segments (18— 21) compressed, with parapodia greatly modified, differing from more anterior parapodia as well as from one another (Fig. 9A, B). No color except for yellow amber-colored setae. Dorsum with transverse ciliated bands, 2 per segment, ex- tending onto elytrophores and dorsal tubercles (Figs. 7A, 8A). Elytra and prominent bulbous elytrophores number 10 pairs (Figs. 7A, D, 8A, C, 9A). Elytra large, covering dorsum, round to oval, last pair elongate-oval, nearly covering posterior modified parapodia (Figs. 8D-—G, 9A, B, E). Elytra delicate, with branched “veins.” Dorsal cirri with short cylindrical cirrophores and long tapered smooth styles extending beyond tips of setae (Figs. 7A, C, E, 8B). Dorsal tubercles elongate and tapered (Figs. 7C, E, 8A). Both elytrophores and dorsal tubercles with delicate arborescent branchiae attached on their lateral sides, with additional smaller group of branchiae on bases of notopodia; branchiae compact with rather long terminal filaments (Figs. 7A, C—E). Branchiae beginning on segment 3 as 2 small groups (Fig. 7A), becoming larger more posteriorly, with long terminal filaments, and continuing to segment 17 as single small group (Fig. 9A). Prostomium bilobed, anterior lobes triangular with delicate frontal filaments; median antenna with bulbous ceratophore in anterior notch, with tapered style and long slender tip; palps stout, tapered, smooth; without eyes (Fig. 7A, B). Tentaculophores with 2 pairs of tentacular cirri, dorsal pair about length of palps, ventral pair slightly shorter; ventrally tentacular segment forming anterior and lateral lips of mouth (Fig. 7A, B). Second or buccal segment forming posterior lip of mouth and bearing biramous parapodia and first pair of elytrophores; notopodia with hood or bract encircling small bundle of notosetae; ventral buccal cirri similar to tentacular cirri, longer than following ventral cirri (Fig. 7A, B). When everted, muscular pharynx showing 5 papillae around opening: 3 dorsal — with detail of tips; G, Subacicular neuroseta, with detail of tips; H, Ventral view left side of segments 10-13, showing segmental papillae. b, notopodial bract. Scales = 1.0 mm for A—C; 0.2 mm for D; 0.1 mm for E-—G; 1.0 mm for H. 462 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON e = D ny Fig. 7. Opisthotrochopodus alvinus, holotype, USNM 80639: A, Anterior end, dorsal view; B, Same, ventral view, style of right buccal cirrus missing; C, Right cirrigerous parapodium of segment 3, anterior view; D, Right elytrigerous parapodium of segment 7, anterior view, acicula dotted; E, Right cirrigerous parapodium of segment 8, posterior view, with detail of branchial tips and neuropodial papillae; F, Four notosetae; G, Supraacicular neuroseta, with detail of part; H, Middle and lower subacicular neurosetae, with detail of part. b, notopodial bract; dT, dorsal tubercle. Scales = 1.0 mm for A, B; 0.5 mm for C—E; 0.1 mm for F-H. VOLUME 98, NUMBER 2 463 Fig. 8. Opisthotrochopodus alvinus. A, paratype, USNM 97255; B-G, I, holotype, USNM 80639; H, paratype, USNM 97254: A, Dorsal view left side of segments 9 (elytrigerous) and 10 (cirrigerous), base of style of dorsal cirrus only shown; B, Right cirrigerous parapodium of segment 16, posterior view; C, Right elytrigerous parapodium of segment 17, anterior view, acicula dotted, D, Right elytron 4 of segment 7; E, Left elytron 8 of segment 15; F, Left elytron 9 of segment 17; G, Right elytron 10 of segment 19; H, Distal end of extended pharynx, ventral view; I, Isolated jaw. b, notopodial bract; dT, dorsal tubercle. Scales = 1.0 mm for A, H; 0.5 mm for B, C; 1.0 mm for D-G; 0.2 mm for I. and 2 ventral; 2 pairs of prominent jaws minutely denticled on inner border (Fig. 8H, I). Parapodia of segments 2 to 17 biramous, with notopodia shorter than neuro- podia, both rami with light yellow amber-colored setae (Figs. 7A—-E, 8B, C). Notopodia with short tapered acicular lobes and, on elytrigerous segments, prom- inent rounded bracts enclosing notosetae dorsally and posteriorly (Figs. 7D, 8C). Notosetae moderate in number, stouter than neurosetae, straight, acicular, ta- pered, smooth or with 1-2 rows of spines (Fig. 7F). Larger neuropodia with long flattened conical presetal lobes with projecting acicular processes; postsetal lobes shorter, rounded; with some micropapillae on distal borders (Fig. 7C—E). Neu- rosetae very numerous, forming fan-shaped bundles. Supraacicular neurosetae longer, with more widely-spaced spines extending to near distal tip, more finely spinous distally (Fig. 7G). Subacicular neurosetae decreasing in length ventrally, finely spinous up to hooked tips, sometimes showing delicate flattened distal part 464 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON VOLUME 98, NUMBER 2 465 (Fig. 7H). Ventral cirri short, tapered, attached on middle of neuropodia (Fig. 7C-E). Parapodia of posterior 4 segments (18-21) modified and directed posteriorly, differing from one another (Fig. 9A, B). Parapodia of cirrigerous segment 18 (Fig. 9A-C) smaller than preceding (Fig. 8C) and following parapodia (Fig. 9E). No- topodium represented by small acicular lobe, dorsal cirrus with cirrophore fused to acicular lobe, with distal style, bundle of short smooth acicular notosetae (Fig. 9D), and delicate, rounded, flattened lamella (1); neuropodium forming small tapered acicular lobe with small ventral cirrus; without neurosetae (Fig. 9C). Modified parapodia of elytrigerous segment 19 lacking both notosetae and neurosetae (Fig. 9A, B, E). Notopodium with short acicular lobe (no), enclosed in large delicate lamella (1), with small elytrophore (ely) and elongate-oval elytron (el) extending posteriorly and covering more posterior parapodia; neuropodium forming tapering acicular lobe (ne) with small ventral cirrus, similar to neuro- podium of segment 18. Parapodia of segment 20 greatly modified and nearly covered dorsally by elon- gated elytra of segment 19 (Fig. 9A, B, F—H). Notopodium represented by achae- tous elongated cylindrical acicular lobe with distal style of dorsal cirrus (Fig. 9F). Neuropodium forming large cylindrical lobe with projecting conical acicular pro- cess and enclosing circle of neurosetae of several types (Fig. 9G): stout reddish amber-colored neurosetae, 2 stout, tapered pointed setae—one long, one short (Fig. 91) and about 14 stout strongly hooked setae (Fig. 9J), few slender smooth neu- rosetae with curved tips (near stout acicular setae; Fig. 9K), and delicate curved spinous setae (Fig. 9L). Ventral cirri lacking. Neuropodia capable of being flared and flattened distally, with neurosetae radiating like spokes of wheel (Fig. 9H). On holotype, neuropodium on left side expanded (Fig. 9A, B, H) and on right side not expanded (Fig. 9A, B, G). On other paratypes, both wheel organs found extended and flared, neither one flared, both only partially exposed or completely withdrawn into body and not visible. Segment 21 with right and left parapodia closely approximated medially and directed posteriorly (Fig. 9A, B, M, N). Notopodium consisting of expanded thin lamella (1) and thickened acicular lobe fused with cirrophore of dorsal cirrus, with distal style (dC); without notosetae. Neuropodium also with expanded thin lamella — Fig.9. Opisthotrochopodus alvinus, holotype, USNM 80639: A, Posterior end, dorsal view, showing left side of segments 14—21 and right side of segments 17-21; parapodial wheel organ of segment 20 expanded on left side, unexpanded on right side; B, Same, ventral view, showing right side of segments 11-21 and left side of segments 20 and 21; C, Right cirrigerous parapodium of segment 18, anterior view, acicula dotted; D, Notoseta from same; E, Right elytrigerous parapodium of segment 19, posterior view, acicula dotted; F, Right notopodium with dorsal cirrus of segment 20, aciculum dotted; G, Two views of right unexpanded neuropodial wheel organ from same, aciculum dotted; H, Left neuropodium of segment 20 showing ventral and posterior views of expanded wheel organ, aciculum dotted; I, Two stout acicular neurosetae from same; J, Three hooked neurosetae from same; K, Two curved smooth neurosetae from same; L, Curved spinous neuroseta from same; M, Right cirrigerous parapodium of segment 21, posterior view, acicula dotted; N, Neuroseta from same. dC, style of dorsal cirrus; el, elytron; ely, elytrophore; 1, notopodial lamella; ne, neuropodial acicular lobe; neL, neuropodial lamella; no, notopodial acicular lobe; py, pygidium; wO, neuropodial wheel organ. Scales = 1.0 mm for A, B; 0.5 mm for C, E-H, M; 0.1 mm for D, I-L, N. 466 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON (neL) and thickened acicular lobe on upper side and joined basally to notopodium; small group of needle-like neurosetae extending slightly beyond tips of neuro- podium, their tips mostly broken (Fig. 9M, N). Ventral cirri lacking. Pygidium (py) consisting of small oval lobe wedged between bases of parapodia of last segment (21), without anal cirri (Fig. 9B). Ventral segmental nephridial areas on bases of neuropodia of segments 4 to 11 low, rounded, with greatly elongated papillae on segment 12, extending posteriorly to segment 15, and with short rounded lamellae on segments 13 to 17 (Fig. 9B). Young specimens with 17 or fewer segments may be confused with Branchi- notogluma hessleri. See below under Branchinotogluma sp. A. Etymology.—The species is named for the submersible Alvin, which has been the means of collecting so many unusual animals. Young of Branchinotogluminae Young specimens of 17 or fewer segments belonging to this subfamily are difficult to separate to species, since the diagnostic posterior segments and ventral segmental papillae or lamellae are not yet developed. Based on their anterior ends, they may be separated into two groups, herein designated as Branchinotogluma sp. A or B, based on the following characters: Branchinotogluma sp. A = Young of B. hessleri or Opisthotrocho-_ | Branchinotogluma sp. B = Young podus alvinus of B. grassslei or B. sandersi Notopodial On all elytrigerous parapodia of On elytrigerous parapodia of seg- bracts segments 2 to 17 ment 2 only Notosetae Both smooth and with 1-2 rows of spines All smooth Arborescent With rather long terminal fila- branchiae ments With short terminal filaments Dorsal tubercles Elongate, tapered Inflated, indistinct Branchinotogluma sp. A Young of Branchinotogluma hessleri or Opisthotrochus alvinus Material examined.—East Central Pacific, from dive of the A/vin along the Galapagos Rift in 1979: ROSE GARDEN, 00°48'15’N, 86°13'28”W: Dive 990, 7 Dec, 2451 m, slurp gun, 1 young (USNM 97341). Pacific Ocean off Western Mexico, 20°50’N, 109°06’W, OASIS, Alvin dives in 1982: Dive 1214, 20 Apr, 2633 m, vestimentiferan wash, 13 young (USNM 97342). Dive 1221-15, 4 May, Riftia and Calyptogena wash, coarse fraction, 4 young (USNM 97343). Branchinotogluma sp. B Young of B. grasslei or B. sandersi Material examined.—East Central Pacific, from dives of the Alvin along the Galapagos Rift in 1979: MUSSEL BED, 00°47'53”N, 86°09'12”W: Dive 991-1, VOLUME 98, NUMBER 2 467 8 Dec, 2490 m, clam bucket wash, 2 young (USNM 97339). GARDEN OF EDEN, 00°47'41”N, 86°07'44”W: Dive 884, 25 Jan, 2482 m, clam bucket with mussels, 1 young (USNM 97337). ROSE GARDEN, 00°48'15’N, 86°13'28”W: Dive 984- 32, 1 Dec, 2451 m, mussel washings, 5 young (USNM 97338). Pacific Ocean off Western Mexico, 20°50’N, 109°06’W, OASIS Alvin dive in 1982: Dive 1214, 20 Apr, 2633 m, vestimentiferan wash, 34 young (USNM 97340). Remarks.—The subfamily Branchinotogluminae agrees with Branchipolynoin- ae Pettibone (1984a) and Branchiplicatinae Pettibone (1985) in having well-de- veloped branchiae, an unusual feature in the Polynoidae. The above three subfam- ilies agree with Macellicephalinae Hartmann-Schréder (Pettibone 1976) and Lepidonotopodinae Pettibone (1983, 1984b) in the structure of the prostomium and tentacular segment in having a median antenna, paired palps, lacking lateral antennae and eyes, with two pairs of tentacular cirri on tentaculophores lateral to the prostomium. Notopodial bracts are rare in the Polynoidae. The notopodial bracts of Bran- chinotogluminae, in the form of oval projections enclosing the notopodial acicular lobes and notosetae on the elytrigerous parapodia only, differ markedly from the notopodial bracts in the Lepidonotopodinae where they are truncate and are found on all the setigerous segments (Pettibone 1983, 1984b). The pharynx with serrated or denticulated jaws is also rare in the Polynoidae but is found in some members of the Macellicephalinae and Bathyedithinae (Levenstein 1971; Pettibone 1976, 1979) and in the Lepidonotopodinae (Pettibone 1983, 1984b). The greatly modified posterior four segments in Opisthotrochopodus alvinus set it apart from all other members of the Polynoidae, especially with the development of extra lamellae and the wheel organs on segment 20, with large acicular and hooked neurosetae. Except for this unusual feature, it agrees with species of Bran- chinotogluma and is placed in the same subfamily. The branchiate subfamilies of Polynoidae from the hydrothermal rift areas may be separated according to the following key: Key to three Branchiate Subfamilies of Polynoidae 1. Prostomium truncate anteriorly, not bilobed, without frontal filaments; median antenna with ceratophore in middle of prostomium. Segments up to 35, first achaetous. Elytra 12 pairs, on segments 2, 4, 5, 7, alternate segments to 23. Dorsal cirri on segments 3, 6, 8, alternate segments to 22 and on up to 12 posterior segments from 24 on. Branchiae beginning on segment 3, flattened elongate sacs, deeply folded and convoluted, attached to flattened elytrophores and dorsal tubercles, both with extra lobes. Elytra large, oval, covering dorsum. Parapodia biramous, both rami with pro- jecting acicular processes. Notopodia without well-developed bracts. Paired palps, tentacular, buccal and dorsal cirri all long. Pharynx with 5 pairs of unequal papillae; 2 pairs of jaws, minutely denticled ................. MeFi acts ty terns oath ani) toed BRANCHIPLICATINAE Pettibone, 1985 (see Branchiplicatus cupreus Pettibone, figs. 1—4, in Pettibone (1985)) — Prostomium bilobed, anterior lobes with minute or filiform frontal fila- ments. Segments 21, first achaetous. Elytra 10 pairs on segments 2, 4, 5, 7, alternate segments to 19. Dorsal cirri on segments 3, 6, 8, alternate 468 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON segments to 20, 21. Elytrophores and dorsal tubercles not flattened, with- out extra lobes. Branchiae arborescent, in 2 main groups lateral to ely- trophores and dorsal tubercles and on dorsal sides of notopodia ....... 2 2. Bilobed prostomium with minute frontal filaments; ceratophore of median antenna indistinct, with short style. Elytra small, leaving middorsum un- covered. Elytrophores and dorsal tubercles low, indistinct. Parapodia sub- biramous. Notopodia small digitiform, without notopodial bracts. No- tosetae few, stout, acicular. Neuropodia short, truncate, without projecting acicular processes. Arborescent branchiae beginning on segment 2, with long terminal filaments. Paired palps, tentacular, buccal and dorsal cirri all short. Pharynx with 5 pairs of jaws entire, not denticled. Associated with deep-sea mussels ....... BRANCHIPOLYNOINAE Pettibone, 1984a (see Branchipolynoe symmytilida Pettibone, figs. 1-8, in Pettibone (1984a)) — Bilobed prostomium with filiform frontal filaments; median antenna with distinct ceratophore in anterior notch. Elytra large, covering dorsum. Ely- trophores prominent, bulbous; dorsal tubercles inflated, indistinct or elon- gate, tapered. Parapodia biramous, both rami with projecting acicular processes. Notopodia with prominent notopodial bracts on elytrigerous segment 2 or on all elytrigerous segments. Notosetae numerous, stout, acicular. Arborescent branchiae beginning on segment 3, compact, with short terminal filaments. Paired palps, tentacular, buccal and dorsal cirri all long. Pharynx with 5 papillae: 3 dorsal and 2 ventral; 2 pairs of jaws mmuteh~adenticled i= BRANCHINOTOGLUMINAE, new subfamily (see Branchinotogluma, new genus; Opisthotrochopodus, new genus) Acknowledgments My thanks go to the members of the Galapagos Rift Biology Expedition in 1979, the OASIS group of Scripps Institution of Oceanography in 1982 to the East Pacific Rise at 21°N, and especially to J. F. Grassle and I. Williams of the Woods Hole Oceanographic Institution for the preliminary sorting and material on which this study is based. My colleague, Kristian Fauchald, kindly reviewed the manuscript. Literature Cited Levenstein, R. J. 1971. [Polychaete worms of the genera Macellicephala and Macellicephaloides (Family Aphroditidae) from the Pacific Ocean. — /n Fauna of the Kurile-Kamchatka Trench.]— Trudy Institut Okeanologii P.P. Shirshov Akademiia Nauk SSSR 92:18-35. [In Russian, English summary]. Pettibone, M. H. 1976. Revision of the genus Macellicephala McIntosh and the subfamily Macel- licephalinae Hartmann-Schréder (Polychaeta: Polynoidae).— Smithsonian Contributions to Zo- ology 229:1-71. 1979. Redescription of Bruunilla natalensis Hartman (Polychaeta: Polynoidae), originally referred to Fauveliopsidae.— Proceedings of the Biological Society of Washington 92(2):384- 388. . 1983. A new scale-worm (Polychaeta: Polynoidae) from the hydrothermal rift-area off West- ern Mexico at 21°N.—Proceedings of the Biological Society of Washington 96(3):392-399. . 1984a. A new scale-worm commensal with deep-sea mussels on the Galapagos hydrothermal vent (Polychaeta: Polynoidae).— Proceedings of the Biological Society of Washington 97(1): 226-239. VOLUME 98, NUMBER 2 469 . 1984b. Two new species of Lepidonotopodium (Polychaeta: Polynoidae: Lepidonotopodinae) from hydrothermal vents off the Galapagos and East Pacific Rise at 21°N.— Proceedings of the Biological Society of Washington 97(4):849-863. 1985. An additional new scale-worm (Polychaeta: Polynoidae) from the hydrothermal rift- area off Western Mexico at 21°N.— Proceedings of the Biological Society of Washington 98(1): 127-149. Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 470-475 THE GENERIC AND SUBFAMILIAL CLASSIFICATION OF THE NAIDIDAE (ANNELIDA: OLIGOCHAETA) R. O. Brinkhurst Abstract.— Examination of the characters used to define genera within the family Naididae leads inevitably to the conclusion that the genus Pristina Ehr. requires subdividing into two genera, Pristina and Pristinella n. gen. Neither classical nor quantitative methods support the division of the family into subfamilies, but the latter support the erection of the genus Pristinella n. gen. During the preliminary stages of an assessment of alternative quantitative ap- proaches to classification of the Oligochaeta all generic characters currently applied in the family Naididae were examined, as was the subfamilial classification. Both of these have been accepted uncritically since they were revised by Sperber (1948), who reviewed the historical basis for that classification. Four subfamilies were recognized. Two are monotypic (Pristininae-Pristina Ehr., Paranaidinae-Paranais Czern.), the third contains two genera (Chaetogastrinae-Chaetogaster von Baer, Amphichaeta Tauber), and the fourth, the Naidinae, contains all other genera and is therefore disproportionately large. While this is not of itself evidence for the unsuitability of the existing classification, it does suggest careful re-examination. An intensive effort to examine quantitative means of doing this is underway; this provisional report is required in order to allow other publications to proceed based on the division of the genus Pristina in particular. Paranaidinae Sperber, 1948 Sperber (1948) established this subfamily based on the unique characteristic of absence of nephridia, even in budding segments. Additional characters cited were a reduction of the number of anterior segments formed on budding from 5 to 4, despite the fact that this is shared with Stephensoniana Cernosvitov, but the latter was held to be “‘otherwise so unlike Paranais that they can hardly be supposed to be especially related to each other.” The male ducts are also said to be unusual, with narrow vasa deferentia, strongly muscular atria and no prostate glands, with other unusual characters in the ejaculatory ducts and clitellum. These are not, in fact, unique to the group (see the presence of the clitellum between the male pores in Piguetiella Sperber, for example) nor are the absence of both hair setae and eyes and the resemblance of dorsal and ventral needles. Since these studies, the genus Wapsa was defined, but a careful revision of North American species has shown that this is synonymous with Paranais (Brinkhurst and Coates 1984) and that both may possess nephridial tissue. Quantitative methods (see below) confirm the synonymy of these two genera and the lack of any clear separation of Paranais from all other genera, particularly Chaetogaster, Amphichaeta, and Homochaeta Bretscher. The closed or even absent nephridia in many species of Paranais, Amphichaeta, and Chaetogaster may be associated with the estuarine habit of most species of VOLUME 98, NUMBER 2 47] the first two. This explanation will not suffice for the latter, though, as this is predominantly a freshwater genus. It may be that Chaetogaster is derived from a common ancestor of all three genera. There is no clear basis for the separation of the Paranaidinae, either by cluster analysis based on overall similarity or by virtue of shared advanced characters (synapomorphy). Chaetogastrinae Sperber, 1948 This subfamily consists of two genera (Chaetogaster and Amphichaeta) that have a strongly lengthened pharynx, (which lacks the dorsal diverticulum of other genera but is connected to the body wall by strong muscle fibers), a short oeso- phagus, a stomach of unusual form, a strongly reduced vascular system and closed nephridia. The nephridial characteristic is now known to be shared with Paranais, as described above. The possible functional reason for the reduction in the vascular system remains obscure, but the other characteristics, along with the reduced prostomium, are clearly associated with the adoption of a predatory mode of feeding. In most aquatic oligochaetes the roof of the pharynx is everted through the mouth in feeding, being returned by strong retractors. The glandular cells are greatly expanded by relocating the cell bodies on the septa of post-oral segments (the pharyngeal or septal glands). The pharyngeal diverticula of most naidid genera appear to be no more than large folds on either side and behind the pad-like pharyngeal roof which contacts the substrate once everted through the mouth (Sperber 1948, plate V fig. 5, plate VI fig. 1). The glandular cells of the pharynx are not often exported to form pharyngeal glands, and the retractor muscles may be poorly developed in the Naididae. The retractor muscles of the pharynx of the Chaetogastrinae are presumably developed from those present in other aquatic oligochaetes, but a comparative study of the pharyngeal structures would be in order to judge from the lack of recent descriptions of these structures. There is no a priori reason to suppose that this pharyngeal modification was independently acquired by these two genera, but this is not in itself perhaps enough to provide the basis for a subfamilial classification. I would prefer to see inde- pendent characters that confirm a separation between these two genera and all other naidids. In fact, in most cluster analyses (see below), Chaetogaster is most closely aligned with Homochaeta Bretscher, which Sperber classified with the Naidinae, and only secondarily groups with Amphichaeta and Paranais. There is no clear evidence of a monophyletic group consisting solely of Chaetogaster and Amphichaeta. Naidinae Lastockin, 1924 According to Sperber (1948), this subfamily was originally established to in- corporate all genera bar Pristina (Pristininae Lastockin, 1924), but was modified by her to exclude the foregoing. Even then, Sperber admits that this subfamily is ““more heterogeneous” than the others. There are no distinct characters that iden- tify this group as a monophyletic assemblage by virtue of shared apomorphies, and they do not cluster out in a single group without the interposition of some or all of the genera excluded from the subfamily. 472 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Pristininae Lastockin, 1924 The single genus in this subfamily is said to be “large and extremely varying, yet curiously homogeneous”’ by Sperber (1948). There follows a recitation of the subfamilial characters, which include a characteristic stomach with intracellular canals and male ducts which, while variable, have narrow atria without prostates with vasa deferentia entering at the top. The genital setae are of peculiar forms not seen elsewhere, according to Sperber. The stomachal canals need to be re-investigated using electron microscopy. The genus Pristina (as Pristina A and B) never appears separated from all other genera in any of the many analyses performed to date, though it is never consis- tently associated with any other genera. As none of the other subfamilies can be supported, there seems little point in returning to the position adopted by Las- tockin, with this subfamily being contrasted to all others. Even a superficial examination of the list of species currently included within Pristina suggests that it has been treated differently from all other genera. The presence or absence of a proboscis is used as a generic character throughout the rest of the family. Beginning with a breakdown based on that sole characteristic, it rapidly becomes clear that other, unrelated, characters are associated with this criterion, leading to the separation of two quite distinct genera as follows: Pristina Ehrenberg, 1828 Type species.—P. longiseta Ehr. Included species. —synclites Steph., peruviana Cern., aequiseta Bourne, longiseta Ehr., proboscidea Bedd., breviseta Bourne, plumaseta Turner, leidyi Smith, amer- icana Cern., macrochaeta Steph. Definition. — Dorsal setae from II, hair setae present, ventral setae progressively differ from anterior to posterior bundles. Proboscis present. Septal glands present in some segments from IIJ-VI. Stomach in VII or VIII with canals. Testes and spermathecae in VII, ovaries and atria in VIII, genital setae on VI, VI and VIII, or VII and VIII with glands. Prostates on the vasa deferentia. Spermathecae present. Cosmopolitan. Pristinella, new genus Type-species.— Naidium bilobatum Bretscher, 1903. Included species.—rosea (Piguet), amphibiotica (Last.), notopora (Cern.), jen- kinae (Steph.), sima (Marcus), longidentata (Harman), menoni (Aiyer), idrensis (Sperber), acuminata (Liang), bilobata (Bret.), osborni (Walton), /ongisoma (Har- man), ? arcaliae (Pop.). The genus is quite similar to the older assemblage associated with the name Naidium Schmidt but because N. /uteum Schmidt, the type-species, is a tubificid, the generic name is not available (Sperber 1948:211). Definition. — Dorsal setae from II, hair setae present, ventral setae progressively differ from anterior to posterior bundles. Proboscis absent. Septal glands in some segments from IIJ—VI. Stomach in VI, VII or VIII, with canals. Testes and sper- methecae in VII, ovaries and atria in VIII. Genital setae absent, or present in VIII in one species. Prostate glands absent. Male pores median in one species. Spermathecae absent. Cosmopolitan. VOLUME 98, NUMBER 2 473 A CHAETOGASTER AMPHICHAETA PARANAIS : HOMOCHAETA ——— NAIS ee OHI NRIS UNCINAIS 4 SPECARIA ARCTEONAIS RIPISTES Ls} STYLARIA 5 ee aa PIGUETIELLA HAEMONAIS BRANCHIODRILUS DERO Bey o N ALLONAIS STEPHENSONIANA PRISTINA 100 75 50 25 0 HOMOCHAETA OPHIDONAIS 1 UNGINAIS NAIS SPECARIA PARANAIS WAPSA 2 PRISTINA A PRISTINA B 3 : CHAETOGASTER AMPHICHAETA 4 SLAVINA B BRATISLAVIA 5 ALLONAIS HAEMONAIS DERO BRANCHIODRILUS - ARCTEONAIS RIPISTES 6 VEJDOVSKYELLA STYLARIA SLAVINA A STEPHENSONIANA PIGUETIELLA 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1 G STEPHENSONIANA : RIPISTES 2 ARCTEONAIS STYLARIA SLAVINA A 3 VEJDOVSKYELLA PIGUETIELLA HOMOCHAETA CHAETOGASTER 4 AMPHICHAETA WAPSA RANAIS PA OPHIDONAIS UNCINAIS SPECARIA NAIS SLAVINA B ALLONAIS 5 HAEMONAIS BRANCHIODRILUS DERO PRISTINA A PRISTINA B BRATISLAVIA Fig. 1. Dendrograms resulting from re-orientation of an evolutionary tree by Sperber 1948(A), and from cluster analyses of a 15 character by 24 taxa matrix by the methods of Ward (B) and Preston (C). There is no quantitative scale for A as the clusters were created qualitatively. 474 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Remarks. — Full synonymies of these genera plus Bratislavia Kosel are discussed by Brinkhurst (1971) and subsequent changes by Harman (1973, 1974, 1982), Harman and McMahan (1975), Harman and Loden (1978), Loden and Harman (1980) and Brinkhurst and Kathman (1983). Other synonymies will undoubtedly be uncovered as recent observations using scanning electron microscopy continue to reveal intraspecific variation in the degree of serration of hair setae, the presence of pectinations in normally bifid setae, and the effects of environmental conditions on setal form. Care should be taken in separating species currently distinguished by such characters (e.g., P. longidentata and P. idrensis). Quantitative Analysis A variety of quantitative methods are being evaluated using a data matrix derived from the characters used to define naidid genera. A 24 x 24 matrix of characters and taxa is used, with Pristina and Slavina divided into two taxa each, and the recent genera Neonais Sokolskaya and Rhopalonais Dzwillo and Grimm excluded by reason of lack of knowledge of the reproductive structures. A con- densed data matrix of only 15 characters was created by eliminating those char- acters thought to be less reliable by virtue of the difficulty experienced in scoring them, or their dependence on other characters. Serrations on the hair setae, for example, may vary in degree of development and their existence depends on the presence of hair setae in the first instance. For the purposes of this discussion only three cluster diagrams will be used (Fig. 1). The first was derived from the evolutionary tree proposed by Sperber (1948). The dendrogram was created by measuring the distances between dichotomies on the evolutionary tree, and so there is no scale of similarity involved (Fig. 1A). The four subfamilies form monophyletic groups, identified as groups 1, 2, 8, and 3-7 respectively. The anal- ysis according to the Ward method (Ward 1963) is illustrated in Fig. 1B, that by the Preston method (small input pair-group cluster, unweighted pair-group mean clustering optimum rotation of dendrogram, Preston’s resemblance equation Preston 1962) in Fig. 1C. Both of these were based on the shorter, more reliable data set. The features of these dendrograms referred to here are consistent in all of the analyses performed to date, though other details vary quite considerably. The points to be emphasized here are that there is no evidence to support the separation of the subfamilies proposed by Sperber, and that, while the two sections of Pristina cluster together, they do so at a lower level of similarity than a great many other genera currently regarded as well established. Examples of the latter would be Dero Oken with Branchiodrilus Mich. (which always pair so closely that a single origin of gills in the family can be postulated) and Ophidonais Gervais with Unicinais Lev. The similarity of Paranais and Wapsa Marcus was rendered total when the latter was shown to be a synonym of the former by Brinkhurst and Coates (1984). These results are obtained with both the Ward and Preston analyses (and others) despite the fact that the first method recognizes all similaarities (0-0, 1-1) whereas the latter only recognizes positive matches (1-1) in a Hennigian manner. The data were polarized with respect to a supposed ancestor in the Preston analysis as required by the method. In a later presentation, the data matrix will be discussed in detail, and the VOLUME 98, NUMBER 2 475 possible new subfamilial groupings derived from a variety of phenetic and cladistic methodologies will be explored. Acknowledgments I am indebted to A. F. Nemec, S. F. Cross, and D. C. Jeffries for statistical analyses and advice, and to R. D. Kathman for reviewing the manuscript. M. Stone typed several drafts of the manuscript. Literature Cited Brinkhurst, R.O. 1971. Jn R. O. Brinkhurst, and B. G. M. Jamieson, Part 2. Systematics. 7. Family Naididae. Aquatic Oligochaeta of the World.—Oliver and Boyd, Edinburgh, XI + 806 pp., pp. 304-443. , and K. A. Coates. 1984. The genus Paranais (Oligochaeta, Naididae) in North America. — Proceedings of the Biological Society of Washington 98(2):303-313. , and R. D. Kathman. 1983. A contribution to the taxonomy of the Naididae (Oligochaeta) of North America.— Canadian Journal of Zoology 61:2307-2312. Harman, W. J. 1973. New species of Oligochaeta (Naididae) with additional distributional records for Oklahoma and Texas.—Southwestern Naturalist 18(2):151-164. 1974. The Naididae (Oligochaeta) of Surinam.—Zoologische Verhandelingen 133:1-36. . 1982. The aquatic Oligochaeta (Aeolosomatidae, Opistocystidae, Naididae) of Central Amer- ica.—Southwestern Naturalist 27(3):287-298. , and M. S. Loden. 1978. Bratislavia unidentata (Oligochaeta: Naididae) a re-description. — Southwestern Naturalist 23(4):541-544. , and M. L. McMahan. 1975. A re-evaluation of Pristina longiseta (Oligochaeta: Naididae) in North America. — Proceedings of the Biological Society of Washington 88(17):167-178. Loden, M. S., and W. J. Harman. 1980. Ecophenotypic variation in setae of Naididae (Oligochaeta). mR. O. Brinkhurst, and D. G. Cook, eds., Aquatic Oligochaete Biology.— Plenum Press IX + 529 pp., pp. 33-40. Preston, F. W. 1962. The canonical distribution of commonness and rarity.—Ecology 43:185-215, 410-432. Sperber, C. 1948. A taxonomical study of the Naididae.— Zoologiska Bidrag fran Uppsala 28:1-296. Ward, J. H. 1963. Hierarchial grouping to optimize an objective function.—Journal of the American Statistical Association 58:236—244. Ocean Ecology Laboratory, Institute of Ocean Sciences, P.O. Box 6000, 9860 West Saanich Road, Sidney, British Columbia V8L 4B2, Canada. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 476-493 DEEVEYINAE, A NEW SUBFAMILY OF OSTRACODA (HALOCYPRIDIDAE) FROM A MARINE CAVE ON THE TURKS AND CAICOS ISLANDS Louis S. Kornicker and Thomas M. Iliffe Abstract. — Deeveyinae, a new subfamily of the Halocyprididae (Ostracoda, sub- order Halocypridina) is proposed for Deeveya spiralis, a new genus and species of troglobitic ostracode from an anchialine cave in the Turks and Caicos Islands, West Indies. The new genus and species is described and illustrated. Danielopol (1972) described the first troglobitic halocyprid ostracode from a marine cave in Cuba. Therefore, the discovery of a second halocyprid in an anachialine cave on the Turks and Caicos Islands, which are a southeast contin- uation of the outer line of the Bahama Islands, is of interest. Although the two caves are relatively close geographically, differences between the specimens from the two caves led to the proposing a new subfamily for the specimens from the Turks and Caicos Islands. Danielopol (1972) referred his species to the Thau- matocyprididae, whereas the present new species is referred to the Halocyprididae. Family Halocyprididae The Halocyprididae comprises five subfamilies, Conchoecinae, Halopyridinae, Archiconchoecinae, Euconchoecinae, and Deeveyinae, the new subfamily de- scribed herein. Deeveyinae, new subfamily Diagnosis. — First antenna with 8 segments. Sixth limb with distal dorsal process on lst exopodial segment. Seventh limb with 3 bristles. Organ of Bellonci bifur- cate. Discussion.—The new subfamily is referred to the Halocyprididae rather than to the Thaumatocyprididae because the furca of Deeveya spiralis is of the halo- cyprid rather than the thaumatocyprid type. None of the other subfamilies of Halocyprididae have members having the combined characters listed in the di- agnosis above. Deeveya, new genus Etymology.—The genus is named for Georgiana B. Deevey. Gender: Feminine. Type-species.— Deeveya spiralis, new species. Distribution.—The type-species from a marine cave in the Turks and Caicos Islands, depth 7 m. Diagnosis.—Carapace without rostrum. First antenna with 8 segments: first segment without bristles; eighth segment bearing 4 bristles. Endopodite of second antenna with 3 segments: first segment without processus mammillaris but with 2 dorsal bristles; second segment with 3 long bristles; small third segment with 2 VOLUME 98, NUMBER 2 477 long bristles. Sixth limb with dorsal process bearing 4 bristles on first exopodial segment of type-species. Seventh limb with 3 bristles. Furca with 7 claws on each lamella, and 1 unpaired dorsal bristle; claws 5—7 bristle-like; all claws separated from lamella by suture. Organ of Bellonci short, bifurcate. Comparisons.—The new genus differs from previously described members of the Halocyprididae in having no rostrum, eight segments on the first antenna, a dorsal bristle-bearing process on the first exopodial segment of the sixth limb, and three bristles on the seventh limb. Deeveya differs from known members of the Thaumatocyprididae in having all furcal claws separated from the lamellae by a suture, three bristles on the seventh limb, a bifurcate organ of Bellonci, and no bristles on the first segment of the first antenna. Some species of the halocyprid genera Halocypris and Halocypria have a minute rostrum. Deeveya spiralis, new species Figs. 1-12 Etymology.—From the Latin spiralis (=coil, twisted), in reference to the two twisted bristles of the mandibular basis. Material.—Turks and Caicos Islands, Caicos Islands, Providenciales Island, The Hole, 30 Oct 1982, coll. Thomas M. Iliffe, specimens collected with suction bottle from 5 to 7 m depths using scuba. Holotype, USNM 193117, adult female; paratype, USNM 193118, adult female. Distribution.—Known only from the anchialine habitats of The Hole, Provi- denciales Island, Turks and Caicos Islands. Habitat.—The Turks and Caicos Islands are a southeast continuation of the outer line of the Bahama Islands. The basic geological and geomorphological setting is generally similar to that of the Bahamas. The Bahama Platform, including the Caicos Bank, is composed of a flat-lying shallow-water carbonate cap thicker than the surrounding ocean is deep (Dietz, Holden, and Sproll 1970). The thickness of this near homogeneous cap indicates that the depositional environment of the Platform—a shallow water situation—must have remained essentially the same since at least the early Cretaceous. A history of subsidence offset by upbuilding of coral-algal carbonates has maintained the plateau at sea level. Providenciales Island is located on the northern edge of the Caicos Bank. The main topographical feature of the island is a line of rounded hills 20 to 40 m above sea level, running parallel to the coastline. These hills are formed from reef-derived eolian carbonates, probably of Pleistocene age. The Hole is a sheer-walled cenote-like pit located near the crest of a line of hills at the western end of Providenciales, 1.1 km from the nearest open water, the south coast. It is about 15 m deep with a 15 m long by 10 m wide lake at the bottom, open to daylight. The bottom of the 6 to 8 m deep lake is completely choked with breakdown and surface debris such that no human-sized cave passages extending off from it were found. In addition to Deeveya spiralis, species observed or collected from the lake include a representative of a new genus of nebaliacean Speonebalia cannoni being described by Bowman, Yager, and Iliffe (1985), a new genus of amphipods being studied by John R. Holsinger, a crab identified as Sesarma (H.) miersii Rathbun by C. W. Hart, Jr., and an uncollected copepod. Other marine caves on Providenciales Island contain the shrimps Barbouria cu- 478 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON bensis and Typhlatya garciai (Buden and Felder 1977), the amphipod Spelaeon- icippe provo (Stock and Vermeulen 1982), a new family of shrimps (Hart and Manning, in preparation), and a new species of stygiomysid (Bowman, Iliffe, and Yager 1984). Although we did not measure salinity in the lake, we expect that, at least in the deeper waters from which the thaumatocyprids were collected, it is at or near open ocean levels (about 35 to 36%c). Salinities presumably measured from the surface waters of other caves on Providenciales were 18 and 19%o (Buden and Felder 1977; Stock and Vermeulen 1982). Under similar conditions in Bermuda caves, a several meters-thick surface layer of 3 to 30%o overlies full salinity waters (Sket and Iliffe 1980; Iliffe, Hart, and Manning 1983). Discussion.—A number of taxa from primarily deep water groups have been recently discovered in shallow water marine caves. Several species inhabiting a sea water flooded lava tube cave in the Canary Islands show affinities to deep sea species (Iliffe et al. 1984). These include Munidopsis polymorpha from the pri- marily bathyal group of the Galatheidae, the amphipod Spelaeonicippe buchi belonging to the mainly abyssal family Pardaliscidae, and the polychaete Gesiella jameensis from the mostly deep water group Macellicephalinae. Similarly, from marine caves in the Caicos Islands, the amphipod Spelaeonicippe provo, a close relative of S. buchi, and a new polychaete being described by Pettibone and co- workers, also from the subfamily Macellicephalinae, may have a deep sea origin. Among the ostracodes, a thaumatocyprid Thaumatocypris orghidani from caves in Cuba (Danielopol 1972, 1976), Deeveya spiralis, and new species having ap- pendages similar to those of D. spiralis from Bermuda caves (Angel and Iliffe, in preparation) are most probably derived from stocks now inhabiting the deep sea. The stability and constancy of the shallow water environment on the Bahama Platform over prolonged geological periods has probably been a significant factor in sustaining relict populations of marine cavernicoles long after their open sea ancestors had become extinct. A large number of marine caves are now known from the Bahama Platform including the famous “‘Blue Holes,”’ submerged circular sink holes of often spectacular dimensions (Warner and Moore 1984). These tidal, seawater-flooded caves have most probably existed on the Platform since the earliest stages of limestone deposition. A number of these Blue Holes have been explored to depths of 100 m or more and probably extend much deeper than that. A characteristic of the Bahama Platform is its steep-walled sides which rapidly drop off into oceanic depths. Strong tidal currents flowing into and out of those caves located near the edge of the Platform could draw deep water organisms or their larvae directly into caves. Subsurface water temperatures in those Bermuda marine caves remote from the sea have been found to remain seasonally constant at near the average year-round temperatures, thus indicating by extension that marine caves in general could have served as refugia for temperature-sensitive species during periods of Pleistocene glaciation (Iliffe, Hart, and Manning 1983). As a result of their age, geological and environmental stability, and proximity and tidal exchange with deep waters, marine caves of the Bahama Platform, including those caves in the Caicos Islands, are indeed highly suitable sites to serve as preserving centers for deep-water species. Description of adult female (Figs. 1-12).—Carapace oval in lateral view except for linear dorsal margin and slightly concave anterior margin (Figs. 1, 4a). Car- VOLUME 98, NUMBER 2 479 Fig. 1. Deeveya spiralis, lateral view of holotype, length 2.87 mm. apace having greatest height just anterior to adductor muscles, greatest length just dorsal to adductor muscles, and greatest width near middle (Figs. 1, 4a—c). Right valve with small tubercle on dorsal margin near posterior end (Fig. 1). Ornamentation (Figs. 1-6): Carapaces appearing reticulate in transmitted or reflected light (Figs. 4, 5), but reticulations mostly within translucent shell wall (Figs. 2, 3). Shell surface smooth except for minute bosses occurring mostly at intersections of walls forming reticulations (Figs. Se, f, 6a—c, e), and for minute, round, shallow fossae occurring mostly in center of low domes within reticulations. Internal reticulations smaller and more numerous where muscles attach to shell (Fig. 2). Internal reticulations mostly with 4 sides but a few with 3 to 6 sides. (SEM micrographs show reticulations to be ubiquitous except in areas of muscle attachments where the shell appears smooth, except for shallow fossae and minute bosses (Figs. 6a—c). Reticulations appearing on surface in SEM micrographs prob- ably occur during freeze-drying prior to taking micrographs when “‘skin”’ of shell contracts more at center of reticulations than at wall-forming reticulations. This interpretation is warranted because under the light microscope, reticulations are present in all areas of shell, not only in areas away from muscle attachments (Figs. 2, 3).) Bristles (Figs. 1, 7a, b): 2—4 very long bristles present along posterior shell margin (Figs. 1, 7a); short bifurcate bristles along anterior and ventral shell margins (Fig. 7b); lateral surface of shell with very few slender short and medium length single bristles. Small bristle bearing long spines present at tip of dorsal tubercle of right valve. Infold (Figs. 7a, b): Broad infold along anterior, ventral and posterior shell margins, narrowest opposite anterior concavity of margin, widest at anteroventral corner. Narrow list present near inner margin of infold, bearing narrow lameller 480 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 2. Deeveya spiralis, stereoscopic pair of part of shell of holotype showing internal reticulations in vicinity of adductor muscles (ends black) and pillow structure of outer surface (upper half of illustration). prolongation with smooth outer edge. Selvage along outer margin of infold with narrow lamellar prolongation with smooth outer edge. Glands: No glandular openings observed on infold. Glandular opening on tip of dorsal tubercle of right valve anterior to minute seta. Muscle attachments (Figs. 1—4a, d): Adductor muscles consisting of about 20 ovoid muscles attaching in elliptical area oriented obliquely just anterior to valve midlength; 3 muscle attachments forming row just anterior and ventral to central adductor muscle attachments (3 muscles may also be adductor muscles); few additional muscles attaching to valves closer to dorsal margin. Internal shell reticulations smaller and more numerous at points of muscle attachments than elsewhere (possible means of strengthening shell at points of muscle attachments). Shell microstructure (Fig. 6f): When calcium carbonate of shell is dissolved by lactic acid, abundant fibers are visible in areas within walls of polygons. Some fibers are visible within broken edge of shell shown in Fig. 6f. Shell size: Holotype, length 2.87 mm, height including tubercle 2.08 mm; para- type, length 2.67 mm, height including tubercle 2.05 mm. First antenna (Fig. 8a): Elongate with 8 segments. First segment with distal lateral spines becoming longer near ventral margin; distal end of lateral side of first segment overlapping proximal part of second segment, especially in vicinity of ventral margin. Second segment with dorsal midbristle bearing short marginal spines; distal end of second segment overlapping proximal end of third segment, especially near ventral margin; distal half of second segment bearing abundant short spines (spines not shown on illustrated limbs). Third segment elongate, with spinous ventral bristle distal to middle, short proximal spines along dorsal margin, and longer spines on medial surface more or less restricted to distal ventral quar- ter. Fourth segment short, with slender dorsal bristle bearing short, faint, marginal spines. Fifth segment shorter than fourth, with long, ventral, terminal, filament- VOLUME 98, NUMBER 2 481 © we b jis . ~ ot, \ > ~) if a OY NA aie MN YN AY + Pi, Wont { ' “ \ i { stat .- eae v . g, €. : a Ws igal LI: wid of \~ 5 A Th. ‘ [3 Fo wed pod 1, a ae Oe 24 ‘ it 4 . is ; Be 5 Oy 2 yn) “ys ‘ iv ee Gi) vey Lt Fe ee ae ae | js a Oe iri i Ok Sy eae Fd Met ey i CF rs ee, . é » rh 4 ; a Playing 9S ; , Fig. 3. Deeveya spiralis, a, Photograph using transmitted light and phase contrast of right valve of holotype showing internal reticulations in vicinity of adductor muscles (dark areas); b, Detail from a. Photographs by Dr. Robert P. Higgins. like bristle bearing widely spaced short marginal spines and minute terminal spine. Sixth segment shorter than fifth, bare. Seventh segment about same length as fourth segment, with 1 short, distal, lateral, spinous bristle near dorsal margin, and 2 long, spinous (spines widely spaced) bristles on terminal ventral pedestal (both bristles longer than bristle of fifth segment; medial bristle about *% length of lateral bristle and filament-like). Eighth segment with long principal bristle (about 2 times length of stem) and 3 filament-like bristles about '2 length of principal bristle, all bristles with widely spaced marginal spines. 482 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 400HM 20K¥V 20K¥ 400M 20Ky¥ e* wet 7 Ay xe - *. | ay ty. = tine, J ye gtecaw Wee, SESE Cee tet Z e Ak Sc Hae we BY Fawn BX ae A ; CY ars Ce Ad) eqs * ws v <9 Ms Vo eG Reis eid reese wi) RA 100FM 20K¥ Fig. 4. Deeveya spiralis, SEM micrographs of left valve of holotype: a, Lateral view; b, Anterior view; c, Ventral view; d, Central area showing places where adductor muscles attach; e, Anterior view in vicinity of anteroventral concavity, from b; f, Detail from e. Second antenna (Fig. 9a): Protopodite with long lateral spines forming 2 groups (1 proximal, 1 distal), and minute medial spines along dorsal margin. Endopodite 3-segmented: First segment with 2 spinous bristles on distal dorsal protuberance; second segment with 4 bristles (1 short, dorsal, spinous bristle and 3 long filament- like bristles with short widely spaced marginal spines), and short spine-like lateral bristle near dorsal margin of terminal ventral pedestal. Third segment with 2 long terminal bristles bearing widely spaced marginal spines (1 of the bristles filament- VOLUME 98, NUMBER 2 483 aN Roeres x7 an Fig. 5. Deeveya spiralis, SEM micrographs of left valve of holotype: a, Posteroventral corner, from Fig. 4a; b, Surface near anterior, from Fig. 4a; c, Surface near anterior showing fracture, from Fig. 4c; d, Surface at ventral margin, from Fig. 4c; e, Detail from a; f, Detail of protuberances shown in e. like). Second and third segments forming right angle with first segment on holotype and paratype. Exopodite 9-segmented: First segment weakly divided into long proximal and short distal parts; proximal part with minute dorsal spines; distal part with slender, bare, medial bristle reaching past distal end of fifth segment. Segment 2 with long bristle bearing ventral spines and natatory hairs; joints 3-8 each with long bristle with natatory hairs; ninth joint with 4 bristles (2 short with marginal spines, 1 medium length with ventral marginal spines, 1 long with ventral 484 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig.6. Deeveya spiralis, SEM micrographs of left valve of holotype; a, Anterior muscle scar forming lower row in Fig. 4d; b, Fossae and minute protuberances near adductor muscles, from Fig. 4d; c, Detail from upper left of a; d, Detail of pore near middle of muscle scar shown in a; e, Detail of protuberance, from middle of Fig. 5b; f, Internal fibers perpendicular to shell surfaces, from fracture shown in Fig. Sc. spines and distal natatory hairs); segments 2—8 decreasing in length distally along stem only slightly; ninth segment about *% length of segment eight; segments 2-8 without basal spines or distal hairs or spines. Mandible (Fig. 10): Coxal endite with proximal (also anterior) distal (also pos- terior) sets of teeth separated by small space (Fig. 10a, b): proximal (anterior) set VOLUME 98, NUMBER 2 485 Fig. 7. Deeveya spiralis, a, Inside view of left valve of holotype showing infold, list on infold, and some muscles (main cluster of adductor muscles indicated by oval); b, Detail from a showing anterior concavity and bifurcate marginal bristles; c, Body of holotype from right side (1, 1st antenna; 2, 2nd antenna; 3, organ of Bellonci; 4, mandible; 5, maxilla; 6, 5th limb; 7, 6th limb; 8, 7th limb; 9, right lamella of furca; 10, hepatic appendage; 11, gut; 12, unknown receptacle; 13, central adductor muscle). consisting of 4 broad teeth plus short, stout, spinous bristle near space; densely packed spines present between each tooth and anterior to anterior tooth and posterior to posterior tooth; slender spines also on medial surface proximal to teeth; distal (posterior) set of coxal teeth consisting of 2 terminal rows with 5 broad teeth in inner row and 6 in outer row (posterior tooth of inner row longer than others). Two lists present proximal and medial to inner row of distal coxal teeth; anterior of these consisting of anterior pectinate bristle-like tooth and pos- terior stout tooth; posterior list consisting of anterior pectinate bristle-like tooth 486 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON eS zs I Y{ sores Fig. 8. Deeveya spiralis, holotype: a, Right 1st antenna; b, Anterior of body from left showing organ of Bellonci, part of left 1st antenna, and left hepatic appendage (with black spots); c, Anterior of body from right showing organ of Bellonci and first two segments of left 1st antenna; d, Upper and lower lips from right side; e, Unknown receptacle removed from right side of posterior part of gut. 487 VOLUME 98, NUMBER 2 \ OM Pree Nr tea, NM lh 1] SN ta SoIMyyp — ray Dou Fig. 9. Deeveya spiralis, holotype: a, Endopodite and part of protopodite of nght 2nd antenna; b, Right maxilla (endites not shown). and long curved tooth with rounded tip. Basal endite with 6 terminal non-serrate cusps (appearing worn); posterior cusp smaller than others and separated from them by space (Fig. 10b, c); posterior margin of endite not separated by suture from second segment but with single proximal bristle and distal, blunt, unringed bristle; anterior margin with single spinous bristle; medial surface with long hairs forming proximal cluster near posterior margin, and also forming distal rows; 488 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Si an eB! 23 2s yy x } basis “/1)) )) yy fy ** AmONY / Fig. 10. Deeveya spiralis, right mandible of holotype; a, Medial view of distal ends of coxa and basis (not under cover slip); b, Basis and endopodite, medial view; c, Distal ends of coxa and basis (under cover slip). lateral surface with 7 bristles (2 stouter than others and twisted together on both limbs of holotype and paratype); proximal lobe on medial surface with 2 hirsute bristles; dorsal margin of basis with long hirsute bristle. Endopodite 3-segmented with first segment about twice length of second and third segments; first segment VOLUME 98, NUMBER 2 489 ss, LH aa a ) d/ N S Das Deeveya spiralis, holotype, limbs of right side: a, 5th limb; b, 6th limb; c, 7th limb; d, Fig. 11. Right lamella of furca and single posterior bristle. with 1 spinous anterior bristle, 1 spinous posterior bristle (behind basis on illus- trated limb), and 5 medial bristles; second joint with medial bristle near distal posterior corner, and 3 terminal bristles at distal anterior corner (1 of bristles stout, claw-like); third joint hirsute medially and along anterior margin, with 5 490 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 12. Deeveya spiralis, holotype: a, Photograph of posterior of body showing unknown receptacle (arrow); b, Phase contrast photograph of part of unknown receptacle showing coils of thin filaments. Photographs by Dr. Robert P. Higgins. distal medial bristles forming row, and 3 stout terminal bristles (middle and anterior of these with short marginal spines and with beak-like tip; posterior bristle with longer spines distally and with linear pointed tip). Maxilla (Fig. 9b): Endites well developed and with numerous bristles (obscure on mounted appendage and not completely shown in illustration). Basis with 1 hirsute dorsal bristle and 2 ventral bristles (1 lateral, 1 medial). Endopodite: First segment with 4 spinous bristles on or near anterior margin, and 5 bristles at distal posterior corner; second segment with 2 stout claws, 6 slender bristles, and long hairs on anterior surface. Fifth limb (Fig. 11a): Epipodial appendage with plumose bristles forming 3 groups, each with 5 bristles. Protopodite and endopodite with total of 28 bristles including 2 claw-like pectinate bristles at ventral margin of knee. Exopodite 3-seg- mented: First segment with 2 distal dorsal bristles (longest of these with minute widely spaced marginal spines, other with long marginal hairs), 2 lateral bristles (distal of these plumose), 1 medial bristle near middle, 4 proximal medial bristles with bases near ventral margin, and 4 distal bristles with bases on or near ventral margin; suture separating first and second segments more defined on medial side; elongate second segment with 4 bristles (1 dorsal, 3 ventral); small third joint with 2 stout pectinate claw-like bristles and 3 slender ringed bristles (bases of slender bristles medial to claw-like bristles). Sixth limb (Fig. 11b): Epipodial appendage with plumose bristles forming 3 groups having 7 bristles in proximal group, 5 in distal group, and 6 in middle group. Protopodite divided distally by suture evident only on medial side; prox- VOLUME 98, NUMBER 2 491 Fig. 13. Deeveya spiralis, holotype: photograph of elliptical organ (heart?) posterior to organ of Bellonci. Note network on surface. Photograph by Dr. Robert P. Higgins. Length of scale bar 0.10 mm. imal segment with 4 bristles on or near ventral margin (all but 1 bristle plumose); distal segment with 2 ventral bristles with short marginal spines and 2 stouter medial plumose bristles. Exopodite 4-segmented: First segment with plumose lateral bristle near middle and 6 plumose bristles on or near ventral margin; process with 4 bristles (longest of these with short, faint, marginal spines, others plumose) present on distal dorsal corner of first segment; second segment with 4 bristles on or near ventral margin (all bristles with short marginal spines); third segment with 3 bristles (2 ventral, 1 dorsal); fourth segment with 2 long, stout, pectinate claw-like bristles, 2 slender bristles ventral to claw-like bristles, and 1 small bristle between and lateral to claw-like bristles. Seventh limb (Fig. 11c): Limb with 1 segment or 2-segmented: First segment elongate, bare; second segment wtih 3 terminal bristles (1 long, 2 shorter, all with short, widely spaced, marginal spines). Caudal furca (Fig. 11d): Each lamella with total of 7 pairs of claws followed by unpaired dorsal bristle; claws 1—4 with faint teeth along dorsal margins; claws 5— 7 bristle-like with teeth along both margins (teeth stouter near midlength); teeth along margins of claw 7 smaller than those on claws 5 and 6; unpaired dorsal bristle about same length as claw 1, and with marginal spines; left lamella of furca of holotype slightly anterior to right lamella, but reverse relationship on paratype; ventral edge of lamellae between claw 7 and unpaired dorsal bristle with minute spines; minute spines also present on distal medial surface of lamellae. Organ of Bellonci (Fig. 8b, c): Well-developed, cone-shaped, bifurcate distally, with tips of branches tapering to point. Lips (Fig. 8b, d): Upper lip with about 6 glandular openings along ventral face and spines along posterior edge. Lower lip with triangular processes on each side and anterior spines. 492 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Posterior of body (Fig. 7c): Evenly rounded, unsegmented. Hepatic appendage (lumen) (Figs. 7c, 8b): Paired, elongate, tapering to narrow opening at esophagus near mouth, and containing many minute brown bodies. Heart: Ellipsoid, just posterior to Bellonci organ of holotype, covered by unusual and unidentified network (Fig. 13) may be a heart. Unknown receptacle (Figs. 7c, 8e, 12): Oval body to right of gut in posterior part of body; with pearly sheen in reflected light; receptacle packed with coils of long thin thread-like filaments. Eggs: Both holotype and paratype with unextruded eggs that vary in size within same specimen. Remarks. —Skogsberg (1920:560) referring to the seventh limb of halocyprids, which generally bears two bristles, noted “only in exceptional cases, in single specimens, are three bristles found.’ In the present species, three bristles were found on both limbs of the holotype and paratype, indicating that three bristles are normal. The basal endite of the mandible of this species bears two stout lateral bristles that are twisted around each other (Fig. 8c). This occurred on all four limbs of the two specimens on hand, suggesting that this unusual morphological character is normal for the species. The “unknown receptacle” was initially thought to be a seminal receptacle and the coiled filaments spermatozoa. In order to verify that the filaments were spermatozoa the receptacle was sectioned with a microtome and the sections stained with hematoxylin and eosin. This treatment failed to stain the filaments, showing them not to be spermatozoa. The receptacle wall has nuclei and appears to be epithelium. No cells or nuclei were present inside the receptacle or within the filaments. The filaments were not identified. It is not known if the receptacle is part of the genitalia, gut, or unrelated to either. Because only two specimens are on hand it can not be ascertained if the organ is present on all specimens of the species, and not a foreign inclusion. Acknowledgments Collection of specimens from caves in the Turks and Caicos Islands was sup- ported in part by a National Science Foundation Grant (BSR 8215672) to Thomas M. Iliffe. The collectors thank Paul and Shirley Hobbs for arranging accommo- dations and, along with Jill Yager, helping with cave location and collections, and Dennis Williams for flights to various islands as well as assisting with field col- lections. We thank Carolyn Gast for rendering the shaded drawing and detail drawing of Deveeya spiralis and Kathryn Schroeder Brown for assisting in prep- aration and inking of appendages of the same species. The SEM micrographs were made by Walter Brown, Smithsonian Institution. We also thank Linda Cullen, Smithsonian Institution Tumor Registry, for sectioning the ““unknown receptacle” of D. spiralis. This paper is contribution number 1016 of the Bermuda Biological Station for Research. Literature Cited Bowman, T. E., J. Yager, and T. M. Iliffe. 1985. Speonebalia cannoni, n. gen., n. sp., from the Caicos Islands, the first hypogean leptostracan (Nebaliaceae: Nebaliidae).— Proceedings of the Biolog- ical Society of Washington 98(2):435—442. VOLUME 98, NUMBER 2 493 , T. M. Iliffe, and J. Yager. 1984. New records of the troglobitic mysid genus Stygiomysis: S. clarkei, new species, from the Caicos Islands, and S. holthuisis (Gordon) from Grand Bahama Island (Crustacea: Mysidaceae).— Proceedings of the Biological Society of Washington 97(3): 637-644. Buden, D. W., and D. L. Felder. 1977. Cave shrimps in the Caicos Islands.— Proceedings of the Biological Society of Washington 90(1):108-115. Danielopol, D. L. 1972. Sur la presence de Thaumatocypris orghidani n. sp. (Ostracoda, Myodo- copida) dans une grotte de Cuba.—Compte Rendu hebdomadaire des Séances de l’Académie des Sciences, Paris 247:1390-1393. 1976. Comparative morphology of the deep-sea Thaumatocypris echinata G. W. Muller, 1906, and the cave species Thaumatocypris orghidani Danielopol, 1972 (Ostracoda, Myodo- copida).— Vie et Milieu 26(1C):9-20. Dietz, R. S., J. C. Holden, and W. P. Sproll. 1970. Geotectonic evolution and subsidence of Bahama Platform.—Bulletin of the Geological Society of America 81(7):1915-1928. lliffe, T. M., C. W. Hart, Jr., and R. B. Manning. 1983. Biogeography and the caves of Bermuda. — Nature 302 (5904):141-142. —, H. Wilkens, J. Parzefall, and D. Williams. 1984. Marine lava cave fauna: Composition, biogeography, and origins.—Science 225:309-311. Sket, B., and T. M. Iliffe. 1980. Cave fauna of Bermuda.—Internationale Revue der Gessampten Hydrobiologie 65(6):87 1-882. Skogsberg, T. 1920. Studies on marine ostracods, 1: Cypridinids, halocyprids, and polycopids.— Zoologiska Bidrag fran Uppsala, supplement, 1:1—784, 153 figures. Stock, J. H., and J. J. Vermeulen. 1982. A representative of the mainly abyssal family Pardaliscidae (Crustacea, Amphipoda) in cave waters of the Caicos Islands.—Bijdragen tot de Dierkunde 52(1):3-12. Warner, G. F., and C. A. M. Moore. 1984. Ecological Studies in the marine blue holes of Andros Island, Bahamas.— Transactions of the British Cave Research Association 1 1(1):30-44. (LSK) Department of Invertebrate Zoology, National Museum of Natural His- tory, Smithsonian Institution, Washington, D.C. 20560; (TMI) Bermuda Biolog- ical Station for Research, Ferry Reach 1-15, Bermuda. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 494-510 QUADRYOPS, NEW GENUS, AND THREE NEW SPECIES OF ARBOREAL DRYOPIDAE (INSECTA: COLEOPTERA) FROM PANAMA AND ECUADOR Philip D. Perkins and Paul J. Spangler Abstract.—A new genus, Quadryops (Coleoptera: Dryopidae), with three new species, QO. chrysosetosus, O. quasimodoi, and Q. obtusetosus are described. Two of the species are from Panama: Q. chrysosetosus from the canopy of a lowland Leuhea seemanni forest (100 meters) and Q. quasimodoi from forest floor litter and wood chips from a montane logging area (1720 meters). The third species, Q. obtusetosus, was collected in Ecuador by sweeping forest vegetation at night (300 meters). The species are illustrated by scanning electron micrographs and line drawings. A key to the species and a tabular comparison of the states of selected taxonomic characters are given. Members of Quadryops, new genus, described below are unique among known Dryopidae in that the tarsi have four articles; all other described dryopid species have five articles. The most obvious diagnostic feature of members of Quadryops, however, is the shape of the pronotum which appears to be inflated due to the lobate discal area (Fig. 3). Development of a median longitudinal depression on the pronotum, and hence division of the pronotal prominence into two lobes, varies among the three species of Quadryops now known. Members of Quadryops have two types of pubescence. Most of the body is clothed in indumentum consisting of aciculate hairlike setae whose density varies among the species. This indumentum abrades moderately easily, leaving distinc- tive punctures (termed micropunctures in the descriptions that follow; Fig. 39). In addition to indumentum, the body has longer, sparser, erect or suberect setae which differ in form among the three species. In Q. chrysosetosus, these setae are acute at the apex, enlarged subapically and have ridges (Fig. 19); erect setae of QO. obtusetosus are blunt and enlarged apically and have ridges (Figs. 6, 21); in the third species currently known, Q. quasimodoi, the holotype is badly abraded and the few remaining erect setae are blunt but not enlarged apically (Fig. 46). These erect setae emerge from punctures much larger than those of the indumentum; punctures (termed macropunctures in the descriptions that follow) are round and flat-bottomed on the elytra of Q. obtusetosus, and setal sockets fit closely around the setae (Figs. 20, 21). The elytral macropunctures of Q. chrysosetosus, contrast- ingly, are round-bottomed and have elongate sockets which would allow forward and backward movement of the setae but restrict side to side motion (Fig. 18). Metathoracic wings are fully developed in the three species of Quadryops. The wings of Q. chrysosetosus and Q. quasimodoi are similar in that both have a well- developed vein 2A,, whereas wings of Q. obtusetosus lack that vein (Figs. 30, 31, 32). The antennae of all three species have 11 articles and well-developed sensilla in both sexes. The sensilla are of two types, simple and dendritic (Figs. 26, 28, VOLUME 98, NUMBER 2 495 29). A comparison of the states of selected characters of the three species is given in Table 1. Members of this genus have been very rarely collected; presently only a single specimen of each species is represented in collections. Two species were collected in arboreal habitats: Q. chrysosetosus was collected in Panama during a pyrethrin fogging experiment in the canopy of a lowland forest of Luehea seemanni (100 meters); Q. obtusetosus was collected in Ecuador by sweeping forest vegetation at night. Quadryops quasimodoi may be a forest floor species, as the single known specimen was collected in Panama by berlese extraction of concentrated forest floor litter and wood chips from a logging area (1720 meters). However, a vast amount of debris from the canopy accumulates on the forest floor of logging areas and it is possible that Q. quasimodoi is an arboreal species like its congeners. Although the habits of dryopid beetles are poorly known, most genera are known to inhabit semiaquatic habitats. However, several genera such as Geoparnus Be- suchet, Sostea Pascoe, Protoparnus Sharp, Oreoparnus Deleve, and Quadryops n. gen. are terrestrial beetles. Because very little is known about dryopids, collectors are urged to record habitats and all biological data possible when the beetles are found. Quadryops, new genus Type-species.— Quadryops chrysosetosus, new species. Description. — Body form oblong, robust, markedly convex dorsally; integument with indumentum of moderately long setae that vary from dense to moderately sparse; dorsally and sometimes ventrally with distinctive suberect setae which may be acute at the apex and thickened subapically, blunt and thickened apically, or blunt and not thickened apically (Figs. 6, 19, 46). Head retractile; eyes well developed, widely separated, pubescent. Clypeus expanded laterally. Labrum par- tially or completely concealed beneath clypeus in dorsal view. Antennomeres 11, with setae and sensilla as illustrated (Figs. 26, 28, 29); head concave between raised and widely separated antennal acetabulae. Labial palpomeres 3 (Fig. 33). Maxillary palpomeres 4 (Figs. 34, 35). Pronotum with discal area very convex; apical *s—*%4 lobate with distinctly or indistinctly developed median longitudinal depression which divides prominence into a lobe on each side (Figs. 5, 36); posterior part of pronotum broad lateral to elevated disc, subtriangular and slightly concave (Figs. 4, 36); lateral margin crenulate, sometimes angulate; prescutellar emargination well developed. Elytra width at base equals basal width of pronotum, parallel-sided, very convex; each elytron with nine more or less developed striae; Margin sinuate in lateral view, basal half depressed to form vertical side of elytron (Fig. 2); epipleuron horizontal, except nearly vertical at subacute elytral apex. Scutellum large, produced anteriorly. Metathoracic wings present. Prosternum long in front of procoxae; process carinate, apex deflexed and inserted into deep mesosternal fovea; short longitudinal carina in front of procoxae on each side of prosternal process (Fig. 41). Trochantin exposed. Mesosternum moderately raised between mesocoxae. Distance separating mesocoxae greater than or less than that separating metacoxae. Metasternum with median longitudinal sulcus which may Or may not be extended forward onto intercoxal process. Metacoxae moderately excavated to receive metatibiae. Legs short; tibiae slightly longer than femora; 496 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Figs. 1-6. Quadryops obtusetosus: 1, Habitus, dorsal view; 2, Habitus, lateral view; 3, Head and prothorax, oblique view; 4, Head, pronotum and anterior region of elytra, dorsal view; 5, Head and prothorax, anterior view; 6, Apex of pronotal seta. tarsus very short, about '4 length of tibia, with four articles, combined lengths of articles 1, 2 and 3 approximately equal to length of article 4 (Fig. 23). Etymology.—Greek quad (four), plus dryops (referring to family name). The tarsi of members of Quadryops have four articles. Quadryops chrysosetosus, new species Figs. 10-19, 22, 26, 28, 32, 35 Type-data.—Holotype male: Panama, Canal Zone, 5.0 mi. NW Gamboa, 09°10'N, 079°45’W, 100 meters; canopy pyrethrin fogging experiment in Luehea seemanni; sample 1—2a, 12 Jul 1976, Montgomery and Lubin collectors. Deposited in the National Museum of Natural History, Smithsonian Institution; type no. 100893. Description. — Holotype male: 2.60 mm long, 1.44 mm wide. Body form oblong, VOLUME 98, NUMBER 2 497 Figs. 7-9. Quadryops obtusetosus: 7, Habitus, ventral view; 8, Prosternum; 9, Antenna and as- sociated head and prosternal structures. markedly convex (Figs. 10, 11). Color brown; dorsum slightly darker than venter and legs. Integument with light brown indumentum; dorsal surface also with suberect setae about 0.16 mm long, each seta enlarged subapically and with pointed apex (Fig. 19). Head: 0.60 mm long, 0.70 mm wide, 0.50 mm wide between eyes. Eyes large, 498 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Figs. 10-13. Quadryops chrysosetosus: 10, Habitus, dorsal view; 11, Habitus, lateral view; 12, Habitus, ventral view; 13, Head, pronotum and anterior region of elytra, dorsal view. hemispherical, pubescent. Frons and base of clypeus transversely confluently con- cave; frons markedly punctate. Clypeus in dorsal view with anterior margin straight in middle '4, arcuate laterally, middle % straight in anterior view; lateral angles deflexed. Labrum small, concealed beneath clypeus in dorsal view; anterior margin VOLUME 98, NUMBER 2 499 shallowly emarginate. Antennomeres 11 (Figs. 26, 27). Last labial palpomere flat, broad; width equal to 2 length. Maxillary palpomeres 2 and 3 each 4 length of palpomere 4 (Fig. 35). Pronotum: 0.60 mm long; width 1.32 mm basally, 0.80 mm apically. Apical %; of pronotal disc markedly inflated, very slightly depressed longitudinally on midline to form lobe on each side. Posterior area lateral to elevated disc triangular, slightly concave; with shiny integument, except alutaceous lateral border. Anterior margin of pronotum with short shelf in front of discal lobes; lateral margin cren- ulate. Short pubescence more developed on reliefs than in area at margin of raised disc. Moderately markedly punctate, especially on inflated disc where punctures separated by 1-2 their width. Integument shiny on disc, dull between disc and lateral margin. Posterior margin trisinuate, contiguous with elytral base, angularly emarginate to receive scutellum; small depression in front of prescutellar emar- gination. Scutellum more markedly angulate anteriorly than posteriorly. Elytra: 1.92 mm long, 1.44 mm wide; parallel sided; markedly declivous laterally to form vertical sides; disc transversely and longitudinally convex. Each elytron with irregular, coarse, dense, setiferous punctures larger than those on pronotal disc; most punctures in 9 shallow, indistinct striae; punctures of striae 7-9 on vertical side of elytron very dense, some separated by narrow ridges. Basal tenth of interval 5 (humeral) irregularly cariniform. Lateral margin, in side view, mark- edly sinuate (Fig. 11). Stria 9 (marginal) well developed. Stria 1 more markedly developed in basal third than on remainder of elytron. Elytral apices subacute. Prosternum: 0.66 mm long; with moderately dense indumentum and sparse, erect, longer hairs (these hairs not expanded subapically as those on dorsum). Prosternal process 0.26 mm long, 0.14 mm wide at base; carinate, carina extended anteriorly onto disc; slightly convex in lateral view; apex inserted into deep me- sosternal fovea. Mesosternum with sides of median fovea raised slightly. Meta- sternum with median longitudinal depression; intercoxal process width less than length, raised slightly above plane of disc, median longitudinal depression well developed; metepisternum without basomedial carina; integument densely in- dumentose, also with large punctures, each puncture with a large suberect seta; large punctures in a marginal row at base of metasternum, absent from small sublateral area in front of marginal row; remainder of metasternum randomly punctate, punctures separated by about 2-3 x their diameter; basal margin lobate on each side of midline. Abdomen: Midline length ratios of sterna: 2/1.2/1/1/3. Intercoxal process length equal to width at base, length about equal to length of remainder of sternum; narrowest separation of metacoxae slightly greater than that of mesocoxae. Ster- num 5 with apical half raised on midline to form low ridge. Integument indu- mentose as metasternum. Macropunctures on intercoxal process as large as and as dense as those on metasternum; remainder of macropunctures on sterna denser and smaller. Each macropuncture with suberect seta. Legs: Densely indumentose. Tarsi densely pubescent beneath; each slightly more than '4 length of respective tibia. Protibia with apical half slightly flattened lat- eromedially, widest near apical third. Metacoxae coarsely punctate similarly to metasternum; metatrochanter globose, size slightly smaller than median third of coxa; metafemur with upper surface of distal end angulate. 500 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Figs. 14-17. Quadryops chrysosetosus: 14, Head and prothorax, anterior view; 15, Head and prothorax, oblique view; 16, Prosternum, ventral view; 17, Prosternum, oblique view. Genitalia: Aedeagus as illustrated (Fig. 22). Female unknown. Distribution. — Presently known only from the type-locality; Canal Zone, Pan- ama. Etymology. —chrysosetosus, from Chrysopidae and setosus. The prominent sub- erect setae of the body are expanded subapically, reminiscent of chrysopid eggs mounted on narrow stalks. VOLUME 98, NUMBER 2 501 Figs. 18-21. 18, Quadryops chrysosetosus, disc of elytron; 19, Q. chrysosetosus, apex of pronotal seta (800 x); 20, Q. obtusetosus, disc of elytron; 21, Q. obtusetosus, pronotal setae. Quadryops obtusetosus, new species Figs. 1-9, 20, 21, 24, 27, 29, 31, 33, 34 Type-data.— Holotype female: Ecuador, Pastaza Province, Ashuara Rio Mac- uma, 10 km from Rio Morona, 300 meters, forest night sweep, 7-16 Jul 1971, coll. B. Malkin. Deposited in the Field Museum of Natural History, Chicago. Description.— Holotype female: 3.20 mm long, 1.76 mm wide. Body form ob- long, markedly convex (Figs. 1, 2). Color brown. Integument with indumentum and prominent, suberect, blunt setae. Head: 0.74 mm long, 0.80 mm wide, 0.56 mm between eyes. Eyes large, hemi- spherical, pubescent. Frons slightly concave between eyes; apex of frons and base of clypeus confluently concave between antennal acetabulae; frontoclypeal suture distinct in middle of concavity; frons with punctures coarse, dense, deep; inter- stices with narrow walls, punctures absent from small median area. Clypeus in 502 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Figs. 22-25. 22, Quadryops chrysosetosus, aedeagus, dorsal and lateral views; 23, Q. quasimodoi, pro- (top), meso- and metatarsi (microslide preparation; pubescence omitted); 24, Q. obtusetosus, Ovipositor, lateral and dorsal views; 25, Q. quasimodoi, ovipositor, lateral and dorsal views. (Scale lines equal 0.1 mm.) VOLUME 98, NUMBER 2 503 he Vf Y/ y KA CR 26 See GT (0CG Figs. 26-29. 26, Quadryops chrysosetosus, antenna (some simple setae omitted); 27, Q. obtusetosus, antenna (all setae omitted); 28, Q. chrysosetosus, antennomere 7; 29, Q. obtusetosus, antennomere 7. (Scale lines equal 0.1 mm.) 504 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON dorsal view with anterior margin nearly straight in middle third, more arcuate laterally; in anterior view, middle % slightly arcuate dorsally; lateral angles deflexed very slightly. Labrum small, concealed beneath clypeus in dorsal view; anterior margin shallowly and broadly emarginate. Antennomeres 11 (Fig. 27). Labial palpomeres 1-3, length ratios 2/5/7 respectively; palpomere 3 flat, broad, width slightly greater than 12 length (Fig. 33). Maxillary palpomeres 2 and 3 of equal length; palpomere 4 equals 2.5 x length of palpomere 3 (Fig. 34). Pronotum: 0.80 mm long; width 0.96 mm apically, 1.52 basally. Apical 74 of disc markedly inflated, longitudinally depressed on midline to form a lobe on each side. Posterior area lateral to elevated disc triangular, concave; punctures less dense than those on disc. Lateral half alutaceous. Basal border very shiny; anterior border in form of narrow shelf in front of discal lobes, shelf widened laterally. Lateral margin sinuate, coarsely crenulate. Integument of inflated disc with dense, deep, coarse punctures; some punctures separated by narrow walls; interpunctal areas dull due to fine, dense indumentum; punctures less dense behind inflated lobes. Posterior margin markedly trisinuate, contiguous with elytral base; prescutellar emargination with small median process in front of which is a small depression. Scutellum ovoid. Elytra: 2.40 mm long, 1.76 mm wide. Lateral margins nearly parallel; markedly declivous laterally to form vertical sides. Disc transversely and longitudinally convex. Each elytron with 9 striae; stria 9 (marginal) well developed; stria 1 more markedly developed in basal fourth than on remainder of elytron; intervals sub- costate; sutural interval raised in basal fourth; intervals 6 and 7 (humeral) confluent and raised in basal tenth; striae on disc with shallow, non-serial, flat-bottomed, setiferous punctures separated by about their diameter and about equal in size to punctures on pronotal disc; punctures of striae 5—9 denser, deeper and coarser than those on disc, some confluent. Integument dull between punctures due to extremely minute spicules (Fig. 20). Lateral margin sinuate in side view (Fig. 2). Elytral apices subacute. Prosternum: 0.90 mm long. With short, dense indumentum and sparse, sub- erect, blunt setae; short basal carina in front of procoxae. Prosternal process 0.40 mm long, 0.16 mm wide at base; markedly carinate, carina extended anteriorly onto disc; markedly convex in apical half; apex inserted into deep mesosternal fovea. Mesosternum with sides of median fovea raised slightly. Metasternum with median longitudinal sulcus confluent anteriorly with irregular fovea at base of intercoxal process; intercoxal process narrow at apex, meso-metasternal suture not apparent; metepisternum with moderately developed basomedial carina. In- tegument indumentose and with large, shallow, setiferous punctures separated by about 1-2 x their diameter; punctures in marginal row at base of metasternum, otherwise random. Basal margin lobate on each side of midline. Abdomen: Midline length ratios of sterna: 2.2/1.2/1/1/3.4. Intercoxal process length equal to basal width; length nearly 7% total length of sternum; narrowest separation of metacoxae slightly greater than that of mesocoxae. Sternum 5 with apical third raised on midline, with narrow apicomedial emargination. Integument moderately sparsely indumentose except in macropunctures which are prominent and shallow; those macropunctures on sterna | and 2 twice size of metasternal punctures, punctures becoming smaller posteriorly; punctures on sternum 5 equal VOLUME 98, NUMBER 2 505 Figs. 30-35. 30, Quadryops quasimodoi, wing; 31, OQ. obtusetosus, wing; 32, QO. chrysosetosus, wing; 33, Q. obtusetosus, labial palpus; 34, QO. obtusetosus, maxilla (setae omitted); 35, QO. chrysosetosus, maxilla (setae omitted). (Scale lines equal 0.1 mm for mouthparts and 1.0 mm for wings.) 506 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Figs. 36-39. Quadryops quasimodoi: 36, Pronotum; 37, Head, dorsal view; 38, Elytra, dorsal view; 39, Micro- and macropunctures of pronotum (900 x). in size to metasternal punctures, punctures not uniformly arranged, separated by 0.5—2 x their diameter; most punctures with prominent, suberect, blunt seta. Legs: Densely indumentose; all segments except tibiae and tarsi with sparse, suberect, blunt setae. Tarsi moderately densely pubescent beneath; each about 14 length of respective tibia. Protibia widest near apex. Metacoxa coarsely punctate similarly to metasternum; metatrochanter globose, size slightly smaller than me- dian third of coxa; metafemur with upper surface of distal end angulate. Genitalia: Ovipositor as illustrated (Fig. 24). Male unknown. Distribution. — Currently known only from the type-locality on the eastern slope of the Andes in southern Ecuador. Etymology. —Latin, obtuse (blunt) plus setosus. This epithet refers to the prom- inent, suberect, blunt setae on the body. Quadryops quasimodoi, new species Figs. 23, 25, 30, 36-47 Type-data. —Holotype female: Panama, Chiriqui Province, ““Barca”’ area, Finca Lerida nr. Boquete, 5650 feet; berlese (B-487), concentrated forest floor litter and VOLUME 98, NUMBER 2 507 Figs. 40-43. Quadryops quasimodoi: 40, Head, ventral view; 41, Prosternum; 42, Mesothorax, metathorax and abdomen, ventral view; 43, Mesothorax and metathorax, ventral view. wood chips in logging area, 14 Mar 1959, coll. H. S. Dybas. Deposited in the Field Museum of Natural History, Chicago. Description.— Holotype female: 3.88 mm long, 1.92 mm wide. Body form ob- long, markedly convex (Figs. 36, 38). Color brown. Integument with indumentum and prominent, suberect setae. Head: 0.80 mm long, 0.98 mm wide, 0.70 mm between eyes. Eyes large, hemi- spherical, pubescent. Frons slightly concave between eyes; apex of frons concave between raised margins of antennal acetabulae. Indumentum sparser on small median area of frons than on remainder; macropunctures moderately coarse and dense, separated by 1-2 their diameter, absent from small median area. La- broclypeal suture subsulcate due to concavity of frontal apex and raised base of clypeus. Clypeus in dorsal view with anterior margin arcuate laterally, less so in middle 3; in anterior view, very shallowly emarginate in middle 4; lateral angles reflexed. Labrum directed ventrad, concealed beneath clypeus in dorsal view, broadly emarginate apically. Antennomeres 11 (Fig. 47). Labial palpomeres 1-3 with length ratios about 2/5/7 respectively; palpomere 3 flat, broad, width slightly 508 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Figs. 44-47. Quadryops quasimodoi: 44, Protarsus; 45, Protarsal claws (450 x); 46, Apex of pro- notal seta (2300 x); 47, Antenna (200 x). greater than length (Fig. 40). Maxillary palpomeres 2 and 3 of equal length; each ¥3 length of palpomere 4 (Fig. 40). Pronotum: 0.80 mm long; width 1.08 mm apically, 1.64 mm basally. Apical ¥4 of disc very markedly inflated, longitudinally depressed on midline to form lobe on each side. Posterior area lateral to elevated disc triangular, slightly concave, punctures less dense than those on disc; lateral border alutaceous, basal border very shiny. Anterior border in form of narrow shelf in front of inflated lobes, shelf widened laterally. Lateral margin extended slightly obliquely to midline from apical angle to midlength then markedly angulate to form wide base, margin coarsely crenulate (Fig. 36). Integument of inflated disc with sparse, round, small punctures separated by 2—5 x their diameter; interpunctal areas with very fine, dense micropunctures of indumentum (abraded on holotype). Posterior margin markedly trisinuate, contiguous with elytral base; prescutellar emargination ar- cuate. Integument in front of emargination raised in form of inverted V. Elytral base diverging anterior of midlength of scutellum. Scutellum spindle shaped; disc raised slightly. Elytra (Fig. 38): 3.00 mm long, 1.92 mm wide. Lateral margins almost parallel; markedly declivous laterally to form vertical sides. Disc transversely and longi- tudinally convex. Each elytron with 9 irregular striae. Stria 1 (sutural) deeply impressed in basal third; basal third of sutural interval markedly raised. Stria 2 nearly obsolete. Striae with irregular, non-serial punctures; punctures of striae 1 and 2 slightly larger than those on pronotal disc; punctures becoming larger and VOLUME 98, NUMBER 2 Table 1.—Character comparison of Quadryops species. 509 chrysosetosus quasimodoi obtusetosus Character (male) (female) (female) Base of clypeus depressed not depressed depressed Lateral angles of clypeus deflexed reflexed deflexed slightly Pronotal punctation Pronotal mid-longitudinal impression Fraction of pronotum lo- bate Median process of pro- notal prescutellar emar- gination Lateral margin of pro- notum Elytral striae Elytral integument be- tween punctures Suberect setae of dorsum Median third of proster- nal carina Metasternal intercoxal process raised above plane of disc Basomedial carina of metepisternum Ratios of lengths of ab- dominal sterna Vein 2A, of wing Size (mm) moderately coarse and dense faintly developed 2/3 absent not angulate moderately devel- oped smooth and shiny hastate straight distinctly absent 2/1.2/1/1/3 present 2.60 x 1.44 fine and sparse well developed % absent markedly angulate moderately devel- oped smooth and shiny obtuse apically angulate slightly well developed 2.3/1.2/1/1/3.3 present 3.88 x 1.92 very coarse and dense well developed 2 present moderately angulate well developed, intervals subcostate asperate and dull obtuse and expanded apically straight distinctly moderately developed 2.2/1.2/1/1/3.4 absent 3.20 x 1.76 deeper laterally; those of striae 5-8 very deep and subconfluent. Intervals 5-8 irregularly subcostate; intervals 6-8 fused in basal fifth to form well-developed humeral ridge; medial margin of interval 6 in form of a low carina. Lateral margin sinuate in side view. Elytral apices subacute. Prosternum (Fig. 41): 0.92 mm long; densely indumentose. Prosternal process 0.46 mm long, 0.20 mm wide at base; markedly carinate, width of carina at midlength subequal to width of shelf separating carina from procoxa; in lateral view, markedly angulate near midlength, this angle about equal to angle formed where prosternal carina joins prosternal disc; apex inserted into deep mesosternal fovea. Mesosternum with posterior margins of median fovea not raised. Meta- sternum with median longitudinal depression well developed on intercoxal pro- cess; intercoxal process width equal to its length; metepisternum with well-de- veloped basomedial carina; disc almost devoid of macropunctures, especially 510 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON laterally; basomarginal groove with macropunctures; basal margin lobate on each side of midline. Abdomen: Midline length ratios of sterna: 2.3/1.2/1/1/3.3. Intercoxal process length slightly greater than basal width, length nearly *%4 total length of sternum. Narrowest separation of metacoxae slightly less than that of mesocoxae. Sternum 5 with apical %4 raised on midline; with narrow apicomedial emargination. In- tercoxal process rugulose, contrasting with smoother discal areas of remaining sterna. Macropunctures of discal areas of sterna 2—4 sparse and small; macro- punctures denser on sternum 5. Legs: Densely indumentose. Tarsi mcderately densely pubescent beneath (Fig. 44); each slightly more than '3 length of respective tibia. Metacoxae coarsely punctate; metatrochanter globose, size slightly smaller than median 4 of coxa; metafemur with upper surface of distal end angulate. Genitalia: Ovipositor as illustrated (Fig. 25). Male unknown. Distribution.—Currently known only from the type-locality in Chiriqui Prov- ince, Panama. Etymology.—The hunchbacked appearance of this species calls to mind the character in Victor Hugo’s novel. Notes.— The holotype is badly abraded and several body parts are disarticulated. Disarticulated parts, including tarsi which were temporarily slide mounted for illustration, are in a microvial attached to the specimen’s pin. Key to the Species of Quadryops 1. Pronotal disc coarsely and densely punctate (Figs. 3, 4); suberect setae of body expanded and blunt apically (Figs. 6, 21); ovipositor as illustrated (Hig 324) SE CuAd On aiserna ct chee ee eae ws Se ean obtusetosus, new species — Pronotal disc not coarsely and densely punctate (Figs. 13, 36); suberect Setactom DodycothenwiSe. .b625%.ciam. suviiaeita. 2nd et ee ee eee 2 2. Lateral margin of pronotum angulate near midlength (Fig. 36); median carina of prosternal process angulate near midlength (Fig. 41); suberect setae of body blunt at apex (Fig. 46); metasternal episternum with well- developed basomedial carina (Fig. 43); ovipositor as illustrated (Fig. 25); SIZESES On Teo 2 ime Chiniquiserovinces Panama. ere teen RE HES Ra ae Omens SUM Lila tes HR oe NE ee a quasimodoi, new species — Lateral margin of pronotum not angulate (Fig. 13); median carina of prosternal process not angulate near midlength (Figs. 16, 17); suberect setae of body acute at apex, expanded subapically (Fig. 19); metasternal episternum without well-developed basomedial carina (Fig. 12); aedeagus as illustrated (Fig. 22); size 2.60 x 1.44 mm; Canal Zone, Panama Raed Vet, Sith yt sre soy odin ek ae baleany alba iS UND PRAT a greene be chrysosetosus, new species Acknowledgments We thank the technicians of the Smithsonian Institution’s Scanning Electron Microscope Laboratory, Susann Braden and Mary-Jacque Mann, for assistance with the micrographs included herein. (PDP) U.S. Department of Agriculture, Custom House, Rm. 801, Boston, Mas- sachusetts 02109; (PJS) Department of Entomology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 511-522 A NEW SPECIES OF BIVIBRANCHIA (PISCES: CHARACIFORMES) FROM SURINAM, WITH COMMENTS ON THE GENUS Richard P. Vari Abstract.—The genus Atomaster, described by Eigenmann and Myers (1927) for a single contained species, A. velox, is placed as a synonym of Bivibranchia Eigenmann. Bivibranchia is characterized by unique elaborations of the portions of the glossopharyngeal and vagus nerves innervating the branchial basket, and of the vagal lobe of the medulla oblongata. Derived modifications of the dorsal portions of the third and fourth pleural ribs, the associated parapophyses and articular fossae, and the intercostal ligaments of the first four ribs distinguish Bivibranchia within the Hemiodontidae. These characters and a variety of unique alterations of the jaws, suspensorium, and gill arches unite the three species of Bivibranchia (sensu lato) as a monophyletic lineage within the Hemiodontidae. Bivibranchia bimaculata, an inhabitant of black acidic waters, is described as new from the Corantijn River drainage system of western Surinam. The species is distinguished from B. protractila and B. velox, the other species in the genus, by the two distinct dark, midlateral spots on the body, one above the pelvic fins and one on the caudal peduncle. Meristic and morphometric differences further distinguish B. bimaculata from its congenerics. One of the most unusual taxa of New World characiforms is the genus Bivi- branchia proposed by Eigenmann (1912:258) for the single contained species, B. protractila, described in the same publication (1912:259) based on specimens from the Essequibo River system of Guyana. That highly modified genus differed from all characiforms known at that time in its markedly protractile upper jaw, fleshy dendritic gill rakers, and a distinctive valvular apparatus on the roof of the mouth (see Eigenmann 1912:pl. 33, figs. 1-5). Subsequently Eigenmann and Myers (1927:565) proposed a new genus and species, Atomaster velox, for a second species of protractile-mouth characiform collected in the Rio Tocantins of Brazil. As noted by those authors, the two genera were very similar other than for the small ctenoid scales of Atomaster which contrasted with the larger cycloid scales of Bivibranchia. Roberts (1974:432), although lacking specimens of Atomaster for comparison with Bivibranchia, reemphasized the evident similarities of the genera and noted that Myers, in a personal communication, questioned the distinctive- ness of the nominal genera. Recent collecting activities in the Corantijn River system of western Surinam yielded a third species of characiform with a protractile mouth. The new species, Bivibranchia bimaculata, inhabits the main river channels, rocky pools, and creeks in the black water systems of that basin. Studies associated with the description of the species led to a reevaluation of the distinctiveness of Bivibranchia and Atomaster. These anatomical investigations revealed a series of additional syn- apomorphies for the species of Bivibranchia (sensu lato). 512 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Methods and materials. —Counts of total vertebrae were taken from radiographs and cleared and stained specimens, and include the four vertebrae of the Weberian apparatus, with the fused PU, + U, of the caudal skeleton counted as a single element. Numbers in parentheses after a vertebral count are the number of spec- imens with that particular count. In counts of the pelvic and median rays, un- branched rays are indicated by Roman numerals and branched rays by Arabic numerals. The ranges for meristic counts are based on the holotype and paratypes, with the value for the holotype indicated in square brackets. Specimens examined for this study are deposited in the following institutions: American Museum of Natural History, AMNH; Academy of Natural Sciences of Philadelphia, ANSP; British Museum (Natural History), BMNH; National Mu- seum of Natural History, Smithsonian Institution, USNM. Specimens cleared and counterstained for cartilage and bone are indicated by CS. In his discussion of the Bivibranchiinae, Roberts (1974:432) noted that Ato- master Eigenmann and Myers (1927) was originally distinguished from Bivibran- chia Eigenmann (1912) by differences in scale form and number, but that the taxa were, at least externally, otherwise very similar, and that Myers (personal com- munication) now questioned the necessity for the recognition of separate genera. Roberts was, however, unable to pursue further the question of generic distinc- tiveness in the absence of comparative material of the single species of Atomaster (A. velox). Examination of a cleared and counterstained specimen of Atomaster shows that the genus shares the distinctive morphological modifications noted by Eigenmann (1912:258-259) and Roberts (1974:432—433) for Bivibranchia. These characters include the valvular processes on the roof of the mouth, the unusual fleshy ridges on the surfaces of the epibranchials and ceratobranchials, the pro- nounced restructurings of the upper jaw and suspensorium which permit pro- nounced protractility of the upper jaw, and the numerous distinctive adaptations of gill arch osteology described by Roberts for Bivibranchia protractila. Those shared derived characters along with the unique synapomorphies described below unite Bivibranchia protractila, Atomaster velox, and Bivibranchia bimaculata, described in this paper, as a phylogenetically monophyletic, and morphologically very distinctive subunit of the family Hemiodontidae. In light of those numerous derived similarities and given the few differences in scale form and number that are the primary distinguishing characters of the nominal genera, I follow Myers’ suggestion and formally place Atomaster Eigenmann and Myers as a synonym of Bivibranchia Eigenmann. Bivibranchia in that more inclusive sense is used throughout the remainder of the paper. Apomorphous modifications of two previously unanalyzed body systems serve further to characterize Bivibranchia and provide additional evidence that the three contained species (protractila, velox, bimaculata) constitute a monophyletic lin- eage within the Hemiodontidae. The first set of these synapomorphies involves the association of the anteriormost pleural ribs with the vertebral column, and the system of ligamentous connections of these ribs to each other. In all other hemiodontids and most characiform outgroups examined, the parapophyses as- sociated with the anterior full pleural ribs are approximately round elements, each of which inserts into a circular or ovoid articular fossa limited to the lateral surface of its respective centrum (see Weitzman 1962:fig. 12). Bivibranchia alternatively has a marked vertical expansion of the articular fossae and parapophyses asso- VOLUME 98, NUMBER 2 513 ciated with the third and fourth full pleural ribs. This restructuring results in distinctly dorsoventrally elongate parapophyses and a parallel vertical lengthening of the associated articular fossae. Correlated with the restructuring of the para- pophyses is the dorsal expansion of the portions of the third and fourth pleural ribs proximate to the vertebral column. A relatively slender bone rounded in cross section is the typical and hypothesized primitive condition of the proximate portion of the shaft of the pleural rib in characiforms. In Bivibranchia this section of the rib is apomorphously expanded dorsally into a transversely aligned, ver- tically triangular plate. This elaboration of the parapophyses, articular fossae, and ribs is similar to that described by Vari (1983:41-42, fig. 37) for the genera Caenotropus and Chilodus of the Chilodontidae. Differences in the identity of the ribs and vertebrae involved in this complex in Bivibranchia on the one hand, and chilodontids on the other raise questions as to the homology of the characters in the involved taxa. In chilodontids the first through third full pleural ribs all show the described modifications, with distinctly greater vertical development of the articular fossa, parapophysis, and proximal portion of the shaft of the rib on the second full pleural rib. In Bivibranchia, in contrast, the first two ribs, their articular fossae and associated parapophyses are unelaborated. Differences in the fourth pleural rib and the associated parapophyses and articular fossae also distinguish the two taxa. Chilodontids retain the conditions of those elements generalized for characiforms, whereas in Bivibranchia the proximal portions of the fourth pleural rib and the associated articular fossae and parapophyses are expanded dorsally to a degree equivalent to that on the third pleural rib. The hypothesis of the homoplasy of these modifications in Bivibranchia and the Chilodontidae is, furthermore, congruent (parsimoniously consistent) with available phylogenetic information about the groups. Outgroup comparisons with- in the Hemiodontidae have not revealed any comparable pleural rib and vertebral alterations in Argonectes, the taxon most closely related to Bivibranchia (Roberts 1974), nor in any of the other examined genera in the family (Anodus, Hemio- dopsis, Hemiodus, Micromischodus). The absence of pleural rib alterations in close relatives of Bivibranchia in conjunction with the closer phylogenetic association of the Chilodontidae with the Anostomidae, Curimatidae, and Prochilodontidae (Vari 1983:46-47) makes it most parsimonious to hypothesize that these simi- larities in Bivibranchia on the one hand, and the Chilodontidae on the other, represent convergencies at the level of the two involved taxa rather than synapo- morphies for a lineage consisting of the Chilodontidae and Bivibranchia. The second derived pleural rib associated modification shared by the three species of Bivibranchia is the presence of well developed intercostal ligaments interconnecting the four anteriormost full pleural ribs. A thick posterodorsally slanting ligamentous band arises from the posterior surface of the first full pleural rib, extends past, but does not contact the medial surface of the second rib, and attaches to the anterior margin of the dorsal portion of the third rib. A second thick intercostal ligament arises from the posterior surface of the first pleural rib ventral of the insertion of the first ligament, extends past the medial surfaces of the second and third ribs, again without direct contact, and attaches to the anterior surface of the fourth pleural rib. Somewhat similar versions of such intercostal connections are present in the other hemiodontid genera examined (Argonectes, Hemiodopsis, Hemiodus, Micromischodus), but in those taxa the intercostal con- 514 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON nections are thin, flattened, strap-like bands rather than the thick cord-like struc- tures characteristic of Bivibranchia. Thus the intercostal bands in those hemio- dontid outgroups are more similar to the broad, thin connective tissue sheet that joins the pleural ribs along their medial surfaces in most characiforms. As such the connections in hemiodontids other than Bivibranchia may represent an in- termediate step in a transition series that resulted in the evolution of the system of well developed intercostal ligaments in that genus. A complex of intercostal ligaments somewhat comparable to those in Bivibranchia was described by Vari (1983:41-—42, fig. 36) in the Anostomidae and Chilodontidae, with a system of thick ligaments in the Chilodontidae most comparable to the pleural rib inter- connections of Bivibranchia. The intercostal ligament complex in chilodontids differs, however, from that in Bivibranchia in having a dorsal ligamentous at- tachment of the first and second ribs not found in Bivibranchia, and in the absence of the ligament attaching the first and fourth ribs that is present in that genus. Furthermore the intercostal ligament in chilodontids that spans the first to third ribs is attached to the medial surface of the second rib, whereas in Bivibranchia the ligaments and the medial surfaces of the intervening ribs are distinctly sep- arated by an intermediate tissue layer. These morphological differences raise ques- tions as to the homology of the intercostal connections in the two groups, Bivi- branchia and the Chilodontidae. Such a hypothesis of the non-homology of the modifications in the two taxa is congruent with the available data on phylogenetic relationships noted above which indicate that Bivibranchia and the Chilodontidae are not each other’s closest relatives. The functional basis for the expansion of the intercostal ligaments and the vertical elongation of the articular fossae, parapophyses, and ribs in Bivibranchia is obscure, but may be correlated with the head-down feeding orientation of mem- bers of the genus (personal observations on B. bimaculata). This feeding position parallels the more pronounced head-down swimming orientation typical of mem- bers of the Chilodontidae which homoplasiously demonstrate comparable mor- phological adaptations. The second set of synapomorphies for the species of Bivibranchia is the series of hypertrophied modifications of the portion of the nervous system associated with the branchial basket. Most noteworthy is the hypertrophy of the nerve com- plex serving the gill arches, and the associated elaboration of the vagal lobes of the medulla oblongata. Among the numerous restructurings of the Bivibranchia branchial basket noted by Roberts (1974:420—421), one of the more conspicuous is the pronounced vertical expansion of the epibranchials and ceratobranchials via thin bony lamina. The surfaces of these expanded gill arch elements are covered by specialized layers of epithelial tissue with “‘. .. a uniform series of prominent finely papillose ridges.” Each ridge is associated with a gill raker, and the ridges on opposing surfaces of the gill arches interdigitate when in contact. The vertical expansion of the epibranchials and ceratobranchials in addition to markedly in- creasing the surface area of the gill arches, also accommodates a change in the form of the central channel of the epibranchials and ceratobranchials of the first three gill arches. On these ossifications, the shallow median groove on the ab- pharyngeal surface of the bone that is typical of characiforms is expanded into a deep central trough. These troughs serve at least in part for the support of a greatly expanded nerve network innervating the anterior portion of the gill arches. Two VOLUME 98, NUMBER 2 515 Fig. 1. Bivibranchia bimaculata, USNM 268203, paratype, gill arches, glossopharyngeal and vagus nerves, and posteroventral portion of neurocranium, left side, lateral view. Abbreviations: C—cerato- branchials (1 to 5); E—epibranchials (1 to 5, 2 and 3 not labelled); N—branches of hypertrophied nerve complex (N, —glossopharyngeal nerve, N, to N,—vagus nerve). Large stippling indicates car- tilage, dashed lines indicate nerves. primary trunks of cranial nerves form this hypertrophied nerve complex. In lateral view (Fig. 1) this complex is seen as a diverging series of large nerves that exit the neurocranium dorsal of the gill arches and pass ventrally into the branchial basket. The nerve complex passes through the wall of the neurocranium via two proximate enlarged foramina on the anteroventral surface of the exoccipital. These expanded apertures are visible as-the enlarged openings in the exoccipitals as illustrated by Roberts (1974:fig. 24) in the ventral view of the skull of Bivibranchia. The enlargement of these foramina is particularly obvious when they are com- pared to the homologous openings in Hemiodus and Argonectes (Figs. 4 and 21 of Roberts). The smaller anterior exoccipital foramen is the point of exit for the anteriormost nerve bundle (Fig. 1, N,) which is part of the ninth cranial (glossopharyngeal) nerve. After exiting the skull, that nerve trunk extends ventrally into the medial margin of the central trough of the first epibranchial, diverges laterally in that element, with a smaller portion of the nerve continuing into the first ceratobran- chial through the epibranchial-ceratobranchial joint. The remaining five major branches of the nerve complex innervating the gill arches all exit from the larger foramen located immediately posterior of the foramen for the glossopharyngeal nerve (Fig. 1, N, to N,), and are all components of the tenth cranial (vagus) nerve. The anteriormost of these (N,) has two subunits. The anterior branch of N, extends to the posterior surface and ventral margin of the first epibranchial, and the larger posterior section of N, enters the medial portions of the central trough of the second epibranchial within which it has a distribution comparable to the primary 516 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON OT Fig. 2. Bivibranchia protractila, USNM 225491, brain, posterior portion, dorsal view, anterior to right. Abbreviations: C—cerebellum; OT—optic tectum; SC—spinal cord; T—telencephalon; VL— vagal lobe of medulla oblongata. nerve branch in the first epibranchial, and with a section continuing into the second ceratobranchial. The next major nerve bundle (N;) has three subsections. The anteriormost branch runs to the posterior surface and ventral margin of the second epibranchial, and the two remaining portions of N; enter the central trough of the third epibranchial, and are distributed in that element and the third cerato- branchial in a pattern comparable to that noted for the primary nerves of the first and second gill arches. The fourth nerve ramus (N,) innervates the posterior surface of the third epibranchial, and the fifth major nerve branch (N;) runs along the posterior portion of the fourth epibranchial with relatively few grossly obvious branches before continuing onto the dorsal surface of the fourth ceratobranchial. The sixth and final section of the complex (N,) extends over the lateral surface of the fourth epibranchial before passing medial to the enlarged cartilaginous fifth epibranchial to innervate the anterior and dorsal surfaces of the fifth ceratobran- chial. The very large nerve bundles innervating the gill arches in Bivibranchia rep- resent a hypertrophy of the glossopharyngeal (IX) and vagus (X) nerve complex relative to the much narrower nerves typical of other characiforms and most teleostean fishes (e.g., Menidia, Bernstein 1970:fig. 18). Information from other groups of fishes indicates that the ninth and tenth cranial nerves consist of both visceral sensory and visceral motor components. It is not possible to determine at present whether the dramatic increase in the size of these nerves in the species of Bivibranchia represents the hypertrophy of only one of these components, or whether both components of those nerves are involved. The question of the degree to which the increased size of the nerves is correlated with an increased degree of proprioperception of the gill arch elements, chemoreception for analysis of potential food items in the pharyngeal cavity, some combination of the above, or perhaps another function is similarly unresolved. Not unexpectedly the dramatic enlargement of the glossopharyngeal (IX) and vagus (X) nerves is reflected in grossly obvious modifications of the central nervous VOLUME 98, NUMBER 2 517 RO” «MN EBC BAG Bae, i in a LY Ore he SD PA ae, ye ae < ae 4 ny eRe iy sg 5 4 « Fig. 3. Bivibranchia bimaculata, new species, holotype, USNM 225974, 80.5 mm SL. system. Moderately developed vagal lobes of the medulla oblongata occur in various ostariophysans (Bernstein 1970:55, fig. 3) and within the Hemiodontidae have been found in both Hemiodus and Hemiodopsis. In Bivibranchia both of the enlarged cranial nerves (IX and X) communicate with the central nervous system at the base of the vagal lobe of the medulla oblongata, and the enlargement of the vagal lobes of the medulla oblongata is carried even further. The vagal lobes in Bivibranchia (Fig. 2) are dramatically enlarged into bulbous structures that are distinctly expanded laterally and extend to the level of the dorsal surface of the cerebellum. More striking is the elaboration of the surface of the markedly enlarged vagal lobes of the medulla oblongata into a series of posterodorsally oriented folds which cover the anterior, dorsal, and lateral surfaces of the lobes. These derived elaborations of this portion of the central nervous system are not apparent in other hemiodontids or other characiforms examined and are conse- quently considered synapomorphies for the three species of Bivibranchia. The functional significance of these vagal lobe elaborations is unknown, although they are presumably correlated with the hypertrophy of the associated glossopharyngeal and vagus nerves. Bivibranchia bimaculata, new species Fig. 3, Table 1 Holotype.—USNM 225974, 80.5 mm standard length (SL), collected by R. P. Vari and L. R. Parenti, 17 Sep 1980, in a rocky pool in the center of the Corantijn River at “Camp Hydro,” Nickerie District, Surinam (approx. 3°42’N, 57°58’W). Paratypes.— All from Nickerie District, Surinam: 12 specimens taken with ho- lotype: ANSP 153656, 1 specimen, 94.5 mm SL; BMNH 1984.10.23:1, 1 spec- imen, 97.8 mm SL; AMNH 55611, 1 specimen, 86.7 mm SL; USNM 268203, 9 specimens, 68.3—94.7 mm SL (1 specimen CS). 1 specimen, BMNH 1981.6.8: 816, 68.2 mm SL, collected by H. M. Madarie, 18 May 1980, in a small creek draining from the right bank of the Corantijn River near Mataway (approx. 4°58.5'N, 57°42'W). 7 specimens, AMNH 54807, Dalbana Creek, 150 m upstream of junction with the Kabalebo River (approx. 4°47'N, 57°26'W). Non-type specimens examined.— All from Nickerie District, Surinam: 7 spec- imens, USNM 225491, same data as BMNH 1981.6.9:816 (1 specimen CS). 49 specimens, USNM 268204, taken with holotype, juveniles (3 specimens CS). 5 specimens, AMNH 54817, small stream entering Kabalebo River, 150 m up- stream of mouth of Dalbana Creek. 1 specimen, AMNH 54835, stream draining 518 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Morphometrics of Bivibranchia bimaculata, new species. Standard length is expressed in mm; measurements | to 10 are proportions of standard length; 11 to 15 proportions of head length. Paratypes (20) Holotype Range Average Standard length 80.5 68.2-97.8 82.69 1. Greatest body depth 0.24 0.22-0.26 0.243 2. Snout to dorsal-fin origin 0.49 0.47-0.51 0.489 3. Snout to anal-fin origin 0.81 0.80-0.84 0.820 4. Snout to pelvic-fin origin 0.56 0.54-0.58 0.561 5. Snout to anus 0.79 0.79-0.82 0.802 6. Origin of rayed dorsal fin to hypural joint 0.53 0.51-0.56 0.544 7. Least depth of caudal peduncle 0.09 0.09-0.10 0.093 8. Pectoral-fin length 0.20 0.19-0.22 0.200 9. Pelvic-fin length 0.19 0.18-0.20 0.189 10. Head length 0.28 0.27-0.30 0.285 11. Snout length 0.34 0.30-0.34 0.313 12. Orbital diameter 0.38 0.34-0.38 0.358 13. Postorbital head length 0.34 0.32-0.36 0.351 14. Interorbital width 0.37 0.34-0.39 0.357 15. Gape width 0.13 0.11-0.13 0.124 into Kabalebo River near Camp Avanavero, about 5 km downstream of DeVis Falls. 14 specimens, AMNH 54858, side channel of Kabalebo River, 1 km south of Avanavero Falls. 13 specimens, AMNH 54926, same locality as holotype. 13 specimens, AMNH 54959, rocky side pool of Corantijn River approximately 378 km from its mouth. 1 specimen, BMNH 1981.6.9:839, stream at km 212 of Amotopo-Camp Geology road, at ““Machine Park.” 1 specimen, USNM 225195, small stream entering Lucie River, 3 km upstream of junction of Lucie and Kabalebo rivers. Diagnosis.—Bivibranchia bimaculata shares with the two other species of the genus a number of modifications of the jaws, branchial apparatus, anterior ribs, vertebral column, glossopharyngeal and vagus nerves, and vagal lobe of the me- dulla oblongata that distinguish the genus within the Hemiodontidae (see dis- cussion above). The presence in Bivibranchia bimaculata of a large spot of dark pigmentation on the midlateral body surface centered slightly posterior of the vertical through the insertion of the posteriormost dorsal fin ray, and of a small darkly pigmented midlateral spot on the caudal peduncle separates the species from B. protractila and B. velox which have plain bodies. The 49 to 55 pored lateral line scales to the hypural joint, 812 or 94 scales in a transverse series above the lateral line to the origin of the dorsal fin, and the possession of cycloid scales distinguish B. bimaculata from B. velox which has 80 or more series of scales to the hypural joint, 12 to 16 scales in a transverse series to the origin of the dorsal fin, and ctenoid scales. The longer pelvic fins (0.18—0.20 of SL), 8 branched anal- fin rays, and typically 11, sometimes 10, branched pelvic-fin rays of B. bimaculata further separate that species from B. protractila in which the pelvic fins are 0.14— 0.17 of SL, and which has 7 branched anal-fin rays, and 9, rarely 10, pelvic-fin rays. VOLUME 98, NUMBER 2 519 58° 57° GUYANA IF SURINAM Fig. 4. Map of the middle portion of the Corantijn River basin region, Surinam and Guyana, showing collecting localities of specimens reported on in this paper (dotted lines depict road systems in area): 1, “Camp Hydro,” type locality; 2, Km 378 of Corantijn River; 3, Lucie River; 4, ““Machine Park’; 5, creek near Mataway; 6, Dalbana Creek; 7, Camp Avanavero and Avanavero Falls. Some squares indicate more than one locality or more than one lot of specimens. See listing of holotype, paratypes, and non-type specimens examined for detailed locality and collection information. Description.— Morphometrics of the holotype and paratypes are given in Table 1. Body slender, slightly compressed laterally. Greatest body depth at origin of rayed dorsal fin. Dorsal profile of body gently curved from tip of snout to caudal peduncle. A slight median keel immediately anterior of origin of dorsal fin. Ventral profile of body smoothly convex from tip of lower jaw to caudal peduncle. Ventral surface of body transversely flattened anteriorly. Head pointed in profile, interorbital region flattened. Fronto-parietal fontanel extensive, extending into rear of ethmoid and onto dorsal portion of supraoccip- 520 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ital. Parietals completely separated, frontals in contact only at epiphyseal bar. Upper jaw longer than lower jaw, highly protractile. Nostrils approximate, anterior Opening round, posterior crescent shaped, partially closed by flap of skin separating nostrils. Eye relatively large. A large, horizontally ovoid “‘adipose eyelid”’ (a thick transparent connective tissue layer) extends from under nostrils posteriorly to opercle, thicker anteriorly; thicker and more developed longitudinally in larger specimens. Adipose eyelid with an ovoid, vertically elongate opening overlying pupil. Lower jaw edentulous, rounded in ventral view, anterior margin fleshy. Upper jaw with a single series of functional teeth. All teeth tricuspidate, 7 on each side of jaw in all cleared and stained specimens examined (30-82 mm SL); teeth becoming progressively larger medially. A single row of partially formed replace- ment teeth internal to functional tooth row. Replacement tooth row embedded in flesh of inner surface of upper jaw. Dermopalatine and ectopterygoid edentu- lous. Fifth ceratobranchial bearing a relatively narrow band of posterodorsally directed teeth along medial and posteromedial borders; teeth along anterior por- tion of band elongate, tricuspidate; those along posterior margin elongate, conical. Fourth and fifth upper pharyngeal tooth-plates with a band of elongate tricuspidate teeth. Gill rakers elongate, with short side processes. Gill rakers extend along surfaces of vertically expanded ceratobranchials and epibranchials. Surfaces of expanded ceratobranchials and epibranchials with series of fleshy ridges aligned nearly perpendicular to their primary axes. Gill arches highly modified, innervated by hypertrophied glossopharyngeal and vagus nerves (see description above). Gill membranes narrowly attached medially to urohyal. Scales cycloid, firm. Pored lateral line scales between supracleithrum and hy- pural joint 49 to 55 [55]. Five to 8 pored lateral line scales extending beyond hypural joint onto base of caudal fin. Scales above lateral line in a transverse series to origin of rayed dorsal fin 8% or 9 [94]. Scales below lateral line in a transverse series to origin of anal fin 5 or 6 [6]. Body squamation extending onto base of caudal fin rays. Axillary process of pelvic fin formed by a single enlarged scale. Vertebrae 38 (20), 39 (3). Rayed dorsal fin obtusely pointed, second unbranched and first branched rays longest, subequal. Dorsal-fin rays 11,9 or i1i1,9 [1i1,9]; when three unbranched rays present, first very short. Adipose dorsal fin of moderate size, unscaled. Anal fin obtusely pointed, anterior branched rays over twice length of posteriormost rays. Anal-fin rays 11,8 or i11,8 [11,8]; when three unbranched anal-fin rays present, first very short. Pectoral fin pointed, reaching two-thirds of distance to vertical through origin of pelvic fin. Dorsalmost rays of pectoral fin correspond to a distinct groove along side of body formed by a connective tissue ridge extending posteriorly from posterior margin of cleithrum. Pelvic fin pointed, reaching slightly over one- half distance to anus. Pelvic-fin rays i,10 or i,11 (typically i,11) [i,11]. Coloration in life.—Overall coloration silvery with a greenish grey shading; silvery coloration more intense on ventral portions of body. Dark midlateral spot on body very obvious; spot on caudal peduncle somewhat masked. Coloration in preservative.— Overall coloration in specimens fixed in formalin and preserved in ethanol light tan. Head darker on dorsal portions, particularly in interorbital region and across parietals. A broad band of scattered chromato- phores extends over dorsal half of opercle. Overall coloration of body darker VOLUME 98, NUMBER 2 521 dorsally. A deep-lying dusky band along lateral line, pigmentation of band more intense posteriorly. A distinct dark, round or horizontally ovoid spot with irregular Margins on midlateral surface of body. Spot extends 8 to 13 scales horizontally and 4 to 7 scales dorsally. Spot centered along or slightly posterior of vertical through insertion of last dorsal-fin ray, and somewhat dorsal of lateral line. A smaller darkly pigmented, round or horizontally ovoid spot on caudal peduncle immediately anterior of hypural joint. Caudal peduncle spot faint in larger in- dividuals, absent in some large specimens. Caudal and rayed dorsal fins dusky, with fin-rays outlined by series of chromatophores. Adipose dorsal fin hyaline. Juveniles with scale margins outlined by series of chromatophores. Midlateral body spot absent in specimens under 20 mm SL. Midlateral caudal peduncle spot not developed in smaller individuals. Etymology.—The specific name, bimaculata, from the Latin bi, two, and mac- ula, spot, refers to the two dark spots on the lateral surface of the body and caudal peduncle. Ecology.—Bivibranchia bimaculata is widely distributed throughout the acid, black waters of the Corantijn River system of western Surinam above the region of tidal influence. It is most common in areas of sandy beaches and in rocky pools, with juveniles often also found in smaller side streams, sometimes a con- siderable distance from the main river channels. This species has been observed travelling in large schools over sandy beaches, evidently feeding on food items which individuals separate out of the substrate by manipulation of mouthfulls of sand. Comparative material examined. —Bivibranchia protractila Eigenmann: GUY- ANA: BMNH 1972.10.17:1378-1397, 15 specimens, Rupununi River; USNM 197104, 13 specimens, Rupununi River; BMNH 1936.4.4:17-18, 2 specimens, Rockstone; BMNH 1911.10.31:484, 2 specimens, Rockstone, paratypes of B. protractila; USNM 66126, 1 specimen, Rockstone, paratype of B. protractila; BMNH 1934.9.12:291, 1 specimen, Mazaruni River, BMNH 1972.7.27:81-109, 27 specimens, Rupununi District, Jacaré; USNM 268205, 5 specimens, Essequibo River (2 specimens CS). BRAZIL,: Mato Grosso: USNM 194302, 1 specimen, Rio Juruena; USNM 194287, 2 specimens, Rio Juruena. Bivibranchia velox Eigenmann and Myers: USNM 268345, 1 specimen, Brazil, Para, Rio Tocantins (CS). Argonectes longipinnis Steindachner: USNM 243224, 2 specimens, Brazil, Ro- raima, Rio Jauaperi (1 specimen CS). Anodus elongatus Spix: USNM 231550, 1 specimen, Peru, Loreto, Rio Ucayali (CS). Hemiodopsis ocellata Vari, USNM 225593, 1 specimen, Surinam, Nickerie District, Corantijn River (CS). Hemiodus species, USNM 231551, 2 specimens, Brazil, Mato Grosso, Rio Arinos (CS). Micromischodus_ sugillatus Roberts, USNM 205527, 1 specimen, Brazil, Para (CS). Acknowledgments Drs. Donn E. Rosen, Gareth Nelson, and Ms. M. Norma Feinberg (AMNH) made available many of the specimens associated within this study. Dr. P. Hum- 922 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON phry Greenwood and Mr. Gordon J. Howes provided facilities for research at the BMNH and access to the series of comparative specimens of Bivibranchia pro- tractila collected by Dr. Rosemary Lowe-McConnell in Guyana. Dr. Michael Goulding of the Museu Paraense “Emilio Goeldi,’”? and Drs. Heraldo Britski, Naércio Menezes, and José Lima de Figueiredo of the Museu de Zoologia da Universidad de Sao Paulo made available comparative materials from the Rio Amazonas drainage basin. I thank all of the above for their generosity and assis- tance. The specimens that served as the basis for the description of the new species were collected during surveys carried out as part of preimpoundment studies associated with the Kabalebo Hydroelectric Project in western Surinam. The assistance of Drs. M. P. Panday-Verheuvel and S. Niekoop in organizing the survey, and Dr. Lynne R. Parenti, Mr. H. M. Madarie, Ms. S. Engel, and Mr. S. Silos in collecting activities is gratefully ackowledged. Mr. Kurt Bruwelheide and Mr. Andrew Gerberich provided technical assistance. Research associated with this project was partially supported by the Neotropical Lowland Research Program of the Smithsonian Institution. Figure 3 was photographed by Mr. T. Britt Gris- wald. This paper benefitted from the comments and suggestions of Drs. Stanley H. Weitzman and Wayne C. Starnes. Literature Cited Bernstein, J. J. 1970. Anatomy and physiology of the central nervous system. Jn Hoar, W. S. and D. J. Randall, eds., Fish physiology, volume 4.—Academic Press, New York, pp. 1-90. Eigenmann, C. H. 1912. The freshwater fishes of British Guiana, including a study of the ecological grouping of species and the relation of the fauna of the plateau to that of the lowlands. — Memoirs of the Carnegie Museum 5:xvii + 1-554. , and G. S. Myers. 1927. A new genus of Brazilian Characin Fishes allied to Bivibranchia. — Proceedings of the National Academy of Sciences 13(8):565-566. Roberts, T. R. 1974. Osteology and classification of the Neotropical characoid fishes of the families Hemiodontidae (iucluding Anodontinae) and Parodontidae. — Bulletin of the Museum of Com- parative Zoology 146(9):41 1-472. Vari, R. P. 1983. Phylogenetic relationships of the families Curimatidae, Prochilodontidae, Ano- stomidae and Chilodontidae (Pisces: Characiformes).— Smithsonian Contributions to Zoology 378:1-60. Weitzman, S. H. 1962. The osteology of Brycon meeki, a generalized characid fish, with an osteo- logical definition of the family.—Stanford Ichthyological Bulletin 8(1):1-77. Department of Vertebrate Zoology (Fishes), National Museum of Natural His- tory, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 523-525 PSYCHROPOTES HYALINUS, NEW SPECIES, A SWIMMING ELASIPOD SEA CUCUMBER (ECHINODERMATA: HOLOTHUROIDEA) FROM THE NORTH CENTRAL PACIFIC OCEAN David L. Pawson Abstract. — Psychropotes hyalinus, new species, is described. The body is colorless and transparent, the dorsal appendage is situated approximately one-third of the body length from the posterior end of the body, the skin is more or less smooth, and contains two types of ossicles in the form of spinose crosses. The species is evidently capable of swimming, as the only known specimen was captured in a trap five meters above the seafloor. The deep-sea Order Elasipodida contains many holothurians that are capable of swimming for varying periods of time. Some species appear to be obligate swimmers, while others can swim for only short periods of time and short distances (Hansen 1975; Pawson 1976, 1982; Pawson and Foell, in press). Hansen (1975) and earlier authors have suggested that within the Family Psychropotidae some species are capable of swimming, and indeed Psychropotes depressa Theel was photographed in the act of swimming by Pawson (1976—reported as ““Euphronides sp.”’). Through the kindness of Drs. Ken Smith and Nancy O. Brown of the Scripps Institution of Oceanography, La Jolla, California, I was sent a specimen of Psy- chropotes which had been captured in a near-bottom trap deployed in abyssal depths in the Pacific Ocean north of Hawaii. The specimen represents a new species, and is described below. I am grateful to Drs. Smith and Brown for allowing me to study this specimen, and to John E. Miller, Harbor Branch Foundation, for reading the manuscript of this paper. The specimen is deposited in the National Museum of Natural History (USNM), Smithsonian Institution, Washington, D.C., U.S.A. Order Elasipodida Theel Family Psychropotidae Theel, 1882 Psychropotes Theel, 1882 Psychropotes hyalinus, new species Fig. 1 Diagnosis.— Body transparent, colorless. Dorsal appendage situated approxi- mately one-third of the body length from the posterior end of the body. Dorsal body-wall ossicles of two types, larger crosses with spinose arms and single spinose apophyses, and smaller crosses with spinose arms. Material examined.—HOLOTYPE USNM E31731, RAMA 2, 2 May 1980 central North Pacific north of Hawaii, 30°05.7'N, 158°44.5’W, tent trap deployed 5 m above seafloor in depth of 5891 meters. 524 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON IN 2 AF Xr 1} 100 urn _, ee Fig. 1. Psychropotes hyalina new species. A, Large four-armed crosses from dorsal body wall; B, 3- and 4-armed crosses from wall of intestine; C, Small 4-armed crosses from dorsal body wall; D, Straight and curved rods from discs of tentacles; E, Portion of testis showing branching genital tubules. Description. — Body of typical ““Psychropotes-shape” (see Heezen and Hollister 1971, figs. 2.56, 2.57, 5.17), fragile, transparent, with light-reflective dermal layer of ossicles. Body approximately 160 mm long and 30 mm in diameter, more or less cylindrical, with flattened ventral surface. Conspicuous dorsal appendage approximately 70 mm long arising from middle of dorsum, 60 mm from posterior end of body. Flat subcircular brim overlying ventrally turned mouth and tentacles. Tentacles 15, non-retractile, with elongate oval terminal discs approximately 8 X 6 mm; long axis of tentacle disc radial in relation to mouth. Edges of discs scalloped, with approximately 25 short, rounded digitiform projections. Mouth circular, 4 mm in diameter, at center of unadorned oral field. Pinkish to violet internal organs clearly visible through body wall. Holotype, male; testis consisting of conspicuous bunch of branching tubules (Fig. 1E) located at extreme anterior end of body cavity. Internal musculature not well developed. Intestine empty. Ossicles in body wall numerous 4-armed crosses of 2 types. Large crosses (Fig. 1A) reaching maximum diameter of 1.0 mm, having 4 curved spinous arms, with spines having no regular arrangement; long unbranched central apophysis carrying prominent spines. Large crosses oriented in body wall with apophyses pointing outwards, forming small contiguous pustules on body wall. Spaces between large VOLUME 98, NUMBER 2 525 crosses occupied by numerous small crosses (Fig. 1C) with slightly curved arms carrying weakly developed spines; central apophysis, when present, small; small crosses average 200 um in diameter. Dorsal appendage with numerous smaller crosses; larger crosses also present but far less numerous. Intestine wall with 3- and 4-armed crosses (Fig. 1B), latter type most common. Arms of crosses more or less straight, with 1 or 2 weakly developed blunt spines; occasionally, short blunt apophysis present. Average diameter of these ossicles 150 wm. Tentacle discs contain straight or curved rods (Fig. 1D) with scattered spines or knobs. Rods greatly variable in length, maximum length approximately 650 wm. Behavior.— As this species was captured some 5 meters above the seafloor, it is apparently capable of actively swimming, but nothing else is known about its living habits. The intestine is completely empty, but it is likely that this animal feeds on the seafloor, perhaps in the same manner as Enypniastes (Pawson, 1982). Remarks.—The presence of a relatively smooth dorsal skin (rather than a skin with warts, each wart containing a giant cross-shaped ossicle), and the location and size of the dorsal appendage place this new species near P. semperiana Theel, 1882, and P. minuta Koehler and Vaney, 1905, in the key to Psychropotes pro- vided by Hansen (1975). Psychropotes hyalinus differs from both species in color (they are dark violet) and in characters of the body wall ossicles. In P. semperiana, the smaller crosses have high central apophyses carrying downcurved hooks, while in P. minuta, the body wall crosses are of only one type, not two, as in P. hyalinus. In having an essentially transparent to whitish body wall, P. hyalinus differs from almost all other species in the genus which usually tend to be violet or purple. According to Hansen (1975) only P. loveni Theel is whitish when it is small (20- 25 mm long) but apparently this species becomes violet as it grows (Theel 1882). In addition, the dorsal appendage of P. /oveni is very close to the posterior end of the body, in contrast to the situation in P. hyalinus. Literature Cited Hansen, B. 1975. Systematics and biology of the deep-sea holothurians. Part 1. Elasipoda.—Galathea Report 13:1-262. Heezen, B. C., and C. D. Hollister. 1971. The face of the deep.—Oxford University Press, New York, vill + 659 pp. Pawson, D. L. 1976. Some aspects of the biology of deep-sea echinoderms. — Thalassia Jugoslavica 12(1):287-293. 1982. Deep-sea echinoderms in the tongue of the ocean, Bahama Islands: a survey, using the research submersible A/vin.— Australian Museum Memoir 16:129-145. ——., and E. J. Foell. [in press]. Peniagone leander new species, an abyssal benthopelagic sea cucumber (Echinodermata: Holothuroidea) from the eastern central Pacific Ocean.— Bulletin of Marine Science. Theel, H. 1882. Report on the Holothurioidea. I.—Report on the Scientific Results of the Voyage of the Challenger, Zoology 4(13):1-176. Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, U.S.A. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 526-532 A NEW SPECIES OF SIPHONORHTS FROM QUATERNARY CAVE DEPOSITS IN CUBA (AVES: CAPRIMULGIDAE) Storrs L. Olson Abstract.—A new species of nightjar, Siphonorhis daiquiri, is described from Quaternary cave deposits in eastern and central Cuba. It is intermediate in size between the other two species in the genus, S. americana of Jamaica, and S. brewsteri of Hispaniola. Because these elusive nocturnal birds are difficult to detect, it is possible that this species is still extant in the ornithologically poorly explored arid regions of eastern Cuba. The distinctive nightjars of the genus Siphonorhis are among the lesser known elements of the Antillean avifauna. The Jamaican species, S. americana, is per- haps the rarest of caprimulgids, being known only from four recent specimens (Olson and Steadman 1977) and now presumed extinct. Nevertheless, S. amer- icana has been recognized practically from the inception of scientific ornithology. It is the Caprimulgus americanus of Linnaeus (1758), the name being intended to compare with C. eurepaeus, the only other species of Caprimulgus then rec- ognized. Linnaeus based his name on Sir Hans Sloane’s description of the “‘small wood-owle,”’ first published by Ray (1678) from Sloane’s notes, and later by Sloane (1707) himself, along with an illustration. Such a bird was unknown to Gosse (1847), the next chronicler of Jamaica’s birdlife, and it was later questioned whether Sloane’s bird had actually come from that island. Cassin (1851) reviewed the whole question in detail, concluding that a caprimulgid, which he referred to the genus Nyctidromus, indeed inhabited Jamaica. It was not until 1859, however, when the collector Osburn succeeded in obtaining two specimens, that the true nature of the species was revealed. With these specimens at hand, Sclater (1861) determined that C. americanus was quite distinct from other caprimulgids and he therefore proposed a new genus, Si- Dhonorhis, for it. Sclater, followed by a number of other authors, used the spelling Siphonorhis americanus, but the generic name is feminine in gender, so the specific name must be rendered americana. Because of the long, bare tarsus, Sclater, too, considered Siphonorhis to be closely related to the mainland genus Nyctidromus. In 1917, at Tubano, in the Dominican Republic, Rollo Beck obtained a single specimen of a new species of caprimulgid with obvious affinities to Siphonorhis. This was described in a new genus by Chapman (1917) as Microsiphonorhis brewsteri. The species is now known to be fairly widely distributed on Hispaniola (Bond 1928a; Wetmore and Swales 1931; Dod 1979). Bond (1928b) regarded Chapman’s characters for Microsiphonorhis to be inconsistent and he placed brewsteri in Siphonorhis. Wetmore (Wetmore and Swales 1931) concurred in this and the genus Microsiphonorhis has since seldom been considered valid. Peters (1940) listed brewsteri as a subspecies of S. americana, but this unsupported view is not tenable (Olson and Steadman 1977). VOLUME 98, NUMBER 2 527 Fig. 1. Distal half of right tarsometatarsus, holotype of Siphonorhis daiquiri, new species (USNM 336506). A, Anterior view; B, Posterior view. 6 x natural size. The absence of the genus Siphonorhis in Cuba has not drawn comment from ornithologists, possibly because several other birds have a similar pattern of distribution, being found in Hispaniola and Jamaica but not in Cuba, viz: cuckoos of the genus Hyetornis, the potoo Nyctibeus griseus, and the swallow Kalochelidon euchrysea. Whatever the cause may be for the distributional patterns of those species, the pattern of Siphonorhis is the result of its having been overlooked in Cuba, for fossil remains from two cave deposits now prove the existence of a hitherto unrecognized species of Siphonorhis on that island. Comparative material examined.—The fossil specimens were compared with Antillean genera of Caprimulgidae and most genera of the Neotropical mainland. No skeletons exist for Caprimulgus cubanensis, which is endemic to Cuba and Hispaniola, but this species is much larger than the fossil form under consideration here. Likewise, there are no entire skeletal specimens of either Siphonorhis amer- icana or S. brewsteri, a lack I was able to make up for in part with bones removed from a skin of S. brewsteri (USNM 354527, Gonave Island, Haiti), including a 528 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON nearly intact skull and mandible, tibiotarsus, tarsometatarsus, distal end of hu- merus, and proximal end of ulna. These were augmented by a humerus of S. brewsteri from a cave deposit in Haiti and humeri of S. americana from cave deposits in Jamaica (see Olson and Steadman 1977), as well as a previously unreported distal end of a tarsometatarsus of S. americana (UF 68055) from a cave deposit near Wallingford, St. Elizabeth Parish, Jamaica. Other taxa used in the comparisons were Chordeiles minor, Caprimulgus noctitherus, C. carolinensis, Nyctidromus albicollis, Phalaenoptilus nuttallii, and Nyctiphrynus ocellatus. Order Caprimulgiformes Ridgway Family Caprimulgidae Vigors Genus Siphonorhis Sclater, 1861 The following characters refer the Cuban fossils to the genus Siphonorhis: tar- sometatarsus very long and slender, with trochleae splayed far apart and the intertrochlear spaces consequently very wide and deep; humerus with bicipital crest somewhat reduced, with distal margin very straight, presenting a squared appearance, with head more pointed and distinctly set off from ventral tubercle, and brachial depression deep. Most authors have not departed from the opinions of Cassin and Sclater that Siphonorhis is closely related to Nyctidromus. However, because Siphonorhis is so distinctive in bill morphology, I previously suggested that the long tarsus shared by these two genera was probably an insufficient indication of close relationship, and that Siphonorhis may represent an ancient relict of an earlier caprimulgid stock that has been isolated in the West Indies (Olson 1978). The structure and proportions of the tarsometatarsus of Siphonorhis are now seen to be quite dif- ferent from those in Nyctidromus (Fig. 2). Furthermore, the skull is also very distinctive in Siphonorhis, with the rostrum being much broader, reflecting its external appearance, and the anterior portions of the pterygoids lacking the dis- tinctive expansion seen in Nyctidromus. I find no reason to consider Siphonorhis and Nyctidromus to be closely allied. Siphonorhis daiquiri, new species Figs. 1-3 Holotype. — Distal half of right tarsometatarsus, collections of the Department of Paleobiology, National Museum of Natural History, Smithsonian Institution, USNM 336506 (Figs. 1, 2b). Collected 31 Oct 1980 by Storrs L. Olson and others. Locality. —‘‘Cueva de Los Indios” (see Anthony 1919), ona hillside overlooking the roadstead of the former port of Daiquiri, on the southern coast of what is now the province of Santiago de Cuba (formerly part of Oriente Province), about 22 km ESE of the city of Santiago de Cuba (19°54.8'N, 75°38.6’W). Chronology. —Quaternary, probably Holocene. Measurements of holotype.—Length of specimen as preserved, 16.4 mm; shaft width above scar for hallux, 1.4 (1.2 in S. brewsteri; 1.9 in S. americana); distal width, 3.9 (3.4 in S. brewsteri). Paratypes.—Topotypes collected by Olson and party in 1980 consist of a left coracoid (USNM 336507), proximal ends of right (USNM 336508) and left (USNM VOLUME 98, NUMBER 2 529 ‘ Fig. 2. Right tarsometatarsi in posterior view. A, Siphonorhis americana (UF 68055); B, S. daiquiri, new species (holotype, USNM 336506); C, S. brewsteri (USNM 354527); D, Nyctidromus albicollis (USNM 344130). 3x natural size. 336509) humeri, and the proximal end ofa left carpometacarpus (USNM 336510). Topotypes collected by H. E. Anthony in 1917 include two right humeri (AMNH 21905, 21906), a right humerus lacking the distal end (AMNH 21907), the prox- imal end of a right humerus (AMNH 21908), a left humerus lacking the proximal end (AMNH 21909), and a right carpometacarpus lacking the minor metacarpal (AMNH 21904). Measurements of paratypes.—Humerus: length, 27.4 mm (24.8 in S. brewsteri; 31.5 in S. americana); shaft width at midpoint 2.0, 2.1, 2.1 (2.1 in S. brewsteri; 2.5 in S. americana); distal width 5.2, 5.6 (5.0 in S. brewsteri; 6.3-6.7 in S. americana). Carpometacarpus: length, 15.6; proximal depth, 4.4, 4.9. Length of coracoid, approximately 15.5. Additional locality and specimens.—In April 1982, E. N. Kurochkin (Paleon- tological Institute, USSR Academy of Sciences, Moscow) showed me the distal end of a humerus and the proximal end of a tarsometatarsus of this species that he collected in ‘““Cueva de Los Fosiles,”’ 28.5 km NE of Camagiiey, Camagiiey Province, Cuba. These were not at hand when the present description was written. The site is approximately 280 km NW of the type-locality. Etymology.—Named for the former port of Daiquiri, at the type-locality. This was the terminus of a railroad from an iron mine that was still operational at the time of H. E. Anthony’s visit in 1917 (Anthony 1919), but has since been aban- doned, there being scarcely a trace of the settlement left. Its name lives on, 530 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. Humeri in anconal view. A, Siphonorhis americana (AMNH 11020); B, S. daiquiri, new species (AMNH 21905); C, S. brewsteri (USNM 336511). 2x natural size. however, as a popular rum drink reputed to have been invented by a cyceonologist (Greek, kykeon, a mixed drink), who named it after his home town. In the present usage it is a noun in apposition. Diagnosis. —Larger than Siphonorhis brewsteri and smaller than S. americana. Scar for hallux more elevated than in the other species of Siphonorhis and shaft of tarsometatarsus proportionately more slender than in S. americana, being more like that in S. brewsteri. Discussion.—The cave deposits at the type-locality were first discovered and excavated in 1917 by H. E. Anthony, who reported on the mammalian remains (Anthony 1919). I visited the site briefly on 31 October 1980 as part of a joint expedition of the Smithsonian Institution and the Instituto de Zoologia, Academia de Ciencias de Cuba. At that time I excavated a small test pit near one wall of the cave, in sediments that proved to be richly fossiliferous. The abundant bird material in the American Museum collections that Anthony obtained was long neglected and for years had been mislabelled as having come from Puerto Rico, a fact wholly incompatible with the species composition of the sample and unequivocally controverted by the discovery of slips of paper amongst the specimens bearing the notation ““Daiquiri.”” The fossil avifauna as a whole from the Daiquiri cave will be treated elsewhere (Olson and W. Hilgartner, MS.). As with most fossils from West Indian cave deposits, no firm dates were as- sociated with the specimens of S. daiquiri. Both the mammalian (Anthony 1919) and avian (Olson and Hilgartner, MS.) faunas at Daiquiri consist mainly of extant species, with the more notable of the large extinct vertebrates of Cuba (e.g., ground sloths, giant owls, flightless cranes, condors) being absent. The species represented are all of a size compatible with their having been prey of the extant barn owl VOLUME 98, NUMBER 2 531 Tyto alba, individuals of which still roost in the cave. The fossils are not heavily mineralized. Although some extinct rodents, the extinct insectivores of the genus Nesophontes, and even some extinct species of bats (Anthony 1919; Woloszyn and Silva 1977) are present in the deposits, their disappearance could have taken place during the Holocene, as known for vertebrates in the Lesser Antilles (Stead- man etal. 1984). At this point, the deposits can only be assigned to the Quaternary, but I consider it likely that they are post-Pleistocene in age. Despite the fact that Siphonorhis daiquiri is known only from fossil remains, it would be premature to regard it as extinct. Being nocturnal and secretive, caprimulgids are often difficult to detect in life, and the species of Siphonorhis may be particularly recondite. As mentioned, Gosse (1847) never met with S. americana on Jamaica, although the species still existed in his time. On the ornithologically well-explored island of Hispaniola, S. brewsteri was not discov- ered until 1917. It eluded Wetmore and most subsequent collectors and was considered rare everywhere except Gonave Island, where Bond obtained a series in 1928 (Wetmore and Swales 1931). Since then, however, it has been shown to be fairly generally distributed in the Dominican Republic (Dod 1979). On Jamaica, S. americana is presumed to be extinct, and it is unlikely to have survived the plague of monogooses (Herpestes) on that island. Mongooses were also introduced to Hispaniola and Cuba but have not achieved the population densities apparent on Jamaica. Siphonorhis brewsteri has persisted on Hispaniola, so there is no good reason why S. daiquiri should be extinct on Cuba. The very arid regions of Cuba east of Daiquiri have been relatively little explored ornithologically. Only in 1959 was a distinctive, isolated population of the sparrow Torreornis inexpectata discovered in the desert-like area near Baitiquiri (Spence and Smith 1961), and it is doubtful that collectors have spent much time in this region at night to hear any nightjars that might be calling. The rugged, arid landscape in this area is reminiscent of that in parts of the American southwest and in the unfrequented reaches away from the coast Siphonorhis daiquiri may still exist. Acknowledgments For assistance in the field I am grateful to my fellow expedition members from the Smithsonian, James F. Lynch and Eugene S. Morton. Our success was due to the support we received from the Instituto de Zoologia, Academia de Ciencias de Cuba, especially from Fernando Gonzales, Hiram Gonzales, Noel Gonzales, and Nicasio Vina. I am grateful to Charlotte Holton, American Museum of Natural History (AMNH), and S. David Webb, Florida State Museum (UF), for lending fossil specimens of Siphonorhis, to William B. Hilgartner for aid in sorting and identifying material from Daiquiri, and to Evgeny N. Kurochkin for permitting me to examine and report on the fossils he collected. J. P. Angle expertly removed the bones from a study skin of S. brewsteri. The photographs are by Victor E. Krantz. Literature Cited Anthony, H. E. 1919. Mammals collected in eastern Cuba in 1917. With description of two new species. — Bulletin of the American Museum of Natural History 41(20):625-643, pl. 35-37. Bond, J. 1928a. A remarkable West Indian goatsucker.— Auk 45:471-474. 532 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 1928b. The distribution and habits of the birds of the Republic of Haiti.— Proceedings of the Academy of Natural Sciences of Philadelphia 80:483-521. Cassin, J. 1851. Notes of an examination of the birds composing the family Caprimulgidae, in the collection of the Academy of Natural Sciences of Philadelphia.— Proceedings of the Academy of Natural Sciences of Philadelphia 5:175-190. Chapman, F. M. 1917. Descriptions of new birds from Santo Domingo and remarks on others in the Brewster-Sanford collection.—Bulletin of the American Museum of Natural History 37: 327-334. Dod, A. S. 1979. The Least Pauraque in the Dominican Republic.—American Birds 33:826-827. Gosse, P. H. 1847. The birds of Jamaica. London, John van Voorst. 447 pp. Linnaeus, C. 1758. Systema naturae. 10th edition. Volume 1. Stockholm, Laurentius Salvius. 824 pp. Olson, S. L. 1978. A paleontological perspective of West Indian birds and mammals. /n Frank B. Gill, ed., Zoogeography in the Caribbean. The 1975 Leidy Medal Symposium.— Academy of Natural Sciences of Philadelphia Special Publication 13: 11 + 128 pp., pp. 99-117. ——., and D. W. Steadman. 1977. A new genus of flightless ibis (Threskiomithidae) and other fossil birds from cave deposits in Jamaica. — Proceedings of the Biological Society of Washington 90(2):447—457. Peters, J. L. 1940. Check-list of birds of the world. Volume 4. Cambridge, Massachusetts, Harvard University Press. 291 pp. Ray, J. 1678. The ornithology of Francis Willughby. London, John Martyn. 441 pp. Sclater, P. L. 1861. List of a collection of birds made by the late Mr. W. Osburn in Jamaica, with notes. — Proceedings of the Zoological Society of London 1861:69-82, plate 14. Sloane, H. 1707. A voyage to the islands Madera, Barbados, Nieves, S. Christophers and Jamaica, with the natural history of the herbs and trees, four-footed beasts, fishes, birds, insects, reptiles, &c. of the last of those islands; to which is prefix’d an introduction, wherein is an account of the inhabitants, air, waters, diseases, trade, &c of that place, with some relations concerning the neighbouring continent, and islands of America. 2 volumes. London, printed for the author. 9 + 274 pp. Spence, M. J., and B. L. Smith. 1961. A subspecies of Torreornis inexpectata from Cuba.—Auk 78: 95-97. Steadman, D. W., G. K. Pregill, and S. L. Olson. 1984. Fossil vertebrates from Antigua: evidence for late Holocene human-caused extinctions in the West Indies.— Proceedings of the National Academy of Sciences U.S.A. 81:4448-4451. Wetmore, A., and B. H. Swales. 1931. The birds of Haiti and the Dominican Republic.— United States National Museum Bulletin 155, 483 pp. Woloszyn, B. W., and G. Silva Taboada. 1977. Nueva especie fésil de Artibeus (Mammalia: Chi- roptera) de Cuba, y tipificacién preliminar de los depésitos fosiliferos Cubanos contentivos de mamiferos terrestres.—Poeyana 161:17 pages, 5 figures. Department of Vertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 98(2), 1985, pp. 533-538 CIRRIPECTES IMITATOR, A NEW SPECIES OF WESTERN PACIFIC BLENNIID FISH Jeffrey T. Williams Abstract. — Cirripectes imitator is described from Taiwan, Japan, and the Oga- sawara Islands. Cirripectes imitator, C. castaneus and C. polyzona frequently have been confused because all have similar color patterns. Cirripectes imitator is ‘distinct among Cirripectes species in having the following combination of char- acters: a broad, expanded flap bearing cirri on each side of the nape; numerous pores at most cephalic sensory pore positions; dorsal-fin rays XII, 14; anal-fin rays II, 15; 7-14 lateral line tubes; lateral line tubes extending posteriorly to or almost to caudal-fin base; adult color pattern consisting of pale spots or bars on a dark background; nuchal cirri 43 or more; and total vertebrae 30. While revising the Indo-Pacific blenniid genus Cirripectes Swainson, I discov- ered an undescribed species from Taiwan, Japan, and the Ogasawara Islands. The similarity of the color pattern of this species to that of two other Cirripectes species has caused workers to make numerous misidentifications. To solve this problem, I have decided to publish the new species description prior to completion of the revision. Counts follow Smith-Vaniz and Springer (1971) as modified by Williams and Maugé (1983) except as follows: number of distinct nuchal cirri bases were counted instead of the free tips; nasal and supraorbital cirri counts include free tips on both left and right sides; gill rakers refer to the total number of rakers on the first arch; nuchal flap is a broadly expanded flap bearing the ventralmost group of cirri on each side of nape; mid-snout pores (Fig. 1, MSP) refer to two pore positions (one each in left and right supraorbital series) in the middle of the snout between the left and right posterior nostrils; extra interorbital pore position (EIP) refers to a pore position (Fig. 1) in both left and right supraorbital series that lies immediately behind and to each side of the supraorbital commissural pore po- sition; pore positions behind nuchal flap (PBN) are those behind the lower part of the nuchal flap (Fig. 1). The following institutional abbreviations are used: BPBM, Bernice P. Bishop Museum, Hawaii; FAKU, Department of Fisheries, Faculty of Agriculture, Kyoto University, Japan; NTUM, National Taiwan University Museum; UF, Florida State Museum, University of Florida; USNM, National Museum of Natural His- tory, Smithsonian Institution, Washington, D.C. Other abbreviations used are: SL, standard length and LL, lateral line. Cirripectes imitator, new species Fig. 2 Holotype.—FAKU 48203 (male: 65.6 mm SL), Ogasawara Islands, Chichi Jima Island (approximately 27°30'N, 142°30’E), Sakiaura; 0.5—S m; 8 Apr 1974. Paratypes.—TAIWAN: BPBM 23228 (12 specimens: 34.0-81.6 mm SL), east 534 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Diagrammatic illustration of cephalic sensory pore system of Cirripectes imitator based on a50 mm SL male (USNM 258316). A, Dorsal view of head; B, Left-lateral view of head. Abbreviations: AN, anterior nostril; EIP, extra interorbital pore position (position on right side indicated); IFO, infraorbital series; MD, mandibular series; MSP, mid-snout pores; PBN, pore positions behind nuchal flap; PN, posterior nostril; POP, preopercular series; SO, supraorbital series; ST, supratemporal series. First and last pore positions indicated for each series. Dashed lines above eyes and across nape indicate bases of cirri not illustrated. coast off San Shien Tai, 0-2 m; NTUM 5777-1 and 2 (2: 78.1—79.1), Su-ao Harbor; NTUM 5780 (1: 42.1), Yeh-jen-chuen (Lan-yu); UF 41606 (1: 41.4), 22°40'N, 121°29’E, Green Island, tidepool; USNM 227979 (2: 80.6—92.9), rocky headland NW of Sha Tao, 1-6 m; USNM 227980 (6: 73.0—90.8), Ch’uan-fan-shih, 4-6 m; USNM 258315 (8: 46.1-57.2) and USNM 258316 (2: 47.7-50.3), 22°40'N, 121°29’E, Green Island, tidepool. RYUKYU ISLANDS: FAKU 48109 (1: 52.0), Okinawa Island, Hedomisaki; FAKU 50400 (1: 71.6) and FAKU 50402 (1: 34.7), Okinawa Island; FAKU 47989 (1: 56.8), Okinawa, Sesoko Jima, 0.5—5 m; FAKU VOLUME 98, NUMBER 2 535 Fig. 2. Typical color patterns of Cirripectes imitator new species. Top, male holotype, FAKU 48203 (65.6 mm SL); middle, female, FAKU 48187 (61.9 mm SL); bottom, young male, FAKU 48478 (31.2 mm SL). [111459] (1: 74.9) and FAKU [111460] (1: 76.2), Yoron Jima, Kanebo; FAKU [111461] (1: 101.4), Okino Erabu Shima. HONSHU, JAPAN: FAKU [111458] (1: 56.8), FAKU 48268 (1: 45.9), FAKU 48269 (1: 51.7), FAKU 48272 (1: 46.0), FAKU 48398 (1: 42.7), FAKU 48399 (1: 38.5), FAKU 48400 (3: 39.5—53.6), 536 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON FAKU 48478 (1: 31.2), FAKU 48963 (1: 28.3), and FAKU 48968 (1: 34.1), Shirahama, 0.5-4 m. JAPAN: USNM 76288 (1: 45.1), no other data. MIYAKE JIMA: BPBM 18983 (1: 75.2), Chodara-ika, 0-2 m. OGASAWARA ISLANDS: FAKU 48184 (1: 52.4), FAKU 48185 (1: 58.6), FAKU 48186 (1: 54.7), and FAKU [111455] (6: 50.6-62.4), Chichi Jima Island, 0.5-5 m; FAKU 48187 (1: 61.9), Nishi Jima Island, 0.5—5 m. Description.— Dorsal fin XI-XIII (XII in 95.2% of specimens), 13-15 (14 in 95.2%); total dorsal-fin elements 26; anal fin II, 14-16 (15 in 95.2%); total pro- current caudal—fin rays 11-14; pelvic fin I, 3-4 (bilaterally I, 4 in 88%, bilaterally I, 3 in 6%, one side I, 3 and other I, 4 in 6%); vertebrae 10 + 20 = 30; last pleural ribs on vertebral centrum 11; last epipleural rib on vertebral centrum 18-21 (19- 20 in 85.5%); anal pterygiophores 1-1-1, 1-1-2, 1-2-1 (1-1-2 in 92.1%); nuchal cirri 40-59 (45-53 in 82.4%); supraorbital cirri 13—46 (15-35 in 82.7%); nasal cirri 9-93 (9-33 in 84.6%); LL tubes 5-14 (8-13 in 94.6%); last LL tube beneath dorsal-fin ray 9 to caudal-fin base (to caudal-fin base in 84.5%); no scale-like flaps along LL; lower lip smooth mesially (plicate laterally); gill rakers 24-27 (based on 22 specimens); pseudobranchial filaments on one side 7—9 (based on 22 spec- imens); premaxillary teeth 192—230 (based on 11 specimens); dentary teeth 84—- 120 (based on 11 specimens); upper lip crenulae approximately 34-50; nuchal cirri in four groups (two on each side) separated at dorsalmost point on nape and about halfway down length of row on each side (gaps less than about 0.5 mm), two groups on one side rarely connected by a low basal membrane, ventralmost group of cirri on each side borne on a broad nuchal flap; adults of both sexes with first dorsal-fin spine slightly longer than second (1—5 mm longer in males, 1-2 mm longer in females); dorsal-fin membrane deeply incised above last dorsal-fin spine; dorsal-fin membrane attached to caudal fin in adults; cephalic pore system complex (numerous pores at most positions; number of pores increases with increasing SL; Fig. 1); mid-snout pores present (Fig. 1); extra interorbital pore position present (Fig. 1); pore positions behind nuchal cirri flap 2 (Fig. 1); male genital papilla with urogenital orifice located basally between two widely separated slender filaments (less than 1.0 mm long) on a fleshy swelling behind anus; testes bulbous, width equals length; maximum size about 100 mm SL. The smallest gravid female (ova about 0.5 mm diameter) examined was 55 mm SL, from the Ogasawara Islands. Some 40-50 mm SL females had large ovaries, but they were granular in appearance and had no large ova. Males mature by about 50 mm SL. No geographic variation was noted for the meristic or morphometric characters examined. Fukao (1984) noted that his Cirripectes polyzona specimens from the Ogasawara Islands (=C-. imitator at that locality) had bright yellow spots or blotch- es in life. I attribute this color variant to populational variation within the species. Color in alcohol.—Color pattern is highly variable, ranging from alternating dark- and pale-brown bars to dark-brown reticulations around pale-brown pupil- sized spots on the body. Adult males tend to be darker overall, with spots co- alescing to form pale bars on body. Females generally exhibit the reticulated pattern, but both sexes can exhibit either pattern. Pupil-sized spots on cheeks, snout, upper lip, and underside of head. These spots are pale brown on darker background or the reverse in either sex. In some specimens, spots on underside of head fuse to form 2 or 3 alternating dark and pale bars across the throat. Dorsal VOLUME 98, NUMBER 2 537 fin with translucent triangular area in anterodorsal part of spinous section, a narrow pale stripe runs through middle of fin to last dorsal spine, remainder of spinous and soft dorsal brown; upper part of caudal fin with translucent triangular area, remainder of fin brown; anal fin brown, tips of rays paler; pectoral and pelvic fins dusky. Rugosities on anal-fin spines of males pale-brown. Nuchal cirri black, other cirri brown. Color in life.— Male (based on fig. 39-27a in Shen 1984, which he referred to as Cirripectes sebae) with alternating brown and bluish-white bars on head and body; those on head broken into irregular pupil-sized spots. Pale section in caudal and spinous-dorsal fins with orange rays; red stripe runs through middle of spinous dorsal. Each nuchal cirrus with pale-yellow band near base. Iris with yellow ring around pupil. Other colors same as in alcohol. Fukao (1984) states that the Oga- sawara specimens have bright yellow spots or blotches on body in life. Females (based on fig. 392—7b in Shen 1984; and plate 87-D in Masuda et al. 1975) similar to male except spots on body are not fused into bars; spots become smaller posteriorly. There is some doubt about the identification of the specimen in fig. 392—7b (Shen 1984). I tentatively identify it as Cirripectes imitator, but females of C. castaneus have a very similar color pattern and this could be a specimen of the latter species. Geographic distribution. — Cirripectes imitator is known to occur from Taiwan northward to Shirahama, Japan, and in the Ogasawara Islands. Etymology.—The specific epithet is derived from the Latin imitor, meaning to mimic, and refers to the similarity of the color pattern to that of Cirripectes polyzona and C. castaneus. Relationships and comparisons. — Cirripectes imitator belongs to the C. fusco- guttatus complex of species, which is not yet resolved, but with which it shares two derived characters: a high number of nuchal cirri (usually 45 or more) and an enlarged nuchal flap on either side of the head. A large nuchal flap is also present in C. auritus and C. kuwamurai (Carlson 1981; Fukao 1984), but the nuchal flaps in these species appear quite different and are probably independently derived. Cirripectes imitator differs from the C. fuscoguttatus complex in having 7-14 (in 98.2% of specimens; | of 56 with 5) LL tubes, whereas the other members of the C. fuscoguttatus complex have 0-4 (in 98.5%; 2 of 69 with 5). The color pattern of C. imitator, with pale spots on a dark background, is the reverse of C. fuscoguttatus, which has dark spots on a pale background. Over some part of its geographic range, Cirripectes imitator is sympatric with eight congeneric species: C. fuscoguttatus Strasburg and Schultz, C. filamentosus (Alleyne and Macleay), C. castaneus (Valenciennes), C. perustus Smith, C. quagga (Fowler and Ball), C. polyzona (Bleeker), C. kuwamurai Fukao, and C. variolosus (Valenciennes). The only species of Cirripectes known to occur in the Ogasawara Islands are C. imitator and C. variolosus, and this is the only locality where their geographic ranges overlap. The reticulated color pattern of C. imitator easily distinguishes it from C. variolosus, which has a brown body and small pale spots (red in life) covering head. In Japan and Taiwan, C. imitator is most frequently confused with C. castaneus and C. polyzona. Cirripectes polyzona (plate 87-C in Masuda et al. 1975; and fig. 392-6 in Shen 1984) differs from C. imitator in having 1 pore position (versus 2) behind lower part of ventralmost group of cirri on each side, 32—42 (in 94.7%; range 32—44) nuchal cirri (versus 43-59 in 96.1%; range 538 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 40-59), ventralmost nuchal cirri not borne on a large flap (versus large nuchal flap present), and I, 3 pelvic-fin rays (versus I, 3-I, 4 or I, 4-I, 4 in 94%). Both sexes of C. polyzona have the barred color pattern, only C. castaneus males are barred, and male C. imitator may have either bars or spots. Cirripectes castaneus differs from both of these species in having relatively few (1-2) pores (versus 3 or more) at most cephalic sensory pore positions. The similarity of these three species is evidenced by the recent treatment of the Japanese Cirripectes species by Fukao (1984). Mr. Fukao made available the specimens used in his analysis and a list of catalog numbers corresponding to his “C. polyzona’’ color pattern types A, B, and C. I found that his type A pattern included female C. castaneus and both sexes of C. imitator, his type B pattern consisted of male C. imitator and one female C. castaneus; and his type C pattern comprised male and female C. polyzona, male and female C. castaneus, and one male C. imitator. In addition to the spotted and barred color patterns, C. castaneus specimens sometimes have a uniform brown head and body (see Fukao 1984, fig. 3-C). This pattern usually is seen after a specimen has been preserved and may be an artifact of the pres- ervation process. Cirripectes imitator differs from the other sympatric species in having the fol- lowing combination of characters: large expanded nuchal flap bearing ventralmost group of nuchal cirri, dorsal-fin segmented rays 14, anal-fin segmented rays 15, numerous pores at most cephalic sensory pore positions, LL tubes present to or almost to caudal-fin base, 43 or more nuchal cirri, and 30 total vertebrae. Acknowledgments For the loan of specimens, I thank J. E. Randall, S. C. Shen, J.-J. Lin, R. Fukao, and H. K. Mok. Mr. R. Fukao kindly provided information on the specimens used in his revision of Japanese Cirripectes. V. G. Springer read and provided helpful comments on an earlier draft of the manuscript. This project was supported in part by National Science Foundation Grant DEB 8207313. Literature Cited Carlson, B. A. 1981. A new Indo-Pacific fish of the genus Cirripectes (Blenniidae, Salariini).— Pacific Science 34(4):407—414. Fukao, R. 1984. Review of the Japanese fishes of the genus Cirripectes (Blenniidae) with description of a new species. —Japanese Journal of Ichthyology 31(2):105-121. Masuda, H., C. Araga, and T. Yoshino. 1975. Coastal fishes of southern Japan. Tokai University Press, 382 pp. Shen, S.-C. 1984. Coastal fishes of Taiwan. Taipei, Taiwan. 190 pp. Smith-Vaniz, W. F., and V. G. Springer. 1971. Synopsis of the tribe Salariini, with description of five new genera and three new species (Pisces: Blenniidae).—Smithsonian Contributions to Zoology 73:1—72. Swainson, W. 1839. The natural history and classification of fishes, amphibians, and reptiles, or monocardian animals. Longman, Orme, Brown, and Longmans, London, vol. 2, 452 pp. Williams, J. T., and L. A. Maugé. 1983. Cirripectes chelomatus, a new species of salarine fish (Pisces, Blenniidae).— Bulletin du Muséum National d’Histoire Naturelle, Paris, serie 4, 5(section A, no. 4):1139-1149. Department of Vertebrate Zoology (Fishes), National Museum of Natural His- tory, Smithsonian Institution, Washington, D.C. 20560. 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 will be published in English (except for Latin diagnosis/description of plant taxa which should not be duplicated by an English translation), with a summary in an alternate language when appropriate. Publication Charges.—Authors will be asked to assume publication costs of page-charges, tabular material, and figures, at the lowest possible rates. Submission of manuscripts.—Manuscripts should be sent to the Editor, Proceedings of the Biological Society of Washington, National Museum of Natural History, Smithsonian Insti- tution, Washington, D.C. 20560. Review.—One of the Society’s aims is to give its members an opportunity for prompt pub- lication of their shorter contributions. Manuscripts will be reviewed in order of receipt by a board of associate editors and appropriate referees. 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. Synonymy of abbreviated style (author, date, page) (except in botanical papers), with full citations of journals and books (no abbreviations) in the Literature Cited is required. The establishment of new taxa must conform with the requirements of the appropriate in- ternational codes of nomenclature. Authors are expected to be familiar with these codes and to comply with them. New species-group accounts must designate a type-specimen deposited in an institutional collection. The sequence of material should be: Title, Author(s), Abstract, Text, Acknowledgments, Literature Cited, Author’s (s’) Address(es), Appendix, List of Figures (entire figure legends), Figures (each numbered and identified), Tables (each table numbered with an Arabic numeral ' and heading provided). Manuscripts should be typed, double-spaced throughout (including tables, legends, and foot- notes) on one side of 844 x 11 inch sheets, with at least one inch of margin all around. Submit a facsimile with the original, and retain an author’s copy. Pages must be numbered on top. One manuscript page = approximately 0.5 printed page. Underline singly scientific names of genera and lower categories; leave other indications to the editor. Figures and tables with their legends and headings should be self-explanatory, not requiring reference to the text. Indicate their approximate placement by a pencil mark in the margin of the manuscript. Illustrations should be planned in proportions that will efficiently use space on the type bed of the Proceedings (12.5 x 20 cm) and should not exceed 24 x 15 inches. Figures requiring solid black backgrounds should be indicated as such, but not masked. Art work will be returned only on request. Proofs.—Galley proofs will be submitted to authors for correction and approval. Reprint orders will be taken with returned proof. Costs.—Page charges @ $60.00, figures @ $10.00, tabular material $3.00 per printed inch. All authors are expected to pay the charges for figures, tables, changes at proof stage, and reprints. Payment of fuil costs will probably facilitate speedy publication. CONTENTS A new species of the colubrid snake genus Liophis from Brazil James R. Dixon The genus Paranais (Oligochaeta: Naididae) in North America Ralph O. Brinkhurst and Kathryn A. Coates Pandion lovensis, a new species of osprey from the late Miocene of Florida Jonathan J. Becker Neotropical Monogenea. 7. Parasites of the Pirarucu, Arapaima gigas (Cuvier), with descriptions of two new species and redescription of Dawestrema cycloancistrium Price and Nowlin, 1967 (Dactylogyridae: Ancyrocephalinae) Delane C. Kritsky, Walter A. Boeger, and Vernon E. Thatcher Four new species of Scalibregmatidae (Polychaeta) from the Gulf of Mexico, with comments on the familial placement of Mucibregma Fauchald and Hancock, 1981 Jerry D. Kudenov A new species of Parotocinclus (Pisces: Loricariidae) from Guyana Robert E. Schmidt and Carl J. Ferraris, Jr. Two new species of Stratiodrilus, S. aeglaphilus and S. pugnaxi (Annelida: Histriobdellidae) from Chile Irma Vila P. and Nibaldo Bahamonde N. Saxipendium coronatum, new genus, new species (Hemichordata: Enteropneusta): the unusual spaghetti worms of the Galapagos Rift hydrothermal vents Keith H. Woodwick and Terry Sensenbaugh Nomenclatural and taxonomic notes on the Pteridophytes of Costa Rica, Panama, and Colom- bia, II David B. Lellinger A review of Cummingsia Ferris (Mallophaga: Trimenoponidae), with a description of two new species Robert M. Timm and Roger D. Price Crystallodytes pauciradiatus (Perciformes), a new creediid fish species from Easter Island Joseph S. Nelson and John E. Randall The voices and relationships of the Chilean frogs Eupsophus migueli and E. calcaratus (Am- phibia: Anura: Leptodactylidae) J. R. Formas A new species of the aquatic beetle genus Dryopomorphus from Borneo (Coleoptera: Elmidae: Larinae) Paul J. Spangler Miocene and Pliocene Molidae (Ranzania, Mola) from Maryland, Virginia, and North Carolina (Pisces: Tetraodontiformes) Robert E. Weems Speonebalia cannoni, n. gen., n. sp., from the Caicos Islands, the first hypogean leptostracan (Nebaliacea: Nebaliidae) Thomas E. Bowman, Jill Yager, and Thomas M. Iliffe Additional branchiate scale-worms (Polychaeta: Polynoidae) from Galapagos hydrothermal vent and rift-area off western Mexico at 21°N Marian H. Pettibone The generic and subfamilial classification of the Naididae (Annelida: Oligochaeta) R. O. Brinkhurst Deeveyinae, a new subfamily of Ostracoda (Halocyprididae) from a marine cave on the Turks and Caicos Islands Louis S. Kornicker and Thomas M. Iliffe Quadryops, new genus, and three new species of arboreal Dryopidae (Insecta: Coleoptera) from Panama and Ecuador Philip D. Perkins and Paul J. Spangler A new species of Bivibranchia (Pisces: Characiformes) from Surinam, with comments on the genus Richard P. Vari Psychropotes hyalinus, new species, a swimming elasipod sea cucumber (Echinodermata: Holo- thuroidea) from the North Central Pacific Ocean David L. Pawson A new species of Siphonorhis from the Quaternary cave deposits in Cuba (Aves: Caprimulgidae) Storrs L. Olson Cirripectes imitator, a new species of western Pacific blenniid fish Jeffrey T. Williams 295 303 314 321 332 341 347 Sil 366 3911 403 411 416 422 439 447 470 476 494 511 523 526 533 THE BIOLOGICAL SOCIETY OF WASHINGTON 1983-1984 Officers President: Donald R. Davis Secretary: Gordon L. Hendler Vice President: Austin B. Williams Treasurer: Leslie W. Knapp Elected Council J. Laurens Barnard Maureen E. Downey Frederick M. Bayer Louis S. Kornicker Isabel C. Canet Storrs L. Olson Custodian of Publications: David L. Pawson PROCEEDINGS Editor: Brian Kensley Associate Editors Classical Languages: George C. Steyskal Invertebrates: Thomas E. Bowman Plants: David B. Lellinger Vertebrates: Richard P. Vari Insects: Robert D. Gordon Membership in the Society is open to anyone who wishes to join. There are no prerequisites. Annual dues of $15.00 include subscription to the Proceedings of the Biological Society of Washington. Library subscriptions to the Proceedings are: $18.00 within the U.S.A., $23.00 elsewhere. 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 Treasurer, Biological Society of Washington, National Museum of Natural History, Smithsonian Instutution, Washington, D.C. 20560. 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. PROC. BIOL. SOC. WASH. 98(3), 1985, pp. 539-543 A NEW SPECIES OF DRAGONET, SYNCHIROPUS RANDALLI, FROM EASTER ISLAND (TELEOSTEI: CALLIONY MIDAE) Guy T. Clark and Ronald Fricke Abstract. —A new species of dragonet, Synchiropus randalli, is described from a male and two females collected at Easter Island. It is most closely related to S. kiyoae Fricke & Zaiser, 1983, from Japan. The new species is characterized by a first dorsal fin that is dusky except for two white blotches on the first membrane in the male, white blotches on the pelvic fin, a band of black and white blotches along the sides of the body, a third spine of the male’s first dorsal fin which is much shorter than the first and second spines, a shorter caudal fin, and a broader interorbital space. Easter Island is of interest to biologists because of its isolated geographic position at the easternmost fringe of Polynesia (27°08’S, 109°23’W). Although relatively little is known of its fish fauna (Allen 1970), it appears to have a high rate of endemics, about 27.5% of the 109 recorded species (Randall 1970, 1973, 1976). The most recent collections are those of Ian Efford and associates during the Canadian Medical Expedition in 1964-1965, and Randall and Allen in 1969. Prior to these only 40 species had been recorded (Allen 1970). Within the material collected by J. E. Randall and G. R. Allen in 1969, were included specimens of a previously undescribed species of the postulus species- group of the dragonet genus Synchiropus Gill, 1860 (Fricke 1981, 1983). Other species of this species-group are S. kiyoae Fricke & Zaiser, 1983, from Japan, S. laddi Schultz, 1960, from the western and central Pacific, S. minutulus Fricke, 1981, from the central Indian Ocean, S. postulus Smith, 1963, from the western Indian Ocean, and S. springeri Fricke, 1983, from Fiji. Methods follow Fricke (1983). The type-material of the new species is deposited in the Bernice P. Bishop Museurn, Honolulu (BPBM), and the National Museum of Natural History, Smithsonian Institution, Washington, D.C. (USNM). Synchiropus randalli, new species Fig. 1 Holotype. —BPBM 6754, male, 22.0 mm SL, Easter Island, west coast off south- ern end of Hanga Roa, sand near rocks 27°09.5’S, 109°27'W, 40 feet (12 m) depth, J. E. Randall and G. R. Allen, 10 Feb 1969. Paratypes.—Same data as holotype. BPBM 26409, 1 female, 20.9 mm SL; USNM 221485, 1 female, 18.6 mm SL. Diagnosis. —A Synchiropus of the postulus species-group with 4 spines in the first dorsal fin, 9 rays in the second dorsal fin, 8 rays in the anal fin, 19-20 pectoral 4-5 fin rays, a preopercular spine formula——— 1, the first dorsal fin of the male dusky (except for two white blotches on the first membrane), with the third spine much 540 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Synchiropus randalli, new species, Easter Island. BPBM 6754, holotype, male, 22.0 mm SL: A, Lateral view; B, Left preopercular spine. BPBM 26409, paratype, female, 20.9 mm SL: C, Lateral view. shorter than the first and second spines, the first dorsal fin of the female higher than the second dorsal fin, the male with light blue blotches on the sides of the head, both sexes with a black and white band along the sides of the body and with white blotches on the pelvic fin, and with the caudal fin 4.5—4.6 in SL. Description. —D, IV; D, viii,1; A vii,1; P, 1,16-17, 11 (totally 19-20); P, 1,5; C (i-11), 1,7, 11 (i-11). Proportions of the holotype and the larger female paratype see Table 1. Body elongate and slightly compressed. Head slightly depressed (females) or slightly compressed (male), 3.5—3.9 in SL. Body depth 6.6—7.7 in SL. Eye diameter 2.5—2.9 in HL. Preorbital length 3.6—3.9 in HL. Interorbital distance 4.2—6.0 in eye. Occipital region with smooth bony plate. Branchial opening sublateral in position. Preopercular spine length 3.4—4.3 in HL; preopercular spine with up- curved main tip, smooth ventral margin, no antrorse spine at its base, and 4—5 ; d 4-5 d ; curved points on its dorsal margin (formula——— 1; see Fig. 1B). Urogenital VOLUME 98, NUMBER 3 541 Table 1.—Proportions of the male holotype and one female paratype of Synchiropus randalli, new species (expressed as hundredths of SL). Holotype Paratype, BPBM 26409 Predorsal (1) length 31.04 29.86 Predorsal (2) length 45.02 41.81 Preanal fin length 52.12 48.16 Prepelvic fin length 27.08 21.79 Head length 25.58 28.57 Body depth 15.16 13.09 Caudal peduncle length 20.89 19.59 Caudal peduncle depth 6.24 5.06 Caudal fin length 21.89 22.02 Eye diameter 9.97 9.94 Preorbital length 7.10 7.36 Urogenital papilla 1.14 = First D, spine length 42.60 15.86 First D, ray length 16.70 11.99 Last D3 ray length 14.02 9.84 First A ray length 11.20 9.46 Last A ray length 11.92 10.08 Pectoral fin length 15.84 14.52 Pelvic fin length 34.36 35.45 papilla elongate in male, 22.5 in HL; not visible in female. Lateral line reaching from preorbital region to end of third branched caudal fin ray (counted from above), with short suborbital and short preopercular branch, as well as short ventral branch above pectoral fin base; lateral lines of opposite sides intercon- nected by commissure across occipital region. Caudal peduncle length 4.8—5.1 in SL. Caudal peduncle depth 16.0-19.8 in SL. First dorsal fin very high in male, first to third spines elongate, with very short filaments, first spine 0.6 in HL, third spine shorter than first and second spines; in female relatively high, first to third spines elongate but not filamentous, longer than first ray of second dorsal fin, first spine 1.8 in HL. Predorsal (1) length 3.22-— 3.35 in SL. Second dorsal fin distally straight (female) or slightly convex (male), first ray in male 1.53 in HL, in female 2.38 in HL. Last ray in male 1.82, in female 2.90 in HL. Rays unbranched except for last which is divided at its base. Predorsal (2) length 2.22—2.39 in SL. Anal fin beginning on vertical through base of second membrane of second dorsal fin. Last anal fin ray in male 2.14 in HL, in female 2.83 in HL. Rays unbranched except for last which is divided at its base. Preanal fin length 1.92—2.08 in SL. Pectoral fin distally convex, reaching back to base of first anal fin membrane. Pelvic fin large, fourth ray elongate, reaching back to base of third anal fin membrane. Pelvic fin length 0.74-0.81 in HL. Pelvic fin connected with mid-base of pectoral fin by membrane. Prepelvic fin length 3.69—4.59 in SL. Caudal fin distally convex. Caudal fin length 4.5—4.6 in SL. Color in alcohol. —Head and body brown, back with few dark brown saddles and dark brown and whitish spots. Ventral portions of body whitish in both sexes. Side of head with dark brown blotches; in male also with ocellate dark streaks which are light blue in fresh specimens. Side of body below lateral line with band 542 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 2.—Comparison between Synchiropus randalli, new species and S. kiyoae Fricke & Zaiser, 1983. S. randalli S. kiyoae Body depth in SL 4.9-6.9 6.6—7.4 Interorbital in eye 4.3-5.9 6.3-24.1 Predorsal (2) length in SL 2.2-2.4 1.8-2.24 Caudal fin length in SL 4.5—4.6 3.0-3.8 Third D, spine (male) much shorter than 2nd spine subequal to 2nd spine First dorsal fin (male) dusky, with light blotches on light, with elongate vertical dark lst membrane streaks and blotches First dorsal fin (female) lst membrane dusky lst membrane whitish, distal one- third dusky Caudal fin with 2 rows of white blotch- without white blotches es Pelvic fin with white blotches without white blotches Anal fin (female) translucent, without blotches with a dark brown blotch distally on each membrane Sides of head (male) with oval light blue blotches with ocellate light blue streaks consisting of alternating black and white blotches. Eye brown, with dark brown spot in its upper rostral section. First dorsal fin dusky in both sexes, in male with two light blotches on first membrane. Second dorsal fin translucent, in male with about three dark brown spots on each ray, and with basal dusky (in fresh specimens blue) blotch on each membrane. Anal fin dusky in male, translucent in female. Caudal fin with 1—2 median horizontal rows of white blotches, and about three vertical rows of dark brown spots on rays. Pectoral fin translucent; pelvic fin with white blotches. Sexual dimorphism. — Males have a higher first dorsal fin than females, the first and second spines being much longer than the third, a longer urogenital papilla, and a different color pattern of the second dorsal and anal fins and of the sides of the head. Distribution. —This species is apparently endemic to Easter Island. No Syn- chiropus of the postulus species-group has been recorded from Pitcairn Island, Ducie Atoll Rapa, or any other islands near Easter Island. It has been collected at a depth of 12 meters, which is about the same depth at which S. kiyoae occurs in Japan (5—13.5 m). Etymology. —This new species is named in honor of Dr. John E. Randall, whose contributions considerably increased our knowledge of the ichthyofauna of Easter Island, and who collected the type-material of the new species. Comparisons.—Synchiropus randalli is compared with the most closely related species, S. kiyoae Fricke & Zaiser (1983:122—128, figs. 1-2; Fricke 1983:603- 608, fig. 185), in Table 2. It differs from Synchiropus laddi Schultz (1960:406- 409, fig. 131; Fricke 1981:124—126, fig. 39; Fricke 1983:608-611, fig. 187) in the proportions of the first dorsal fin spines in the male (in S. /addi second spine subequal to third, first spine shorter), in the first dorsal fin which is lower than the second dorsal fin, and in the general color pattern (S. /addi: completely pale, except for the eye and two ventral black spots on each side of the body). Syn- chiropus randalli can be distinguished from S. minutulus Fricke (1981:119-123, VOLUME 98, NUMBER 3 543 fig. 38; Fricke 1983:624—627, fig. 192) by the same characters as from 5S. laddi. Synchiropus postulus Smith (1963:560, fig. 7, pl. 86E; Fricke 1981:116-118, fig. 37; Fricke 1983:658—660, fig. 203) is distinguished by the proportions of the first dorsal fin spines of the male (second and third spines longest, first spine shorter), by the presence of long filaments in that fin, by the first dorsal fin of the female which is lower than the second dorsal fin, and by the general color pattern. Synchiropus springeri Fricke (1983:673-677, fig. 208) differs in the proportions of the first dorsal fin of the female (which is lower than the second dorsal fin), the bright rose pink occipital region, and the color pattern of the head, the body, and the pelvic fin. The two species of the ocellatus species-group of the genus Synchiropus occurring in the central Pacific, S. ocellatus (Pallas, 1770) (Fricke 1983:635-642, fig. 197, east to Marquesas Islands and Pitcairn) and S. morrisoni Schultz, 1960 (Fricke, 1983:630-—635, figs. 195-196, east to Marshall and Fiji islands), differ from S. randalli in having eight rays in the second dorsal fin, seven : 1-2 : rays in the anal fin, preopercular spine formulae of——— 1, and completely dif- ferent color patterns. Acknowledgments We thank R. S. Birdsong of Old Dominion University for reviewing the manu- script, and J. E. Randall (BPBM, Honolulu) for the loan of specimens. Literature Cited Allen,G. R. 1970. Two new species of frogfishes (Antennariidae) from Easter Island. — Pacific Science 24:517-522. Fricke, R. 1981. Revision of the genus Synchiropus (Teleostei: Callionymidae). J. Cramer, Braun- schweig. 194 pp. . 1983. Revision of the Indo-Pacific genera and species of the dragonet family Callionymidae (Teleostei). J. Cramer, Braunschweig. x + 774 pp. , and M. J. Zaiser. 1983. A new callionymid fish, Synchiropus kiyoae, from the Izu Islands, Japan.— Japanese Journal of Ichthyology 30(2):122—-128. Gill, T. N. 1860. On the genus Callionymus of authors.— Proceedings of the Academy of Natural Sciences of Philadelphia, (1859):128—130. Randall, J. E. 1970. Easter Island, an ichthyological expedition.— Oceans 3(3):49-59. 1973. Expedition to Pitcairn.—Oceans 6(2):12-21. —. 1976. The endemic shore fishes of the Hawaiian Islands, Lord Howe Island, and Easter Island. Colloque Commerson 1973.—ORSTOM Traveaux et Documents 47:49-73. Schultz, L. P. 1960. Family Callionymidae. Jn Schultz, L. P., et al., Fishes of the Marshall and Marianas Islands. 2.— Bulletin of the United States National Museum 202(2):397—410. Smith, J. L. B. 1963. Fishes of the families Draconettidae and Callionymidae from the Red Sea and Western Indian Ocean.— Rhodes University, Ichthyological Bulletin 28:547-564, pls. 83-86. (GTC), 533 New York Ave., Norfolk, Virginia 23508; (RF), Staatliches Natur- historisches Museum, D-3300 Braunschweig, Federal Republic of Germany. PROC. BIOL. SOC. WASH. 98(3), 1985, pp. 544-553 A NEW SPECIES OF BULLFINCH (AVES: EMBERIZINAE) FROM A LATE QUATERNARY CAVE DEPOSIT ON CAYMAN BRAC, WEST INDIES David W. Steadman and Gary S. Morgan Abstract. —A new species of bullfinch, Melopyrrha latirostris, is described from latest Pleistocene to early Holocene cave deposits on Cayman Brac, West Indies. This species is larger in its cranial dimensions than other species of West Indian finches. Melopyrrha latirostris represents one of many species of vertebrates that is known on Cayman Brac only from fossils. Specimens referred to M. nigra taylori, which occurs today only on Grand Cayman, were recovered from the same stratigraphic levels as M. Jatirostris. The Cayman Islands are three small islands in the Caribbean Sea, about halfway between Cuba and Jamaica (Fig. 1). In 1965, Dr. Thomas H. Patton, then of the University of Florida, excavated several deposits of fossiliferous sediment from limestone caves on Cayman Brac, the easternmost island. From the richest of these sites, known herein as Patton’s Fissure, thousands of vertebrate fossils were recovered from deposits of presumed Holocene age. These fossils, reported first by Patton (1966), are discussed in detail by Morgan (in press). Patton’s Fissure is located in the village of Spot Bay, 3 km west of the northeast point of Cayman Brac at 19°45’N and 79°45’W. It is in the side of a cliff about 15 m above sea level and 250 m inland from the northern coast. Patton’s Fissure is about 50 m long, a maximum of 4 m wide at the base, and trends east to west, parallel to the cliff face. A layer of unconsolidated sediments 1—2 m deep covers the entire fissure. These sediments consist of buff to reddish-colored silts and clays, angular limestone fragments, land snail shells, and bones of small verte- brates. Three holes were excavated in Patton’s Fissure, one of which (Hole 1) produced a significant amount of bone. Hole | was approximately 2 m square by 1.6 m deep. The stratigraphy of Patton’s Fissure is as follows: Layer 1 (0-20 cm) contains abundant bone, including both extinct endemic mammals as well as introduced species such as Rattus that indicate a post-Columbian age. Layers 2— 4 (20-80 cm) are sparsely fossiliferous, but contained no introduced species or evidence of human occupation, thus indicating a pre-Columbian age for these and all deeper layers. Layers 5—7 (80-140 cm) are extremely rich in both land snail shells and bones of small vertebrates. Most of the bird fossils described in this paper, and the great majority of all vertebrate fossils from Cayman Brac, are from these three layers. Layers 8—9 (140-160 cm) contain few bones and many are either covered with a calcareous precipitate or are contained in an indurated breccia. Solid limestone was encountered below Layer 9. Unfortunately, neither Patton’s original field notes nor the results from several radiocarbon ages determined in the late 1960’s for Patton’s Fissure are available, although Patton (pers. comm.) refers the age of the lower levels of this site to the early Holocene. This age is reasonable based upon preservation of the fossils and VOLUME 98, NUMBER 3 545 FLORIDA NES) Sroneat anaco GRAND BAHAMA s ELEUTHERA a. —_— p Ye ATLANTIC OCEAN ANDROS. ou SAN SALVADOR GREAT, EXUMA LONG CROOKED ACKLINS MAYAGUANA SS eBencaicos ‘ ° { GREAT INAGUA ZTORTUE 2 MEXICO Geozumer LITTLE CAYMAN CAYMAN BRAC i=) HAITI ) GRAND CAYMAN 5 CNET © DOMINICAN ¢ REPUBLIC RIC ye || JAMAICA Vee NGUI nae ) Uf ST Eusratius/“% @anticua Ki y MEvIEg OMONTSERRAT GUADELOUPE OMARIE GALANTE Aeommuica HONDURAS Sant CARIBBEAN SEA SS UARTINGUE »PROVIDENCIA ST. VINCENT): {Si BARBADOS 3 GRENADINES y ARUBA SAN ANDRES fee cuRAGAO e Ocrenava BUBONAIRE ABLANQUILLA TOBAGO. MARGARITA a: 2 (yaa 2 TORTUGA S PANAMA, COLOMBIA VENEZUELA ‘ Fig. 1. The West Indian Islands. upon comparisons with fossil faunas from other West Indian caves. Recently we dated three samples of land snails from Patton’s Fissure, using the single-species methodology of radiocarbon age determination on land snails developed for Ja- maican species (Goodfriend and Hood 1983; Goodfriend and Stipp 1983). Our samples of Caymanian snails (Hemitrochus caymanensis) should provide a fairly accurate estimate of the age of Patton’s Fissure, for H. caymanensis is an arboreal snail that does not feed on the ground (F. G. Thompson, pers. comm.). Therefore, this species should incorporate little if any “dead carbon”’ into its shell through ingestion of limestone. The age determinations are (in years BP, with lab number): 11,180 + 105 (Layer 5, SI-6518); 13,230 + 135 (Layer 7, SI-6519); and 13,850 + 135 (Layer 9, SI-6520). These concordant results represent maximum ages, de- pending upon the level at which the dated snails had incorporated environmental carbonate into their shells during life. The radiocarbon data suggest an age of latest Pleistocene or earliest Holocene for the fauna from Layers 5—9 of Patton’s Fissure. The fauna from Patton’s Fissure includes the extinct capromyid rodents Cap- romys and Geocapromys and the insectivore Nesophontes, as well as several living species of bats that no longer occur on Cayman Brac. Based upon Minimum Number of Individuals, lizards dominate the fauna of Patton’s Fissure (67%), followed by mammals (25%, most of which are Nesophontes), and birds (8%, not including unidentified passerines). 546 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON The avian fossils from Patton’s Fissure are mainly of small passerines that remain incompletely studied. Conspicuous among these passerine fossils are nu- merous cranial elements of a finch that is much larger than TJiaris olivacea, the only emberizine known historically from Cayman Brac. We describe these fossils as representing two sympatric forms of Melopyrrha, of which one is extinct and the other survives only on Grand Cayman, a larger island 130 km west of Cayman Brac. Systematic Paleontology Class Aves Order Passeriformes Family Fringillidae Subfamily Emberizinae Genus Melopyrrha In possessing the following characters, the series of fossils from Patton’s Fissure may be referred to Melopyrrha rather than to the closely related West Indian emberizine genera Tiaris, Loxipasser, Loxigilla, Euneornis, or Melanospiza. (De- scriptive terminology follows Baumel et al. 1979; fossil specimens are deposited in the Vertebrate Paleontology Collections of the Florida State Museum [UF], while modern skeletal specimens are from the National Museum of Natural His- tory, Smithsonian Institution [USNM]. Maxilla.—In lateral aspect, more arched (curved) along both the dorsal and ventral surfaces (most closely approached by Loxipasser); relatively broad medial bar of Os nasale; presence of a small but distinct foramen in medial portion of the lateral bar of Os nasale, near the dorsal margin of the nares; relatively much shorter and stouter than in Euneornis campestris or Melanospiza richardsoni. Mandible.—In dorsal aspect, distal end of pars symphysialis less pointed; in lateral aspect, dorsal surface of dentary more curved than in all except Jiaris bicolor and Loxipasser anoxanthus; mandibular foramen relatively small; overall much stouter and more “‘finch-like”’ than in Euneornis campestris. Quadrate. — Except for differences in size, it is difficult or impossible to distinguish individual quadrates among the six closely related genera of West Indies finches mentioned above. The fossil quadrates from Cayman Brac differ from those of the only other similarly-sized nine-primaried oscine in the fossil deposit (the tanager Spindalis zena; Thraupinae) in forming an obtuse angle in lateral aspect between the processus quadratojugalis and condylus squamosus, this angle being more nearly 90° in Spindalis. In describing Melopyrrha taylori from Grand Cayman, Hartert (1896) doubted the distinctness of Melopyrrha from other (unspecified) genera of finches. Standard check-lists (i.e., Bond 1956, Paynter 1970, AOU 1983) recognize all six of the emberizine genera discussed above, although we believe that most or all of these genera can be accommodated in an expanded genus 7J7varis Swainson, 1827, on the basis of plumage and osteology. These finches represent an unrecognized evolutionary radiation within the West Indies, in many ways comparable to that of emberizines in the Galapagos Islands. This West Indian emberizine radiation and its systematic ramifications have not yet been fully documented, so we will describe the new species from Cayman Brac in the genus Melopyrrha rather than in Tiaris. VOLUME 98, NUMBER 3 Fig. 2. Maxillae and skulls of fossil and modern Melopyrrha. Lateral aspect in left column, dorsal aspect in right column. A, M. nigra nigra male, USNM 321962, Cuba; B, M. n. taylori male, USNM 554483, Grand Cayman; C, M™. Jatirostris, holotype, UF 23011, Cayman Brac; D, M. Jatirostris, paratype, UF 61022, Cayman Brac. Scale bar = 1 cm. 548 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Melopyrrha latirostris, new species Figs. 2, 3 Holotype. —Complete maxilla, UF 23011, from Hole 1, Layer 7 of Patton’s Fissure, Spot Bay, Cayman Brac, Cayman Islands. Collected by T. H. Patton during the summer of 1965 (exact date unknown). Paratypes. — All from Hole 1, Layer 7 of Patton’s Fissure. 12 complete or nearly complete quadrates, UF 61008-61019; 9 complete or partial maxillae, UF 61020- 61028; 13 incomplete mandibles, UF 23012, 61029-61042. Referred material. — All from Hole 1 of Patton’s Fissure. Layer 2— partial max- illa, UF 23016. Layer 4— partial maxilla, UF 23015. Layer 5—complete or nearly complete quadrates, UF 61001, 61003, 61005. Layer 6—partial mandible, UF 23013. Diagnosis. —Larger than Melopyrrha nigra, especially in width of maxilla and height of mandible (Table 1, measurements A, B, D, F, G). Nares relatively small compared to size of entire maxilla. In lateral aspect, ventral surface of mandible nearly straight (VW. nigra with a distinctly obtuse angle at junction of Os suran- gulare and Os dentale). In lateral aspect, dorsal surface of Os dentale relatively straight. Os dentale proportionately long relative to length of entire mandible. Etymology. —From the Latin /atus, broad, and rostrum, bill or snout. The name latirostris is regarded as a noun in apposition. Discussion Evolution. — Melopyrrha nigra, the only living species in the genus, occurs today on Cuba and the Isle of Pines (M. n. nigra) and Grand Cayman (M. n. taylori). Melopyrrha latirostris is much closer in size to M. n. taylori than to M. n. nigra (Table 1), and on this basis it is likely that MW. /atirostris evolved from a population of M. n. taylori or its immediate progenitor that became isolated on Cayman Brac. Nevertheless, several of the fossils from Hole 1 of Patton’s Fissure are much too small to be referred to M. /atirostris and are similar in size to modern spec- imens of M. n. taylori (Table 1). These smaller specimens, which we refer to M. n. taylori, include a mandible (UF 61045) from Layer 5, and a mandible (UF 61043) and a quadrate (UF 61006) from Layer 6. From these same layers are six other specimens that are intermediate in size between M. Jatirostris and M. n. taylori (a maxilla [UF 23014], a mandible [UF 61046], and two quadrates [UF 61002, 61004] from Layer 5, and a mandible [UF 61044] and quadrate [UF 61007] from Layer 6). We cannot say with certainty whether these last specimens represent very small female individuals of M. /atirostris, or very large male in- dividuals of M. n. taylori, or hybrids between the two species. No specimens of M. n. taylori were recovered from Layer 7, the most fossil- iferous layer collected. This fact suggests that M. /atirostris was already established on Cayman Brac before M. n. taylori colonized (or re-colonized) the island. The intermediate specimens suggest that genetic interchange may have occurred be- tween M. Jatirostris and M. n. taylori at that time, and the two specimens of M. latirostris from Layers 2 and 4 suggest that this species may have outlived its congener, only to disappear as well sometime in the Holocene. We regard MM. latirostris as a full species rather than a subspecies of M. nigra because of its VOLUME 98, NUMBER 3 549 Fig. 3. Mandibles and quadrates of fossil and modern Melopyrrha. Dorsal aspect of mandibles in left column, lateral aspect of quadrate in right column. A, M. nigra nigra male, USNM 321962, Cuba; B, M. n. taylori male, USNM 554483, Grand Cayman; C, M. n. taylori-M. latirostris intermediate fossil, UF 61046, Cayman Brac; D, M. latirostris paratype, UF 23012, Cayman Brac; E, ™. /atirostris paratype, UF 61008, Cayman Brac. Scale bar = 1 cm. sympatry with M. n. taylori and because of its very large size; it is larger relative to M. n. taylori than the latter is to M. n. nigra (see ratios in Table 1). The maxilla, quadrate, and mandible of M. Jatirostris are broader and more massive than in other West Indian finches. All or nearly all of the diagnostic 550 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Measurements (in mm) of fossil and modern Melopyrrha, giving mean, sample size (in parentheses), and range. UF catalogue numbers are given in parentheses for individual fossils referred either to M. nigra taylori or to M. n. taylori-M. latirostris intermediates. A Maxilla: minimum Maxilla: minimum Maxilla: length from width of medial bar width of lateral bar nares to tip of Os of Os nasale of Os nasale premaxillare M. nigra nigra 1.3 (7) 0.5 (8) 6.1 (7) Cuba, males 1.1-1.7 0.4—0.7 5.8-6.5 M. n. taylori 1.7 (7) 0.6 (8) 7.0 (8) Grand Cayman, males 1.5-1.9 0.5-0.8 6.7-7.4 M. n. nigra 1.4 (2) 0.6 (2) 5.8 (2) Cuba, females 1.2-1.6 0.6 5.7-5.9 M. n. taylori 1.5 (5) 0.6 (5) 6.5 (5) Grand Cayman, females 1.4-1.7 0.5-0.6 6.3-6.7 M. latirostris 2.5 (10) 1.2 (9) 8.2+ (3) 2.2-2.8 1.1-1.5 8.0+—-8.4+ M. n. taylori — — — fossils, Cayman Brac Intermediate fossils, 2.1 (UF 23014) 0.9 (UF 23014) 7.0+ (UF 23014) Cayman Brac Ratio of mean in M. n. nigra to that in 0.76 0.83 0.87 M. n. taylori (males) Ratio of mean in M. n. taylori males to 0.68 0.50 0.86 or less mean in M. Jatirostris ““+°° after a value for certain fossil specimens means that the measurement of a slightly damaged specimen approaches to within 0.4 mm or less the actual value of the measurement if the specimen had been undamaged. characters of M. /atirostris are associated allometrically with its large size. The large, rounded maxilla of M. /atirostris is reminiscent of that found in Geospiza crassirostris, a frugivorous emberizine finch from the Galapagos. The maxilla of M. latirostris, is more powerfully built than that of G. crassirostris, especially in the nasal region, so M. Jatirostris may have subsisted on a mixed diet of fruit and seeds. Alternatively, M. /atirostris may have been mainly a seed-eater, for its larger bill would have permitted it to take a variety of seeds. Further speculation on the feeding habits of M. /atirostris awaits better documentation of the feeding habits of living M. nigra. The only report we have found on this topic is Johnston’s (1975:300) for M. n. taylori of Grand Cayman, where insects and unidentified seeds were found in the stomachs of four different birds. Zoogeography. —The two species of Melopyrrha from Cayman Brac are most closely related to M. n. nigra from Cuba, reflecting the dominant zoogeographic pattern seen in the extinct and living vertebrate fauna from the island. In the combined vertebrate fauna from five cave deposits excavated on Cayman Brac, 17 (81%) of the 21 species of known zoogeographic affinities are conspecific with or are derived from Cuban species, whereas the remaining four species have Jamaican affinities (Morgan, in press). Several physical and biological factors favor Cuba over Jamaica as a source area for most of the vertebrate fauna of Cayman Brac. These include: the considerably larger area, and longer coastline of Cuba; VOLUME 98, NUMBER 3 551 Table 1.—Extended. F Mandible: G I D maximum Mandible: height at H Mandible: Maxilla: E height of junction of Mandible: maximum length of Quadrate: maximum Mandible: Os sur- Os dentale and width of entire cotyla total width total length angulare Os surangulare Os dentale lateralis height 5.6 (8) 18.2(5) 3.9(7) 2.7(8) 8.0 (7) 2.2(8) 5.2 (7) 5.0-6.1 17.4-18.6 3.5-4.4 2.3-3.1 7.8-8.5 2.1-2.4 5.0-5.3 6.3 (7) 19.6 (5) 5.0 (8) 3.4 (8) 9.5 (6) 2.6 (8) 5.6 (8) 6.1-6.6 19.2-20.1 4.5-5.3 3.2-3.6 9.2-9.9 2.4-2.7 5.5-5.9 5.5 (2) 172@Q 38@) 240 7.4 (2) DOO? 50) 5.4-5.6 17.1-17.3 3.8-3.9 2.4 7.3-7.5 2.0 4.9-5.2 5.8 (5) 18.6(2) 4.4(4) 3.1 (4) 8.5 (3) 2.4(5) 5.4 (4) 5.7-5.9 18.3-18.9 4.3-4.5 2.9-3.2 8.2-8.7 2.2-2.6 5.3-5.5 8.3+ (8) 23.4 (1) 6.5 (2) 4.7 (8) 11.4 (8) 3.0 (4) 7.1 (15) 7.6+-8.8 23.4 6.4-6.6 4.1-5.0 10.3-12.3 2.8-3.2 6.6-7.5 — — — 3.1 (UF 61043) 8.7(UF 61045) — 6.1 (UF 61006) 3.2 (UF 61045) 7.0+ (UF 23014) — = 3.8 (UF 61044) — = 6.4 (UF 61002, 3.9 (UF 61046) 61004, 61007) 0.89 0.93 0.78 0.79 0.84 0.85 0.93 0.76 or less 0.84 0.77 0.76 0.83 0.87 0.79 the closer proximity of Cuba to the Cayman Islands during Pleistocene glacial intervals; Cuba’s greater species diversity; and, today’s prevailing currents favor Overwater dispersal from Cuba rather than Jamaica. Cayman Brac is almost equidistant (200 km) from Cuba to the east and northeast and from Jamaica to the southeast, and is separated from both islands by oceanic depths in excess of 1000 m, eliminating the possibility of land bridges during the late Tertiary. How- ever, during periods of lower sea level in Pleistocene glacial intervals, Cuba would have extended to within 100 km of Cayman Brac as the extensive carbonate bank areas along its southern coast became exposed. The lack of evidence for a land connection leaves overwater dispersal as the only means by which Cayman Brac could have received its vertebrate fauna. Based on the low percentage of endemic species and the absence of generic level endemism on Cayman Brac, we believe that the majority of the fauna arrived during the Pleistocene. Extinction. —Forty species of vertebrates have been identified from Holocene cave deposits on Cayman Brac (8 species of reptiles, 23 of birds, and 9 of mam- mals), the great majority of which are from Hole | of Patton’s Fissure (Morgan, in press). Of these 40 species, 17 (11 species of birds and 6 of mammals) no longer occur on Cayman Brac. Of these 17 species, six are still found on Grand Cayman, seven no longer occur in the Cayman Islands but exist elsewhere in the West Indies, and four (including Melopyrrha latirostris) are extinct species known only SoZ PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON from the Cayman Islands. The stratigraphy and chronology of the five caves excavated on Cayman Brac are not known well enough to determine precisely when the majority of these 17 species disappeared from the island. Fossils of all six species of mammals now extinct on Cayman Brac have been collected in caves from surface remains that are believed to be less than 500 years old based upon the presence of Rattus. On the other hand, only one of the 11 extirpated species of birds, Puffinus [herminieri, has been recovered from these same surface layers, while the remaining 10 species are known only from the pre-Columbian strata in Hole 1 of Patton’s Fissure. From the data available, we cannot determine whether these 10 species, which include both species of Melopyrrha, disappeared from natural causes before AD 1500 or were extirpated as a result of extensive habitat disturbance by post-Columbian peoples. There is no evidence of aboriginal occupation of any of the Cayman Islands (Hirst 1910; Richards 1955), so all Holocene habitat alteration can be attributed to post-Columbian settlers. Thus it is conceivable that the extinction of either or both forms of Melopyrrha on Cayman Brac was an historic event, but we need a refinement of the chronology of the upper sediments at Patton’s Fissure or other fossil sites before the chronology of extinction of Melopyrrha on Cayman Brac can be resolved beyond “probably late Holocene.” Nevertheless, the stratigraphic evidence suggests that M. /atirostris may have swamped out M. n. taylori through interbreeding. That two congeneric finches could co-inhabit an island as small as Cayman Brac is not extraordinary, for until recently a parallel situation existed on St. Kitts in the Lesser Antilles. Two species of bullfinches, Loxigilla noctis and L. porto- ricensis grandis, occurred on St. Kitts until several decades ago when L. p. grandis apparently became extinct (Olson 1984). St. Kitts is not much larger than Cayman Brac in area, but is much higher in elevation, supporting lush forest in the volcanic highlands. However, Olson points out that L. p. grandis may have evolved in the lowlands of St. Kitts, so habitat diversity may have played little if any role in permitting the two species of Loxigilla to co-exist on St. Kitts. Although the limestone forest of Cayman Brac has a low canopy height and a low species diversity today, birds elsewhere in the West Indies tend to be relatively abundant in both absolute numbers and numbers of species in arid habitats (Kepler and Kepler 1970; Pregill and Olson 1981). Thus we see no reason why the prehistoric forests of Cayman Brac could not have supported two or more species of con- generic finches. Acknowledgments We profited from discussions with G. A. Goodfriend, S. L. Olson, and F. G. Thompson. R. Stuckenrath provided radiocarbon ages. We thank V. E. Krantz for the photographs, and M. E. Parrish for their help in labelling. E. M. Paige drew Fig. 1. Comments by R. I. Crombie, H. F. James, and S. L. Olson improved the manuscript. Literature Cited American Ornithologists’ Union. 1983. Check-list of North American birds, 6th ed. Allen Press, Inc., Lawrence, Kansas, 877 pp. Baumel, J. J., A.S. King, A. M. Lucas, J. E. Breazile, and H. E. Evans (eds.). 1979. Nomina Anatomica Avium. Academic Press, London, 664 pp. VOLUME 98, NUMBER 3 553 Bond, J. 1956. Check-list of birds of the West Indies. Academy of Natural Sciences of Philadelphia, 214 pp. Goodfriend, G. A., and D. G. Hood. 1983. Carbon isotope analysis of land snail shells: implications for carbon sources and radiocarbon dating.— Radiocarbon 25:8 10-830. , and J. J. Stipp. 1983. Limestone and the problem of radiocarbon dating of land-snail shell carbonate.— Geology 11:575-577. Hartert, E. 1896. Description of a new finch from the West Indies.—Novitates Zoologicae 3:257. Hirst, G. S. S. 1910. Notes on the history of the Cayman Islands. P. A. Benjamin Manufacturing Company, Kingston, Jamaica, 412 pp. Johnston, D. W. 1975. Ecological analysis of the Cayman Island avifauna.— Bulletin of the Florida State Museum, Biological Sciences 19(5):235—300. Kepler, C. B., and A. K. Kepler. 1970. Preliminary comparison of bird species diversity and density in Luquillo and Guanica forests. Pp. E-183-186 in H. T. Odum, ed., A tropical rain forest: a study of irradiation and ecology at El Verde, Puerto Rico. U.S. Atomic Energy Commission, Division of Technical Information, Oak Ridge, Tennessee. Morgan, G. S. [In press.] Late Pleistocene and Holocene fossil vertebrates from the Cayman Islands, West Indies.— Bulletin of the Florida State Museum, Biological Sciences. Olson, S. L. 1984. The last St. Kitts bullfinch (Loxigilla portoricensis grandis) and the extinction of its race.— Bulletin of the British Ornithologists’ Club 104(4):121-123. Patton, T. H. 1966. Occurrence of fossil vertebrates on Cayman Brac, B. W. I.—Caribbean Journal of Science 6(3—4):181. Paynter, R. A., Jr. 1970. Subfamily Emberizinae. Pp. 3-214 in R. A. Paynter, Jr., ed., Check-list of birds of the world, vol. 13. Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts. Pregill, G. K., and S. L. Olson. 1981. Zoogeography of West Indian vertebrates in relation to Pleistocene climatic cycles.— Annual Review of Ecology and Systematics 1981, 12:75-98. Richards, H. G. 1955. The geological history of the Cayman Islands.—Notulae Naturae 285:1-11. (DWS) Department of Vertebrate Zoology (Birds), Smithsonian Institution, Washington, D.C. 20560. Present address: New York State Museum, 3140 Cul- tural Education Center, Albany, New York 12230; (GSM) Florida State Museum, University of Florida, Gainesville, Florida 32611. PROC. BIOL. SOC. WASH. 98(3), 1985, pp. 554-560 REMASELLUS, A NEW GENUS FOR THE TROGLOBITIC SWIMMING FLORIDA ASELLID ISOPOD, ASELLUS' PARVUS STEEVES Thomas E. Bowman and Boris Sket Abstract. —Asellus parvus Steeves, 1964, from Ten Inch Cave, Alachua Co., Florida, is assigned to’a new genus, Remasellus, characterized by a simple per- eopod 1, broad natatory pereopods 2-7 fringed with long setae, and male pleopod 1 without retinacula. New records are reported from the Split Spring Cave system, Wakulla Co., Florida, and the Peacock Spring cave system, Suwanee County, Florida. Its behavior is unusual for an asellid in that in addition to walking on the substrate like an ordinary asellid, it also swims weakly in the open water. Preceding the Eighth International Congress of Speleology at Western Kentucky University, Bowling Green, Kentucky, 18-24 July 1981, an International Cave Diving Camp was held in Florida in which the second author participated. In two of the caves a swimming isopod was collected that proved to be Asellus parvus Steeves, 1964. The natatory pereopods 2-7, not described or illustrated by Steeves, together with other features have led us to place it in the new genus described below. Remasellus, new genus Diagnosis. —Mandible with 3-segmented palp. Maxilla 1, inner lobe with 5 apical spines. 2 maxilliped oostegite with apical plumose setae. Pereopod | simple, not subchelate; pereopods 2—7 natatory/ambulatory, with broad segments fringed with closely spaced setae. 6 pleopod 1 without retinacula. 6 pleopod 2 endopod with small inner and outer basal apophyses, tip with several processes. 2 pleopods 2 not overlapping medially. Uropod exopod very short; endopod long (é only?). Type-species.—Asellus parvus Steeves, 1964. Etymology. —Remus = oar (Latin) + Asellus, referring to the shape and func- tion of pereopods 2-7. Remasellus parvus (Steeves) Figs. 1-4 Asellus parvus Steeves, 1964:450—451, figs. 10-14.—Fleming, 1973:295 (in list), 299 (in key). Material. —FLORIDA: Alachua Co., Ten Inch Cave, 5 mi N of Newberry, leg. R. D. Warren: 9 Jun 1961, 1 6, 1 2, USNM 111142, holotype and allotype (not separated in vial); 4 Jun 1961, 1 9, USNM 111140; 20 Jan 1962, 1 46, 3.3 mm, USNM 111141.— Wakulla Co., Split Sink cave system, leg. Boris Sket and Wayne Marshall, Jul 1981, 1 6, 5.5 mm, USNM 213321; 4 9, 4.7, 6.0, 7.0, 8.7 mm, USNM 213332.—Suwannee Co., Peacock Spring cave system, leg. Boris Sket and Wayne Marshall, Jul 1981, 1 6, 2.7 mm, 2 2, 1.9, 2.35 mm, USNM 213323. VOLUME 98, NUMBER 3 555 Fig. 1. Remasellus parvus, A-H, J—L, from Split Spring specimens; I, from Peacock Spring é: A, Habitus, 7.0 mm 2; B-H, Pleura of pereonites 1-7, 7.0 mm @; I, Antenna 1, 2.7 mm 4; J, Left mandible, palp omitted; K, Right mandible, distal segment of palp omitted; L, Incisor of right mandible. 556 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON B Mf " i \ Fig. 2. Remasellus parvus, A-C, F—G, from Split Spring specimens; D-E, from Peacock Spring 3: A, Maxilla 1; B, Maxilliped, 2; C, Endite of same, anterior; D, Pereopod 1; E, Pereopod 2; F, Un- determined pereopod, distal segments; G, Pereopod 3 or 4. VOLUME 98, NUMBER 3 557 Collections by Sket and Marshall were made with a “squeeze bottle” (Iliffe and Sket, in press). Although the holotype and allotype were not separated, it can be safely assumed that Steeves intended the male to be the holotype since Steeves always selected males for holotypes. Curiously, Steeves stated that two allotypes were deposited, an impossibility since by definition there can be only a single allotype, and he failed to list the additional 6 and 2 although they were identified by him and are clearly topotypes. Description. —Blind, unpigmented. Body about 3 x as long as wide, widening from head to pereonite 3; pereonites 3—7 subequal in width. Head about 1.5 x as wide as long; anterior margin shallowly concave; postmandibular lobes incon- spicuous. Coxae visible dorsally on all pereonites. Telson about 0.8 x as wide as long; lateral margins slightly convex; posterior margin slightly concave on either side of truncate caudomedial lobe. Antenna 1, flagellum with short proximal segment and 3 distal segments pro- gressively longer; esthete formula 1-1. Antenna 2 broken off at base in all speci- mens. Mandibles with 4-cuspate incisors and lacinia; spine-row with 13 and 15 spines in left and right mandibles. Maxilla 1 with 5 and 13 apical spines on inner and outer lobes. Maxilliped with 5 retinacula; endite densely setose; epipod with few setae. Pereopods of Split Spring specimen all detached and free in vial. Pereopods of Peacock Spring specimens absent except attached pereopods 1-2 in 6. Detached pereopods identified by comparison with those of topotype é (Fig. 4A-E). Pereopod 1 slender, simple; ischium and merus with 3—4 long naked setae on anterior margin; carpus very short; propus slightly longer than ischium, with a few setae on both margins; dacty] slightly shorter than propus, biunguiculate, with 2-3 spines on posterior margin. Pereopod 2 much longer than pereopod 1, with long setae on anterior margin of ischium, merus and carpus, and on posterior margin of carpus and propus; shorter setae on anterior margin of propus; propus slightly more than twice as long as dactyl. Pereopods 3-6 successively longer, with flat expanded segments; basis usually with several broom setae on | margin; merus, carpus, and propus with rows of setae on both margins; dactyl with 3-5 spines on flexor margin. Pereopod 7 much longer than pereopods 5-6, segments longer and narrower; carpus and propus with long setae on both margins. 6 pleopod 1 endopod narrowly oval, about 1.6 x as long as protopod; apex and distal third of lateral margin armed with short non-plumose setae. 6 pleopod 2 protopod unarmed; exopod proximal segment with | lateral seta, distal segment obovate with 14 setae on apical margin and distal half of lateral margin; endopod with small basal apophyses, tip with short broad cannula, broadly rounded caudal process excavated centrally, and short recurved pointed lateral process. 2 pleopod 2 with straight medial margin; lateral margin diverging gradually to widest point of appendage slightly beyond midlength where setae begin, then converging to narrow apex; 14 plumose setae on lateral and apical margins. Other pleopods in poor condition, not examined. 558 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. Remasellus parvus, A-H, M, from Split Spring specimens; K—L, from Peacock Spring ¢: A, Pereopod 3 or 4; B, Pereopod 5 or 6; C, Pleopod 1, 6; D, Pleopod 2, 6; E-L, Pleopod 2, 6, endopod tip in different aspects; M, Pleopod 2, 2°. VOLUME 98, NUMBER 3 559 Fig. 4. Remasellus parvus, 6 topotype: A, Pereopod 1; B, Pereopod 2; C, Pereopod 4; D, Pereopod 5; E, Pereopod 7; F, Pleopod 2; G, Pleopod 2, endopod tip. Uropod with narrow protopod and endopod subequal in length, latter with cluster of long apical setae; exopod very short, less than one-seventh as long as protopod. The single uropod found appears to be from a ¢; in other asellids such unequal uropod rami are present only in the 6; rami of the 2 are nearly equal. Remarks. —Some differences in detail can be seen between the type-locality specimens (Fig. 4) and those from the other localities (Figs. 1-2). Moreover, we are not absolutely certain that the much smaller specimens from Peacock Spring are identical with the larger Split Spring specimens. When adequate material becomes available the question of conspecificity should be carefully studied. Relationships. —Remasellus is quite distinct from all other genera of Asellidae. Its apomorphic character states, simple pereopod | and natatory pereopods 2-7, and 6 pleopod 1 without retinacula, are not found in combination in other genera of the family. The reduced basal apophyses of 6 pleopod 1, also apomorphic, are found in some species of Caecidotea, the only other genus of Asellidae known 560 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON from Florida. Three species of Caecidotea are known from Florida: C. racovitzai australis Williams, 1970, and C. obtusus Williams, 1970, both epigean, and the troglobitic C. hobbsi (Maloney, 1939). None of these is significantly similar to Remasellus parvus. Habitat. —The Split Sink cave system, near Tallahassee, is a corridor occupied by a subterranean river, accessible through several collapse sinks. Remasellus was fairly common on the walls of one of the large drowned halls. The walls were covered by a compact black deposit that was tattered and fragile. The Peacock Spring cave system is one of the largest drowned cave systems surveyed thus far (Exley and DeLoach 1981). Remasellus was found in one of the side branches where the bottom was partly covered with silt deposits. Behavior. —The Split Sink specimens behaved like the usual asellid while on the rock walls. When removed form the walls to the water column, they swam slowly with what appeared to be metachronal movements of the pereopods and slight lateral sigmoid movements of the body. The body was in a vertical position with the head held upwards and the uropods downwards close to one another. The Peacock Spring specimens were seen in the water about 5 cm above the silt, also in a vertical position with the head up and the uropods down. A thin trail of silt was present between the uropods and the bottom. Acknowledgments We are grateful to Wayne Marshall, who dove with the second author and called his attention to the presence of swimming asellids. Literature Cited Exley, S., and N. DeLoach. 1981. The world’s longest underwater cave.— Proceedings of the Eighth International Congress of Speleology 1:16-17. Fleming, L. E. 1973. The evolution of the eastern North American isopods of the genus Asellus (Crustacea: Asellidae).— International Journal of Speleology 5:283-310. Iliffe, T. M., and B. Sket. [in press]. A simple device for repetitive collections of macrofauna by divers. Maloney, J. O. 1939. A new cave isopod from Florida.— Proceedings of the United States National Museum 86(3057):457—459. Steeves, H. R., III. 1964. The troglobitic asellids of the United States: the Hobbsi group.— American Midland Naturalist 71(2):445-451. Williams, W. D. 1970. A revision of the North American epigean species of Ase//us (Crustacea: Isopods).— Smithsonian Contributions to Zoology 49:1—80. (TEB) Department of Invertebrate Zoology (Crustacea), NHB 163, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560; (BS) Institute za Biologijo, Univerze v Ljubljani, Askerceva 12, P.O. Box 141/3, 61001 Ljubljana, Yugoslavia. PROC. BIOL. SOC. WASH. 98(3), 1985, pp. 561-563 EUPLEURODON PERUVIANUS (RATHBUN, 1923); A SPECIES OF CRAB NEWLY RECORDED FROM CHILE (CRUSTACEA: DECAPODA: BRACHYURA) Pedro Baez R. Abstract. —Five specimens of the rarely collected crab Eupleurodon peruvianus were collected near Iquique, Chile (20°12’S, 70°10’W), during winter, 1977 and 1978. Two of them were found in the stomach contents of the labrid fish, Pi- melometopon maculatus; the other three specimens were collected from the red alga, Corallina officinalis chilensis, in the rocky intertidal zone. These records represent an extension of the known latitudinal range of this crab which now reaches from Salinas, Ecuador, to Iquique, Chile, and adds a genus and species to the 15 species of Majidae recorded from Chilean waters. Brachyuran decapods of the genus Eupleurodon Stimpson, 1871, are small crabs, which have been collected only from the intertidal zone of the Eastern Pacific. They have a carapace with depressed surface, its anterolateral angles being re- markably prominent, tooth-shaped, and projected forward subparallel to the body axis; the abdomen has only 5 free segments in both sexes and the ambulatory appendages are strongly prehensile, provided with crenate articles. The genus is closely related to Epialtus. There are three species of Eupleurodon: E. trifurcatus Stimpson, extending from Cape San Lucas, Baja California, Mexico (Garth 1958) to Chimbote, Pert (Chirichigno 1970); E. rathbunae Garth, 1958, collected only on Hood Island, Galapagos Archipelago and E. peruvianus (Rathbun) whose distribution ranges from Salinas, Ecuador to Iquique, Chile. The latter species was transferred from Epialtus (see Rathbun 1923, 1924) to Eupleurodon by Garth (1958) who stated that the differences used to distinguish the two forms as different species were sexual, with Eupleurodon peruvianus being the female and Epialtus peruvianus the male of the same species, Eupleurodon peruvianus. Material. —A male and 4 females of Eupleurodon peruvianus were examined (Table 1): the male (MNHN D-10858) and an ovigerous female (MNHN D-10859) were captured by Prof. Ratl Soto M., in the rocky intertidal zone of Iquique in samples of Corallina officinalis chilensis during winter, 1977. The other 3 females were collected by Lic. Héctor R. Fuentes in Jun—Jul 1978 in the vicinity of Huaiquique beach: an ovigerous female (MNHN D-10860) obtained from inter- tidal seaweed samples; and two others, found in stomach contents of a labrid fish, Pimelometopon maculatus Pérez, captured between 3 and 20 m depth (Fuentes 1978). Diagnosis. —(Modified from Garth 1958:243). Male: Hepatic width surpassing branchial width; rostrum elongate and bilobed; orbital arch making an obtuse angle with rostral border. First pleopod concave, with a thumb-shaped process, and opposed horny pointed tip. Female: branchial width surpassing hepatic width; rostrum short, its point 562 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Morphometric data of Eupleurodon peruvianus (Rathbun, 1923). Measurements in (mm). Sex Males Females MNHN MNHN MNHN Catalogue number # D-10858 # D-10859 D-10860 Carapace length 9.50 12.49 7.30 8.84 (rostrum fractured) Carapace length to rostral base — 8.69 — 7.59 7.90 Hepatic width 6.45 7.74 5.25 7.19 7.10 Branchial width 6.45 8.64 5.45 7.49 7.26 Rostral length 2.70 3.78 1.80 1.62 — Rostral width 1.50 1.82 1.00 1.43 1.63 Chela length 3.40 — 1.70 (r) 2.52 (r) 2.70 Chela height = — — 1.13 1.08 Chela dactylus length 1.60 — 1.00 1.25 1.50 Abdomen length (folded) — 4.08 — 5.90 6.30 Abdomen width — 2.56 — Voz 7.05 Length second leg 8.50 — — 8.41 8.23 third leg — 7.57 — 7.70 4.59 fourth leg — 8.56 — 6.07 6.20 fifth leg 4.50 — — 5.85 4.71 # = Taken from Garth (1958). MNHN-D = Museo Nacional de Historia Natural, Chile; coleccion de Crustaceos Decapodos. r = right. divided; prominent pre-orbital lobe. Merus of external maxilliped with border entire or slightly V-shaped in both sexes. Discussion. —The morphological characters of these specimens coincide in gen- eral terms with the description of Eupleurodon peruvianus (Rathbun). Neverthe- less, they differ from it, particularly the male, in which the sulcus that separates the hepatic and branchial regions is tiny, and the interorbital sulcus is not very acute. The left hepatic lobe is fractured, so it is not possible to see if this sector is curved. The branchial lobe is slightly smaller than the hepatic lobe and its sides are lightly concave. The rostrum is remarkably bifurcate, thick, and provided with 2 lobes at the tip; it is longer than broad, with parallel sides in the proximal 7/3. The fingers of the chelae fit well although they will not close completely. The upper side of the merus of the first and the second ambulatory legs is not laminated as it appears in Garth’s (1958) description; it is difficult to see the 2 or 3 teeth described. On the other hand, the female shows variations in the merus of the maxillipeds which when closed are slightly V-shaped at the anterior and internal angle; on the chelae the propodal finger is not strongly twisted, and the dactylus has a large basal tooth situated nearly at the midlength of its cutting border. The observed morphological variations are generally differences in growth stage rather than shape. The ovigerous females have about 700 eggs, those of MNHN D-10859 in an early stage of development, without ocular spots; MNHN D-10860 has eggs with ocular spots. The specimens of Eupleurodon peruvianus from samples of Corallina officinalis chilensis were collected together with specimens of another majid, Acanthonyx VOLUME 98, NUMBER 3 563 petiveri H. Milne Edwards; the two specimens found by Fuentes (1978) in the stomach contents of Pimelometopon maculatus indicate that E. peruvianus is only an occasional prey item in the diet of this third level carnivorous fish within the food chain of the upper sublittoral. The present finding of the rarely collected Eupleurodon peruvianus constitutes a new record of this genus and species for the carcinological fauna of Chile (Re- tamal 1981), and an extension of its known latitudinal distribution from Salinas, Ecuador, to Iquique, Chile. Sixteen species of Majidae are now known from Chile. Acknowledgments My thanks are due to Mr. Raul Soto M., Departamento Ciencias del Mar, Instituto Profesional de Iquique, and to Professor Héctor R. Fuentes, Departa- mento de Investigaciones Marinas, Universidad del Norte, Sede Coquimbo, Chile. Literature Cited Chirichigno, F. N. 1970. Lista de crustaceos del Peri (Decapoda y Stomatopoda) con datos de su distribucion geografica.—IMARPE, INFORME (35):1-95. Fuentes, H. R. 1978. Biologia del Pejeperro en el sublitoral de Iquique (Pimelometopon maculatus (Pérez 1885) (Pisces: Labridae)). Informe Técnico, Universidad del Norte, Iquique, 24 pp. [typed] Garth, J.S. 1958. Brachyura of the Pacific Coast of America, Oxyrhyncha.— Allan Hancock Pacific Expeditions 21(1):1—xii + 1-499; (2) tables and plates: 501-854, pls. A-Z, Z,-Z,, 1-55. Rathbun, M. J. 1923. New species of American spider crabs.— Proceedings of the Biological Society of Washington 36:71-74. . 1924. Newspecies and subspecies of spider crabs. — Proceedings of the United States National Museum 64:1-5. Retamal, M. A. 1981. Catalogo ilustrado de los crustaceos decapodos de Chile.—Gayana, Zoologia (44):1-110. Seccion Hidrobiologia, Museo Nacional de Historia Natural, Casilla 787, San- tiago, Chile. PROC. BIOL. SOC. WASH. 98(3), 1985, pp. 564-570 A NEW CRAYFISH OF THE GENUS ORCONECTES FROM THE LITTLE WABASH RIVER SYSTEM OF ILLINOIS (DECAPODA: CAMBARIDAE) Lawrence M. Page Abstract. —Orconectes stannardi, new species, is endemic to the Little Wabash River system of Illinois and appears to be most closely related to O. propinquus. Orconectes iowaensis Fitzpatrick is relegated to the synonymy of O. propinquus. During a decade-long (1972-1982) survey of the decapods of Illinois (Page, in press), an undescribed species of Orconectes was discovered. It appears to be restricted to the Little Wabash River system of southeastern Illinois and brings to two (with O. illinoiensis Brown, 1956) the number of crayfishes known to be endemic to Illinois. Earlier investigators (Rietz 1912; Brown 1955) apparently collected the new Orconectes, but referred it to O. propinquus (Girard, 1852) or to O. indianensis (Hay, 1896). Based on data accumulated during the recent survey of Illinois, neither O. propinquus nor O. indianensis occurs in the Little Wabash River system. Orconectes stannardi, new species Fig. 1 Diagnosis. —Body and eyes pigmented. Rostrum concave with margins mod- erately thickened, subparallel, slightly converging medially; terminating in spines; median carina extending onto long acumen. Areola 29.1—33.3 (mean = 30.2, n = 14) percent of total length of carapace, 3.6—6.7 (mean = 5.0) times as long as wide with 3 to 4 punctations across narrowest part. One large cervical spine on each side of carapace. Postorbital ridges well developed, grooved dorsolaterally and terminating in large spines. Suborbital angle weakly developed. Antennal scale broadest slightly distal to midlength, 2.3-3.0 (mean = 2.7) times as long as wide. Ischia of third pereiopods only of form I male with hooks overreaching basioischial articulation. Chela with 2 rows of tubercles along mesial margin of palm; small tufts of setae over dorsal surface; dorsal surfaces of fingers with well-defined longitudinal ridges. First pleopods of form I male symmetrical, extending to bases of second pereiopods when abdomen flexed. First pleopod of form I male with shoulder on cephalic surface at base of central projection; central projection cor- neous, strongly tapered to hooked tip; mesial process slightly shorter, non-cor- neous, tapered to rounded tip, with spur about midlength on caudal surface. Annulus ventralis immovable, subrhomboidal; cephalic half with medial trough and 2 caudally directed protuberances overhanging centrally located fossa; sinuate sinus extending from fossa to caudal edge. Holotypic male, form I: Body somewhat depressed, abdomen narrower than thorax (12.4 and 14.3 mm). Greatest width of carapace greater than depth at caudodorsal margin of cervical groove (14.0 and 11.1 mm). Areola 5 times longer VOLUME 98, NUMBER 3 565 (9.1 mm) than wide (1.8 mm) with 4 punctations across narrowest part; length of areola 31.6 percent of length of carapace. Rostrum excavated dorsally with submarginal and scattered punctations, median carina; thick margins terminating in corneous spines. Acumen terminating in upturned corneous spine reaching nearly to end of antennular peduncle. Postorbital ridge well developed, grooved dorsolaterally, terminating in prominent corneous spine. Suborbital angle poorly developed; branchiostegal spine small. Cervical spine large and corneous; hepatic area tuberculate; dorsal and branchiostegal areas of carapace punctate. Abdomen longer than carapace (31.4 and 28.8 mm). Cephalic section of telson with one movable and one immovable spine in each caudolateral corner. Basal podomere of uropod with spine extending over mesial ramus. Lateral ramus of uropod with median and submedian ridges. Lateral ramus with moderately large movable spine submarginally at caudolateral corner. Mesial ramus of uropod with prominent median ridge terminating in premarginal spine. Dorsal margin of telson and uropods lightly setiferous. Cephalic lobe of epistome spatulate with thickened cephalolateral margins; no cephalomedian projection. Antennal scale broadest subdistally; thickened part terminating in large corneous spine. Mesial margin of palm of right (left is regenerated) chela with primary row of 9 tubercles and secondary row of 8 smaller tubercles on dorsal surface lateral to primary row; distoventral surface of palm with 2 large tubercles at base of dacty]l; dorsal surface of palm covered with many small setal tufts. Propodus with lateral base of finger impressed dorsally, less so ventrally; dorsal and ventral surfaces with submedian ridges flanked by setiferous punctations; opposable margin with row of 6 tubercles, fourth from base largest, along proximal half of finger. Dorsal and ventral surfaces of dactyl with median longitudinal ridges flanked by setiferous punctations; opposable margin with row of 4 tubercles, first and fourth from base largest, on proximal half; mesial surface with 2 rows of tubercles on proximal half, one row on distal half. Fingers with distal tubercles small and inconspicuous, terminating in large corneous tips. Carpus with deep oblique furrow dorsally; mesial surface with one tubercle proximally and large procurved spine near midlength. Dorsodistal surface of left merus with 3 spines (right with 2); ventral surface with 1 large spine laterally and mesial row of 7 tubercles, some corneous, decreasing in size proximally; row terminating on large corneous spine. Ischium with | small corneous-tipped tu- bercle on ventromesial margin. Hook on ischium of third pereiopod only; hook simple, overreaching basiois- chial articulation and not opposed by tubercle on basis. First pleopods (see ““Di- agnosis” for description) barely reaching bases of second pair of pereiopods when abdomen flexed. Allotypic female: Differing from holotype in following respects: areola consti- tuting 30.9 percent of length of carapace (27.8 mm) and 4.3 times longer than broad. Left chela with mesial surface of palm bearing 8 tubercles in primary and 7 in secondary row (short tertiary row distally). Propodus with opposable margin bearing row of 6 tubercles, third from base largest. Dactyl with opposable margin bearing 8 tubercles, most proximal ones largest. Left merus with 2 spines dorsally; ventral surface with no spine laterally, mesial row of 8 tubercles. Sternum between third and fourth pereiopods narrowly V-shaped. Postannular PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 566 Orconectes stannardi: a, Dorsal view of carapace; b, Dorsal view of chela; c, Annulus Fig. 1. ventralis; d, Antennal scale; e, Mesial view of first pleopod of form I male; f, Lateral view of first pleopod of form I male; g, Lateral view of first pleopod of form II male. Drawings are composites based on type-specimens. VOLUME 98, NUMBER 3 567 Table 1.—Measurements (mm) of Orconectes stannardi. Holotype Allotype Morphotype Carapace: Entire length 28.8 27.8 32.4 Postorbital length 21.5 20.1 24.6 Width 14.0 13.1 15.1 Height 11.1 10.4 12.6 Areola: Width 1.8 2.0 2.1 Length 9.1 8.6 10.6 Rostrum: Width 3.9 3.4 4.4 Length 9.3 8.7 10.4 Chela, left (except Holotype): Length, palm mesial margin 7.8 5.5 5.8 Palm width 10.7 Ve7 8.2 Length, lateral margin 26.9 18.2 24.5 Dactyl] length 16.4 11.1 15.0 Abdomen: Width 12.4 13.5 12.8 Length 31.4 28.0 30.8 sclerite three-fourths as wide as annulus ventralis (described in Diagnosis). First pleopods of female uniramous, barely reaching annulus when abdomen flexed. Morphotypic male, form II: Differing from holotype in following respects: areola constituting 32.7 percent of length of carapace (32.4 mm) and 5.0 times longer than broad. Mesial surface of palm of left chela with 6 tubercles in primary and 5 in secondary rows; propodus bearing 12 tubercles, second and fourth from base largest, on opposable margin. Dactyl with opposable margin bearing 12 tubercles, second and third from base largest. Merus with one spine dorsally, no spine ventrally, and mesial row of 5 tubercles. Hook on ischium of third pereiopod much reduced, not overreaching basiois- chial articulation. First pleopod of uniform texture; both terminal elements straight, noncorneous. Spur of mesial process undeveloped, represented by small acute bulge. Size.—The largest specimen examined is a 32.5 mm-CL form II male. The largest female is 31.9 mm. Form I males (n = 21) range from 15.8 to 28.9 mm CL. Color. —Cephalothorax and abdomen olive-green to light brown and heavily speckled with dark brown dorsally, white ventrally. Side of carapace light green, subtended by darker green stripe. Caudal edge of carapace and first abdominal tergum crossed by dark brown band. Posterior edge of each abdominal tergum with thin red line. Chelae light brown with numerous dark brown specks; each finger with red tip and subdistal black band. Type-locality. —Little Wabash River at Secondary Road 719, 6km NNW Louis- ville (TSN, R6E, Sec. 33 SW), Clay County, Illinois. Immediately east of the river and secondary road junction is a large northward bend in the river coming within 10 m of the road; the type-specimens were collected from the east (downstream) 568 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON side of the bend in a large slab riffle. At normal level the river bed at the riffle is about 5 m wide. Disposition of types. —The holotype, allotype, and 10 paratypes (4 form I males and 6 females) collected with the holotype and allotype on 6 Nov 1983 are deposited at the Illinois Natural History Survey; 11 paratypes (USNM 209119; 5 form I males and 6 females) collected at the same time are deposited at the National Museum of Natural History, Smithsonian Institution. The morphotype and 18 paratypes (1 form I male, 12 form II males, and 5 females) collected at the type-locality with the morphotype on 17 May 1984 are deposited at the Illinois Natural History Survey; 10 paratypes (USNM 2091 20; all form II males) collected at the same time are deposited at the National Museum of Natural History, Smithsonian Institution. Etymology. —Named in honor of Dr. Lewis J. Stannard, Entomologist Emeritus of the Illinois Natural History Survey, in recognition of his outstanding contri- butions to the study of Thysanoptera and to the conservation of natural habitats. Range and specimens examined. —Orconectes stannardi has been found only in the Little Wabash River system of southeastern Illinois. Recent (since 1973) localities in addition to the type-locality (all Littlhe Wabash River proper) are 2 mi N Louisville, Clay Co.; 1 mi E Clay City, Clay Co.; and 2 mi S Carmi, White Co. Rietz (1912) and Brown (1955) recorded localities for O. propinquus and O. indianensis in the Little Wabash River system that almost certainly were based on misidentified O. stannardi. A recent survey of the crayfishes of Illinois (Page, in press) suggests that O. propinquus, O. indianensis and O. stannardi occur allopatrically, and that neither O. propinquus nor O. indianensis occurs in the Little Wabash River system. Unfortunately, the collections made by Rietz and Brown cannot be located and re-examined. Although all collections of O. stannardi that have been made since 1973 are from the Little Wabash River proper, some of the collections of Rietz (1912) and Brown (1955) were from tributaries. Collections of Rietz (1912) were made in Big Muddy Creek between Richland and Clay counties, and in Skillet Fork at Wayne City, Wayne Co.; the former was identified by Rietz as O. propinquus and the latter as O. indianensis. Collections cited by Brown (all as O. propinquus) but which must have been of O. stannardi were made in Dismal Creek, 244 mi N Iola, Clay Co.; Salt Creek, 1 mi SE Effingham, Effingham Co.; Blue Point Creek, 2% mi S Shumway, Effingham Co.; Lost Fork, 12 mi E Omega, Marion Co.; branch of Little Wabash River, 1'4 mi SW Trowbridge, Shelby Co.; and Little Wabash River, 4 mi NE Shumway, Effingham Co. Habitat. — Most individuals were found in shallow riffles composed of large flat stones. A few were found in deeper water but always in association with stones or accumulations of sticks and other debris. The preference of O. stannardi for rocky riffles is typical of the propinquus group of Orconectes, the members of which typically hide under stones and among debris. Life-history notes. — All five collections of O. stannardi presently available (from Nov 1973, Nov 1983, and May 1984) contain form I males, and one (May 1984) contains two ovigerous females. All 20 males collected in November are form I; only one of 25 males collected in May is form I. The two ovigerous females VOLUME 98, NUMBER 3 569 collected in May 1984 were 20.7 mm and 28.9 mm CL, and carried 124 and 184 eggs, respectively. Eggs average 1.9 mm in diameter. Relationships. — Orconectes stannardi appears to be most closely related to O. propinquus from which it differs in having, on the first pleopod of the form I male, a caudal spur on the mesial process and a shoulder on the cephalic surface at the base of the central projection; a longer rostrum with margins less converging anteriorly and a less prominent carina; and a narrower areola. Orconectes stan- nardi is endemic, and possibly autochthonous, to the Little Wabash River system. The region drained by the Little Wabash was glaciated during the Illinoian (ca. 100,000 y.b.p.) but not the Wisconsinan (ca. 10,000 y.b.p.) advances, suggesting that the species originated sometime within the past 100,000 years. Fitzpatrick (1967) postulated that some speciation within the O. propinquus group was as- sociated with the Wisconsin glaciation. Recognized members of the O. propinquus group, as defined by Fitzpatrick (1967, 1968) and Fitzpatrick and Pickett (1980) are O. propinquus, O. erichson- ianus (Faxon, 1898), O. illinoiensis, O. jeffersoni Rhoades, 1944, O. sanborni (Faxon, 1884) (including O. s. erismorphorous Hobbs and Fitzpatrick, 1962, O. obscurus Hagen, 1870, O. virginiensis Hobbs, 1951, O. iowaensis Fitzpatrick, 1968, and O. kinderhookensis Fitzpatrick and Pickett, 1980. O. stannardi can be added to this group, but O. iowaensis should be removed. Fitzpatrick (1967, 1968) described O. iowaensis as endemic to Iowa and a sister species to O. propinquus. Features distinguishing O. iowaensis from O. propinquus were stated to be the truncate or spatulate mesial process of the first pleopod of the form I male and the more prominently sculptured annulus ventralis. The mesial process and annulus ventralis of O. propinquus were described, respec- tively, as tapering to an acute tip and being wider than long. Orconectes iowaensis was recognized by Hobbs (1972, 1974) and by Phillips and Reis (1979), who extended its known range into southern Minnesota. In his report on the crayfishes of Iowa, Phillips (1980) again recognized O. iowaensis but commented on the difficulty of separating O. iowaensis from O. propinquus. In studying Illinois crayfishes, I have been obliged to search, in the northwestern part of the state, for populations assignable to O. iowaensis and to look for evidence of intergradation between O. iowaensis and O. propinquus. No population of O. lowaensis was found in Illinois and comparisons of Illinois samples of O. pro- pinquus and O. iowaensis from the Maquoketa and Volga rivers, Iowa (including paratypes USNM 117970 and 117971), reveal that O. iowaensis does not differ from Illinois populations of O. propinquus; i.e., the mesial process of the first pleopod of form I male is not more truncate or spatulate, and the annulus ventralis is not narrower and more prominently sculptured. Inasmuch as Fitzpatrick (1967) noted the absence of ““morphologically distinct geographic races”’ within O. pro- pinquus, Illinois samples can be considered typical of the species. O. 1owaensis 1s relegated to the synonymy of O. propinquus. Acknowledgments I am indebted to Kevin S. Cummings and Charles S. Swanson for assistance in collecting specimens, and to Horton H. Hobbs, Jr., John D. Unzicker, and Donald W. Webb for helpful comments on the manuscript. 570 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Literature Cited Brown, P. L. 1955. The biology of the crayfishes of central and southeastern Illinois. Doctoral Dissertation, University of Illinois. 158 pp. . 1956. Anewcrayfish of the genus Orconectes from Illinois (Decapoda, Astacidae).— American Midland Naturalist 56:163-167. Faxon, W. 1884. Descriptions of new species of Cambarus, to which is added a synonymical list of the known species of Cambarus and Astacus. — Proceedings of the American Academy of Arts and Science 20:107-158. 1898. Observations on the Astacidae in the United States National Museum and in the Museum of Comparative Zoology, with descriptions of new species. — Proceedings of the U.S. National Museum 20(1 136):643-694. Fitzpatrick, J. F., Jr. 1967. The Propinquus group of the crawfish genus Orconectes (Decapoda: Astacidae).— Ohio Journal of Science 67:129-172. 1968. A new crawfish of the genus Orconectes from Iowa.—American Midland Naturalist 79:507-512. , and J. F. Pickett, Sr. 1980. A new crawfish of the genus Orconectes from eastern New York (Decapoda: Cambaridae).— Proceedings of the Biological Society of Washington 93:373-382. 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 American Astacidae.—Illustrated Catalogue of the Museum of Comparative Zoology at Harvard College 3:1—109. Hay, W. P. 1896. The crawfishes of the state of Indiana. Pages 475-507 in 20th Annual Report of the Department of Geology and Natural Resources of Indiana. Hobbs, H. H., Jr. 1951. Anewcrayfish of the genus Orconectes from southeastern Virginia (Decapoda, Astacidae).— Virginia Journal of Science 2:122-128. . 1972. Crayfishes (Astacidae) of North and Middle America.— U.S. Environmental Protection Agency Biota of Freshwater Ecosystems Identification Manual 9:1-173. 1974. A checklist of the North and Middle American crayfishes (Decapoda: Astacidae and Cambaridae).—Smithsonian Contributions to Zoology 166:1—-161. , and J. F. Fitzpatrick, Jr. 1962. A newcrayfish of the Propinquus group of the genus Orconectes from the Ohio drainage system in West Virginia (Decapoda, Astacidae).— Proceedings of the Biological Society of Washington 75:207—214. Page, L. M. [In press.] The crayfishes and shrimps (Decapoda) of Illinois. —Illinois Natural History Survey Bulletin. Phillips, G.S. 1980. The decapod crustaceans of Ilowa.— Proceedings of the lowa Academy of Science 87:81-95. , and L. A. Reis. 1979. Distribution and ecology of Orconectes iowaensis Fitzpatrick and Orconectes rusticus (Girard) in Minnesota. — Journal of the Minnesota Academy of Science 45: 18-19. Rhoades, R. 1944. The crayfishes of Kentucky, with notes on variation, distribution, and descriptions of new species and subspecies. — American Midland Naturalist 31:111—149. Rietz, N. M. 1912. Ecological relations of the crawfishes of Illinois.—B.S. Thesis, University of Illinois, 86 pp. Illinois Natural History Survey, 607 E. Peabody Drive, Champaign, Illinois 61820. PROC. BIOL. SOC. WASH. 98(3), 1985, pp. 571-573 NEW RECORDS OF CARIDEAN SHRIMPS IN THE GULF OF CALIFORNIA, MEXICO Mary K. Wicksten and Michel E. Hendrickx Abstract. —New records are provided for six species of caridean shrimps in the Gulf of California. Alpheus splendidus is reported for the first time from the eastern Pacific. Veleronia laevifrons and Periclimenes soror are noted for the first time from the Gulf of California. The pandalids Pantomus affinis and Heterocarpus vicarius are reported for the first time from off Sinaloa. Alpheus websteri has been taken off San Carlos, Sonora. Recent collecting in the Gulf of California has resulted in the capture of caridean shrimps not reported there previously, or found outside of their recorded ranges. During a research cruise of the B/O E/ Puma of the Instituto de Ciencias del Mar y Limnologia, Universidad Nacional Autonoma de México, two series of speci- mens of Pandalidae were found off the coast of Sinaloa, on the eastern side of the Gulf of California. Alex Kerstitch, University of Arizona, collected shrimps by SCUBA diving along the coast of Sonora, to the north of Sinaloa. The specimens have been deposited at the Estacion Mazatlan and the Allan Hancock Foundation, University of Southern California. Family Palaemonidae Periclimenes soror Nobili Periclimenes soror Nobili, 1904:232.—Bruce, 1976:299-306, figs. 1—4.—Castro 1982:11. (See Bruce 1976 for an extensive synonymy.) Previous records. — Widespread in the Indo-West Pacific region: Red Sea, Indian Ocean, Australia, Malaysia, tropical Pacific east to Hawaii, Taboga Island, Panama (Bruce 1976); Gorgona Island, Colombia (Castro 1982). Material examined. —SONORA: Bahia San Carlos, 20 m, on sea star Mithrodia bradleyi Verrill, 28 Dec 1982, A. Kerstitch, 2 specimens.— Punta Doble, 20 m, under sea stars M. bradleyi, Oreaster occidentalis Verrill, Nidorellia armata (Gray) and Astrometis sertulifera Xantus; 25 Jun 1983, A. Kerstitch, 24 specimens. Remarks. — Periclimenes soror previously has been found in association with 16 species of sea stars, including O. occidentalis, M. bradleyi, and N. armata (Bruce 1976; Castro 1982). This is the first report of Astrometis sertulifera as a host of P. soror. In his field notes, Alex Kerstitch reported that “‘about 75% of the larger sea stars may have 2—4 shrimps apiece” at Punta Doble. Veleronia laevifrons Holthuis Veleronia laevifrons Holthuis, 1951:199-—201, pl. 63, figs. fm. Previous records. -ECUADOR: off Cabo San Francisco, 4 m.—off La Playa, Bahia Santa Elena, 4-13 m. GALAPAGOS: Gardner Bay, Isla Espanola (Hood Island), 7 m (Holthuis 1951). 572 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Material examined. —SONORA: Bahia San Carlos, 20 m, among Parazoanthus sp., 28 Dec 1982, A. Kerstitch, 1 specimen.—Isla Peruano, 10 m, on gorgonians Eugorgia aurantica Verrill and Muricea californica Aurivillius, 28 Jun 1983, A. Kerstitch, 27 specimens. —Isla Candolero, near Guaymas, 20 m, on Parazoanthus sp., 2 Jan 1984, A. Kerstitch, 17 specimens. Remarks. —Photographs of living animals by A. Kerstitch show the shrimp resting along the main axis of a branch of E. aurantica, which they match well in color. Family Alpheidae Alpheus websteri Kingsley Alpheus websteri Kingsley, 1880:416.— Wicksten 1983:42—43. (See Wicksten 1983 for a more extensive synonymy.) Previous records. —Eastern Atlantic, Caribbean, southern Gulf of California to Galapagos (Wicksten 1983). Material examined. -SONORA: Morro Colorado, 10 m, among rocks and Sargassum sp., 1 Jan 1984, A. Kerstitch, 1 specimen. Alpheus splendidus Coutiére Alpheus splendidus Coutiére, 1897:235.— Banner and Banner 1982:56-—60, fig. 12. Previous records. —Red Sea, Seychelles, Indonesia, Malaysia, Thailand, Phil- ippines, and Hong Kong (Banner and Banner 1982). Material examined. -SONORA: San Carlos, 5 m, under rock, 8 Jul 1983, A. Kerstitch, 1 specimen. Remarks. — Alex Kerstitch took a color photograph of the living shrimp. The color pattern matches the description provided in French by Coutiére (1897)—a slender bright yellow band from rostrum to the telson, bordered with two brown bands, the rest of the body reddish except for two straight white lines contiguous to the brown bands; pincers light orange. Family Pandalidae Pantomus affinis Chace Pantomus affinis Chace, 1937:116—-118, fig. 13.—Wicksten and Méndez 1983:82— 85, pl. E, fig. 31.—Wicksten 1983:21. Previous records. —Bahia Santa Inez, Baja California, western side of Gulf of California, Mexico.—Golfo de Dulce, Costa Rica.—southwest of Islas Lobos de Tierra, Peru (Wicksten and Méndez 1983). Material examined. — Off coast of Sinaloa, 25°45'N, 109°34’W, trawled, 91 m, silty bottom, 12 May 1982, B/O E/ Puma, 56 specimens including 2 ovigerous females. Heterocarpus vicarius Faxon Heterocarpus vicarius Faxon, 1893:203; 1895:148-151, pl. 40, figs. 1-1b. —Chace 1937:118.—Méndez 1981:100-101, pl. 41, figs. 296-299. VOLUME 98, NUMBER 3 573 Previous records. —Isla de San Ildefonso and Punta Gorda, Baja California, Bay of Panama, Pacific coast of Colombia, Peru (Méndez 1981). Material examined. — Off coast of Sinaloa, 24°57'N, 108°41'W, otter trawl, 75 m, sand, 3 May 1982, B/O E/ Puma, 58 specimens. Literature Cited Banner, D. M., and A. H. Banner. 1982. The alpheid shrimp of Australia. Part III. The remaining alpheids, principally the genus Alpheus, and the family Ogyrididae.— Records of the Australian Museum 34(1):1-357. Bruce, A. J. 1976. Periclimenes soror Nobili, a pontoniin shrimp new to the American fauna, with observations on its Indo-West Pacific distribution.— Tethys 8(4):299-306. Castro, P. 1982. Notes on symbiotic decapod crustaceans from Gorgona Island, Colombia, with a revision of the eastern Pacific species of Trapezia (Brachyura, Xanthidae), symbionts of scler- actinian corals.—Anales del Instituto de Invertebrados Marinos, Punta de Betin—Santa Marta, Colombia 12:9-17. Chace, F. A., Jr. 1937. The Templeton Crocker Expedition. VII. Caridean decapod Crustacea from the Gulf of California and the west coast of Lower California.— Zoologica 22(2):109-138. Coutiére, H. 1897. Note sur quelques alphéidés nouveaux ou peu connus rapportés de Djibouti (Afrique Orientale).— Bulletin Muséum National d’Histoire Naturelle, Paris 3(6):233—-236. Faxon, W. 1893. Reports on the dredging operations off the west coast of central America to the Galapagos, to the west coast of Mexico, and in the Gulf of California, in charge of Alexander Agassiz, carried on by the U.S. Fish Commission steamer “Albatross” during 1891, Lieut. Commander Z. L. Tanner, U.S.N., commanding. VI. Preliminary descriptions of new species of Crustacea.— Bulletin of the Museum of Comparative Zoology, Harvard University 24(7): 149-220. —. 1895. 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 “Albatross” during 1891, Lieut. Commander Z. L. Tanner, U.S.N., commanding. 15. The stalk-eyed Crustacea.— Museum of Comparative Zoology Memoir, Harvard University 18: 1-292. Holthuis, L. B. 1951. A general revision of the Palaemonidae (Crustacea Decapoda Natantia) of the Americas. I. The subfamilies Euryrhynchinae and Pontoniinae. — Occasional Papers of the Allan Hancock Foundation 11:1—332. Kingsley, J. S. 1880. Ona collection of Crustacea from Virginia, North Carolina and Florida with a revision of the genera of Crangonidae and Palaemonidae.— Proceedings of the Academy of Natural Sciences of Philadelphia 31(3):383-—427. Méndez, M. 1981. Claves de identificacion y distribucion de los langostinos y camarones (Crustacea: Decapoda) del mar y rios de la costa del Peru.— Boletin del Instituto del Mar del Pera—Callao 5:1-170. Nobili, G. 1904. Diagnoses préliminaires de vingt-huit espéces nouvelles de Stomatopodes et Déca- podes Macroures de la Mer Rouge. — Bulletin Muséum National d’Histoire Naturelle, Paris 10: 228-238. Wicksten, M. K. 1983. A monograph on the shallow water caridean shrimps of the Gulf of California, Mexico.— Allan Hancock Monographs in Marine Biology 13:1-59. , and M. Méndez. 1982. Nuevos registros de camarones carideos en el Pert.— Boletin de Lima 25:75-89. (MKW) Department of Biology, Texas A&M University, College Station, Texas 77843; (MEH) Estacion Mazatlan, P.O. Box 811, Mazatlan, Sinaloa, Mexico. PROC. BIOL. SOC. WASH. 98(3), 1985, pp. 574-590 CALANOID COPEPODS (DIAPTOMIDAE) FROM COASTAL LAKES, STATE OF RIO DE JANEIRO, BRAZIL Janet W. Reid Abstract. —Three species of diaptomid copepods (Calanoida) were collected from the plankton of four coastal lakes, State of Rio de Janeiro, Brazil. Noto- diaptomus iheringi (Wright) is redescribed based on topotype specimens in the collection of the National Museum of Natural History and on specimens from two of the Rio lakes. The female and male of ““Diaptomus”’ azureus, new species, and the male of ““D.”’ fluminensis, new species, are described. In the course of a survey of physical, chemical, and faunistic characteristics of lakes and lagoons along the coast of the State of Rio de Janeiro, Brazil (Esteves et al. 1983), samples of zooplankton from 14 lakes were taken by F. A. Esteves in January and September, 1983, and confided to me for determination of species present. Adult diaptomid copepods occurred in samples from 4 lakes. Notodiap- tomus iheringi (Wright) occurred in Lagoa do Campelo and Lagoa da Saudade; ‘“‘Diaptomus”’ azureus, new species, in Lagoa Comprida; and “D.” fluminensis, new species, in Lagoa Iodada (Coca-Cola). Wright (1935) supplied only a partial description of N. iheringi. The incompleteness of extant descriptions of many species of South American diaptomids is seriously impeding the understanding of interspecific and intergeneric relationships (Brehm 1958; Brandorff 1976). A redescription of this species from specimens collected by Wright from the type- locality and deposited in the collections of the National Museum of Natural History, as well as from the Rio specimens is presented. The female and male of ““D.” azureus and the male of “D.” fluminensis are described also. Some physical and chemical characteristics of the Rio lakes appear in Table | (data supplied by F. A. Esteves). Notodiaptomus iheringi (Wright) Figs. 1-28 Diaptomus theringi Wright, 1935:223-226, pl. 1, fig. 4, pl. 2, figs. 3, 5-11.— Wright, 1936:80-81.— Wright, 1938:562. Notodiaptomus theringi. — Kiefer, 1936:197-198, figs. 3—4.—Cipolli and Carval- ho, 1973:97, 98, 101, tab. 2.—Brandorff, 1976:616-617, fig. 2.—Sendacz and Kubo, 1982:69-71, 85-86, figs. 25-29. Material examined. —1 2 (National Museum of Natural History —USNM 79542) topotype (after Wright’s label, ““from the type-locality, Agcude Puxinama, near Campino Grande, State of Parahyba, Brazil’’), S. Wright coll.; 3 2 and 3 6(USNM 216065), 3 2 (Museu de Zoologia da Universidade de Sao Paulo—MZUSP 6191), 4 6 (MZUSP 6192), 4 copepodites (MZUSP 6193), all alcohol-preserved, and 1 2, dissected (author’s collection), from Lagoa da Saudade, State of Rio de Janeiro, VOLUME 98, NUMBER 3 575 Table 1.—Physical and chemical characteristics of four lakes inhabited by diaptomid copepods, State of Rio de Janeiro, Brazil (data supplied by F. A. Esteves). Lake L. do Campelo L. da Saudade L. Comprida L. Todada (Coca-Cola) SarmpiinetdatesMeminO GS amamnKOS/ 6S aan NIN OL/S3) MENT OU/CSUMMIURSR MONO SANRINN IRE GMMNNO9/8300 Del Area (km?) 9.85 0.79 0.11 0.12 Max. depth (m) 1.5 Dn) 3.0 1.5 Water color Clear Black Black Black Secchi depth (m) 1.50 — 2.0 0.75 0.40 0.50 pH 7.45 9.0 6.6 4.4 5.3 5.6 % sat. O, 86 110 93 79 70 79 Conductivity (wS cm!) 440 645 410 530 4700 4520 Chlorophyll (ug 1) 10.42 4.10 1.20 0.67 — 8.47 Ca?* (mg 1") 11.07 — 4.68 6.30 19.43 Na* (mg 1"') 53.8 — 57.0 85.9 561.9 — Cl (mg 1"') 78.0 — 94.4 123.3 997.5 — CO,2- (mg I!) 0.012 0.101 7.0 x 10-4 — 1.2 x 10-5 6.1 x 10-4 NO; (ug 17!) 10.7 6.5 7.6 7.3 10.2 25.0 PO, (ug 1!) 0.8 5.9 0.6 0.5 1.25 — Brazil, 21°42'S, 41°20'W; 10 2 and 10 6 (USNM 216066), 10 2 (MZUSP 6194), 10 6(MZUSP 6195) and 56 copepodites (MZUSP 6196), 22 2 and 14 é (collection of F. A. Esteves), all alcohol-preserved, and 1 2 and 1 4, dissected (author’s collection), from Lagoa do Campelo, State of Rio de Janeiro, 21°40'S, 41°11’W, all collected 28 Jan 1983. Description. — Female: Length (including caudal rami) of Agude Puxinama spec- imen 1.11 mm; mean length of 10 Lagoa da Saudade specimens 1.34 mm (range 1.24—1.40 mm); mean length of 10 Lagoa do Campelo specimens |.23 mm (range 1.16-1.30 mm). Body widest at Ist pediger in dorsal view. Suture between 4th and 5th pedigers complete but faint dorsally. 3rd and 4th pedigers in Rio de Janeiro specimens each with | to 3 rows of spinules near posterior border, single row on 4th pediger continuing laterally (Figs. 1—4). 5th pediger produced on each side in small wing, ending in single spine; left wing directed laterally (Fig. 3), right wing dorsally (Fig. 4); inner corner of left wing slightly expanded (Fig. 2). Urosome of 3 segments, most of 2nd segment covered by genital segment and hardly visible. Genital segment about 1.5 x as long as rest of urosome (including caudal rami), slightly expanded anterolaterally, right expansion slightly larger than and anterior to left expansion, each expansion with laterally directed spine. Right posterolateral margin of genital segment slightly produced. Genital segment sad- dle-shaped in lateral view. Genital opening with conical process on either side (Fig. 5). Inner margins of caudal rami haired (Fig. 2). Rostral points acute (Fig. 6). 1st antenna reaching well past caudal rami; seg- ments 11 and 13-21 each with | seta; complete armature as in Fig. 7. 2nd antenna with normal setation; bases of terminal setae on exopod distinctly separated from segment 7 (Fig. 8). Mandible also with normal setation (Fig. 9); dentition of gnathal lobe in ventral to dorsal order as follows (nomenclature after 576 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fleminger 1967): apical and subapical teeth pointed. Medial teeth 4 in number, ventralmost a rounded lobe with tiny acute cusp, dorsal 3 medial teeth bicuspidate. Basal teeth 3 in number, ventralmost tooth tricuspidate, 2 dorsalmost teeth slen- der, bicuspidate (Figs. 13, 14). Maxillula, maxilla and maxilliped as in Figs. 10-12 respectively; distal lobe at lst basipod of maxilliped with only 3 setae. Swimming legs with normal complement of spines and setae; setation of leg 4 identical to leg 3 (Figs. 15-17). Leg 2 with Schmeil’s organ on posterior surface of 2nd endopod segment (Figs. 16, 18). Leg 5 (Figs. 19-22) stout, 1st basal segment with prominent posteroventrally directed process tipped with stout spine. Posterodistal angle of 2nd basal segment produced in rounded process; lateral seta reaches to distal third of Ist exopod segment. Ist exopod segment unarmed, about 2 x longer than broad. 2nd exopod segment with short spine lateral to base of 3rd exopod segment, about as long as 3rd segment (Fig. 21); claw stout, curved slightly inwards, middle part of each margin spinulose. 3rd exopod segment distinct from 2nd exopod segment, slightly longer than wide, bearing 2 terminal spines, inner spine about 2.3 x longer than outer. Endopod of | segment, slightly constricted near base, reaching past mid- length of lst exopod segment, bearing 2 long spines on either side of group of hairs on oblique inner distal margin (Fig. 22). Male: Mean length (including caudal rami) of 7 Lagoa da Saudade specimens 1.16 mm (range 1.14—1.18 mm); mean length of 10 Lagoa do Campelo specimens 1.06 mm (range 1.02—1.12 mm). Body widest at 2nd pediger in dorsal view. Suture between 4th and 5th pedigers complete but faint; posterior border of 4th pediger with row of spinules on each side. Each side of 5th pediger produced into pos- teriorly directed wings, each tipped with spine; left spine directed posteriorly, right spine directed obliquely outward. Urosome of 5 segments, curved slightly to right in most preserved specimens. Genital segment asymmetrical, right side expanded posteriorly over next urosomal segment. Inner margins of caudal rami haired (Fig. 23). Right lst antenna with socketed spines on segments 8 and 12; spines on segments 10, 11, 13, 15, and 16 without basal articulation (Fig. 24). Spines on segments 10 and 11 of equal length, parallel to axis of antenna; spine on segment 10 overlapping base of spine on segment 11. Spine on segment 13 large, with notched tip. Spine on segment 16 very small. Antepenultimate segment with narrow hya- line membrane (Fig. 25). Armature of left 1st antenna identical to that of female. Structure and armature of 2nd antenna, maxillula, maxilla and maxilliped as in female. Dentition of gnathal lobe of mandible similar to female except dorsal 3 medial teeth and middle basal tooth each with additional tiny cusp on ventral surface (Fig. 26). Structure and armature of swimming legs as in female. 5th legs agree in essential details with the descriptions of Wright (1935) and of Sendacz and Kubo (1982). Exopod of left leg 5 with widely separated proximal and distal hairy pads covered posteriorly by hyaline membrane extending to tip of wide, serrated distal process. Proximal pad in Lagoa do Campelo and Lagoa da Saudade specimens undivided. Spinous proximal process extending slightly beyond distal process (Figs. 27, 28). Remarks. —Wright reported the mean length of 20 females to be 1.18 mm, with a range of 1.11—1.27 mm, and the mean length of 20 males to be 1.03 mm, with VOLUME 98, NUMBER 3 577 Figs. 1-7. Notodiaptomus theringi, female from Lagoa do Campelo: 1, Habitus, lateral; 2, Urosome, dorsal; 3, Urosome, left lateral; 4, Urosome, right lateral; 5, Genital area, ventral; 6, Rostrum, ventral; 7, 1st antenna. Scale a, Fig. 1; Scale b, Figs. 2—4, 7; Scale c, Figs. 5, 6; each scale = 100 um. VOLUME 98, NUMBER 3 579 a range of 0.97-1.11 mm; thus the Rio specimens are within these ranges or slightly larger. Female specimens from Rio lakes differ from Wright’s in several details: distal border of 4th pediger not dorsally elevated in some specimens; distal border of 3rd as well as 4th pediger with fine spinules; Ist antennae slightly longer (in Wright’s specimens these reached to the end of the caudal rami); spinous process on basipod segment | of leg 5 stouter and exopod segment 1 somewhat broader; and urosome segment 2 never large, a feature which occurred “rarely” in Wright’s specimens. The spination of the right 1st antenna of the male agrees with Wright’s verbal description. A few of Wright’s specimens bore a short curved spur on the ante- penultimate segment of the right 1st antenna; no Rio males bore such a spur. Sendacz and Kubo (1982) showed a divided proximal hairy pad of exopod of left 5th leg in specimens from Sao Paulo. Neither Wright (1935) nor Sendacz and Kubo (1982) noted the serration of the distal process of the exopod of left leg 5. The dentition of the gnathal lobe in the dissected female from the type-locality (Fig.-14) differs in several respects from that of the Rio specimens, having the subapical tooth very broad and blunt, the cusps of the medial and basal teeth more rounded, and the notch between the apical and subapical teeth somewhat deeper. Ecology. —Notodiaptomus theringi was recorded by Wright (1935) from res- ervoirs, lakes and ponds in the States of Paraiba and Pernambuco; the type- locality is an acude (reservoir). Cipdlli and Carvalho (1973) encountered the species in lakes near the Guama and Capim Rivers and in the main stream and small side branches of the Tocantins River, State of Para. Sendacz and Kubo (1982) recorded it from two reservoirs in the drainage basin of the Paraiba do Sul River, State of Sao Paulo. These reservoirs were warm (median temperature 23.3°C), with high concentrations of nutrients and chlorophyll (median 40.1 mg: 1~'), and water of high transparency. In Para, this species occurred only in white and clear waters, but never in black (high in humic acids). Lagoas do Campelo and da Saudade resemble each other in being relatively transparent, of moderate conductivity, with pH close to neutral. Notodiaptomus iheringi appears to have a much broader ecological range than postulated by Wright (1935), who found it only in the arid interior of the Brazilian northeast, and in none of the coastal lakes in that area. ““Diaptomus”’ azureus, new species Figs. 29-59 Material examined. —1 2, holotype (MZUSP 6197) and 1 6 paratype (MZUSP 6198), dissected on slides; 4 2 (MZUSP 6199), 10 6 (MZUSP 6200) and 140 copepodites (MZUSP 6201), alcohol-preserved; 4 2 and 10 6 (USNM 216067), _— Figs. 8-18. Notodiaptomus iheringi: 8-13, 15-18, female from Lagoa do Campelo: 8, 2nd antenna; 9, Mandible; 10, Maxillula; 11, Maxilla; 12, Maxilliped; 13, Mandible, gnathal lobe; 14, Mandible, gnathal lobe, female from Acude Puxinama; 15, Leg 1, anterior; 16, Leg 2, posterior; 17, Leg 4, posterior; 18, Leg 2 endopod, lateral. Scale a, Figs. 8-12, 15-18; Scale b, Figs. 13, 14; each scale = 100 um. 580 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Figs. 19-28. Notodiaptomus iheringi: 19-22, female from Lagoa do Campelo: 19, Right leg 5, anterior; 20, Right leg 5, posterior; 21, Leg 5, exopod segment 3; 22, Leg 5, endopod. 23-28, Male from Lagoa do Campelo: 23, Urosome, dorsal; 24, Right lst antenna, segments 7-17; 25, Right Ist antenna, segments 23-25; 26, Mandible, gnathal lobe; 27, Legs 5, posterior; 28, Left leg 5 exopod, posterior. Scale a, Fig. 23; Scale b, Figs. 19, 20, 24, 25, 27; Scale c, Figs. 21, 22, 26, 28; each scale = 100 pm. VOLUME 98, NUMBER 3 581 Figs. 29-36. ““Diaptomus” azureus, female: 29, Habitus, dorsal; 30, Urosome, dorsal; 31, Urosome, left lateral; 32, Urosome, right lateral; 33, Urosome, ventral; 34, Rostrum, lateral; 35, 1st antenna; 36, 2nd antenna. Scale a, Fig. 29; Scale b, Figs. 30—33, 35; Scale c, Figs. 34, 36; each scale = 100 um. 582 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Figs. 37-48. ‘“Diaptomus” azureus, female: 37, Mandible; 38, Mandible, gnathal lobe; 39, Max- illula; 40, Maxilla; 41, Maxilliped; 42, Leg 1, anterior; 43, Leg 2, posterior; 44, Leg 3, posterior; 45, Leg 2 endopod, lateral; 46, Left leg 5, posterior; 47, Right leg 5, lateral-oblique; 48, Leg 5, endopod. Scale a, Figs. 37, 39-47; Scale b, Figs. 38, 48; each scale = 100 um. VOLUME 98, NUMBER 3 583 alcohol-preserved; 3 2 and 11 4, collection of F. A. Esteves, alcohol-preserved; 10 6, alcohol-preserved, and 1 2 and 1 4, dissected on slides, author’s collection; all paratypes; all from Lagoa Comprida, State of Rio de Janeiro, 21°17’S, 41°39’W, collected 26 Jan 1983. Description. —Female: Mean length (including caudal rami) of 10 specimens 1.06 mm (range 1.0—1.15 mm). Body widest at Ist pediger in dorsal view. Suture between cephalic segment and Ist pediger indistinct dorsally; 4th and Sth pedigers fused dorsally, separated laterally by indistinct suture. 5th pediger produced pos- terolaterally into asymmetrical wings: left wing small, rounded, ending in small dorsally directed spine and bearing second smaller spine on posterolateral margin; right wing larger, directed latero-obliquely, ending in stout spine and bearing smaller spine on posterodorsal margin (Figs. 29-32). Urosome 3-segmented, 2nd segment partly telescoped into and covered dorsally by genital segment, appearing fused with 3rd segment ventrally. Genital segment about 1.2 x as long as rest of urosome, asymmetrically expanded anteriorly with left side slightly more expanded than right, each expansion bearing a laterally directed spine slightly posterior to broadest part of segment. Posterodorsal margin of genital segment produced. This segment slightly saddle-shaped dorsally, ventral surface rugose posterior to genital opening; area of genital opening produced ventrally, with rounded process on each side of opening (Figs. 30-33). Inner margins of caudal rami haired. Rostral points acute (Fig. 34). Ist antenna reaching well past caudal rami; segments 11 and 13-21 each with 1 seta; complete armature as in Fig. 35. 2nd antenna with normal setation, terminal setae on exopod distinctly separated from segment 7 (Fig. 36). Mandible as in Fig. 37; dentition of gnathal lobe as follows: apical and subapical teeth pointed; medial teeth 3 in number, each with 2 principal cusps and 1 small ventral cusp; basal teeth 3 in number, ventralmost 2 each with 2 principal cusps and 1 ventral and | dorsal smaller cusp; dorsalmost basal tooth slender, with margin smooth (Fig. 38). Maxillula, maxilla and max- illiped as in Figs. 39—41 respectively; distal lobe of 1st basipod of maxilliped with 3 setae. : Segmentation and armature of swimming legs 1—4 normal for genus, leg 4 identical to leg 3; Schmeil’s organ present on 2nd endopod segment of leg 2 (Figs. 42-45). Leg 5, posterior surface of Ist basipod with small spinous process. 2nd basipod with short lateral seta; posterodistal margin slightly produced. Ist exopod segment about 2.5 longer than wide. 2nd exopod segment with spine on distolateral corner reaching to midlength of 3rd exopod segment; claw curved slightly inward, each margin finely toothed. 3rd exopod segment distinct from 2nd exopod seg- ment, slightly longer than wide, inner terminal seta about 5 x longer than outer terminal seta. Endopod indistinctly 2-segmented, reaching to midlength of inner margin of lst exopod segment, bearing 2 short spines and group of fine hairs on rounded oblique distal margin (Figs. 46-48). Male: Mean length (including caudal rami) of 10 specimens, 0.93 mm (range 0.90-0.96 mm). Body widest at 2nd pediger. Suture between 4th and 5th pediger complete. 5th pediger produced into small obliquely directed wings each ending in small spine; right wing slightly larger. Genital segment asymmetrical, right side slightly extended posteriorly. Inner margins of caudal rami haired (Figs. 49-51). 584 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Rostral points acute (Fig. 52). Right 1st antenna with socketed spines on seg- ments 8 and 12; spine on segment 10 about two-thirds length of spine on segment 11, these spines parallel to each other and directed obliquely outwards from axis of antenna. Spine on segment 13 large, acute, reaching midlength of segment 14. Spine on segment 15 slightly longer than spine on segment 16. Antepenultimate segment with smooth hyaline membrane (Figs. 53, 54). 2nd antenna and mouthparts similar to those of female. Medial and basal teeth of mandibular gnathal lobe slightly blunter than in female, with at most | sec- ondary small cusp; dorsalmost basal tooth serrate (Fig. 55). Segmentation and armature of swimming legs as in female. Right leg 5, 1st basal segment with posterior expansion ending in short spine. 2nd basal segment almost as broad as long; outer margin with seta at distal quarter; inner margin convex, smooth. Exopod, excluding claw, nearly twice as long as basipod; lst segment about 1.3 x longer than broad, distolateral corner produced. 2nd exopod segment broadest at midlength, 2.5 x longer than lst segment; strong, smooth lateral spine at distal quarter of outer margin, extending almost at right angle to axis of segment and about 1.3 x longer than greatest width of segment. Terminal claw slightly longer than exopod, evenly curved, teeth of inner margin increasing in size distally. Endopod short, located on inner posterodistal margin of and distinct from 2nd basal segment; slightly longer than broad and bearing fine hairs on distal margin (Figs. 56, 57). Left leg 5 (Figs. 56, 58, 59) reaching proximal third of 2nd exopod of right leg. lst basal segment slightly longer than wide, with small conical process on posterior surface. 2nd basal segment about 1.5 x longer than broad, broadest just beyond midpoint, with seta at distal quarter of outer margin. Proximal and distal hairy pads of exopod separate, proximal pad incompletely divided with proximal part slightly more expanded. Proximal expansion of distal pad with numerous fine hairs, separated by sinus from distal part which bears short spinules. Distal process about as long as 2nd exopod segment, with serrate hyaline flange on each side. Proximal process slender, acute, distinct from 2nd exopod segment, slightly longer than distal process. Endopod of 1 segment, conical, reaching midlength of 2nd exopod segment, with fine hairs and 1 spine on rounded distal margin. Color of formalin-preserved female and male specimens light tan with rostral area, margins of pedigers and thoracic wings, caudal rami and setae, 1st antennae, maxillipeds, swimming legs and area of genital opening colored dark blue (Figs. 31, 32). Some color persists in alcohol preservation with chromatophores in caudal rami and caudal setae remaining visible (Fig. 30). Etymology. —The specific epithet azureus is derived from the Old French “‘azur,”’ blue. Remarks. —‘“‘Diaptomus”’ azureus resembles members of the genus Notodiap- tomus in several respects, but differs in having the spines of segments 10 and 11 of the right antenna of the male angled outwards rather than parallel to the axis of the antenna; and in several details of the right leg 5 of the male, primarily the lack of a pointed process on the anterior surface of the 2st exopod segment; the blunt, not conical endopod; and the lack of a prominence on the inner border of the 2nd exopod segment. The left exopod of the left leg 5 of the male also differs from those of Notodiaptomus spp. in the structure of the distal pad and in having the proximal process extending past the distal process. VOLUME 98, NUMBER 3 585 Figs. 49-59. “Diaptomus” azureus, male: 49, Urosome, dorsal; 50, Anterior urosome, right lateral; 51, Anterior urosome, left lateral; 52, Rostrum, lateral; 53, Right lst antenna, segments 8-17; 54, Right lst antenna, segments 23-25; 55, Mandible, gnathal lobe; 56, Legs 5, posterior; 57, Right leg 5, basipod and expanded view of endopod, anterior; 58, Left leg 5, exopod and endopod, posterior; 59, Left leg 5, exopod and endopod, anterior. Scale a, Figs. 49-51; Scale b, Figs. 52-54, 56, 57; Scale c, Figs. 55, 58, 59; each scale = 100 um. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 586 me pee —<—$<——==> To ——s = VOLUME 98, NUMBER 3 587 Since “D.” azureus closely resembles no other presently known South American diaptomid, it is provisionally assigned to ““Diaptomus’’ sensu lato. It is distin- guishable from all other species by details of the structure of the right 1st antenna and 5th legs of the male, and of the thoracic wings, urosome and Sth legs of the female. ““Diaptomus”’ fluminensis, new species Figs. 60-82 Material examined. —1 6, holotype (MZUSP 6202), dissected on slides; 1 3 (MZUSP 6203) and 2 Stage V copepodites (MZUSP 6204), alcohol-preserved; and 1 6, collection of F. A. Esteves, dissected on slides; all paratypes; all from Lagoa Iodada (Coca-Cola), State of Rio de Janeiro, 22°27'S, 41°51'W, collected 14 Sep 1983. Description. — Male: Length (including caudal rami) of type-specimen 1.02 mm; length of adult paratypes 0.94 mm and 0.96 mm. Body widest at Ist pediger (Fig. 60). Suture between 4th and Sth pedigers incomplete dorsally. 5th pediger slightly elevated dorsally, produced laterally into short wings, each ending in elongate, obliquely directed spine, left wing slightly larger than right. Genital segment asym- metrical, right side produced over succeeding segment and bearing small poste- riorly directed spine. Right margins of succeeding urosomal segments also slightly produced posteriorly. Inner margins of caudal rami haired (Figs. 60-63). Rostral points acute (Fig. 64). Right 1st antenna with socketed spines on seg- ments 8 and 12. Spines on segments 10 and 11 slender, subparallel to axis of antenna, each reaching past insertion of succeeding spine; spine on segment 11 with notched tip. Spine on segment 13 slightly stouter, subparallel to axis of antenna, reaching nearly to midlength of segment 14, with notched tip. Spine on segment 16 very small. Antepenultimate segment with narrow hyaline membrane (Figs. 65, 66). Left 1st antenna with armature as in Fig. 67; segments 11 and 13-21 each with 1 seta. 2nd antenna, mandible, maxillula and maxilliped appear to have normal setation for genus (Figs. 68-72 respectively), though maxillula was broken during dissection. Terminal setae on exopod of 2nd antenna distinctly separated from segment 7. Gnathal lobe of mandible with following dentition: apical tooth point- ed, subapical tooth blunt; ventralmost medial tooth bicuspidate, dorsal 2 medial teeth more acute, each with additional minor cusp; 3 basal teeth increasingly acute dorsally, each with 1 major and 2 minor cusps. Notch between apical and subapical teeth deep (Fig. 73). Structure and armature of swimming legs normal for genus; leg 2 with Schmeil’s organ on posterior surface of 2nd endopod segment. Setation of leg 4 identical to that of leg 3 (Figs. 74-77). —_— Figs. 60-73. “Diaptomus” fluminensis, male: 60, Habitus, dorsal; 61, Urosome, dorsal; 62, An- terior urosome, right lateral; 63, Anterior urosome, left lateral; 64, Rostrum, ventral; 65, Right Ist antenna, segments 7-16; 66, Right lst antenna, segments 23-25; 67, Left 1st antenna; 68, 2nd antenna; 69, Mandible; 70, Maxillula (broken); 71, Maxilla; 72, Maxilliped; 73, Mandible, gnathal lobe. Scale a, Fig. 60; Scale b, Figs. 61-63, 67; Scale c, Figs. 64-66, 68-72; Scale d, Fig. 73; each scale = 100 pm. 588 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 74 Lil Figs. 74-82. “Diaptomus” fluminensis, male: 74, Leg 1, anterior; 75, Leg 2, posterior; 76, Leg 3, anterior; 77, Leg 2 endopod, lateral; 78, Legs 5, posterior; 79, Right leg 5, lateral; 80, Right leg 5, endopod; 81, Left leg 5, exopod and endopod, posterior; 82, Left leg 5, exopod, anterior. Scale a, Figs. 74-79; Scale b, Figs. 80-82; each scale = 100 um. Sth legs (Figs. 78-82) stout, 1st basipods broader than long, each with posterior conical projection ending in posterolaterally directed spine, right projection and spine larger than left. Right 2nd basipod also stout, broader than long, inner margin convex with several posteromedial cuticular thickenings, outer margin with short seta at distal third, anterodistal margin produced. Right exopod about 1.5x length of basipod, Ist exopod segment slightly longer than broad, outer margin about twice as long as inner margin, outer and inner distal corners pro- duced. 2nd exopod segment broadest at distal quarter, slightly less than twice as long as broad, with cuticular thickenings on inner proximal corner, distal quarter of inner margin, and posterior surface. Smooth lateral spine at distal third of outer margin, bent slightly at proximal quarter. Terminal claw slightly longer than exopod, directed posterodorsally, recurved at tip, inner margin finely serrate. Endopod broadly triangular, of 1 segment, inserted on produced anterodistal VOLUME 98, NUMBER 3 589 margin of 2nd basal segment, reaching almost to mediodistal corner of 1st exopod segment, bearing subterminal row of fine hairs and | stout claw on outer margin. Left 2nd basipod segment slightly longer than broad, both margins slightly convex, inner proximal corner produced into rounded knob, short seta proximal to outer distal corner. Left endopod broadly triangular, weakly 2-segmented with suture most apparent on anterior surface, reaching midlength of distal pad of left exopod; blunt tip with terminal group of fine hairs and | spine on mediodistal corner. Both segments of left exopod distinctly separated, pad on posterodistal corner of lst exopod segment partly divided, haired on anterior surface. 2nd exopod segment, inner margin with crescentic pad set with spinules and anter- odistal rounded pad with many long hairs on anterior surface. Distal process continuous with 2nd exopod segment, broadly triangular with tip recurved out- wards, margins serrate. Slender proximal process about as long as and inserted slightly anterior to distal process. Color of formalin-preserved specimens light tan. Etymology. —The specific epithet is derived from the Brazilian term “‘flumi- nense,”’ a native of the State of Rio de Janeiro. Remarks. —Though many characters of “D.” fluminensis agree with those of the rather loosely defined genus Notodiaptomus, several details of the Sth legs such as the shape of the right 1st exopod, the 2-segmented left endopod and the placement of the proximal process of the left exopod preclude its inclusion in this genus. Principal differences from other species include the details of the spination of the right 1st antenna and the form of the distal process of the exopod of the left 5th leg. It is provisionally assigned to ““Diaptomus”’ sensu lato. Discussion As M. S. Wilson (1951) pointed out, knowledge of the structure and armature of mouthparts, particularly the maxilliped, the Ist and 2nd swimming legs, the armature of the Ist antennae of the female and of the left 1st antenna of the male is necessary for satisfactory comparison of species or genera of diaptomid cope- pods. Unfortunately these features-have been described for only a few South American diaptomids. Lack of comprehensive knowledge of the morphology of most species has contributed to the present confusion in which about 20 species have not been assigned to genera (or subgenera), remaining in ““Diaptomus”’ sensu lato (Brandorff 1976). Nor has it been possible to construct a complete key to the identification of all species (Brehm 1958). Therefore I have included descriptions of mouthparts and other characters of uncertain taxonomic utility. For instance, the extent to which details of the den- tition of the mandibular gnathal lobe may vary with sex, age or genetic differences has not been investigated. The dentition of the lobe in N. theringi differs from N. cearensis as redescribed by Bowman (1973) in having the cusp of the ventralmost medial tooth produced in a separate acute tooth; the margin ventral to the sub- apical tooth produced into a distinct lobe; and most teeth tricuspidate rather than bicuspidate as in N. cearensis. The setation of the lst basipod of the maxilliped is similar in both species. Wilson (1951) noted that Schmeil’s organ (Schmeil 1896) is lacking in both sexes of Nordodiaptomus alaskaensis and in females of Nordodiaptomus siberien- sis, but is present in males of the latter species. This organ, of unknown function, 590 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON occurs in both sexes of Notodiaptomus venezolanus deeveyorum Bowman (1973), N. cearensis as redescribed by Bowman (1973), and Pectenodiaptomus caperatus (Bowman, 1979), as well as in the three species described here. Acknowledgments I thank Dr. Francisco de A. Esteves for providing the samples of zooplankton and the data in Table 1, as well as for his kind hospitality during a working visit to the laboratorio de Limnologia, Departamento de Ciéncias Bioldgicas, Uni- versidade de Sao Carlos, Sao Paulo. Drs. Thomas E. Bowman and Carlos Eduardo F. da Rocha and an anonymous reviewer made helpful comments on the manu- script. The Conselho Nacional de Desenvolvimento Cientifico e Tecnolégico (CNPq) and the Fundagao de Amparo para Pesquisa do Estado de Sao Paulo (FAPESP) provided support for the collections and water analyses. The U.S.—Latin American Cooperative Science Program, Division of International Programs, National Sci- ence Foundation, provided a travel grant to the author. Literature Cited Bowman, T.E. 1973. Two freshwater copepods from Barrancas, Venezuela: Notodiaptomus cearensis (Wright) and N. venezolanus deeveyorum, new subspecies (Calanoida: Diaptomidae).—Pro- ceedings of the Biological Society of Washington 86:193-202. 1979. Amsterdam Expeditions to the West Indian Islands, Report 5. Notodiaptomus cap- eratus, a new calanoid copepod from phreatic groundwater in Barbuda (Crustacea: Diaptomi- dae). —Bijdragen tot de Dierkunde 49:219-226. Brandorff, G.-O. 1976. The geographic distribution of the Diaptomidae in South America (Crustacea, Copepoda).— Revista Brasileira de Biologia 36:613-627. Brehm, V. 1958. Bemerkungen zu einigen Kopepoden Siidamerikas.—Sitzungsberichte der Oster- reichischen Akademie der Wissenschaften, Abteilung 1, 167:139-171. Cipolli, M. N., and M. A. J. Carvalho. 1973. Levantamento de Calanoida e Cyclopoida (Copepoda, Crustacea) das aguas da regiao do Guama, Capim e Tocantins, com nota sobre a fauna acom- panhante.— Papéis Avulsos de Zoologia 27:95-110. Esteves, F. A., R. Barbieri, I. H. Ishii, and A. F. M. Camargo. 1983. Estudos limnoldgicos em algumas lagoas costeiras do Estado do Rio de Janeiro.—Anais do III Seminario Regional de Ecologia (Universidade Federal de Sao Carlos, Brazil):25-38. Fleminger, A. 1967. Taxonomy, distribution and polymorphism in the Labidocera jollae group, with remarks on evolution within the group (Copepoda: Calanoida).— Proceedings of the United States National Museum 120:1-61. Kiefer, F. 1936. Uber die Systematik der Siidamerikanischen Diaptomiden (Crust. Copep.).—Zoolo- gischer Anzeiger 116:194—200. Schmeil, O. 1896. Deutschlands freilebende Siisswasser-Copepoden. III Teil: Centropagidae. Erwin Nagele Verlag, Stuttgart. Pp. 1-143. Sendacz, S., and E. Kubo. 1982. Copepoda (Calanoida e Cyclopoida) de reservatorios do Estado de Sao Paulo.—Boletim do Instituto de Pesca 9:51-89. Wilson, M. S. 1951. A new subgenus of Diaptomus (Copepoda: Calanoida), including an Asiatic species and a new species from Alaska.—Journal of the Washington Academy of Sciences 41: 168-179. Wright, S. 1935. Three new species of Diaptomus from northeast Brazil.—Annais da Academia Brasileira de Sciéncias 7:213-233. . 1936. Preliminary report on six new species of Diaptomus from Brazil.—Annais da Academia Brasileira de Sciéncias 8:79-85. . 1938. Distribuicao geographica das espécies de Diaptomus na América do Sul.—Livro Jubilar do Prof. Travassos 3:561—566. Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 98(3), 1985, pp. 591-597 NEW SUBSPECIES OF THICK-BILLED VIREO (AVES: VIREONIDAE) FROM THE CAICOS ISLANDS, WITH REMARKS ON TAXONOMIC STATUS OF OTHER POPULATIONS Donald W. Buden Abstract. — Vireo crassirostris stalagmium is described as a new subspecies from the Caicos Islands (southern Bahamas) and is compared with all other subspecies of V. crassirostris. The name Vireo crassirostris alleni Cory is resurrected for the Cayman Islands population. The Thick-billed Vireo (Vireo crassirostris) is a common resident in scrublands and xeric to mesic woodlands throughout most of the Bahamas, though apparently absent from the Turks Bank at the extreme southeastern end of the archipelago. This species is present elsewhere in similar habitats on Ile de la Tortue (off the northern coast of Haiti), on Grand Cayman, Little Cayman, and Cayman Brac (islands approximately equidistant from Cuba and Jamaica), and on Old Provi- dence Island (=Providencia), including Santa Catalina Island (in the southwestern Caribbean)—the Little Cayman population may be recently extirpated (Diamond 1980). Vireo crassirostris has been recorded casually in Florida (American Or- nithologists’ Union 1983), presumably vagrant from the Bahamas. Bond (1956) and Blake (1968) recognized three subspecies: Vireo crassirostris crassirostris (Bryant) in the Bahamas and the Cayman Islands, V. c. tortugae Richmond on Tortue, and V. c. approximans Ridgway on Old Providence. Ac- cording to Bond (1965) approximans differs from the other subspecies mainly in voice; also, the mandible tends to be paler (more yellow) in the specimens from Old Providence. Chromatically, V. c. tortugae is the most distinctive subspecies. Specimens from Tortue are darker (more brown or buff) on the venter than are those from elsewhere in the range, though one or two (of nearly 200 examined) from the Bahamas approach this condition. Morphological variation among Ba- haman and Cayman populations is more complex. Cory (1886) described Vireo alleni as a new species based on material collected on Grand Cayman. In 1887, Ridgway proposed the name Vireo crassirostris flavescens for relatively bright yellow examples of Thick-billed Vireos in the Bahamas that he found exclusively on Rum Cay and Conception Island. He reported (Ridgway 1887, 1904) that flavescens and the nominate form both oc- curred together on Eleuthera, Cat Island, Green Cay, and Inagua. Cory (1892) treated V. c. flavescens as a synonym of V. c. alleni giving the range of the latter as Grand Cayman and Cayman Brac and the following Bahama Islands: Berry Islands, Bimini Islands, Eleuthera, Rum Cay, Conception Island, Mayaguana, Inagua. Todd and Worthington (1911) reported that specimens of V. crassirostris in the southern Bahamas are more yellow ventrally than are those from the northern islands, and that geographically intermediate samples include “puzzling series of 592 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON intergrades . . . as well as typical examples of each form occurring well within the area of predominance of the other.”’ They treated V. c. flavescens as a synonym of V. c. crassirostris, but retained the name V. alleni, at least tentatively, for the Cayman population. Bangs (1916) followed Todd and Worthington (1911) in regarding V. c. flavescens as a synonym of V. c. crassirostris and included the Cayman Islands within the range of this subspecies stating that characters sup- posedly distinguishing alleni from crassirostris (browner back, shorter outer pri- mary, broader wing bars) all prove illusive. I have examined new material from the Bahamas together with other samples from throughout the range of this species; mensural data are given in Table 1. All measurements are in millimeters. Wings were measured flattened against a ruler and the comparisons involving wing and tail measurements (for all samples) are limited to specimens collected in the months January through April. Specimens from the Cayman Islands tend to have darker bills than do those from elsewhere in the range of the species. The Cayman birds usually have a dark brown culmen that appears almost black in many individuals with dark pigment especially prominent on the proximal half of the bill; the tomial region is paler, more gray than brown or black. In Bahaman specimens, the culmen is medium brown with tan and/or reddish tones, especially toward the tip; the coloration of the tomial region is similar to that of the Cayman birds. The contrast in coloration between the mandibular tomial region and the darker parts of the gonys and rami is greater in Cayman birds than in Bahaman examples. Among MCZ specimens of Vireo crassirostris, individuals with the darkest bills are from Little Cayman. Examples from Grand Cayman and Cayman Brac are, for the most part, intermediate in bill color between Little Cayman and Bahaman specimens. In my opinion, the difference in bill color justifies treating V. crassi- rostris in the Cayman Islands as members of a separate subspecies. The name Vireo crassirostris alleni Cory is available for this population. Except for the measurement tail length in females, the specimens from Grand Cayman average slightly smaller in bill, wing, and tail measurements than do those from Little Cayman and/or Cayman Brac (Table 1). The pattern of distribution of color variants among 190 Bahaman specimens of V. crassirostris that I examined (Figure 1) essentially accords with the descrip- tion given by Todd and Worthington (1911). The whitest birds are from the northwesternmost islands generally, whereas individuals with the greatest amount of yellow are from the central and southern Bahamas. Individuals that are rela- tively more yellow on the venter also are relatively more yellow-green (less olive, brownish-green, or grayish-green) on the dorsum. Distribution of these “‘color-phases,”’ however, is not consistent geographically; many samples, especially among those from the more central part of the archi- pelago, show much variation between (in some cases including) the two extremes in coloration. Also, “yellowish” individuals are well-represented in samples from the Bimini and Berry islands in the northwestern Bahamas where white-ventered individuals predominate. Todd and Worthington (1911) cite other examples of color extremes found outside of the areas they usually occur. I follow these authors in treating V. c. flavescens as a synonym of the nominate subspecies, as did Bangs (1916), Hellmayr (1935), and Bond (1956). Todd and Worthington (1911) did not discuss mensural variation in V. cras- s ————=—S—S Wine on =~ Y EL LOW Oo ENC: BOS rs) A B C D E = G H | J K Le Ot Oe OF oe Sot Ot oe es Fig. 1. Distribution of 190 specimens of Vireo crassirostris in four categories of ventral coloration ranging from predominately white (1) to predominately yellow (4) given as males/females for each sample; localities as follows: A = Little Bahama Bank, B = Andros, C = New Providence, D = Eleu- thera, E = Green Cay, F = Exumas, G = Cat Island, H = Long Island, I = Rum Cay, J = Crooked- Acklins + Mayaguana, K = Great Inagua, L = Caicos Islands. 593 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 594 — — = t'b-0'r Gielen ¢1S-7 9r 0°79-0'°09 () Ir (1) 601 (1) L’8h (1) 0°19 () TP (bp) LT (S) S'6r (Ss) 0°19 dUSPIAOId PIO 8r-S Pp SCLC Ul 1 9b-l bP = 6b 9TI-E TI 0°6b-S 8b 0°€9-0°79 (S) 9b (9) SIT (Z) 1st (1) 0°09 (LY (S) OCI (€) LSP (€) L°79 oelg ueWIARD = = oa = 6 b-£'P Salad = = _ = = = (9) 9b (S) 611 = = UBWARD OINTT €v-I'v 811-7 Ol 9°Lb-0'°SP 0'09-0°8S 8'b-E'b 811-801 O'rS-0'Sb 0°€9-0'09 (D7Pr (S) 80 (Z) €9r (Z) 0°6S (S) SP (9) E11 (pb) 18h (bp) O19 uewke) puein Tr-lb 8°OI-L'01 L'9b-€ EV 0°09-0'8S l'b-6'€ VII 01 v'6b-0'°SP 0°€9-0'°19 (2) 7v (Z) 8°01 (Z) O'S¥ (Z) 0°65 (7) 0'v (9) 601 (S)7LP (€) 0°79 onyI0 |, v'-6'€ 7 TI-8'O1 8°6b-0° Tr 0°19-0°8S S'b-0'b 71-901 6 9b-b TP 0°'19-0'LS (L) €'v (6) STI (9) b br (€) 0°6S (Str (81) STI (€1l) Lbr (6) b'6S yueg soore) Lb-'b €7I-S Ol l'8b-7 br 0'9-0'09 0'S-€'p ¢ 71-9 01 € 0S-€°Lv 079-019 (8) 9° (pI) SII (01) $'9r (bv) 8°79 (6) Lv (i) eu (S) 8°8r (€) €°79 enseuy 1eoIH = = = = = O'EI-I' IT 7 8b-8°Sb 0'v9-0'°79 = (1) 871 (1) 8°LP (1) 0°79 (Z) 9b (bp) 171 (€) OLP (€) 0'€9 SUIPPOY-P2x001D WAG OCIA 8b-S br 0'v9-0'09 == OPS VI 1'0S-9 6b 0'¥9-0°€9 (S) br (9) STI (€) 6°9r (bp) O19 (2 Sr (€) S11 (€) 6'6r (€) L°€9 Aed wing = = v'6b-8 SP 0°€9-0'79 8'b-S' Pb acl O'1S-7 Sb 0°S9-0°09 (€) 9p (E) LUI (Z) 9 Lb (€) $79 (Lv (L) SII (8) SSP (8) €°€9 seuinxg Sr'L TCI-C II 0'6b-7 9b 0'79-0'°8S = - GMSOM? 0'¥9-0°€9 (€) 9b (E) LI (Q) 7 LY (€) O19 (St COZral (Z) 7 6b (Z) S°€9 Aed uae1h a = = = 8 P-E'P 8° TI-S II I CSO? 0°99-0'19 (D Lv (1) O71 (1) 8°8r (1) 0°79 (9) 9b (9) V°ZI (9) b'6r (9) €'€9 1) 8'b-E'b Gila Gs lil ¢'0S-8'9r 0'9-0'09 L-7P Ose 1M SIS-E Lb 0°$9-0°09 (OI) SP (€1) OCI (6) 68h (01) 0°79 (CDS (61) ICI (OI) 70S (OI) 8°79 raeli plats) (| 8-7 LP CLO pIs-S Sp 0°€9-0'09 O'S-€'r NO iit 9°7S-I bb 0°S9-0'09 (SDS (ZZ) SII (LI) 1'8¢ (11) L°19 (pI) 9b (9@) LIT (07) $ 6b (91) $79 JOUSPIAOIg MON = 911-601 = = SP-E'b 071-801 0'7S-I 6b 0'79-0°79 = (ZEIT (1) p'9r = (_) rv (ETI (Z) 9°06 (7) 0'€9 soipuy 0'S-€'r 6 CI-T II bv IS-79r = 6 b-£'P 8 7IV 01 0°7S-0'LP 0°99-0°79 (01) 9+ (Ep) AP (01) 7 8r (1) 0°79 (91) 9b (61) 81 (11) b'6r (01) 8°€9 yueg 917 yidop [it yisuey Ia yisugy [te L yisug] Bury yidop I'd yisugy Id yisugy [rey yisusy SUT Aytpeoo7T so[ewloy sole “SIOJOUIT] [IU UT S}JUSWOINSeOW [[e ‘judy ysno1) Arenuesl Jo syJUOW oY) SULINp A[UO pa}da][09 susuUIIOAdsS Jo oIe s}UDWIOINSeOUT [Ie] Pue BUIAA “SPULIS] PUILYeY oIe (yURg SOdIeD YsnoIy) YURg 9]11I]) SOMI[eoO] [] IsIY 9y ‘sotoads oy} JO BBULI OY} INOYZNOIY) WOT S71JSOAISSDAI OAALA JO SO|AUILS OQ] UT SJUWOINSeOUT INO} 1OF (Z MOI) DSU PUL (] MOI) 9ZIS a[dWIes puke ‘URI —"] 29IGe_L VOLUME 98, NUMBER 3 595 sirostris and no marked differences in measurements among Bahaman samples of this species were reported by Ridgway (1904). However, many of Ridgway’s samples were extremely small; measurements of only one male and three females from Inagua comprised his mensural data on specimens from the southern Ba- hamas. The means and sample sizes for the measurements wing length, tail length, bill length, and bill depth in Bahaman specimens that I examined are given in Table 1. There are no noteworthy differences among 10 of the 11 samples. How- ever, with the exception of the measurement bill length, individuals of both sexes from the Caicos Islands average consistently smaller than those of any other Bahaman sample. The Caicos population may be known by the name. Vireo crassirostris stalagmium, new subspecies Holotype. —LSUMZ 70838, adult male, Caicos Islands, North Caicos, along Flamingo Pond Road, east of Kew, collected 29 Feb 1972 by D. W. Buden. Characters. —In comparison with the nominate subspecies, stalagmium differs by its smaller measurements of wing length, tail length and bill depth; differences are most marked in males. Nine male stalagmium range from 57.0 to 61.0 mm in wing length, whereas 63 male crassirostris (from throughout the Bahamas) range from 60.0 to 66.0 mm in this character. The 10 samples of male crassirostris each averages over 61.0 mm in wing length and the one sample of male stalagmium averages 59.4 mm. The sample of male stalagmium averages 44.7 mm in tail length, whereas each of 10 samples of male crassirostris averages greater than 47.0 mm in this character. In coloration, specimens of stalagmium tend to resemble many examples of crassirostris from the central and southern Bahamas, but in comparison with many of the individuals from the northern islands, they have more yellow pigment, especially on the venter. In wing and tail measurements, examples of stalagmium average also smaller than do those of all other subspecies (tortugae, alleni, approximans), at least in males. Individuals of stalagmium differ from those of tortugae further in having (on the average) longer bills and in having more yellow on the venter with less buff color on the upper part of the breast and along the sides of the neck and throat. Examples of stalagmium tend to have paler bills than do those of alleni. Range. —Known only from the Caicos Islands in the extreme southeastern Bahamas. Recorded from West Caicos, Providenciales, Bay Cay, Water Cay, Pine Cay, Parrot Cay, North Caicos, Middle Caicos, East Caicos, and South Caicos, and doubtlessly occurring on other of the many islands of the Caicos Bank. Etymology. —From the Latin, stalagmium, a golden pearl pendant or earring, in allusion to the yellow ventral coloration of this subspecies and to the location of the Caicos Bank relative to the other islands in the Bahama chain; a noun in apposition. Specimens Examined Vireo crassirostris crassirostris. -BAHAMA ISLANDS: Grand Bahama, AS (1M, 1F), FMNH (2M, 1F), MCZ (2M, 2F); Strangers Cay, MCZ (1M); Great Abaco and Little Abaco, AS (1M), FMNH (5M, 6F), USNM (6M, 1F); Elbow Cay, MCZ (1M); South Bimini, AS (1M, 2F); Berry Islands, MCZ (2M, 1F); Andros, AS (3M, 2F), MCZ (2M, 1F), USNM (1M); New Providence, LSUMZ 596 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON (1M), MCZ (24M, 20F, 4?), USNM (8M, 8F, 1?): Eleuthera (including Current Island), AS (6M, 3F), MCZ (4M, 1F), USNM (11M, 8F); Cat Island, USNM (6M, 1F, 1?); Exuma Cays—Cistern Cay, MCZ (1F), Farmer’s Cay, MCZ (1M), Exuma (=Great Exuma?), MCZ (2M, 1F), Roseville, Great Exuma, MCZ (2M), Anna’s Tract, off Exuma, MCZ (1M), Brigantine Cay, MCZ (1M), Conch Cut Cay, MCZ (2M), Green Turtle Cay, MCZ (1M), Refuge Cay, MCZ (1M); Green Cay, USNM (2M, 1F); Long Island, USNM (3M, 2F); Rum Cay, AS (1F), USNM (3M, 5F, 1?); San Salvador, AS (1M); Ragged Islands, Hog Island, USNM (1M); Crooked Island, LSUMZ (1M, 1F); Acklins Island, LSUMZ (2M), USNM (1M); West Plana Cay, USNM (1M); Mayaguana, LSUMZ (2M, 2F, 1?); Great Inagua, FMNH (8M, 12F), MCZ (3M, 2F); Little Inagua, USNM (1F); no other locality, MCZ (12). Vireo crassirostris stalagmium. —CAICOS ISLANDS: Providenciales LSUMZ (5M, 1F, 1?), USNM (1M, 1F); North Caicos, LSUMZ (5M, 1?); Middle Caicos, LSUMZ (1?), USNM (2F); East Caicos, LSUMZ (2M, 2F, 1?); South Caicos, MCZ (1M, 17); no other locality, FMNH (5M, 3F). Vireo crassirostris tortugae. —-ILE DE LA TORTUE, AS (2M), MCZ (1M, 1F), USNM (3M, 1F). Vireo crassirostris alleni. —CAYMAN ISLANDS: Grand Cayman, LSUMZ (3M, 1F, 12), MCZ (3M, 4F); Little Cayman, MCZ (6M); Cayman Brac, FMNH (4M, 2F), MCZ (2M, 2F). Vireo crassirostris approximans. -OLD PROVIDENCE, MCZ (5M, IF, 1F?). Acknowledgments For making comparative material available, either on loan or during visits to museums, I thank the curators and supporting staff of the following collections: Albert Schwartz Collection, Miami Florida (AS), Field Museum of Natural History (FMNH), Louisiana State University Museum of Zoology (LSUMZ), Museum of Comparative Zoology (MCZ), National Museum of Natural History (USNM). I thank Albert Schwartz for reviewing the manuscript and Raymond A. Paynter, Jr. and Alison Pirie for receiving and storing borrowed material at the MCZ and for their assistance during many visits. This study is based in part on a doctoral dissertation completed at Louisiana State University. Literature Cited American Ornithologists’ Union. 1983. Check-list of North American birds. 6th edition. American Ornithologists’ Union, Washington, D.C. 877 pp. Bangs, O. 1916. A collection of birds from the Cayman Islands.— Bulletin of the Museum of Com- parative Zoology 60:303-320. Blake, E.R. 1968. Family Vireonidae. Pp. 103-138 in R. A. Paynter, Jr., ed., Check-list of birds of the world. Vol. 14. Museum of Comparative Zoology, Cambridge, Massachusetts. Bond, J. 1956. Check-list of birds of the West Indies. 4th edition. Academy of Natural Sciences, Philadelphia. 214 pp. . 1965. Tenth supplement to the check-list of birds of the West Indies (1956). Academy of Natural Sciences of Philadelphia. 16 pp. Cory, C. B. 1886. Descriptions of thirteen new species of birds from the island of Grand Cayman, West Indies.— Auk 3:497-S01. . 1892. Catalogue of West Indian birds. [Privately published by the author]. Alfred Mudge and Son, Printers, Boston. 163 pp. VOLUME 98, NUMBER 3 597 Diamond, A. W. 1980. Ecology and species turnover of the birds of Little Cayman.— Atoll Research Bulletin 241:141-164. Hellmayr, C.E. 1935. Catalogue of birds of the Americas. Part 8.— Field Museum of Natural History, Zoological Series 13(8):1-541. Ridgway, R. 1887. A manual of North American birds. J. B. Lippincot Co., Philadelphia. 631 pp. 1904. The birds of North and Middle America.—Bulletin of the United States National Museum 50(3):1-801. Todd, W. E. C., and W. W. Worthington. 1911. A contribution to the ornithology of the Bahama Islands.— Annals of the Carnegie Museum 7:388-464. Worcester Science Center, Harrington Way, Worcester, Massachusetts 01604. PROC. BIOL. SOC. WASH. 98(3), 1985, pp. 598-603 THE REINSTATEMENT OF BATHYPOLYPUS FAEROENSITS (RUSSELL, 1909) (OCTOPODA: BATHYPOLYPODINAE) Ronald B. Toll Abstract. —Polypus faeroensis Russell, 1909, is removed from the junior syn- onymy of Bathypolypus arcticus (Prosch, 1849) and is retained in Bathypolypus as a separate species. Bathypolypus faeroensis is further characterized and partially redescribed based on a female specimen from the Denmark Strait. The cephalopod type-material collected by the Fishery Cruiser Goldseeker and described by Russell (1909, 1922) is determined to be no longer extant and a neotype is designated for P. faeroensis. The systematic relationship of Bathypolypus Grimpe, 1921, and Benthoctopus Grimpe, 1921, is briefly discussed. Russell (1909) described Polypus faeroensis based on two males and one female collected at a single trawl station by the Goldseeker in the Faeroe Channel. Russell separated this species from P. arcticus Prosch, 1849, by the shape of the mantle, length of the arms relative to the body, relative length of the hectocotylized portion of the right third arm, and dermal ornamentation. In a supplementary paper based on the same material Russell (1922) repeated his earlier description of P. faeroen- sis, gave illustrations of the radula, hectocotylus, and dermal ornamentation, and included a photograph of the dorsal aspect of the whole animal. Robson (1932) tentatively placed P. faeroensis into the synonymy of Bathy- polypus arcticus (Prosch, 1849). He noted, however, that the form described by Russell represented one extreme end of the range of variation with regard to total length, shape of the mantle, surface sculpture, and web morphology. Jaeckel (1958) regarded P. faeroensis as a varietal form of B. arcticus. Kumph (1958) suggested that the peculiarities represented by P. faeroensis are within the range of variation for B. arcticus based on his examination of 178 specimens and retained it in synonymy. While examining the unidentified octopod holdings of the National Museum of Natural History I encountered a large specimen resembling Bathypolypus from the Denmark Strait that could not be attributed to either B. arcticus or B. proschi Muus, 1962. Further study demonstrated that it was attributable to P. faeroensis and that a reevaluation of that taxon was required. Measurements and indices are as defined by Roper and Voss (1983). Bathypolypus faeroensis (Russell, 1909) Figs, 15 2 Polypus faeroensis Russell, 1909:446; 1922:5, pls. 1, 2, figs. 1, 4-6.—Robson, 1932:287. Bathypolypus faeroensis.—Grimpe, 1921:300; 1922:40; 1925:93.—Robson, 1927: 252, fig. 7.—Muus, 1962:11. Octopus faeroensis. —Robson, 1926:1330. VOLUME 98, NUMBER 3 Table 1.—Measurements (in mm) of Bathypolypus faeroensis (Russell, 1909). ML— 73 Sn— 4.1 MW— 49 WD*A— 55 HW— 30 B— 60 61 AL— L R C— 51 51 I— 163 145+ D— 48 50 Il— 156 151 E— 4 II— 142 132 gills— 9/10 IV— 133 124 FL— 26.7 AW— 10.2 TL— 24 * Approximate values. Bathypolypus arcticus (pars).— Robson, 1932:286.—Kumph, 1958:13 (non Prosch, 1849). Bathypolypus arcticus var. faeroensis. —Jaeckel, 1958:565. Material examined. -NEOTYPE, | female (gravid), ML = 73 mm, FFS Wal- ther Herwig 630/73, 67°21.5'N, 23°30’W, 480-485 m, 140’ net, 9 Sep 1973, Institut fir Seefischerei und Zoologisches Museum der Universitat Hamburg (on extended loan to National Museum of Natural History). Description. —The mantle is ellipsoid in outline, bluntly pointed posteriorly and widest posterior to its midpoint (MWI 67.1) (Table 1). The head is considerably narrower than the mantle and brachial crown and is separated from both by a weak constriction (HWI 41.1) (Fig. 1a). The mantle aperture extends about one half of the circumference of the mantle. The funnel is moderately large (FLI 36.5) and stout. It is free for slightly less than half of its length. The funnel organ is VV-shaped and acutely pointed basally. The inner and outer limbs are subequal in length. The outer limbs are broader basally than the inner ones (Fig. 2a). The web is damaged but appears shallow (WDI 37.8) and has the formula B.A.C. = D.E. The arms are long (ALI 66.5), moderately stout basally (AWI 14.0), and taper to short acute points. The arm order is J.II.IJI.1V. The web extends dorsally and ventrally down the arms almost to the tips. The biserial suckers are small (SNI 5.6) and well separated along the midportion of several arms; the 2 sucker rows are particularly widely placed, with suckers only slightly elevated. The gills are relatively small with nine primary lamellae on the outer demibranch of one, 10 on the other. The viscera were partially dissected. The ovary is massive and completely occupies the ventro-posterior third of the mantle cavity (Fig. 2b). Proximally the oviducts are short and appear to connect by separate pores with the ovarian membrane. The oviducal glands are large, broader than long and completely darkened. Distally the oviducts are large, stout, and curved to form an inverted J-shape. The tips are expanded into a conical swelling. The ovarian eggs are large (x = 18 mm), with a smooth surface and a series of longitudinal lines (Fig. 2c). The eggs appear to be mature. Midway along the esophagus is an enlarged crop that bears a prominent anterior 600 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Bathypolypus faeroensis, neotype: a, Dorsolateral whole view; b, Radula. diverticulum (Fig. 2d). The liver is large and bulbous. When squeezed it exuded a greenish-amber colored oil. The radula and mandibles are as figured (Figs. 1b, 2e—f). Neither is distinctive. The color in alcohol is rust brown to purple with prominent, irregularly sized, subcircular, tan colored markings over the entire dorsal surfaces of the mantle, head, and brachial crown. These spots are loosely organized into a pattern of circlets of smaller ones surrounding a central larger one. The smaller spots are sometimes associated with a small, conical papilla. A single, large, ocular cirrus is located above each eye. It is covered with smaller, irregular warty protuberances. Smaller wart-like papillae form an incomplete cir- clet around the small eyes. Discussion. —While most closely resembling the genus Bathypolypus, the pres- ence of a large and well developed diverticulum of the crop eliminated the pos- sibility that the present specimen could be attributed to Bathypolypus arcticus (Prosch, 1849), B. proschi Muus, 1962, both from the North Atlantic, or any other recognized congener. The combination of a well developed supraocular cirrus and unicuspid rachidian further eliminated its placement in Benthoctopus. VOLUME 98, NUMBER 3 601 Fig. 2. Bathypolypus faeroensis, neotype. a, Funnel organ; b, Ovary; c, Ovarian egg; d, Esophagus with diverticulum; e, Upper mandible; f, Lower mandible. A broader review of the literature,. stimulated by this unusual combination of characters, resulted in comparison of the present specimen to Bathypolypus faero- ensis (Russell), previously placed in synonymy of B. arcticus by Robson (1932) and Kumph (1958). I inquired about the location of Russell’s cephalopod type- material taken by the Goldseeker at the Royal Scottish Museum (RSM) in the hope of obtaining the type-series of Polypus faeroensis for comparison with the present specimen. Dr. D. Heppell, curator of Mollusca, informed me that this material was no longer at the Aberdeen Marine Laboratory and had never been transferred to the RSM “although many of the Goldseeker marine invertebrates are housed here” (pers. comm., 18 Mar 1983). The types of P. faeroensis and those of two teuthoids described by Russell (1909: Brachioteuthis bowmani and Taonidium pfefferi) must be assumed to be lost or destroyed. In the absence of type-material, Russell’s accounts of P. faeroensis were critically evaluated for relevance to the present specimen. Morphometric comparison was difficult because of the small number of specimens in the syntypic series (three) and the difference in size between the present specimen (73 mm ML) and Russell’s syntypes (40, 42, 48 mm ML). The more valuable characters including the relative length of the arms, length of the arms with respect to total length, web depth, web 602 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON order, funnel size and sucker size are, nonetheless, compatible. Unfortunately Russell did not report on the crop morphology. The most substantial basis for comparison is Russell’s description and photograph (Russell 1922, fig. 1) of the distinctive “papillary areas” found on the dorsal surface of the mantle. These are nearly identical in size, coloration, and distribution to the Herwig specimen. Based on these comparisons the present specimen is attributed to Polypus faeroensis Russell and is designated as the neotype. Kumph (1958) stated that the form represented by faeroensis was within the range of variation he determined for Bathypolypus arcticus based on his exami- nation of 178 specimens. A review of his data, however, indicates that differences do exist. Kumph reported that the LLI, based on a total of 96 animals, ranged from 9.5 to 49.2, thereby encompassing the value of 12.2 for the larger and sexually mature syntype of faeroensis (ML = 42 mm). Only a single specimen of Kumph’s had a lower value (9.5). This specimen was the smallest that he had examined (ML = 7.8) and it is certainly immature. Kumph reported that the range of LLI for mature males is 18.1 to 44.8 with a mean of 32.9. The two species can, therefore, further be distinguished on the basis of ligula length. I concur with Robson’s (1932:293) suggestion that Russell’s description of the spermatophores is erroneous. Specifically Russell confused broken or pinched off segments of the sperm rope as individual spermatophores and referred to the actual whole spermatophore as a ““membranous bag.”’ Also, his assessment of the distinctiveness of the radula of faeroensis appears to be based on an unusual orientation of the radular teeth as seen in his figure (Russell 1922, fig. 4) and not on actual morphology. Muus (1962) reported on several large specimens of B. arcticus (TL = 145-220 mm) with multicuspid rachidian teeth and small ligulas (LLI 9.3-16.9). Based on Kumph’s (1958) extensive comparative study, apparently unknown to Muus, it is unlikely that these specimens are correctly identified. Based on the generic definitions given by Robson (1927, 1932) Polypus faeroen- sis cannot be accommodated by either Bathypolypus, because of the size of the ligula and the presence of a crop diverticulum, or Benthoctopus, because of the unicuspid condition of the rachidian teeth and presence of supraocular cirri. Wirz (1955) noted that “‘La stricte séparation des deux genres Benthoctopus et Bathy- polypus faite par Robson en 1932 est sans aucun doute injustifiée, étant donné le trés petit nombre de charactéres différents.’’ Creation of a new genus for the inclusion of faeroensis, placement of Benthoctopus in the synonymy of Bathypoly- pus, or redefinition of these two genera would be premature at this time and only add to the current confusion. In lieu of these alternatives, the writer chooses provisionally to maintain faeroensis as a separate species in Bathypolypus until these genera can be reviewed and their relationship established. Acknowledgments Dr. D. Heppell, Royal Scottish Museum, kindly undertook the search for the types of P. faeroensis and made information available to me concerning their location. Dr. C. F. E. Roper read the manuscript and made many helpful sug- gestions. This work was supported by a Postdoctoral appointment to the Division of VOLUME 98, NUMBER 3 603 Mollusks by the Smithsonian Office of Fellowship and Grants and a faculty re- search grant from the University of the South. This support is gratefully acknowl- edged. Literature Cited Grimpe, G. 1921. Systematische Ubersicht der Nordseecephalopoden.— Zoologischer Anzeiger 52(12/ 13):297-305. 1922. Systematische Ubersicht der Europaischen Cephalopoden.—Sitzungberichte der Na- turforschenden Gesellschaft zu Leipzig 45—48:36-52. 1925. Zur Kenntnis der Cephalopodenfauna der Nordsee.— Helgolander Wissenschaftliche Meeresuntersuchungen 16(3):1-125. Kumph, H.E. 1958. Astudy ofthe Bathypolypus arcticus-bairdii-lentus-obesus complex of the North Atlantic (Cephalopoda, Octopoda). Master’s thesis, University of Miami, 135 pp. Jaeckel, S. G. A. 1958. Cephalopoden.—Die Tierwelt der Nord-und Ostsee.—Lieferung 37, 9(3): 479-723. Muus, B. J. 1962. Cephalopoda. The Godthaab Expedition 1928.— Meddelelser Om Gronland 81(5): 4-23. Robson, G. C. 1926. The deep-sea Octopoda.— Proceedings of the Zoological Society of London 50: 1321-1356. . 1927. Notes on the Cephalopoda. IIT. On the anatomy and classification of the North Atlantic species of Bathypolypus and Benthoctopus.—Annals and Magazine of Natural History (9) 20: 249-263. 1932. A monograph of the Recent Cephalopoda Pt. II. The Octopoda (excluding the Oc- topodinae). London: British Museum, 359 pp., 79 text figs., 6 pls. Roper, C. F. E.,and G. L. Voss. 1983. Guidelines for taxonomic descriptions of cephalopod species. — Memoirs of the National Museum of Victoria 44:48-63. Russell, E.S. 1909. Preliminary notices of the Cephalopoda collected by the Fishery Cruiser ““Gold- seeker,” 1903-1908.—Annals and Magazine of Natural History (8) 3:446—455. . 1922. Report on the Cephalopoda collected by the Research Steamer “‘Goldseeker,”’ during the years, 1903—1908.—Fisheries, Scotland Scientific Investigations 1921, III (Feb. 1922), 45 pp., 25 figs. Wirz, K. 1955. Bathypolypus sponsalis (P. et H. Fischer) espéce commune dans la partie ouest de la Méditerranée.— Vie et Milieu 6(1):129-147. Department of Biology, University of the South, Sewanee, Tennessee 37375. PROC. BIOL. SOC. WASH. 98(3), 1985, pp. 604-610 REDESCRIPTION OF THE NEPHTYID POLYCHAETE AGLAOPHAMUS MINUSCULUS HARTMAN, 1965 Takashi Ohwada Abstract. —The holotype of Aglaophamus minusculus has been redescribed to prevent the establishment of new species that will later have to be synonymized. Examination of the holotype has revealed discrepancies from the original de- scription in two taxonomically important characteristics. The holotype has been found to have at least 12 pairs of interramal cirri rather than about six pairs as previously reported and only one kind of furcate setae, not two. The furcate setae with tines of distinctly unequal length, which were regarded as a second kind of furcate seta in the original description have been identified as damaged forms which originally had tines of almost equal length but one of which had broken near the base. The generic allocation of the present neotenic species in Ag/aopha- mus 1s reviewed on the basis of findings on the morphology of juvenile nephtyids, and a description of the proboscis is provided, based on specimens from off Florida. Aglaophamus minusculus, one of the smallest species of nephtyid polychaete so far reported, was originally described by Hartman (1965) as having about six pairs of interramal cirri and two kinds of furcate setae on both the notopodia and neuropodia among the postacicular fascicle of long and numerous setae. Dr. R. W. Virnstein and Ms. M. A. Capone of the Harbor Branch Foundation, Fort Pierce, Florida, collected approximately 1500 specimens of small nephtyid poly- chaetes from depths of about 200 m off Fort Pierce, on the east coast of Florida, from June 1978 to June 1979. Two hundred and thirty specimens from the Indian River Coastal Museum, Fort Pierce, were sent to the author for identification by Ms. Capone. Except for one Nephtys specimen, all of the specimens seemed to fit the description of Aglaophamus minusculus with the exception that they have 12-17 pairs of interramal cirri and only ‘one kind’ of furcate setae. The holotype of Aglaophamus minusculus, collected from a depth of 200 m on the continental slope off New England, is deposited in the Allan Hancock Foun- dation, Los Angeles, California. This specimen was examined and found to have at least 12 pairs of interramal cirri and only ‘one kind’ of furcate setae. Since the distribution pattern of interramal cirri along the body and the setal types are important taxonomic characteristics of Nephtyidae, it is felt that a redescription of the holotype is necessary to prevent future descriptions of species that might eventually prove to be synonymous. There is no description of the proboscis in the original description; a description of the proboscis of the Florida specimens is provided here. The following description is based on the holotype (AHF-Poly 0784). Although its proboscis is retracted, no dissection of the holotype for examination of the proboscis and parapodia was attempted to avoid further damage to the specimen since it is macerated and three parapodia are missing. VOLUME 98, NUMBER 3 605 Aglaophamus minusculus Hartman, 1965 Fig. la—e Description. —The holotype (AHF-Poly 0784), which is complete, is 4.0 mm long with 28 setigers. The body is broad for its length and 0.6 mm wide excluding setae at the eighth setiger, its broadest point. The proboscis is completely retracted and eyes are absent. The prostomium is longer than wide, the anterior corner is nearly right-angled, with the front and sides almost straight. The first and second antennae are of equal length and tapering (Fig. la). They are close together on the anterior corner of the prostomium with the second pair on the ventral side slightly behind the first pair on the dorsal side (Fig. 1b). Each antenna is about half as long as the prostomium is wide. The first parapodium is reduced and directed forward (Fig. 1a). An acicular lobe supported by a curved-tipped aciculum is recognized in the first right neu- ropodium whereas it is missing in the left one. In the first neuropodium, only finely serrated capillary setae are present. A dorsal cirrus is present as a digitate or somewhat pyriform papilla arising backward from midway along the outer surface of the notopodium. The ventral cirrus is similar, but slightly longer than the dorsal one, arising anterolaterally from the outer surface of the neuropodium. The anterior edge of the second setiger on the ventral side forms the lower lip of the mouth, and the lateral lips are formed by the first setiger (Fig. 1b). From the second setiger, parapodia are similar, and the preacicular and post- acicular lobes are very small. The acicular lobes are much longer, nearly conical and very pointed, and are supported by acicula whose tips curve dorsad in the notopodia and ventrad in the neuropodia. The curved tips of the acicula are covered by a thin sheath. The dorsal cirri are tapered and in the anterior parapodia are pyriform, while in the middle and posterior parapodia they are often slightly constricted at the base. The ventral cirri are smaller and digitiform. Both dorsal and ventral cirri are present from the first to last (28th) setiger. Possibly due to the age and small size of the specimen, most of the interramal cirri are macerated and the posterior interramal cirri are in especially bad con- dition. Short and plump interramal cirri are present from the ninth setiger on the right side of the body, and at least from the tenth setiger on the left side (the ninth left parapodium is missing in the holotype). The interramal cirri are distinct and already almost fully developed in the ninth right and tenth left parapodia. The size of the interramal cirri increases slightly up to 13th to 14th setiger on both sides and then appears to decrease only slightly until the last few setigers. Even in the most developed interramal cirri, they are not developed enough to tell whether they are involute or recurved. They are short and almost straight. On the right side, the interramal cirri are recognizable up to the 21st setiger, with the last two too macerated to tell whether they are reduced or broken off near the base. The interramal cirrus of the 19th right parapodium appears to be broken off at midlength. On the left side, interramal cirri are also recognized up to the 21st setiger, but the interramal cirrus of the 20th left parapodium is very small and that of the 21st left parapodium, even though distinct, is vestigial. This may be due to the macerated condition of these interramal cirri. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON = = o a) 606 200 ym VOLUME 98, NUMBER 3 607 The 29th segment, the pygidium, has no parapodium but acicula can be ob- served to project slightly over the epidermis of the pygidium and are covered by a thin sheath. One slender anal cirrus is present ventral to the anus. Four types of setae are present in both the notopodia and neuropodia, 2 types in the preacicular fascicle and the other 2 in the postacicular fascicle. In the preacicular fascicle there are barred (laddered) setae and capillary setae showing very fine serration. The 2 types of setae are of about the same length. The majority of preacicular setae are barred, and a few capillary setae are present at the upper and lower end of the fascicle. The barred setae are thicker at the base than the capillary setae. In the right notopodia, one capillary seta first appears at the upper end of the preacicular fascicle by at least the sixth setiger and at the lower end at the seventh setiger, and | to 3 (usually 2) capillary setae are present at each end of the fascicle except for the last 4 setigers in the case of the upper end and the last several setigers in the case of the lower end. In the left notopodia, capillary setae first appear at the seventh setiger, 1 at the upper end and 2 at the lower end. Posterior to the seventh setiger, 1 to 3 (usually 2) capillary setae are present at each end of the preacicular fascicle, except for the last 7 setigers in the case of the upper end and the last 6 setigers in the case of the lower end. In the right neuropodia, capillary setae are recognized at the upper end by at least the 13th setiger, and at the lower end 2 capillary setae first appear at the eighth setiger, and | to 2 setae are present at each end of the fascicle except for the last 8 setigers in the case of the upper end and the last 5 setigers in the case of the lower end. At the lower end of the right neuropodia, a maximum of 3 capillary setae are recognized. In the left neuropodia, capillary setae are recognized at the upper end by at least the 10th setiger and at the lower end one capillary seta first appears at the eighth setiger. Thereafter 1 to 2 setae, usually 2 at the lower end, are present at each end except for the last 8 setigers. In the postacicular fascicle more numerous capillary setae of very fine serration and a few furcate (lyrate) setae are found. These capillary setae are slender and 1.5 to 2 times as long as the preacicular setae. Only one kind of furcate seta is present among the capillary setae, and they are longer than half the length of the preacicular barred setae, but less than the total length of the latter. These furcate setae have tines of slightly unequal length, and both tines are spinulose on the inner margins. In the right notopodia, | furcate seta first appears in the fourth setiger and 2 to 4 (usually 3) furcate setae are present except in the case of the last 4 setigers. In the left notopodia, one furcate seta first appears at the fifth setiger and 2 to 4 furcate setae are present except in the case of the last 2 setigers. In the neuropodia, they first appear at the sixth setiger, 2 on the right neuropodium and one on the left. In the right neuropodia, | to 2 (usually 2) furcate setae are recognized except — Fig. 1. Aglaophamus minusculus: a, Anterior end, dorsal view, proboscis completely retracted; b, Same, ventral view; c, Ninth left parapodium, anterior view; d, 16th left parapodium, anterior view; e, Furcate seta from 16th right parapodium (a, b, holotype (AHF-Poly 0784); c, d, from Florida (ORIUT-BEPL-M8407-1-1003); e, from Florida (ORIUT-BEPL-M8407-1-201 1). 608 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON at the last setiger, but many of the setae observed on the neuropodia are broken. In the left neuropodia, 2 to 3 furcate setae are present except in the case of the last 3 setigers. Remarks. —The furcate setae with two tines clearly of unequal length, only the longer one of which was spinulose on the inner margin, were regarded as a second sort of furcate seta in the original description (Hartman 1965; p. 90 and PI. 13, fig. d). Such furcate setae are recognized in some parapodia of the holotype, but in this redescription, these setae are identified as broken forms which originally had tines of almost equal length (see discussion below). Discussion The original description refers to the presence of two kinds of furcate setae in each postacicular fascicle. This, however, is not the case in the holotype. Furcate setae with tines of distinctly unequal length are observed to be present in some parapodia, but they occur intermittently along the length of the body, whereas furcate setae with tines of almost equal length (Fig. le) are recognized in nearly all parapodia except for the first few anterior and last few posterior setigers. I mounted furcate setae from the Florida specimens on slides and under a compound microscope I broke them deliberately by pressing down on the coverglass. In many cases, the broken setae looked like those with tines of distinctly unequal length described by Hartman (1965). In the present paper, therefore, the former type of furcate setae are regarded as damaged forms that originally had tines of almost equal length but one of which had broken near the base. It is, however, not known whether this breakage occurred during life or only after the specimens were fixed. Whatever the case, it is not reasonable to include damaged forms in the description of setal types. Nearly all of the Florida specimens have 27 to 30 setigers and are 4 to 5 mm long (the holotype has 28 setigers and is 4.0 mm long). In the holotype and the Florida specimens, interramal cirri first appear on the seventh to tenth setiger (ninth in the holotype) and they are present until the 20th to 24th setiger (21st in the holotype). Twelve to 17 pairs of interramal cirri are present (13 in the holotype), not about six pairs as stated in the original description. While the interramal cirri of the holotype and the Florida specimens are ob- served to be short, plump and almost straight, the figure of the 16th parapodium in the original description (Pl. 13, fig. b) shows an interramal cirrus which ap- parently bends inward at midlength, and is not plump. Since the holotype has the 16th parapodia on both sides, it appears that the figure might not have been drawn from the holotype. As dissection of the holotype was not conducted, the ninth left and 16th left parapodia of one of the Florida specimens (ORIUT-BEPL- M8407-1-1003) which are identical with those of the holotype are shown in Fig. le, Gl The interramal cirri are not well enough developed to tell whether they are involute or recurved. The interramal cirri of juveniles of Nephtys caeca and N. polybranchia have a slightly involute appearance during the process of develop- ment to adult recurved interramal cirri (Ohwada, unpublished observation). In a neotenic species, therefore, a slightly involute appearance for the interramal cirri does not negate the choice of Nephtys, which has recurved interramal cirri, as the correct genus for the species (Ohwada 1985). VOLUME 98, NUMBER 3 609 Contrary to Fauchald (1968, 1977) who described the tip of the aciculum in Nephtys as straight and blunt, curved acicular tips have been recognized in some Nephtys (Ohwada 1985). In N. caeca, acicula were observed to change from curved to blunt-tipped form as the juvenile grew (Ohwada 1983). With growth, the shaft of the aciculum thickened without thickening of the curved tip and, as a result, the curved tip became practically negligible in size compared to the thick straight shaft. Thus, the presence of curved acicular tips in a neotenic species does not exclude Nephtys as the correct genus for the species. In the present paper, the placement of the present neotenic species within Aglaophamus is determined rather by the presence of interramal cirri, furcate setae and the first pair of an- tennae. Micronephtys lacks interramal cirri (Fauchald 1968, 1977), Nephtys lacks furcate setae (Day 1967) and Inermonephtys lacks the first pair of antennae (Fau- chald 1968). No description of the proboscis is given in the original description, and in the present study dissection of the holotype was not attempted. In the Florida spec- imens with everted proboscises, the proboscis was observed to have 20 bifid subequal papillae and two simple ones in the terminal region surrounding a dorsoventral slit, 10 bifid papillae on each side and a single simple papilla in both the middorsal and midventral position. The subterminal papillae are arranged in 22 longitudinal rows consisting of several conical papillae that decrease in size toward the base of the proboscis. No middorsal unpaired subterminal papilla is present. In the original description, Hartman (1965) considered the present species to have affinity with Aglaophamus malmgreni (Théel) on the basis of similarities in the morphology of the parapodium. In the present study, this view is supported by the similarities in morphology of the anterior end of the body between these two species (Fig. la; for A. malmgreni see Fauchald 1963, fig. 1, F) Ohwada (1985) suggests the possibility that the similarities in morphology of the prosto- mium and the first setiger may indicate systematic closeness. Twenty-six of the Florida specimens (ORIUT-BEPL-M8407-1-1001-—1004, 2001-2011, 6-4001—4007, 9001-9004) are preserved in the Ocean Research In- stitute, University of Tokyo. Acknowledgments The author wishes to express his gratitude to Dr. Susan J. Williams, Allan Hancock Foundation, Los Angeles, for the loan of the holotype of Aglaophamus minusculus. The author also wishes to express his appreciation to Prof. Masuoki Horikoshi of the Ocean Research Institute, University of Tokyo, for his helpful suggestions. Dr. David D. Swimbanks of the Ocean Research Institute, University of Tokyo, kindly corrected the English of the manuscript. Literature Cited Day, J. H. 1967. A monograph on the Polychaeta of Southern Africa. Part I. Errantia. British Museum (Natural History), London, Publication No. 656, 458 pp. Fauchald, K. 1963. Nephtyidae (Polychaeta) from Norwegian waters.—Sarsia 13:1-32. . 1968. Nephtyidae (Polychaeta) from the Bay of Nha Trang, South Viet Nam.— Naga Report, Scientific Results of Marine Investigations of the South China Sea and the Gulf of Thailand, 610 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 1959-1961, 4(3):1-33. University of California, Scripps Institution of Oceanography, La Jolla, California. . 1977. The polychaete worms. Definitions and keys to the orders, families and genera. Science Series 28, Natural History Museum of Los Angeles County in conjunction with the Allan Hancock Foundation, University of Southern California, 188 pp. Hartman, O. 1965. Deep-water benthic polychaetous annelids off New England to Bermuda and other North Atlantic areas.— Allan Hancock Foundation Occasional Paper 28:1—378. Ohwada, T. 1983. The identification of juveniles of nephtyid polychaete and their ecology with spe- cial reference to settling seasons. M.Sc. Thesis, University of Tokyo, 45 pp. 1985. Prostomium morphology as a criterion for the identification of nephtyid polychaetes (Annelida: Phyllodocida) with reference to the taxonomic status of Aglaophamus neotenus. — Publications of the Seto Marine Biological Laboratory 30 (1/3):55-60. Ocean Research Institute, University of Tokyo, 15-1, 1-Chome, Minamidai, Nakano-ku, Tokyo, 164 Japan. PROC. BIOL. SOC. WASH. 98(3), 1985, pp. 611-614 PINNOTHERES JAMEST SYNONY MIZED WITH P. RETICULATUS (DECAPODA: BRACHYURA) Timothy M. Green Abstract. —The pea crab Pinnotheres jamesi is synonymized with P. reticulatus. Drawings of entire specimens are provided for the first time. The known range is extended from the tip of Baja California to the Pacific coast of Costa Rica. A host is recorded. Upon reviewing several vials of pinnotherid crabs from the Muséum National D’Histoire Naturelle, Paris, I discovered that Pinnotheres jamesi Rathbun, 1923 (male) is identical with P. reticulatus Rathbun, 1918 (female). Both were taken from the host bivalve Polymesoda inflata Keene, 1971. Pinnotheres reticulatus was described from a female only, while P. jamesi was described from a male. There may be great sexual dimorphism in the genus Pinnotheres (Rathbun, 1918) which was taken into account in synonymizing these two species. Simple line drawings of both male and female P. reticulatus are provided for the first time. These include dorsal, ventral, and frontal views, as well as chelae and third maxillipeds. Comparison of the third maxillipeds is based partly on illustrations of these structures in the original descriptions. The known range of P. reticulatus is extended from San Josef Island and Pi- chilinque Bay, Baja California, to the Pacific coast of Costa Rica. There are no published records of occurrences between these two points. Pinnotheres reticulatus Rathbun Jeng, I, 2 Pinnotheres reticulatus Rathbun, 1918:93—94, pl. 21, fig. 1 & 2.—Glassell, 1934: 301, list.—Schmitt, McCain, and Davidson, 1973:83, list. Pinnotheres jamesi Rathbun, 1923:625-626, pl. 9, fig. 1 & 2; text figs. 1 & 2.— Glassell, 1934:301, list.—Silas and Alagarswami, 1967:1200—1223, catalogue. — Schmitt, McCain and Davidson, 1973:50, list. Type-locality. — Gulf of California: off San Josef Island, Lower (Baja) California, Mexico. Recorded range. —San Josef Island and Pichilinque Bay, Baja California. Material examined. —Pacific coast, Costa Rica (no further data) from host bivalve Polymesoda inflata Keene, 1971, A. Castaing, 2 females, 1 male. Holotype female of P. reticulatus, USNM 18217, San Josef Island, Baja California (25°02'15"N, 110°43'30”"W), 17 fathoms. Holotype male of P. jamesi, USNM 57005, Pichilinque Bay, Baja California, by electric light. Remarks. —In the description of P. jamesi, Rathbun mentioned similarity of the third maxillipeds to those of P. reticulatus, but dismissed any other similarities and the possibility of their being the same species. I have compared the Paris 612 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Sop ‘jc Ig Fig. 1. Pinnotheres reticulatus. A, 3rd maxilliped male; B, 3rd maxilliped female; C, Front, male; D, Front, female; E, Right chela, male; F, Right chela, female. Museum specimens with the type-specimens of those species. Similarities in the third maxillipeds among these specimens are based partly on published drawings of those structures, as they are missing from the P. reticulatus holotype (Rathbun, 1918, 1923). Additional drawings of the third maxillipeds are provided here. The dactyl, propodus, and carpus are identical in the specimens compared. There is a tuft of fused or densely packed, sickle-shaped, setae on the tip of the dactyl that extends well past the tip of the propodus. This feature is not seen in Rathbun’s illustrations, but is present on the specimens I compared (Fig. 1A, B). The general shape of the two sexes is similar (Fig. 2A, B, C, D). The front, VOLUME 98, NUMBER 3 613 Fig. 2. Pinnotheres reticulatus. A, Male, dorsal view; B, Female, dorsal view; C, Female, ventral view; D, Male, ventral view. abbreviated in the female and pronounced in the male, is similar in both (Fig. 1C, D). Leg shapes and length ratios are similar, relative lengths from first to last in the male being 2:3:1:4 and 2:4:3:1 in the female. The only major difference is that the fourth walking leg of the female is longest due to difference in the dactyl length and shape. Shapes of the limbs are similar, the male’s being more stout and the female’s more elongate. The chelae are sexually dimorphic (Fig. 1E, F), but between sexes are similar to each other in the carpal and meral segments (Fig. 2). The reticulated pattern of the chelae is not evident in the Paris Museum specimens, but is in the holotype female. There is no reticulated pattern on the chelae of male P. reticulatus. Acknowledgments I would like to thank Dr. Mary K. Wicksten for obtaining specimens from Paris, for type-specimens from the USNM, for verification of identification, and for editorial assistance. I also thank Dr. Daniéle Guinot for the loan of specimens from the Muséum National D’Histoire Naturelle. Literature Cited Glassell, S. A. 1934. Affinities of the brachyuran fauna of the Gulf of California.—Journal of the Washington Academy of Sciences 24(7):296-302. Rathbun, M. J. 1918. The grapsoid crabs of America.— Bulletin of the U.S. National Museum 97, 461pp., 161 pl. 614 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON 1923. The brachyuran crabs collected by the U.S. Fisheries Steamer ‘Albatross’ in 1911, chiefly on the west coast of Mexico.—Bulletin of the American Museum of Natural History 48(2):619-637, pl. 24-34. Schmitt, W. L., J. C. McCain, and E. S. Davidson. 1973. Decapoda I, Brachyura I. Family Pin- notheridae. /n Gruner, H. E., and L. B. Holthuis, eds., Crustaceorum catalogus. Den Haag. W. Junk B. V. 160 pp. Silas, E. G., and Alagarswami, K. 1967. On an instance of parasitation by the pea-crab (Pinnotheres sp.) on the black water clam [Meretrix casta (Chemnitz)] from India, with a review of the work on the systematics, ecology, biology and ethology of pea-crabs of the genus Pinnotheres La- treille.— Marine Biology Association of India, Symposium Series 2(3):1161—1227. Department of Biology, Texas A&M University, College Station, Texas 77843- B25 er PROC. BIOL. SOC. WASH. 98(3), 1985, pp. 615-621 FRESHWATER SHRIMPS FROM VENEZUELA III: MACROBRACHIUM QUELCHI (DE MAN) AND EURYRHYNCHUS PEMONTL N. SP.: (CRUSTACEA: DECAPODA: PALAEMONIDAE) FROM LA GRAN SABANA Guido Pereira Abstract. —Macrobrachium quelchi (De Man), not collected since the original description, is redescribed and the color pattern is given. Euryrhynchus pemoni, new species, the fourth species in the genus, is described. The material comes from La Gran Sabana, a large highland plateau in the southeast Venezuela (6°15’N, 60°1'W), in the headwaters of rivers that discharge into the Orinoco River basin. Freshwater palaemonid shrimps represent a large group of crustaceans that inhabit brackish and freshwater habitats throughout the tropics. American species living in brackish water and rivers discharging directly into the sea are, at present, well known (Holthuis 1952; Chace and Hobbs 1969). However, in South America, especially in the Amazon region, there are many tributary streams with almost no connection with the estuarine zone of the main river. Shrimp species from these rivers are not well known, and recent surveys have yielded several new species (Tiefenbacher 1978; Rodriguez 1982; Kensley and Walker 1982; Pereira, in press). Several species of freshwater shrimps from southern Venezuelan rivers are not dependent on saline water for their development (Rodriguez 1981; Pereira 1982). La Gran Sabana, located in this zone, represents an altiplane (1200 m above sea level) consisting of a vast system of creeks and streams (of both black and clear waters) which drain into the Caroni River. This black-water river is one of the main tributaries of the Orinoco River basin. This paper describes two interesting palaemonid shrimps from La Gran Sabana. Macrobrachium quelchi was known only from the original description in 1900 from the the Upper Mazaruni River (Esequibo River basin) Guyana. The other shrimp represents a new species of Euryrhynchus, a fairly homogeneous group of shrimps with an Amazonian distribution (Tieffenbacher 1978). Abbreviations tl and cl are used for total length and carapace length respectively, measured from tip of rostrum to tip of telson, and posterior orbital margin to posterior edge of cephalothorax. MBUCV, Museum of Biology, Central University of Venezuela. USNM, National Museum of Natural History, Smithsonian Institution. Macrobrachium quelchi (De Man) Figs, | 233.4 Palaemon (Macrobrachium) quelchi De Man, 1900:57, pl. 6, figs. 1-8. Material examined. —MBUCV (XI-2111) 3 6, 2 2; stream tributary of River Kama. La Gran Sabana, Bolivar State, Venezuela (6°15'N; 60°1’W), 29 Mar 1983, 616 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Macrobrachium quelchi: A, Cephalothorax, lateral view; B, Cephalothorax, detail; C, Scaph- ocerite; D, Telson; E, Telson, detail; F, First pereiopod; G, Third pereiopod; H, Fifth pereiopod. coll. Guido Pereira and Alfredo Paolillo; MBUCV (XI-2108) 6 6, 3 2; MBUCV (XI-2112) 3 6; same locality. 27 Mar 1983, coll. Guido Pereira and Alfredo Paolillo. Description. —Rostrum straight, apex reaching distal margin of scaphocerite. Upper border bearing 6—8 regularly distributed teeth, first (proximal) always be- hind posterior limit of orbit. Lower margin with 2—3 teeth (commonly 2). Carapace smooth. Scaphocerite 3 times longer than wide. Abdomen smooth, posteroventral angle of fifth pleuron not acute. Sixth abdominal segment 1.25 times length of fifth, and 0.6 times length of telson. Telson, with 2 pairs of dorsal spines, situated at 7%, and % of its length from base. Posterior margin with acute apex and bearing two pairs of lateral spines, inner pair overreaching median apex, and 18-20 plumose setae between the inner spines. First pereiopods slender, overreaching scaphocerite by 3 of distal carpus. Palm cylindrical, about 1.1 times length of dactyl. Carpus 3.2 times length of palm and 1.2 times length of merus. Second pair of pereiopods subequal in shape and length, major leg with distal part of merus reaching anterior border of scaphocerite. Fingers short and strong, gaping when closed; both fingers with conspicuous teeth. Dactyl with a strong tooth at distal third and row of 3—4 small teeth on proximal third. Fixed finger with strong tooth at midlength and row of 3—4 small teeth behind. Fingers without tubercles. Palm cylindrical, 3.5 times longer than high, about 1.8 times length of dactyl, VOLUME 98, NUMBER 3 617 Fig. 2. Macrobrachium quelchi: A, Second pereiopod (5 x); B, Second pereiopod, detail (10 x). with numerous longitudinal rows of short spines, those on lower surfaces longer. Carpus 0.75 times length of palm and 1.2 times length of merus, spinulation pattern as in palm. Ischium with only few ventral spines. Third pair of pereiopods with dactyl reaching distal border of scaphocerite. Propodus about 3.2 times length of dactyl and 1.75 times length of carpus, with longitudinal row of 15-17 spines on inner margin. Fifth pair of pereiopods with dactyl reaching border of scapho- cerite. Propodus about 4.4 times length of dactyl and 1.7 times length of carpus, with longitudinal row of 15-17 spines on inner margin. Size. —The largest male measures 52.7 mm tl and 15 mm cl. Females smaller than male. Fecundity. —One ovigerous female, 50 mm tl and 13 mm cl with 15 oval eggs (2.5 mm largest diameter). Color. — Adult males with a background pale green color, with non-uniform dark green spots scattered around the body in no definite pattern. Pereiopods 3- 5 pale green, with 6-7 dark green horizontal stripes, 1 on ischium, 2—3 on merus Fig. 3. Macrobrachium quelchi: lateral view (1.5 x). 618 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON a ie . ae y Fig. 4. Macrobrachium quelchi: dorsal view (1.5 x). and 3 on carpus. Second pair of legs, generally dark green (almost black), with 2 narrow stripes proximally and distally on merus. Some specimens with a clear green area on dorsal surface of propodus. Remarks. —Two syntypes of M. quelchi, on loan from the British Museum, were studied. There are no major differences between these specimens from Guy- ana and those found in Venezuela. The palm of the second leg is more inflated in the syntypes, but these animals are larger, more robust and probably older than those from Venezuela. VOLUME 98, NUMBER 3 619 CLE) d Udide, J Fig. 5. Euryrhynchus pemoni, male paratype: A, Anterior part, dorsal view; B, Antennule; C, Antenna; D, Telson and Uropods; E, Appendix masculina; F, First pleopod; G, Second pereiopod; H, First pereiopod; I, Third pereiopod; J, Fifth pereiopod. Subfamily Euryrhynchinae Holthuis, 1951 Euryrhynchus pemoni, new species Fig. 5 Material examined. — Holotype ¢ tl 15.5 mm; cl 5.3 mm. USNM 216240, Creek in La Gran Sabana road, between Sn. Rafael Town and Kama Fall, Venezuela (6°15'N, 60°1’W), 27 Mar 1983, coll. Guido Pereira and Alfredo Paolillo. Para- types: 3 2 USNM 216241; 1 6,2 2 USNM 216242; 1 6 USNM 216243, same data as holotype. 2 6, 42 MBUCV (XI-1980). Creek near Chirimata indigenous town. La Gran Sabana, Venezuela (6°15'N, 60°1'W), 27 Mar 1983, coll. Guido Pereira and Alfredo Paolillo. Description. —Rostrum depressed, triangular, ending acutely, failing to reach anterior border of eyes. Antennal spines conspicuous, placed slightly above lower orbital angle, tip reaching beyond posterior margin of cornea. Pterygostomian angle forward produced and sharp, failing to reach beyond tip of rostrum. Ab- domen smooth, pleura of first 5 somites rounded. Sixth somite 1.6 times as long as fifth, 0.75 times length of telson. Telson broad, with 2 pairs of dorsal spines; anterior pair lying distally on first 3 and posterior pair on beginning of distal '. Posterior pair longer than anterior pair, placed farther from lateral margin of telson. Posterior margin wide, rounded, with 2 pairs of lateral spines, inner larger than outer; 18—22 plumose setae around margin. Eyes with distinctly pigmented cornea. Antennule peduncle, trisegmented, anterolateral angle of first segment produced into sharp pointed process reaching distal part of second segment. Latter 620 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON shorter than third segment, with anterolateral angles produced. Third segment shortest, with 3 antennular flagella, 2 long and slender, with numerous segments; third flagellum broader, short, with 4 similar segments; distal segment with nu- merous setae on apex. Scaphocerite slightly longer than antennular peduncle, about ¥; longer than wide. Outer margin straight, with anterolateral spine slightly over- reaching lamella. Mouth parts typical of genus (cf. Holthuis 1966). First pereiopod, slender, reaching with 74 of carpus beyond scaphocerite, smooth, with only scattered hairs. Fingers 0.9 length of palm, without teeth on cutting edges. Carpus 2.5 times length of palm, about same length as merus. Second legs equal or subequal, reaching with '3 of carpus beyond scaphocerite, smooth, with only scattered hairs. Fingers 2 length of palm; cutting edges smooth except for small tooth basally on both; dactylar tooth less prominent and situated distal to that on fixed finger. Carpus 0.75 length of palm. Merus 1.2 times length of carpus, with prominent mesial spine situated anteromesially on lower surface. Third leg reaching with distal propodus to border of scaphocerite; dactyl bifid, bearing 2 distinct spines on anterior third of outer margin. Propodus 3.1 times length of dactyl, bearing longitudinal row of 9-11 spines on posterior margin. Carpus twice length of dactyl and 0.5 length of propodus. Merus same length as propodus. Fifth pereiopod reaching with dactyl to border of scaphocerite. Dactyl bifid, propodus 3.2 times length of dactyl, with longitudinal row of 9 spines on posterior margin. Carpus, 0.5 times length of propodus; merus same length as propodus. Female: Similar to male but second legs not so strong. Size. —Largest male 15.5 mm tl and 5.3 mm cl; largest female 18.5 mm tl and 6.9 mm cl. Fecundity. —One ovigerous female, 16.6 mm tl and 5.5 mm cl had 11 eggs (oval in shape) with longest diameter 1.7 mm. Another female 15.2 mm tl and 5.3 mm cl has the pleura of the second abdominal segment especially enlarged, forming a brood pouch (sensu Holthuis 1966) containing only one egg with a well developed larva. This egg measured 1.9 mm largest diameter. Etymology. —The name “‘pemoni” is derived from the word “‘Pemon,”’ a general name for the indigenous people of this region. Remarks. —The present species is closely related to E. burchelli Calman, and E. wrzeniowskii Miers. Euryrhynchus pemoni can be differentiated by the second legs which have no spines on the carpus and only one on the merus, while there are three spines in E. burchelli, one on the carpus and two on the merus, and no spines in E. wrzeniowskii (sensu Tiefenbacher 1978). Recently, Kensley and Walk- er (1982) compared gonopod morphology of the species of Euryrhynchus. The gonopod of E. pemoni resembles that of EF. burchelli in that the lateral row of longitudinal spines does not reach the apex, but the present species has 3 rows of longitudinal spines compared to 2 in E. burchelli. Acknowledgments I wish thank Lic. Alfredo Paolillo, for his generous assistance during the field work; Dr. Brian Kensley, Dr. Marjorie Reaka, Dr. Thomas E. Bowman and Loren Coen, for critical review of the manuscript; and Dr. Marjorie Reaka and Dr. Raymond Manning for providing me with working space. This paper was partially financed by a CONICIT grant, S1-1259. VOLUME 98, NUMBER 3 621 Literature Cited Chace, F. A., Jr., and H. H. Hobbs, Jr. 1969. The freshwater and terrestrial decapod crustaceans of the West Indies with special reference to Dominica.— U.S. National Museum Bulletin 292:1- 258. Holthuis, L. 1952. A general revision of the Palaemonidae (Crustacea, Decapoda, Natantia) of the Americas. II The subfamily Palaemoninae. — Allan Hancock Foundation Occasional Papers 12: 1-396. 1966. A collection of freshwater prawns (Crustacea, Decapoda, Palaeomonidae) from Ama- zonia, Brazil collected by Dr. G. Marlier.— Bulletin de l’Institute des Sciences Naturelles de Belgique 42:1-11. Hulbert, B., G. Rodriguez, and N. Dos Santos. 1981. Aquatic biota of tropical South America. Part 1. Arthropoda. San Diego University Press, 323 pp. Kensley, B., and I. Walker. 1982. Palaemonid shrimps from the Amazon basin, Brazil (Crustacea, Decapoda, Natantia).—Smithsonian Contributions to Zoology 362:i11+ 28 pp. Pereira, G. 1982. Los camarones del genero Macrobrachium (Decapoda, Palaemonidae) de Vene- zuela. Taxonomia y distribucion.—Trabajo de Ascenso. Universidad Central de Venezuela, Facultad de Ciencias, 227 pp. . (In press). Camarones de agua dulce de Venezuela II: Nuevas adiciones en las familias Atydae y Palaemonidae (Crustacea, Decapoda).— Acta Biologica Venezuelica. Rodriguez, G. 1982. Freshwater shrimps (Crustacea, Decapoda, Natantia) of the Orinoco River basin and the Venezuelan Guayana.— Journal of Crustacean Biology 2:378-391. Tiefenbacher, L. 1978. Zur Systematik und Verbreitung der Euryrhynchinae (Decapoda, Natantia, Palaemonidae).—Crustaceana 35:177-189. Universidad Central de Venezuela, Instituto de Zoologia Tropical, Aptdo. 47058, Caracas 1041-A, Venezuela. Present address: Department of Zoology, University of Maryland, College Park, Maryland 20742, U.S.A. PROC. BIOL. SOC. WASH. 98(3), 1985, p. 622 REVIEW OF THE FORAMINIFERAL GENUS ORBIGNYNELLA SAIDOVA, 1975 Drew Haman Abstract. — Orbignyella Saidova, 1975, is rejected as a junior objective synonym of Neoglabratella Seiglie and Bermudez, 1965. Saidova (1975) established the genus Orbignynella to accommodate Recent Pacific Ocean discorbid specimens. She selected Discorbis wiesneri Parr, 1950, as the type-species for this genus. Orbignynella was retained as a valid genus and included in a systematic review of Cainozoic benthic Foraminifera (Saidova 1981). Discorbis wiesneri Parr, 1950, had, however, been selected earlier by Seiglie and Bermudez (1965) as the type-species for their new genus Neoglabratella. The species Discorbis wiesneri Parr, 1950, should be retained in the genus Neoglabratella Seiglie and Bermudez, 1965, and the genus Orbignynella Saidova, 1975, rejected as a junior objective synonym of Neoglabratella Seiglie and Ber- mudez, 1965 (ICZN Art. 61b). Acknowledgments Chevron Oil Field Research Company is acknowledged for publication per- mission. Literature Cited International Commission on Zoological Nomenclature. 1964. International Code of Zoological Nomenclature, adopted by the XV International Congress of Zoology, 2nd ed. International Trust for Zoological Nomenclature, London, 176 pp. Saidova, K. M. 1975. Bentosnye foraminifery Tikhogo Okeana.—P. P. Shirshov Institute of Ocean- ology, Academy of Sciences of the U.S.S.R., Moscow, 875 pp. . 1981. Osovremennom sostoyanii sistemy nadvidovykh taksonov Kainozskikh bentosnykh foraminifer.—P. P. Shirshov Institute of Oceanology, Academy of Sciences of the U.S.S.R.., Moscow, 73 pp. Seiglie, G. A., and P. J. Bermudez. 1965. Monografia de la familia de foraminiferos Glabratellidae. — Geos (Caracas) 12:15-93. Chevron Oil Field Research Company, P.O. Box 446, La Habra, California 90631. PROC. BIOL. SOC. WASH. 98(3), 1985, pp. 623-626 STUDIES IN NEOTROPICAL SENECIONEAE IV. NEW TAXA IN SENECIO AND CABRERIELLA Jose Cuatrecasas Abstract. —Two new species of Senecio (Compositae) from Peru are described. Senecio dolichodoryius from Chachapoyas belongs to section Dendrophorbium, and its relationship with the other members of the section are discussed; S. kingsbishopii is closely related and differentiated from S. flacciifolius Wedd. Also, a transfer of S. oppositicordius to Cabreriella is made. Senecio dolichodoryius Cuatr. sp. nov. Frutex ad 3 m alta caule ramisque viridibus medullosis. Rami terminales valde foliosi erecti vel ascendentes subteretes moderate striati intus medullosi extus pilis minutis patulis vel antrorsis plus minusve copiosis muniti et cum velo tenuiter membranaceo hyalino fragile ad modum squamarum partiale deciduo. Folia al- terna crasse coriacea petiolata. Laminae longi-lanceolatae basi paulo angustatae subacutae vel subobtusae, vel praecipue distales acute cuneatae, ad apicem longi- attenuatae acutissimae, (16-)19-38 x (2.6—-)2.9-7.6 cm, ratio (4.2)5—6.5(—7.3) : 1; margine argute dentatae dentibus triangularibus subacutis 0.7—1 mm longis cum minoribus 0.3—0.5 mm longis alternantibus, 2—4 inter se distantibus, apice calloso- incrassato submucronato (glanduloso); adaxiale virides tactu laeves, minutissime scrobiculatae, costa subplana bene notata nervis secundariis fere obsoletis, su- perficie glabra sed velo hyalino fragile irregulariter secedenti, ad costam sparsis pilis minutis glandulosis; abaxiale itidem squamoso-velatae visu argute discolore nervatae, costa valde prominenti striata laevi glabra sed hyalino-squamosa, nervis secundariis patulo-ascendentibus angustis sed prominentibus prope marginem curvatis tenuioribusque anastomosantibus, nervis alternantibus intermediis bre- vioribus, venulis minoribus in reticulum valde conspicum prominulum minute polygonalem anastomosatis; nervatio omnis brunnescens pseudo-pilosula cum squamis piliformibus irregularibus plus minusve tecta; alveolae pallidae planae laeves cereae. Petiolus 2—4.7 cm longus robustus sursum subplanus deorsum canaliculatus, abaxiale striatus, basi triangulare ampliatus et amplectens, ubique velato-squamosus. Synflorescentiae monotelicae terminales corymboso-paniculatae, 15—30 cm lon- gae, foliosae floribundae folia suprema paulo excedentes. Rami alterni robus- tiusculi striati proximales saepe valde longi ascendenti-fastigiati ceteri erecto- patenti omnes hyalini-squamosi et plus minus sparseque minute crassi-pilosull. Ramuli numerosi ramificati striolati minute puberuli pilis subpatulis vel ascen- dentibus, ultimi teneri sed erecti. Pedicelli graciles longiusculi 5-30 mm longi recti vel paulo flexuosi. Folia subtendentia proximalia quam folia sterilia simi- lissima, sursum gradatim breviora; bracteae in ramis secundariis inferne foliaceae sursum bracteosae, lanceolatae in ramulis ultimis lanceolato-lineares acutissimae minute puberulae 5—1.5 mm longae. Capitula cylindracea 9-11 mm longa radiata circulo ligulari expanso 16-20 mm diam. Involucrum cylindricum vel cylindro-campanulatum 6.5-7.5 mm altum 5-6 mm diametro, explanatum 14-16 mm diam. Phyllaria 13-12 in capitulo; 624 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON exteriora 6—7 mm longa 1.2—1.5(—2) mm lata, oblonga sursum attenuata acuminata apice acutissimo dorso crasso margine anguste membranaceo ad apicem papilloso- pilosulo papillis longiusculis saepe pyriformibus; interiora 5.5—6 mm longa 2—4 mm lata elliptica apice angustata acutaque dorso valde protrudo crasso marginibus late hyalino-membranaceis, omnia apice papilloso-pilosulo excepto glabriuscula. Bracteolae calyculares 4-6 herbaceae lineares acutae parce puberulae pilis minutis crassis apice papillosae, 1-4 mm longae, ad basim involucri et apicem pedicelli affixae. Receptaculum planum glabrum 3(—4) mm diam. alveolatum alveolis pen- tagonalibus moderate profundis marginibus ad angulos elevato-dentatis. Flores radii ligulati 11-13. Corolla lutea 10-12 mm longa glabra, tubo 3.5—4.5 mm longo angusto ad faucem annulare incrassato, lamina |.8—2.5 mm lata elliptica vel oblongo-elliptica basim attenuata apice obtusa minute 2—3-denticulata, con- spicue 4-nervata nervis croceis notatis, adaxiale superficie mamillata. Stylus ramis circa 1.5 mm longis, faucem corollae paulo excedentibus lineis stigmaticis crassis conniventibus apice obtuso cum papillis crassioribus. Flores disci hermaphroditi 20-36. Corolla lutea tubulosa glabra 6.5—7.5 mm longa, tubulo 3-3.5 mm, limbo tubuloso distale leviter ampliato, commisurale nervato, lobis oblongo-triangularibus mediale cum vena conspicua 1-1.2 mm longis apice incrassato papillosoque papillis crassis obtusis seu pryiformibus. An- therae 2.2-2.4 mm longae appendice apicale oblonga, lobis basi anguste brevi- terque auriculatis, cellulis endothecialibus oblongis in parietibus lateralibus seriate noduliferis. Collum 0.4—0.5 mm longum cum filamento valde crassius plus mi- nusve lageniforme dilatatum cellulis subquadratis multiseriatis. Stylus 7-7.5 mm longus ramis circa 1.5 mm recurvis marginibus lineis stigmatibus duobus valde crassis adaxiale conniventibus, apice obtuso cum papillis crassulis paulo dilatato. Ovaria circa 1.5 mm longa oblongo-ellipsoidea glabra conspicue 5-costata costis ductis resiniferis lutescentibus. Pappus albus 4.5—5 mm longus pilis strictis sca- bridis acutis 2—3-seriatis basi connatis. Type. —Peru, Depto. Amazonas, Pcia. Chachapoyas: middle eastern Calla-Calla slopes, near km 411-416 of Leimabamba-Balsas road, 3100-3250 m, herb 1-3 m tall, locally frequent; rays and disc yellow; 11 Jul 1962, J. J. Wurdack 1306 (US, holotype). Other collections.— Peru, Depto. Amazonas, Pcia. Chachapoyas: Cerros Calla- Calla, 18 km above Leimabamba on road to Balsas, 3000 m, open forest below road ca. 1 km downhill from km 410, swampy, single stems 1'2—2 m, rays golden yellow, 16 Jun 1964, Hutchison & Wright 5684 (UC, paratype); 11 kms along road ascending mountain SE of Chachapoyas, 8700 ft., large shrub 3 m tall with rank odor, 20 Jan 1983, Robert M. King & L. Earl Bishop 9261 (US, paratype); 58 km NE of Balsas on road to Chachapoyas, ca. 3400 m, shrub to 2 m, flowers yellow, 5 Jan 1979, M. Dillon & B. L. Turner 1734 (US). Remarks. —Senecio dolichodoryius belongs to section Dendrophorbium and is characterized by the combination of its leathery large, long-lanceolate, petiolate, minutely bidenticulate leaves which are adaxially smooth, by the radiate capitula of 13-12 medium-sized (6-7 x 1.8—4 mm) phyllaries, and by the similar size of disk corollas (7-7.5 mm). The branchlets and terminal inflorescences are very shortly puberulous with the cuticular velum or film covering almost every veg- etative part at least at the earlier stage. This fragile film becomes fragmented and VOLUME 98, NUMBER 3 625 partially caducous and disintegrates, producing a kind of irregular, spreading pseudo-pubescence on the abaxial surface of the leaves. The other species of the section can be easily distinguished from the new species as follows. The Bolivian species S. tabacifolius Rusby and S. cabrerae Cuatr. (including S. myrianthus Klatt, a discarded later homonym), have involucres with eight phyllaries and the first has, in addition, tuberculate, scabrous leaves. The Peruvian S. castanaefolius DC, and S. submultinervis Cuatr. and the Ecuadorian S. ingens Benoist have the involucre with 9-10 smaller phyllaries (3.5—5.5 mm). Two other Bolivian species, B. conocephalus Cabrera and S. longilinguae Cuatr., can also readily be separated by the 8-merous involucre, in addition to their sessile leaves. Among the majority of species with 13 (12-14) phyllaries there are the Bolivian S. biserrifolius Ktze. and S. ayapoyensis Cuatr. and the Peruvian S. sandemanii Cuatr. and S. tergopurpureus Cuatr. which have sessile leaves. The Colombian S. moscopanus Cuatr., S. sibundoyensis Cuatr. and suazaensis Cuatr. have leaves with broadly auriculate bases. Senecio tipocochensis Domke and S. balsapampae Cuatr. from Ecuador, S. fortunatus Cuatr. from Peru and S. arboluco Cuatr. from Colombia have involucres only 2.5—3.5 mm long. Conversely, Senecio lloensis of Ecuador is characterized by larger capitula with phyllaries 10-9 mm long. Of S. chingualensis Cuatr. and S. dielsii Domke, with 6—7(—8) mm long phyllaries, the first has rather thinner, broader and pubescent leaves and broad phyllaries, while the second possesses thicker, stronger branchlets and leaves with a scabrous upper surface, thicker capitula, harder phyllaries and overall white lanugineous, arachnoid indument. The remaining species S. pururu Cuatr. (Ec- uador), S. silvani Cuatr. (Col.), S. goodspeedii Cuatr. (Peru), S. multinervis Sch. Bip., S. unduavianus Cuatr. (Bolivia), and S. yalusay Cabrera (Peru) have rather smaller heads with phyllaries 4-5 mm long, and, besides some other par- ticular features, all have a more or less abundant lanate or arachnoid indument of thin, long, entangled trichomes which is completely lacking in S. dolichodoryius. Few other species should probably join this section Dendrophorbium, and further study is necessary to improve the knowledge of its above-mentioned taxa. Many of those species are still only partially known through fragmentary specimens lacking complete mature vegetative leaves and branches. Much research is also needed on floral characters in order to help to define the limits of the sections and the closely related genera, like Pentacalia, into which the section Dendrophor- bium might well be merged. Senecio kingbishopii Cuatr. sp. nov. Suffrutex ad 1.5 m altus profuse ramosus et foliatus aspectu cinerascens. Rami ramulique alterni copiosi argute angulati vel striati densiuscule subadpresseque lanugineo-arachnoidei cinerei. Folia alterna cum lamina sessili crassiuscula lan- ceolato-oblonga apice acuta basi plus minusve ampliata auriculato-amplectenti, margine repando denticulata dentibus minutis mucroniformibus patulisque plana vel interdum revoluta; 40-60 x 4-13 mm; adaxiale tenuiter lanugineo-arach- noidea tantum costa magis lanata notata; abaxiale crasse densiuscule subadpres- seque lanata pilis tenuissimis intricatis, costa infra indumentum prominenti striata sed leviter conspicua cum vestimento tecta, reliquis nervis obsoletis. 626 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Inflorescentiae terminales corymbiforme paniculatae valde ramosae inferne foliatae et bracteosae, ramis ramusculisque striatis lanuginoso-cinereis. Folia sub- tendentes vel bracteae proximales hastato-lanceolatae, 5—2.5 cm longae 12—8 mm latae. Ramuli ultimi et pedicelli (1-8 mm longi) rigidi striati paulo arachnoidei plus copiosis pilis pluricellaribus crassiusculis tortuoso-flexuosis brunnescentibus valde conspicuis. Bracteae et bracteolae distales breves lineares acutae. Capitula discoidea cylindracea 10-11 mm alta circa 6 mm diametro. Involu- crum circa 9 mm longum, expansum 1 7—19 mm diam.., viridi-grisaceum. Phyllaria 13—14(—-17) in capulto tenuiter herbaceo-membranacea exteriora 8—9 x 1.2-1.5 mm, interiora 8 x 2 mm ad marginem late scariosa, omnia oblonga cum apice subite acutato et papilloso-piloso, abaxiale pilis albis strictissimis lanugineo- arachnoideis et pilis crassiusculis brunnescentibus flexuoso-tortuosis circa 1.5 mm longis valde motatis. Bracteolae calyculi 4—6 lineares vel anguste triangulares 2.5— 5 mm longae etiam pubescentes. Receptaculum 2-—2.5 mm diametro planum alveolatum alveolis pentagonalibus moderate profundis marginibus ad angulos dentibus triangulari-acutis. Flores omnes hermaphroditi 27—45 in capitulo. Corolla lutea 6.5—7 mm longa glabra, tubulo circa 2.5 mm, limbo tubuloso sursum gradatim leviterque ampliato, lobis triangulare oblongis 0.8—1 mm longis apice marginibusque paulo incrassatis et minute papillosis. Antherae circa 2 mm basi attenuatae obtusaeque, appendice apicale oblonga 0.2—0.3 mm longa. Collum album crassum 0.5—0.7 mm longum cellulis quadratis longiseriatis. Ramuli styli curvati adaxiale lineis duobus stig- maticis marginalibus crassiusculis, apice obtuso minute papilloso cum papillis marginalibus longioribus patentibus radiatis. Ovarium circa 1.5 mm longum ob- longum plus minusve conspicue 10-nervatum. Papus albus 6 mm longus pilis uniseriatis strictis acutis strigulosis. Type. —Peru, Depto. Cajamarca: 62 km NE of Cajamarca along the road to Celendin, 10,000 ft. elevation; shrub '2 m tall, flowers yellow, 9 Jan 1983, Robert M. King & L. Earl Bishop 9144 (US, holotype). Remarks. —Senecio kingbishopii is closely related to S. flaccidifolius Wedd. from which it differs by the auriculate, amplectant leaves and by the lanate and arachnoid indument with additional thicker, brownish, flexuose, pluricellular tri- chomes which are conspicuous on branchlets and inflorescences. Senecio albotectus Cuatr. Feddes Repert. 55(2/3):130. 1953. This species had been erroneously transferred by the author to Pentacalia (Phy- tologia 49(3):252. 1981). It is to be kept in the genus Senecio and its section Latiflorus, as it was suggested in the original publication (1.c. pag. 131). Cabreriella oppositicordia (Cuatr.) Cuatr., comb. nov. Senecio oppositicordius Cuatr. Feddes Repert. 55(2/3):142-143. 1953. Type. —Grant 10862 (F, holotype; US, isotype). Department of Botany, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 98(3), 1985, pp. 627-629 A NEW ENTOCYTHERID OSTRACOD OF THE GENUS DACTYLOCYTHERE Arnold W. Norden and Beth B. Norden Abstract. —A new species of entocytherid ostracod, Dactylocythere scotos, is described from Pennsylvania and Maryland. Comparisons are made with its most closely related cogeners, D. crawfordi and D. phoxa. While surveying the entocytherid ostracods of Maryland, an undescribed species belonging to the genus Dactylocythere was found at a single locality in the moun- tainous western portion of the state. Subsequently, Dr. Horton H. Hobbs, Jr., obtained additional specimens of this ostracod from crayfish collected by A. E. Ortmann in Beaver County, Pennsylvania, in 1905. We thank Dr. Hobbs for allowing us to include the Pennsylvania material in this description, for critically reading this manuscript, and assisting us in other ways too numerous to mention. Dactylocythere scotos, new species Fig. 1 Description. —Male: Eye pigmented, situated about one-fifth shell length from anterior end. Shell (Fig. 1d) ovate but ventral margin slightly concave, greatest height distinctly posterior to midlength, posteroventral prominence lacking. Mar- ginal setae sporadic around entire shell, most abundant along posterodorsal mar- gin. Sternal spine not discernible. Copulatory complex (Fig. 1a) with peniferum tapering to acute apex directed anteroventrally; peniferal groove generally narrow but widening slightly toward tip. Penis situated approximately at base of ventral one-third of peniferum, distal to base of clasping apparatus. Finger guard relatively thin, almost straight, and with rounded distal extremity entire. Clasping apparatus (Figs. la, b) L-shaped with vertical ramus thickened and with massive, angular shoulder on postaxial border; preaxial border of horizontal ramus with four evenly spaced teeth, fourth tooth extending as ridge obliquely across ramus toward postaxial border; apex with two denticles. Accessory groove simple, extending dorsally to level of dorsal margin of spermatic loop. Female: Eye pigmented, situated about one-fourth shell length from anterior end. Shell (Fig. le) ovate, with greatest height distinctly posterior to midlength. Submarginal setae disposed as in male. Genital apparatus (Fig. 1c) with prominent, strongly curved J-shaped rod and flared amiculum supported by U-shaped thickenings. Size.—The lengths (in um) of ten males range from 457 to 492, average 471; the heights range from 265 to 293, average 279; corresponding measurements of ten females are 473 to 522, average 487, and 268 to 299, average 292. Type-locality. —Baden, Beaver County, Pennsylvania; from Cambarus (Lacuni- cambarus) diogenes Girard, 1852, coll. A. E. Ortmann, 26 Jul 1905. Disposition of types. —The holotypic male and allotypic female are deposited 628 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Dactylocythere scotos: a, Copulatory complex of male; b, Clasping apparatus of male; c, Genitalia of female; d, Shell of male; e, Shell of female. (Scales in mm.) in the National Museum of Natural History, Smithsonian Institution, USNM 210690. Paratypes are in the Smithsonian Instutition, and the collection of the authors. Range. —In addition to the type-locality, Dactylocythere scotos has also been found on Cambarus diogenes taken from burrows around Deep Creek Lake at McHenry, Garrett County, Maryland, by A. Norden and B. Norden on 16 June 1976. Host. —The host at both known localities was Cambarus diogenes. Also present at the Maryland locality were Cambarus (Cambarus) bartonii bartonii (Fabricius, 1798), Procambarus (Ortmannicus) acutus acutus (Girard, 1852), and Orconectes obscurus (Hagen, 1870). Although large series of each of these three species were examined, no D. scotos were recovered. Entocytherid associates. —One Donnaldsoncythere donnaldsonensis was re- covered from collections containing D. scotos. Relationships. — Dactylocythere scotos appears to have its closest affinities with D. crawfordi Hart, 1965, and D. phoxa (Hobbs, 1967). The most striking similarity is in the clasping apparatus of the male in which the horizontal ramus is much VOLUME 98, NUMBER 3 629 longer than the short, heavy vertical ramus which bears a conspicuous shoulder on the postaxial margin. Similar also is the simple finger guard and the length of the accessory groove, which reaches the dorsal extremity of the spermatic loop. It and D. phoxa differ from D. crawfordi in possessing a heavier vertical ramus of the clasping apparatus and a subtriangular ventral part of the peniferum. It differs from D. phoxa in that the proximal tooth on the preaxial margin of the horizontal ramus is not digitiform, rather little different in size and shape from the second tooth, and the peniferal groove opens anteriorly rather than apically. Etymology. —From the Greek, scotos, meaning darkness, in reference to the apparent restriction of this species to the burrowing crayfish, C. diogenes. Literature Cited Hart, C. W., Jr. 1965. New entocytherid ostracods and distribution records for five midwestern states.— Transactions of the American Microscopical Society 84(2):255-259. Hobbs, Horton H., Jr. 1967. A new genus and three new species of ostracods with a key to the Genus Dactylocythere (Ostracoda: Entocytheridae).— Proceedings of the United States National Mu- seum 122(3587):1-10. (AWN) Maryland Natural Heritage Program, C-3, Tawes State Office Building, Annapolis, Maryland 21401; (BMN) Department of Entomology, University of Maryland, College Park, Maryland 20742. PROC. BIOL. SOC. WASH. 98(3), 1985, pp. 630-635 A NEW MARINE GENUS OF THE MAERA GROUP (CRUSTACEA: AMPHIPODA) FROM BELIZE James Darwin Thomas and J. L. Barnard Abstract. —A new genus and species, Dumosus atari of the Maera group from coral rubble at Belize is described. It is related to Gammarella and is probably cryptic. The eyes are reduced. Dumosus differs from Gammarella in the loss of medial setation on the maxillae, reduction of article 3 on the mandibular palp, loss of article 2 on the outer ramus of uropod 3, and in the loss of major spination on the plates of maxilla | and the maxilliped; in contrast to Gammarella, Dumosus retains the plesiomorphic uropod 3 typical of the Maera-group. Legend for Figures Capital letters in illustrations are explained in the following list; lower case letters to the right of capital letters or in the body of an illustration are explained also in the following list; lower case letters to the left of capital letters are provided for subsidiary figures to note illustrated specimens listed in “‘Material.”’ For each page of figures one main specimen is called “unattributed” and lacks letter des- ignation. B, body; C, coxa; D, dactyl; G, gnathopod; I, inner plate; L, labium; M, mandible; O, outer plate or ramus; P, pereopod; R, uropod; S, maxilliped; T, telson; U, labium; V, palp; W, pleon; X, maxilla; Y, gill; 0, opposite; r, right. Dumosus, new genus Diagnosis. —Head and antennae of Elasmopus-form, thus rostrum small, an- teroventral antennal sinus small but notch absent, antenna | of medium length, article 2 nearly as long as 1, article 3 much shorter, accessory flagellum 3-articulate; antenna 2 very short, slightly exceeding article 2 of antenna 1, flagellum very short, not longer than article 5 of peduncle, 4-articulate. Prebuccal mass weakly extended anteriorly, upper lip rounded below. Mandib- ular incisors toothed, laciniae mobiles and rakers present on both sides, molar triturative, with large seta on right, small seta on left, palp article 1 scarcely elongate, article 2 naked, article 3 about 70 percent as long as 2, slender, linear, with 1 D-seta, 2 E-setae. Lower lip with fleshy inner lobes and well developed mandibular lobes. Inner plate of maxilla 1 with 1 apical, 1 subapical medial seta, outer plate with 7 spines, palp article 1 elongate, armament of right and left palps asymmetrical. Plates of maxilla 2 slender, inner with one medial seta. Inner plate of maxilliped lacking thick spines, bearing only setae, outer plate with medial and apical spines, palp poorly setose, article 3 with apical hook, dactyl stubby, with medium nail. Coxae of ordinary length, poorly setose, coxa 1 quadrate, unproduced anteriorly, coxae 1-2 with posteroventral tooth-notch, coxa 3 slightly the narrowest of coxae 1-4, posterodorsal excavation shallow, coxa 5 much shorter than 4, lobes of coxae 5-6 shallow. Simple gills on coxae 2—6; female unknown. VOLUME 98, NUMBER 3 631 Gnathopod 1 small, of melitid form, carpus and propodus subequally long, carpus not lobed, propodus with stiff posterior setal-spines, palm oblique, weakly sculptured. Male gnathopod 2 greatly enlarged, metacarpus with posterodistal sharp tooth (thus almost metacarpochelate), carpus short, strongly lobate, hand large, pyriform, palm and hind margin continuous, palm undefined, armed with few spines and many long apically curved bulbar setae, dactyl much shorter than false palm. Pereopods 3-4 ordinary, slender, pair of locking spines asymmetrical, dactyl with 2 main inner setules near base of nail, no outer tooth. Pereopods 5-7 of short form but increasingly elongate from 5 to 7, of reverted form, article 2 on pereopods 5-6 of narrow pyriform shape with weak posterior sinuosity, postero- ventrally lobate, posterior serrations weak to moderate respectively, remaining articles slender; pereopod 7 with shield-like article 2 bearing medium castellations posteriorly. Pleopods well developed, peduncle long, rami equally long. Pleon unarmed dorsally. Epimera diverse, third dominant, epimeron 1 with subventral ridge, naked ventrally, with small posteroventral tooth, epimeron 2 with lateral ridge, one facial spine, large posteroventral tooth; epimeron 3 nearly straight behind, with medium tooth and posteroventral serrations below main tooth, 1 facial spine. Uropod | with strong basofacial spine, strong apicolateral spine; uropods 1-2 with long apical spines, outer ramus of uropod 1 lacking marginal spines, other rami with few marginal spines. Uropod 3 small, aequiramous, not exceeding uropods 1-2, peduncle short, rami lanceolate, sharp, inner only with tiny apical armament, outer similar apically also with 2 sets of lateral armaments. Telson short but longer than broad, deeply cleft, each apex bifid, with one spine and setule, each lobe with subapical lateral setule set. Type-species. —Dumosus atari, new species. Etymology. —Dumosus from “covered with thorn bushes”’ referring to gnatho- pod 2, and atari, a Carib Indian word for star. Relationship. —The present genus differs from Maera in the Elasmopus-like male gnathopod 2 lacking defined palm. As far as we know, no species of Maera has a shield-like article 2 on pereopod 7, nor the degree of diversity of pereopods seen in this species. This genus differs from Elasmopus in the linear article 3 of the mandibular palp, with lanceolate and poorly armed rami of uropod 3. It bears close resem- blance to Lupimaera Barnard and Karaman (1982) (based on Maera lupana J. L. Banard, 1969) but the rami of uropod 3 are slender and poorly armed, coxa 5 is short, and article 2 of pereopods 5-7 is diverse, unlike Lupimaera. Unlike Meximaera Barnard (1969), the new genus has enlarged and non-female- like gnathopod 2, poorly armed rami of uropod 3, poorly armed mandibular palp, and diverse pereopods 5-7. A superficial resemblance occurs between this genus and the Caribbean cave genus Paraweckelia Shoemaker (1959). Although Paraweckelia appears less spe- cialized in the presence of more medial maxillary setae, non-diverse pereopods 5-7, long and spiny rami of uropod 3, and longer antennae, Paraweckelia is more specialized in the apomorphic telson. Because of the poorly developed rami of uropod 3 and diverse pereopods 5-7, Dumosus cannot be ancestral to hadziids and weckeliids. 632 Fig. 1. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Dumosus atari, new species, holotype, male “x.” VOLUME 98, NUMBER 3 633 C3 C1 C4 Fig. 2. Dumosus atari, new species, holotype, male “‘x.” The shield-like pereopod 7 is comparable to Gammarella Bate (see Chevreux & Fage 1925 as Pherusa [=Nuuanu, =Cottesloe| and Tabatzius McKinney and Barnard (1977), from the Caribbean Sea, but those genera differ from our new genus in the fully setose medial margins of the maxillae and the non-falcate article 3 of the mandibular palp lacking most D-setae. It is possible to build a plesio- morphic-apomorphic sequence of evolution from the base stock of Gammarella to Dumosus through loss of medial maxillary setae, reduction in mandibular palp article 3, loss of article 2 on the outer ramus of uropod 3, loss of major spines on inner plate of maxilliped, loss of several spines on the outer plate of maxilla 1 and retention of the primitive elongate inner ramus of uropod 3. The two genera (and the similar Tabatzius) share similar gnathopods, telson, pereopods, and many similarities in body form, head, antennae, mandibles, maxillipeds and pleon. 634 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 3. Dumosus atari, new species, holotype, male ‘“‘x.” There is a superficial resemblance of Dumosus to Ceradocopsis (see Barnard and Karaman 1982), from the southern hemisphere, in the shortness of uropod 3, form of head, antennae, gnathopods, mandibles, maxilliped, uropods 1—2 and telson, but Ceradocospis has more medial maxillary setae, less shield-like pereopod 7 and bears an article 2 on the outer ramus of uropod 3. Dumosus atari, new species Figs. 1-3 Description of unique male, 1.96 mm.—Eyes composed of ochre-brown pig- ment mass with several clear, mostly anterior ommatidia extending outward, one detached gangliar ommatidium dorsally. Flagellum of antenna 1 shorter than articles 2—3 of peduncle combined. Article 4 of antenna 2 with weak apical hump; gland cone large, basal article small. VOLUME 98, NUMBER 3 635 Mandibular incisors with articulate accessory tooth on medialmost edge, right and left rakers 3, with interrakers. Inner plate of maxilliped with scythe-like ventral coupling hook. Article 4 of gnathopod | pubescent; dactyls of gnathopods with apical nail and inner setules. Pereopod 3 like illustration of pereopod 4 but slightly larger. Peduncle of pleopods elongate, usually with one lateral seta, 2 coupling hooks, rami extending equally, as long as peduncle, 5-articulate. Peduncle of uropod 2 with one apicomedial spine. Holotype. —USNM 195138, male, 1.96 mm. Unique specimen with left per- eopod 5 missing. Type-locality. —Carrie Bow Cay, Belize, 24 Jun 1982, formalin wash of coral rubble from back reef, 0.5—1.0 m, coll. J. D. Thomas. Remarks. —We have searched for this species again at Belize several times and in many other Caribbean localities to no avail; its clearly distinct generic position, and the excellent condition of the specimen makes it possible to describe. Acknowledgments We would like to thank the Scholarly Studies Program, administered by As- sistant Secretary for Science, Dr. David Challinor, Smithsonian Institution, for support during this study. The first author was also supported by NSF grant DEB8121128. Dr. Klaus Riitzler, Director of the Smithsonian Western Atlantic Mangrove Project (SWAMP) kindly provided field facilities in Belize. Mike Car- penter of the Smithsonian also provided the first author with invaluable assistance in the field. We would like to thank Linda B. Lutz, Mobile, Alabama, for inking the plates. Literature Cited Barnard, J. L. 1969. Gammaridean Amphipoda of the rocky intertidal of California: Monterey Bay to La Jolla.— United States National Museum Bulletin 258:1—230, figs. 1-65. , and G.S. Karaman. 1982. Classificatory revisions in Gammaridean Amphipods (Crustacea), part 2.— Proceedings of the Biological Society of Washington 95(2):167—187, fig. 1. Chevreux, L., and L. Fage. 1925. Amphipodes.—Faune de France 9:1—488, figs. 1-438. McKinney, L. D., and J. L. Barnard. 1977. A new marine genus and species of the Nuuanu-group (Crustacea, Amphipoda) from the Yucatan peninsula.— Proceedings of the Biological Society of Washington 90:161-171, 3 figs. Shoemaker, C. R. 1959. Three new cave amphipods from the West Indies.—Journal of the Wash- ington Academy of Sciences 49:273-283, figs. 1-4. (JDT) Newfound Harbor Marine Institute, Rt. 3, Box 170, Big Pine Key, Florida 33043; (JLB) Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 98(3), 1985, pp. 636-643 SCALPELLOID BARNACLES FROM THE UPPER CRETACEOUS OF SOUTHEASTERN NORTH CAROLINA Victor A. Zullo and Norman F. Sohl Abstract. —Zeugmatolepas sp. and Virgiscalpellum inornatum, new species, Oc- cur in the Upper Cretaceous (Campanian) Black Creek Formation at Donoho Creek Landing, Cape Fear River, North Carolina. These are the first barnacles to be described from the Cretaceous of the Atlantic Coastal Plain. Capitular plates of two scalpelloid barnacles were collected from a medium- to coarse-grained sand near the top of the Upper Cretaceous (Campanian) Black Creek Formation at Donoho Creek Landing on the Cape Fear River in Bladen County, North Carolina (Fig. 1). One series of small plates, tentatively identified as carinae, rostra and lower latera, is from an undetermined species of Zeug- matolepas Withers. The only previous North American record of Zeugmatolepas is that of Hattin (1982) from the Smoky Hill Chalk Member of the Niobrara Formation in western Kansas. Otherwise, Zeugmatolepas is known from Europe, the South Georgia Islands, western South America and western Australia. The larger and more abundant plates from Donoho Landing include carinae, scuta and terga of a new species of Virgiscalpellum Withers. This new species is similar to V. gabbi (Pilsbry) from the Ripley and Prairie Bluff Formations of Tennessee, Mississippi and Alabama, but is readily distinguished on ornamen- tation, location of umbones, and shape of plates. Virgiscalpellum is known only from Cretaceous rocks in western Europe and the Americas. Although Cretaceous barnacles have long been known from the Western Interior and Gulf Coast regions of North America (see Collins 1973; Hattin 1977), the Black Creek specimens constitute the first Cretaceous barnacle records for the Atlantic Coastal Plain. Stratigraphic Setting The barnacles were obtained during an investigation of the Black Creek-Peedee formational contact along the Cape Fear River (Sohl and Christopher 1983). The Black Creek Formation of Campanian age (Exogyra ponderosa Zone) is overlain disconformably by the Peedee Formation of Maestrichtian age (E. costata Zone). At Donoho Creek Landing (Fig. 2), units | through 4 represent the Black Creek Formation and contain E. ponderosa erraticostata (Stephenson), E. costata spi- nosa Stephenson and Flemingostrea pratti (Stephenson), indicating assignment to the uppermost part of the FE. ponderosa Zone. Barnacle plates are abundant in unit 2 (USGS localities 31796, 31868), and are associated with a nearshore, marine megafossil assemblage of epifaunal suspension feeders. Locality description. —U.S. Geological Survey (Mesozoic Invertebrate Collec- tion) localities 31796, 31868. Black Creek Formation, unit 2, Donoho Creek Landing, milepost 50.25 on west bank of Cape Fear River, Bladen County, North VOLUME 98, NUMBER 3 637 AS Donoho Creek Landing \ Council Fig. 1. Location of Donoho Creek Landing on the Cape Fear River, North Carolina. Carolina 34°28'26"N, 78°24'40”W. Medium- to coarse-grained, subrounded to subangular, poorly sorted, fossiliferous sand. N. F. Sohl, collector. Systematic Paleontology Subclass Cirripedia Burmeister, 1834 Order Thoracica Darwin, 1854 Suborder Lepadomorpha Pilsbry, 1916 Superfamily Scalpelloidea (Pilsbry) Zevina, 1980 Family Scalpellidae Pilsbry, 1916 Subfamily Calanticinae Zevina, 1978 Genus Zeugmatolepas Withers, 1913 Zuegmatolepas species indet. Figs. 3a—m Material. —Nine carinae, 15 rostra and 2 lower latera, USGS locs. 31796, 31868. Disposition of specimens.—Hypotypes USNM 382778 through 382783 and hypotype lots 382784 and 382785 are deposited in the Division of Paleobiology, U.S. National Museum of Natural History, Washington, D.C. Discussion. —A group of plates, ranging in size from 3.0 to 6.5 mm, appear to be from the lower whorls of a species of Zeugmatolepas. This genus is known to range from the Jurassic (Callovian) through the Paleocene (Danian), and is best known from numerous species in western Europe (Withers 1928, 1935, 1947). The Donoho Creek Landing specimens, although well-preserved, are difficult to relate to specific plate types and, in the absence of species-diagnostic terga and scuta, are insufficient for species recognition. The carinae (Figs. 3a—b, f—g) range in size from 4.0 to 5.5 mm in height, are thick, and extremely narrow (five times higher than wide). These plates bear some resemblance to carinae of the European Campanian species Z. cretae (Steenstrup) 638 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON DONOHO CREEK LANDING 31796, 31868—/ :~ tea Lithic symbols Medium- to coarse- Sand grained sandstone Medium-grained Gravel sand Medium- to coarse- grained sand r Clay Medium- to very coarse grained Burrows sand Fig. 2. Stratigraphic section of the column exposed at Donoho Creek Landing, showing barnacle localities in unit 2. Contact of Black Creek Formation with overlying Peedee Formation is at base of unit 5. Section modified from Sohl and Christopher (1983). VOLUME 98, NUMBER 3 639 Fig. 3. Zeugmatolepas species: a—b, Oblique interior and exterior views of carina, hypotype USNM 382778; c—d, Exterior and oblique interior views of rostrum, hypotype USNM 382779; e, Rostrum of Fig. 3c overexposed to show fine radial striae; fg, Oblique interior and exterior views of carina, hypotype USNM 382780; h-i, Exterior and interior views of rostrum, hypotype USNM 382781; j, Rostrum of Fig. 3h overexposed to show fine radial striae; k, Oblique side view of lower latus, hypotype USNM 382782; l—-m, Side and exterior views of lower latus, hypotype USNM 382783. All illustra- tions x7. figured by Withers (1935, pl. 3, fig. 15, pl. 4, fig. 1). The tectum is narrow and in lateral view is nearly straight except for the apical part which is curved inwardly. The plate in cross-section is acutely V-shaped, with no apparent differentiation between tectum and parietes. In tectal view the carina is widest at or below the midline. The apex is acute and the basal margin is narrowly rounded. The lateral edges of the plate are thickened, especially in the upper half, but the inner surface is open from base to apex. A few specimens bear faint, oblique growth lines on the inner surface near the apex. The rostra (Figs. 3c-e, h-j) are considerably thinner and broader than the carinae, and range in size between 3.0 and 6.0 mm in height. These rostra are remarkably similar to those of the European Cenomanian species Z. mockleri Withers (Withers 1935, pl. 2, figs. 9a—b). The rostra are sub-diamond-shaped, broadly V-shaped in cross-section, and bear a low, rounded, apico-basal ridge that broadens basally. In lateral view the plates are nearly straight to slightly convex. The apex is acute and the basal margin is truncate or narrowly rounded. The outer surface is ornamented by prominent growth ridges cut by fine, longi- tudinal striae on either side of the apico-basal ridge (Figs. 3e, j). The edges of the plate are slightly thickened, the inner surface is broadly concave, and the inner apical region bears oblique growth lines. Two plates (Figs. 3k—m) appear to represent lower latera, particularly the type 640 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON depicted for the European Lower Cretaceous species Z.(?) hausmanni (Koch and Dunker) by Withers (1935, pl. 3, figs. 5t-u). These plates are 4.0 to 5.0 mm in height, very narrow, are infilled in their upper halves, and are ornamented by prominent, imbricating growth ridges. It is likely that the Black Creek Zeugmatolepas is a new species, but in the absence of upper whorl plates, specific identification is not warranted. Subfamily Arcoscalpellinae Zevina, 1978 Genus Virgiscalpellum Withers, 1935 Virgiscalpellum inornatum, new species Figs. 4a—n Holotype. —Carina, U.S. National Museum no. 382786. Type-locality. —Upper Cretaceous, Black Creek Formation, Donoho Creek Landing, Bladen County, North Carolina, USGS loc. 31796. Diagnosis. —Carina narrow, angular, with prominent umbo removed from apex and located on upper two-fifths of plate; surface of plate without radial ornament. Umbo of scutum markedly subcentral, one-third distance from rostral angle; surface ornamented by prominent growth ridges crossed by fine striae radiating from umbo. Tergum narrow, slightly curved towards carinae, with low, rounded, apico-basal ridge. Description. —Carina (Figs. 4a—e) narrow in tectal view, thin in lateral view, obtusely angulate (130°-140°); largest specimen 8.0 mm in length; position of prominent umbo variable, but always below apex and never more than a distance equal to two-fifths length of plate; narrow, arched tectum not differentiated from broad, parallel-sided or moderately divergent parietes; parietes separated from moderately divergent intraparietes by obscure, low ridge; carina in tectal view broadest in lower half, with breadth of plate increasing between umbo and acutely rounded basal margin; carina narrowest near middle (waist), becoming slightly broader with nearly parallel sides to acutely rounded apex; interior of plate never infilled, broadly open between apex and base, except at waist; exterior surface without ridges or striae radiating from umbo, and ornamented solely by prominent growth ridges. Scutum (Figs. 4f-k) up to 10.0 mm in height, twice as high as wide, subrectan- gular, moderately thick, with markedly subcentral umbo located on the occludent margin one-third the distance from the rostral angle; occludent margin straight to slightly concave above umbo, strongly convex below umbo to the acutely rounded rostral angle; a deep, narrow furrow, sometimes bordered inwardly by a low ridge, extends from the umbo to a point on the tergal margin one-quarter the distance from the acutely angulate apex to the broadly rounded basitergal angle; the intersection of the furrow with the tergal margin is marked by a sharp pro- jection above the otherwise straight to slightly convex tergal margin; the basi- carinal margin is straight to slightly convex and about equal in length to the tergal margin; exterior surface ornamented by prominent growth ridges and in the upper half of the plates, by fine, obscure, and irregularly-spaced striae radiating from the umbo; the interior of the scutum bears a large, deep, adductor muscle pit bordered on its upper, occludent margin by a prominent, straight ridge; the oc- VOLUME 98, NUMBER 3 641 Fig. 4. Virgiscalpellum inornatum: a—b, Side and tectal views of carina, holotype USNM 382786; c—d, Oblique side and tectal views of carina, paratype USNM 382787; e, Oblique side view of carina, paratype USNM 382788; f—g, Exterior and interior views of scutum, paratype USNM 382789; h-1, Exterior and interior views of scutum, paratype USNM 382790; j—k, Exterior and interior views of scutum, paratype USNM 382791; 1, Exterior of broken tergum showing radial threads, paratype USNM 382792; m-n, Interior and exterior views of broken tergum, paratype USNM 382793. All illustrations x7. 642 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON cludent border often bears a small, shallow pit (for males?) immediately below the umbo. Tergum (Figs. 4l—n) relatively thin, narrow, about 3 times higher than wide, the largest specimen (broken) 7.8 mm in height; apico-basal ridge broad, low, rounded; occludent margin broadly convex; carinal margin slightly concave; apex worn in all specimens, but probably acute; basal angle, based on growth lines, narrowly rounded; a narrow, sharp ridge extends from the apex to the scutal margin halfway between the apico-basal ridge and the occludent margin; 1 or 2 fine longitudinal threads may be present between the 2 main ridges. Material. —Nine complete and 18 fragmentary carinae; 11 complete and 8 fragmentary scuta; 5 partial terga from USGS loc. 31796. Disposition of types. —Holotype USNM 382786, paratypes USNM 382787 through 382793, and paratype lots USNM 382794 through 382796 are in the Division of Paleobiology, U.S. National Museum of Natural History, Washington, IDC. Etymology. —The specific name inornatum is from the Latin inornatus, mean- ing unadorned, and refers to the absence of external radial ornament on the carina. Discussion. — Virgiscalpellum inornatum most closely resembles V. gabbi (Pils- bry, 1933) from the Ripley and Prairie Bluff Formations in Tennessee, Mississippi and Alabama and particularly, V. gabbi gabbi as described by Collins (1973). The new species is distinguished by its carina which lacks external radial ornament and has its greatest breadth in the lower rather than the upper half of the plate; by its scutum whose umbo is well below the center of the occludent margin and whose shape is more rectangular than triangular; and by its tergum that is some- what broader, less curved towards the carina, and bears a better developed apico- basal ridge. The tergum of V. inornatum approaches that of Virgiscalpellum sp. figured by Collins (1973, pl. 5, fig. 13) from the upper Ripley Formation in Mississippi, in breadth and in development of an apico-basal ridge, but differs in having a concave carinal margin and a more attenuated apical half. Virgiscalpel- lum inornatum is most readily distinguished from European species of Virgis- calpellum and from V. paitense (Pilsbry and Olsson, 1951) from the Upper Cre- taceous Tortuga Formation of Peru by the absence of radial ornament on the carina and the subdued nature of the ornament on the scuta and terga. Virgis- calpellum heteroplax (Pilsbry and Olsson, 1951) and V. euglyptum (Pilsbry and Olsson, 1951), also from the Tortuga Formation, are known only from scuta. The shape of the scutum of V. inornatum differs markedly from that of V. heteroplax, and the position of the scutal umbo in V. inornatum is considerably lower on the occludent margin than the centrally-placed umbo of V. euglyptum. The absence of radial ornament on carinae of V. inornatum is not the result of abrasion during transport and deposition. Such abrasion is present on the edges of the plates, but the preservation of growth ridges and umbones on several carinae clearly indicates that wear did not alter external ornamentation significantly. The plates of both V. inornatum and Zeugmatolepas sp. show signs of sorting. In the case of the former species, none of the plates is more than half the size of the larger plates of V. gabbi figured by Collins (1973), suggesting that the Black Creek examples are from young adults. Similarly, none of the larger, upper whorl plates where found for the Black Creek Zeugmatolepas. The genus Virgiscalpellum ranges from the Aptian (Withers 1947, 1953) through VOLUME 98, NUMBER 3 643 the Maestrichtian (Withers 1935), with the greatest known species diversity being in Maestrichtian time. In North America Virgiscalpellum is now known from the upper Campanian and the Maestrichtian. Literature Cited Collins, J. S. H. 1973. Cirripedes from the Upper Cretaceous of Alabama and Mississippi, eastern Gulf region, U.S.A., I. Palaeontology.— Bulletin of the British Museum, (Natural History), Geology 23(6):351-380. Hattin, D. E. 1977. Articulated lepadomorph cirripeds from the Upper Cretaceous of Kansas: family Stramentidae.— Journal of Paleontology 51(4):797-825. . 1982. Stratigraphy and depositional environment of the Smoky Hill Chalk Member, Niobrara Chalk (Upper Cretaceous) of the type area, western Kansas. — Kansas Geological Survey Bulletin 225:1-108. Pilsbry, H. A. 1933. An unusual Cretaceous cirriped.—Science 77:283-284. ,and A. A. Olsson. 1951. Tertiary and Cretaceous Cirripedia from northwestern South Amer- ica.— Proceedings of the Academy of Natural Sciences of Philadelphia 103:197—210. Sohl, N. F., and R. A. Christopher. 1983. The Black Creek-Peedee formational contact (Upper Cretaceous) in the Cape Fear River region of North Carolina.— U.S. Geological Survey Profes- sional Paper 1285:1-37. Withers, T. H. 1928. Catalogue of fossil Cirripedia in the Department of Geology, Vol. I. Triassic and Jurassic.— British Museum (Natural History), London, 154 pp. 1935. Catalogue of fossil Cirripedia in the Department of Geology, Vol. II. Cretaceous. — British Museum (Natural History), London, 534 pp. —. 1947. In O. Wilckins, Palaeontologische und geologische Ergebnisse der Reisen von Kohl- Larsen nach Sud-Georgien.—Senckenburg Naturforschung Gesellschaft, Frankfurt am Main, Abhandlungen 474:1-66. 1953. Catalogue of fossil Cirripedia in the Department of Geology, Vol. III. Tertiary.— British Museum (Natural History), London, 396 pp. (VAZ) Department of Earth Sciences, University of North Carolina at Wil- mington, Wilmington, North Carolina 28403; (NFS) United States Geological Survey, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 98(3), 1985, pp. 644-654 A NEW SPECIES OF PSILOGOBIUS FROM THE INDO-PACIFIC WITH A REDESCRIPTION OF PSILOGOBIUS' MAINLANDI (PISCES: GOBIIDAE) Ronald E. Watson and Ernest A. Lachner Abstract. —The genus Psilogobius Baldwin is redescribed and now contains two species, P. mainlandi from the Hawaiian Islands and P. prolatus, new species, from the Great Barrier Reef, Australia and the Cocos-Keeling Islands, eastern Indian Ocean. Pronounced sexual dimorphism occurs in both species. In addition to the genital papilla, males have longer fins, longer filamentous first dorsal fin spines, longer upper jaws than the females, and differ in coloration. Psilogobius prolatus differs from P. mainlandi in having dark pigmentation on the upper jaw, the AITO sensory pore usually present, a small patch of tiny scales on upper operculum, pectoral fin rays less numerous, modally 15, longer dorsal, anal and caudal fins, a longer upper jaw and in certain sexual dichromatic marks or bars. Psilogobius mainlandi was originally described by Baldwin (1972:125) as mono- typic from specimens collected by G. B. Mainland in 1939 at Kaneohe Bay, Oahu, Hawaiian Islands. Mainland named the new species Paroxyurichthys edmondsoni and described it in his master’s thesis that was never published. The name Pa- roxyurichthys edmondsoni was published by C. H. Edmondson (1946); however, Baldwin (1972:125) noted that there was no description or illustration and that the name constituted a nomen nudum. The first author, while examining recently captured Psijlogobius from Hawaii, thought he had a new species. Further study revealed that these specimens were P. mainlandi but were not easily recognized since many important diagnostic features were overlooked or vague in Baldwin’s account. A redescription of P. mainlandi from the Hawaiian Islands revealed a new species of Psilogobius from Australia and the eastern Indian Ocean. Methods Methods follow those of Lachner and Karnella (1980), and Lachner and McKinney (1974, 1978), except where indicated. The pterygiophore formula of the spines of the dorsal fin in relationship with the underlying vertebrae is after Lachner and McKinney (1974:875). Measurements were made with dial callipers in mm, and counts were taken from the left side unless noted otherwise. Mea- surements of fins follow Hubbs and Lagler (1958:25) except for the caudal fin which is measured from the central hypural base to the posterior tip of the longest ray. All proportional measurements are expressed to the nearest tenth of a percent of the standard length (SL), except for certain tabular data. Osteological aspects of the study were accomplished with radiographs, and cleared and alizarin stained specimens. In the descriptive accounts the values given for meristic characters ere followed VOLUME 98, NUMBER 3 645 by the frequency in parenthesis; the holotypic value is italicized. Data in the material examined section include the following: type status (if any), catalogue number, number of specimens, sex, SL in parenthesis (range if more than one specimen), specific locality, date collected, and collector. Abbreviations and terminology used to designate cephalic sensory pores follow Lachner and McKinney (1974:865): NA, nasal; AITO, anterior interorbital; PITO, posterior interorbital; AOT, anterior otic; SOT, supraotic; IT, intertemporal; POP, preopercular. The following abbreviations are used to designate institutions and collections cited: AMS, Australian Museum, Sydney; ANSP, Academy of Natural Sciences, Philadelphia; BPBM, Bernice P. Bishop Museum, Honolulu; NIMA, Northern Territory Museum, Darwin, Australia; USNM, National Museum of Natural History, Smithsonian Institution, Washington, D.C. Psilogobius Baldwin Fig. | Psilogobius Baldwin, 1972:125 (type-species: Psilogobius mainlandi Baldwin, 1972: 126 by original designation). A small, tropical, marine, Indo-Pacific goby, body elongate, subcylindrical and somewhat compressed; branchiostegals 5; 6 spines in first dorsal fin; first dorsal fin free from second dorsal fin; pterygiophores 3 (22110); vertebrate (including urostyle) 10 + 16 = 26; gillrakers on outer arch 3 + 1 + 7 or 8. Ctenoid scales well developed on caudal peduncle, becoming progressively smaller anteriorly; ctenii in a single row, well developed, spike-like, and equal or subequal on each scale (regenerated scales may lack cteni1); a patch of tiny scales sometimes present on upper portion of operculum; nape, breast and belly naked. Teeth conical, recurved; upper jaw teeth in 4 or 5 rows anteriorly, and 2 rows laterally; lower jaw teeth in 3 or 4 rows anteriorly, and 2 rows laterally; teeth absent on vomer and palatines. Cephalic sensory pore system is limited to anterior oculoscapular canal, pores NA, PITO, and AOT always present. Gill opening moderate. Sexual dimorphism present in coloration and morphology. Psilogebius mainlandi Baldwin leg, Z Psilogobius mainlandi Baldwin, 1972:126 (type-locality: Coconut Island, Kaneohe Bay, Oahu, Hawaiian Islands). Material examined. —59 specimens, 32 males, 20 females and 7 juveniles or unsexed; size range 8.5-37.5 mm SL, largest male 37.5, largest female 34.3, smallest gravid female 22.6. Holotype. -USNM 206174, male (30.9), Coconut Island, Kaneohe Bay, Oahu, Hawaiian Islands, 9 May 1968, W. Baldwin and J. Richards. Paratypes. —BPBM 10862, female (34.3), Coconut Island, Kaneohe Bay, Oahu, Hawaiian Islands, 1 Apr 1968, J. E. Randall et al.; BPBM 10864, 26 (22.4—35.5), 17 males, 9 females (5 males cleared and alizarin stained), Kahaluu, Kaneohe Bay, Oahu, Hawaiian Islands, 7 Dec 1967, W. Baldwin et al.; BPBM 10865, 5 (10.0-20.3), 1 male, 2 females, 2 juveniles, Coconut Island, Kaneohe Bay, Oahu, Hawaiian Islands, 12 Jul 1968, W. Baldwin et al.; BPBM 22624, 4 unsexed (8.5-— 646 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Ba oe of oo oe? 09° 25 00 a e aceese FOF Base SSeaea a 29 ° Seo aor 220000000" (Fe) Soo foje) 8° 9Q9000 Fig. 1. Diagrammatic illustration of Psilogobius showing cephalic sensory pore system and cuta- neous sensory papillae distribution on head. The arrow indicates the connection of the gill membrane with the isthmus. 23.5), Kaneohe Bay, Oahu, Hawaiian Islands, 13 May 1939, G. B. Mainland; USNM 206175, 21 (19.3—31.3), 12 males, 8 females, 1 unsexed, same collection data as holotype; USNM 202527, 1 male (37.5), same collection data as BPBM 22624. Diagnosis. —Cephalic sensory pore AITO usually absent; pectoral fin rays mod- ally 17; dorsal, anal and caudal fins and length of upper jaw shorter than in Psilogobius prolatus; patch of tiny scales on upper operculum absent; no dark pigmentation on upper jaw; males lack narrow, vertical silvery trunk bars; females have dark oblique band extending from notch between dorsal fins downward to anal aperture; silvery vertical trunk bars on females somewhat narrower than those on P. prolatus. Description. —Dorsal rays VI-9 (1); VI-I,10 (55); VI-I,11 (1); anal rays 1,8 (1); 1,9 (54), 1,10 (2); pectoral rays 15 (1), 16 (20), 17 (27), 18 (2), 19 (1),; counts for 1 specimen taken from right side due to deformed left pectoral fin; pelvic rays I,5 (57); segmented caudal rays 11 (1), 15 (4), 16 (13), 17 (34), 18 (2), 19 (1). First dorsal fin higher than second dorsal; first dorsal spines filamentous, second, third and fourth spines longest; third spine may reach base of tenth ray of second dorsal fin in mature males; origin of second dorsal fin over anus. Depressed VOLUME 98, NUMBER 3 647 Fig. 2. Psilogobius mainlandi, paratype, 34.3 mm SL, female, Hawaiian Islands, BPBM 10862. pectoral fin reaches to or beyond anus, posterior margin more or less rounded. Pelvic fins joined to form a cup-like oblong disk not adherent to body; pelvic fin frenum well developed between pelvic spines; pelvic fin rays branched with tips free; membrane between fifth medialmost rays complete along length of fin. Caudal fin oblong, equal to or shorter than head. All scales ctenoid, extending from trunk to slightly onto caudal fin base; scales small, largest in area of hypural base; scales smallest in area of pectoral fin base. Ctenii on scales of hypural base range from 4 to 9 per scale with 8 or 9 radii; scales on midline of trunk between second dorsal and anal origin with 3 or 4 cteni and 4 to 7 radii; scales in area of pectoral fin base with 1 to 3 cteni and 3 to 4 radii. Scales in horizontal series highly variable, 32 to 78; their size and spacing vary considerably. Head subcylindrical and somewhat compressed; bony interorbital very narrow with eyes nearly touching; snout short, less than diameter of eye; isthmus narrow, extending ventrally to base of pectoral fin. Mouth terminal, lower jaw not pro- truding and slightly oblique; upper jaw reaches posterior margin of eye in females and juveniles; well beyond posterior margin of eye in large males. Cutaneous sensory papillae pattern poorly to moderately developed on head. Two or 3 transverse rows, and | or 2 longitudinal rows on cheek (see Fig. 1). No sensory ridges or barbels. Cephalic sensory pore system variable; AITO pore seldom present; IT pore often present (see Table 3). Sexual dimorphic characters well developed. Male genital papilla tube-like, tapering and pointed at tip, about 4 times length of its base. Female genital papilla short, bulbous; about as long as length of its base. All fins of adult males longer than in females. Filamentous spines of first dorsal fin and upper jaw longer in males than females. Dimorphic differences not always obvious in small males. Color of preserved specimens. —Trunk with 8 or 9 dark lateral spots that orig- inate behind pectoral base and extend to hypural base. Spots tend to alternate in size with smallest at or near hypural base. Six to 8 dark spots dorsally on trunk extending from first dorsal fin to end of caudal peduncle. Trunk in females with 6 or 7 silvery, vertical, narrow bars separated by faint dusky areas; silvery bars 648 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON not meeting ventrally, and usually not extending to dorsal fins; bars absent in males. Females with dusky band between origin of second dorsal fin and anus. Branchiostegal membrane of males always dark; immature males show this char- acteristic before other external sexual characteristics develop; females lack bran- chiostegal membrane pigmentation. First dorsal fin with light spots and dark patches; second dorsal fin with dark spots arranged in oblique rows. Posterior tips of pectoral and pelvic fins, lower third of anal and caudal fins dusky; markings more intense on males. Peritoneum silvery. Color in life. —Following description from color slide of a 34.3 mm SL female specimen (BPBM 10862). Overall body color light brown; arched band from near end of upper jaw to above opercle; 7 small blue-white spots on lower preopercle; 4 patches of melanophores on cheek; lower *%4 of opercle orange, 7 small blue- white spots near preopercular margin, 8 small blue-white spots near posterior margin of opercle; patch of melanophores in center of opercle; dusky patch at upper pectoral base and one above anus, a small dark spot between these; 5 prominent, narrow, vertical white bands extending from behind pectoral base posteriorly to origins of first dorsal and anal fins. Weak, narrow white band originates above second anal ray and extends almost to second dorsal fin base; 7 dark spots with orange tinge in center, originating at weak white band on mid- trunk and extending to caudal fin base; spots alternate in size. Dorsal surfaces of body and upper trunk with scattered dark spots extending from just below first dorsal fin to upper caudal fin base; 2 or 3 blue-white spots found between these dorsolateral spots; lower half of first dorsal fin with oblique bands that alternate from brown to blue-white, upper half of fin brown; lower 74 of second dorsal fin with oblique alternating brown and blue-white bands (similar to first dorsal fin); upper half of fin brown, its outer margin blue; upper third of caudal fin with weak alternating brown and blue-white bands, lower portion translucent; pelvic spines whitish, rest of fin orange with scattered melanophores, posterior tip dark; anal fin clear next to body progressively darkening distally. Ecology. —Specimens examined were reported from marine habitats up to depths of 15 meters, over sand and silt in shallow water, in sandy areas around living reef. This species may require waters of sheltered areas that are little affected by wave activity. Psilogobius mainlandi is found in association with an alphaeid shrimp; this association has been reported in Cryptocentrus-like gobies of the genera Amblye- leotris, Cryptocentroides, Cryptocentrus, Ctenogobiops, Eilatia, Flabelligobius, Mahidolia, Stonogobiops, Tomiyamichthys, Vanderhorstia, and Yongeichthys (Hoese and Randall 1982; Hoese and Steene 1978; Klausewitz 1974; Polunin and Lubbock 1977; Yanagisawa 1978). This relationship may exist in Myersina as well (Akihito and Meguro 1983). Remarks. — Baldwin (1972) overlooked important characters and briefly treated others. He noted taking scale counts in his methods section, but none were given. The illustrations (1972, Fig. 1) of scales show ctenii to be wedge shaped, whereas they are actually spike-like. He stated that the scales are embedded, but we found the scales to be firmly attached anteriorly, while they may fall out posteriorly. Baldwin gave a vertebral count of 10 + 15 = 25, but did not say if this includes the urostyle. Our examination of radiographs and cleared and stained specimens VOLUME 98, NUMBER 3 649 Fig. 3. Psilogobius prolatus, paratype, 31.0 mm SL, female, Cocos-Keeling Islands, ANSP 131147. shows the count to be 10 + 16 = 26, including the urostyle. Baldwin (1972:126, Figs. 3, 4) described and showed the POP pores as present on the preopercle margin, but an examination of 72 specimens of both species showed the POP pores and the associated canal to be lacking. He stated that the third and fourth spines of the first dorsal fin are longest, but we find the second and third spines of the first dorsal fin are longest as confirmed by Baldwin’s illustration (1972, Fig. 4). He described the branchiostegal membrane as dusky to dark; our examination shows this to be a sexual dichromatic feature and that the membranes are un- pigmented in females. He mentioned the abdominal bars and noted that they are occasionally absent, whereas the bars are absent in all males in preservation. Psilogobius prolatus, new species Fig. 3 Material examined. — 13 specimens totalling 8 males and 5 females; size range 27.5—36.1; largest male 36.1; largest female 35.0; no gravid material. Holotype. — AMS 1.20578-002, 1 male (34.2), patch reef, One Tree Island, Cap- ricorn Group, Great Barrier Reef, Queensland, Australia, western Pacific Ocean, 23°30’S, 152°05’'E, 30 Nov 1973, B. C. Russell et al. Paratypes. — AMS 1.20578-007, 2 (33.2, 35.4), 1 male, 1 female, same collection data as holotype; ANSP 131145, 1 male (33.7), West Island, Cocos-Keeling Is- lands, eastern Indian Ocean, 12°9’00”S, 96°50'15”E, 15 Mar 1974, W. F. Smith- Vaniz et al.; ANSP 131147, 5 (29.6—36.1), 2 males, 3 females, West Island, Cocos- Keeling Islands, eastern Indian Ocean, 12°7’40”S, 96°51'35”E, 18 Mar 1974, W. F. Smith-Vaniz et al.; USNM 265099, 2 (32.8, 34.5), 1 male, 1 female, same 650 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 1.—Meristic, morphometric and color differences between the two species of Psilogobius. Proportional values include number of specimens, range in parenthesis, followed by mean value. Character Number of pectoral fin rays Small scales on upper operculum AITO sensory pore Length of upper jaw in male Length of upper jaw in female Length of anal fin in male Length of anal fin in female Length of pectoral fin Length of second dorsal fin Length of caudal fin Length of head P. mainlandi modally 16-17 absent usually absent 29 (12.5-17.3) 15.0 20 (10.3-15.2) 13.5 13 (37.7—45.4) 42.1 9 (36.3—41.4) 39.6 22 (21.2-26.0) 23.4 22 (40.3—47.5) 43.6 22 (24.3-34.1) 29.6 22 (29.3-33.1) 31.6 P. prolatus modally 15 present usually present 8 (16.0-18.5) 16.8 5 (14.3-16.6) 15.1 8 (43.4-51.2) 47.0 5 (38.7—44.6) 42.4 13 (19.4—24.4) 21.6 13 (42.8-49.9) 45.9 13 (29.4—40.4) 34.2 13 (28.2-32.4) 30.3 Upper jaw pigmentation absent posteriorly darkened; weak in female Branchiostegal membrane dark; male only dusky; male only Silvery trunk bars in male; in preservation absent 5-6 Silvery trunk bars in female, in preservation 6-7 6-7 Dusky vertical bar from notch between present; female only absent dorsal fins downward to anal aperture collection data as ANSP 131147; USNM 266378, 2 males (27.5, 34.3), same collection data as holotype. Diagnosis. —Cephalic sensory pore AITO usually present; pectoral fin rays mod- ally 15; dorsal, anal and caudal fins and length of upper jaw longer than in P. mainlandi; small patch of tiny scales on upper portion of operculum; some dark pigmentation on upper jaw; males with 5—6 narrow, vertical, silvery trunk bars; females lack dark oblique band on middle of trunk that extends from notch of dorsal fins to anal aperture; silvery vertical trunk bars on females wider than in P. mainlandi. Other differences between these two species are shown in Tables 1-5. Description. — Dorsal rays VI-I,10 (73); anal rays I,9 (77), 1,10 (2); pectoral rays 14 (1), 15 (12); pelvic rays I,5 (13); segmented caudal rays 15 (1), 16 (1), 17 (77). First dorsal fin higher than second dorsal fin; spines filamentous with third spine longest. Depressed pectoral fin reaching to or slightly short of anus, posterior margin rounded. Pelvic fins joined to form oblong cup-like disk not adherent to body; pelvic frenum well developed; pelvic fin rays branched with tips free; mem- brane between fifth (medialmost) rays of pelvic fins complete along entire length. Caudal fin oblong, equal to or slightly longer than head. See Tables 1, 2 and 5 for fin differences in P. mainlandi and P. prolatus. Table 2.—Number of pectoral fin rays in species of Psilogobius. Pectoral fin rays Species 13 14 15 16 17 18 19 P. mainlandi | 19 25 y 1 P. prolatus 1 12 VOLUME 98, NUMBER 3 651 Table 3.—Number, distribution and variation in occurrence of cephalic sensory pores in species of Psilogobius based on 22 specimens of P. mainlandi and 13 of P. prolatus (n = number of specimens; 0 = pore absent; 1-1 = pore present on left and right sides; see ““Methods” for names of pores). Cephalic sensory pores Species N NA AITO PITO SOT AOT IT P. mainlandi 10 1-1 0 1 0-0 1-1 0-0 10 1-1 0 1 0-0 1-1 1-1 2 1-1 1 1 0-0 1-1 1-1 P. prolatus 7 1-1 1 1 0-0 1-1 1-1 3 1-1 0 1 0-0 1-1 1-1 p} 1-1 1 1 1-1 1-1 1-1 1 1-1 1 1 0-0 1-1 0-0 Head subcylindrical and somewhat compressed; bony interorbital region very narrow with eyes nearly touching; snout short, less than diameter of eye; isthmus narrow, extending ventrally to base of pectoral fin. Mouth terminal, lower jaw not protruding and slightly oblique; upper jaw normally reaching beyond posterior margin of eye in female, and well beyond posterior margin of eye in male. See Tables 1 and 4 for head differences between P. mainlandi and P. prolatus. Various proportional measurements for P. prolatus are presented in Table 6. Cutaneous sensory papillae pattern very similar to P. mainlandi. Cephalic sen- sory pore system variable with AITO and IT often present and SOT usually absent (see Fig. 1 and Table 3). All scales ctenoid, extending from trunk slightly onto caudal fin; scales largest at hypural base; scales becoming progressively smaller anteriorly and not imbri- cate. Scales with single row of ctenii; scales at hypural base with 6 or 7 ctenii and 9 to 11 radii per scale; scales on trunk between second dorsal fin and anal fin origin with 6 or 7 ctenii and 6 to 8 radii, scales at pectoral base with 1 to 3 ctenii and 3 to 5 radii. Small patch of tiny scales above operculum, ctenil prominent. Color of preserved specimens. — Five prominent dorsolateral dark spots extend- ing from behind pectoral base to hypural base, smaller intermediate spots between Table 4.—Morphometric differences in species of Psilogobius in percent of the standard length (M = male; F = female). Upper jaw length Species Sex 10 11 12 13 14 15 16 17 18 19 P. mainlandi M 3 9 14 6 1 F 1 — 1 5 6 3 P. prolatus M 3 4 — 1 F 2) 2 = 1 Head length 28 29 30 31 32 33 34 35 P. mainlandi 1 2 7 7 5 P. prolatus 1 1 7 3 1 652 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Table 5.—Fin lengths in species of Psi/logobius in percent of the standard length. Pectoral fin Species 19 20 21 22 23 24 25 26 P. mainlandi 2 5 3 8 3 1 P. prolatus 1 2 4 3 — 3 Second dorsal fin 39 40 41 42 43 44 45 46 47 48 49 50 P. mainlandi 1 1 2 2 5 3 3 2 3 P. prolatus 1 2 3 2 2 2 —_ 1 Anal fin Sex 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 P. mainlandi M 1 1 1 Dp, 2 ie ae ES] F 1 — 1 — 5 yD P. prolatus M 1 1. = 2° = 2- = wl 1 F (o——— + + |) evens || 1 1 Caudal fin 27 28 29 30 31 32 33 34 35 36 37 38 39 40 P. main- landi 3 4 m 6 5 1 _ 1 P. prolatus 1 1 1 _ D 2 2 2 — — 1 1 in some specimens; 5 to 7 narrow vertical, silvery bands on trunk from behind pectoral base to above anal spine; silvery bands extending close to dorsal fin base, not connecting on belly. Dark spots dorsally vary in size and number; 7 small spots from above pectoral fin base to below spine on second dorsal fin; four spots below base of second dorsal fin, evenly spaced and extend length of fin. Posterior portion of upper jaw dusky or finely pigmented, pigmentation weak in females; branchiostegal membrane dusky only in males, but not as dark as in P. mainlandi. First dorsal fin filaments dusky, especially first spine; pectoral fins faintly dusky ventrally; anal fin dark distally; pelvic fins dark posteriorly. Basal half of first and second dorsal fins with few oblique markings; markings not as intense as in P. mainlandi. Color in life. —The following description is from a color slide of a 31.0 SL female paratype (ANSP 131147). Overall background coloration light, cream; head almost white with many tiny orange bands arranged irregularly over cheek; opercle whitish with brown spots in middle portion; 6 narrow white bands (much wider than in P. mainlandi) extending from behind pectoral fin base to above anus; large dark pigment patches scattered among all 6 bands; weak white band (shorter than anterior bands) originating above second anal ray; 5 brownish orange spots near mid-trunk region from behind pectoral fin base to caudal fin base; 2 dark and 2 whitish bands on first dorsal fin; lower half of second dorsal fin with mottled oblique bands alternating whitish and brownish; outer margin of second dorsal fin bluish-white; pelvic fins light brown; anal fin opaque. Ecology. —Specimens from Australia and Cocos-Keeling Islands were collected over white sands near reefs at depths ranging from 2 to 7.5 meters. There are no VOLUME 98, NUMBER 3 653 Table 6.—Proportional measurements of type-specimens of Psilogobius prolatus expressed as a percentage of standard length. Data for paratypes include 7 males, 5 females, the range of measurements enclosed in parenthesis, followed by mean value. Holotype Paratypes Male Male Female Standard length (mm) 34.2 (27.5—36.1) (29.6-35.0) Head length 30.4 (29.5-32.4) 31.2 (28.2-31.2) 30.3 Head depth 14.9 (13.7-16.0) 14.7 (13.8-15.2) 14.4 Head width 14.9 (11.3-13.7) 12.6 (12.0-14.3) 13.1 Snout length 3.8 (2.5-4.8) 3.7 (2.8-5.4) 4.4 Predorsal length 37.4 (35.6-38.4) 36.8 (35.1-36.9) 35.9 Preanal length 55.1 (56.2-60.7) 58.4 (56.5-62.4) 60.2 Body depth at anal origin 14.0 (11.6-14.4) 13.2 (12.0-14.9) 12.9 Caudal peduncle length 14.0 (12.5-16.3) 14.6 (12.6-15.2) 14.0 Caudal peduncle depth 8.5 (7.7-9.0) 8.4 (7.7-9.4) 8.3 Upper jaw length 16.4 (16.0-18.5) 16.9 (14.3-16.6) 15.1 Pectoral fin length 20.2 (19.4—24.4) 21.8 (20.6—24.1) 22.1 Pelvic fin length 21.3 (21.6—24.2) 22.8 (20.6—24.6) 22.2 Second dorsal fin length 43.7 (44.0—49.9) 46.6 (42.8—48.0) 45.4 Anal fin length 46.0 (43.4-51.2) 47.2 (38.7—44.6) 42.4 Caudal fin length 34.2 (29.4—40.4) 35.5 (29.7-35.2) 32.6 observations of this species in life. It is probable that P. prolatus will be found in association with an alphaeid shrimp as with P. mainland. Etymology. —The specific name prolatus from the Latin for extended or elon- gated, refers to the longer jaw and longer dorsal, anal and caudal fins of this species when compared with Psilogobius mainlandi. Relationships. — Psilogobius appears to be closely related to Cryptocentrus and Mahidolia; each have transverse and longitudinal rows of cutaneous sensory papillae over cheek; dorsal fin counts of VI-I,10 and anal fin I,9 are also shared. Both species of Psilogobius, Cryptocentrus filifer, and Mahidolia mystacina were found to have sexual dimorphic jaw lengths and silvery peritoneums. Cryptocen- trus callopterus, C. leptocephalus, C. lutheri, C. singapurensis, C. strigilliceps, and Yongeichthys criniger possess blackish peritoneums and Amblyeleotris perioph- thalmus, A. randalli, and Ctenogobiops crocineus have whitish peritoneums. It is of interest that species with black peritoneums showed practically no sexual di- morphism in jaw lengths. Psilogobius appears to differ from Amblyeleotris, Cryptocentroides, Cryptocen- trus, Ctenogobiops, Mahidolia, Myersina, Stonogobiops, Tomiyamichthys, Van- derhorstia, and Yongeichthys in possessing only the anterior oculoscapular canal of the cephalic sensory pore system, while these other genera have greater elab- orations of the canal system, usually the presence of the AT, PT, and POP pores. The anterior oculoscapular canal in Psilogobius extends from the SOT to the IT rather than from or near the AOT to the IT as is common for many Cryptocentrus- like gobies. Acknowledgments We thank the following individuals for information and assistance: A. Y. Su- zumoto (BPBM); B. C. Russell (NTMA); E. N. Gramblin and S. L. Jewett (USNM). 654 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON We thank E. O. Murdy (USNM) for his review and comments of an earlier manuscript. Special thanks are extended to W. F. Smith-Vaniz (ANSP) for the loan of type-material, the use of a color slide and the gift of paratypes to the USNM;; to D. F. Hoese (AMS) for the loan of type-material, the gift of paratypes to the USNM, and for direction during discussions of Cryptocentrus-like gobies; and to J. E. Randall (BPBM) for the loan of type-material, slides, and negatives and for his continuous support during the course of this and other goby studies. Literature Cited Akihito, P., and K. Meguro. 1983. Myersina nigrivirgata, a new species of goby from Okinawa Prefecture in Japan.—Japanese Journal of Ichthyology 29(4):343-348. Baldwin, W. J. 1972. A new genus and new species of Hawaiian gobiid fish. — Pacific Science 26(1): 125-128. Edmondson, C. H. 1946. Reef and shore fauna of Hawaii.—Special Publication of the Bernice P. Bishop Museum 22:1-381. Hoese, D. F., and J. E. Randall. 1982. Revision of the gobiid fish genus Stonogobiops. —Indo-Pacific Fishes, Bishop Museum Press 1:1-18. , and R. Steene. 1978. Amblyeleotris randalli, a new species of gobiid fish living in association with alphaeid shrimps.— Records of the Western Australian Museum 6(4):379-389. Hubbs, C. L., and K. F. Lagler. 1958. Fishes of the Great Lakes region.— University of Michigan Press, Ann Arbor. 213 pp. Klausewitz, W. 1974. Eilatia latruncularia n. gen. n. sp. und Vanderhorstia mertensi n. sp. vom Golf von Aquaba.—Senckenbergiana biologisch 55(4/6):205—212. Lachner, E. A., and 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—127. , and J. F. McKinney. 1974. Barbuligobius boehlkei, a new Indo-Pacific genus and species of Gobiidae (Pisces), with notes on the genera Callogobius and Pipidonia. —Copeia 4:869-879. , and 1978. A revision of the Indo-Pacific fish genus Gobiopsis with descriptions of four new species (Pisces: Gobiidae). Smithsonian Contributions to Zoology 262:1—S2. Polunin, N. V. C., and R. Lubbock. 1977. Prawn-associated gobies (Teleostei: Gobiidae) from the Seychelles, western Indian Ocean: Systematics and ecology.—Journal of the Zoological Society of London 183:63-101. Yanagisawa, Y. 1978. Studies of the interspecific relationships between gobiid fish and snapping shrimps. I. Gobiid fishes associated with snapping shrimps in Japan.— Publication of the Seto Marine Biological Laboratory 24(4/6):269-325. (REW) 166th MI Company, VHFS, Warrenton, Virginia 22186. (EAL) De- partment of Vertebrate Zoology (Fishes), National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. PROC. BIOL. SOC. WASH. 98(3), 1985, pp. 655-656 VARICHAETADRILUS MINUTUS (BRINKHURST, 1965) NEW COMBINATION FOR PSAMMORYCTIDES (?) MINUTUS (OLIGOCHAETA: TUBIFICIDAE) Ralph O. Brinkhurst Abstract. —Psammoryctides minutus is transferred to Varichaetadrilus because of the resemblance of the penis sheaths and penial setae to those of members of that genus. Psammoryctides minutus was tentatively assigned to the genus Psammoryctides when first discovered (Brinkhurst 1965) and was listed incertae sedis in that genus by Brinkhurst (1971). The doubt about its generic placement was due to the lack of material which prevented a complete description of the male ducts. A dissection revealed one fragment of the male duct, which suggested that after the atrium narrowed to form the ejaculatory duct, a second enlarged section existed before the duct narrowed a second time to enter the base of the penis. This pattern resembles the form of the atria in other North American species of Psammoryc- tides reviewed by Loden (1978), and so it was assigned to the subgenus Spencerius by that author. This subgenus includes all of the North American species, the exception, Psammoryctides barbatus (Grube, 1861), is clearly an introduction to the St. Lawrence River from Europe. The major obstacle to acceptance of this decision is the absence of the char- acteristic long, thin, spermathecal setae and the presence instead of penial setae. These penial setae are wider than normal ventrals (8 wm), and longer (147 wm versus 71—85 wm). They are very short distally, but have clearly bifid distal ends. Such penial setae have been described in Varichaetadrilus (Brinkhurst 1981, Brinkhurst and Kathman 1983). Examination of the drawings of the penis of P. minutus, independently con- firmed by Loden, show a small penis sheath set on the distal end of a large penial mass. This too is reminiscent of the form of the penis in Varichaetadrilus, in which the main body seems to be erectile. The apparent difference in the terminal portion of the atrium and ejaculatory duct between minutus and the ducts of Varichaetadrilus species could be explained by stretching during the dissection process. Confirmation of the generic position of this species requires acquisition of new material of this scarce species, known only from one lake in Alberta and Lake Tahoe, but it is clearly much more appropriately placed in Varichaetadrilus than in Psammoryctides. While minutus is quite similar to V. pacificus (Brinkhurst, 1981), the latter has more hair and pectinate setae anteriorly, the hair setae are found in postclitellar bundles, and there are more posterior ventral setae per bundle. The penis sheaths of pacificus are spool-shaped as opposed to conical. Varichaetadrilus nevadana (Brinkhurst, 1965), transferred to this genus from Isochaetides (Brinkhurst, 1981), lacks genital setae but is otherwise quite similar 656 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON to minutus. There are distinct differences in setal form and number, however. This species is restricted to Lake Tahoe. Another species with distinct setae, V. israelis, 1s restricted to Lake Tiberias, Israel. Literature Cited Brinkhurst, R. O. 1965. Studies on the North American aquatic Oligochaeta, II: Tubificidae.— Proceedings of the Academy of Natural Sciences, Philadelphia 117:117—-172. 1971. Jn Brinkhurst, R. O., and B. G. M. Jamieson, eds. Aquatic Oligochaeta of the world. Oliver and Boyd, Edinburgh, xi + 806 pp. Part 2. Systematics. 8. Family Tubificidae, pp. 444— 625. —. 1981. A contribution to the taxonomy of the Tubificinae (Oligochaeta, Tubificidae).— Proceedings of the Biological Society of Washington 94:1048-1067. , and R. D. Kathman. 1983. Varichaetadrilus, anew name for Varichaeta Brinkhurst, 1981, non Speiser, 1903, (Diptera) with a description of a new species V. fulleri. —Proceedings of the Biological Society of Washington 96:301—306. Loden, M. S. 1978. A revision of the genus Psammoryctides (Oligochaeta: Tubificidae) in North America.— Proceedings of the Biological Society of Washington 91:74-84. Ocean Ecology Laboratory, Institute of Ocean Sciences, P.O. Box 6000, Sidney, British Columbia V8L 4B2, Canada. PROC. BIOL. SOC. WASH. 98(3), 1985, pp. 657-671 NEW SPECIES OF FROGS FROM BORACEIA, SAO PAULO, BRAZIL W. Ronald Heyer Abstract. —Four new species of frogs are described from Boracéia, the University of Sao Paulo’s field station in the Atlantic forests of Brazil: Hyla hylax, Eleu- therodactylus randorum, E. spanios, and Physalaemus franciscae. In order to allocate Eleutherodactylus names, the status of Eupemphix bolbodactyla, Bas- anitia gehrti, B. lactea, Hylaplesia nigriventris, Eleutherodactylus unistrigatus hol- ti, and E. venancioi are briefly discussed. Lectotypes are designated for Basanitia lactea and Hylaplesia nigriventris. Eleutherodactylus holti is recognized as a valid species from Itatiaia, Rio de Janeiro, Brazil. A summary of our knowledge of the frogs of Boracéia is being prepared (Heyer, Rand, Cruz, Peixoto, Nelson, in prep.). In the process of identifying the frog fauna from this locality in the Atlantic forests of coastal Brazil, several species have been studied that lack names. The purpose of this paper is to describe these new species before our larger report is published. A new species of Hylodes is the subject of a separate report (Heyer and Cocroft, in prep.). Species Descriptions Carlos Alberto G. da Cruz, Oswaldo Peixoto, and I sorted out three members of the Hyla circumdata group from specimens collected at Boracéia: Hyla astar- tea Bokermann, 1967, Hyla circumdata (Cope, 1867), and a new species, to be known as Hyla hylax, new species Fig. | Holotype. —MZUSP 59937, adult male from Brazil: Sao Paulo; Boracéia, ap- proximately 23°38’S, 45°50'W. Collected by A. Stanley Rand, 4 Nov 1983. Paratopotypes. —MZUSP 2357, 2535-41, 4030-31, 30875-78, 30892-94, 30901-03, 34571, 34576-77, 37801-03, 54499-501, USNM 208744, 208747, 208749, 208754, 208757, 208764—71, 208776, 208779, 208781-82, 243466-74, taken by various collectors on various dates. Diagnosis. —Hyla hylax has in common a well developed prepollex and pos- terior face of thigh with dark vertical stripes with Hyla astartea, H. carvalhoi, H. circumdata, H. izecksohni, and H. nanuzae; these two characteristics distinguish this cluster of frogs from all other Hyla of eastern Brazil. The posterior thigh stripes are straight and bold, with some stripes connecting dorsally with stripes on the front of the thigh in H. hylax; the stripes are faint or discontinuous in astartea, carvalhoi, and nanuzae. Hyla hylax further differs from astartea and nanuzae in size, hylax having a SVL range 55-64 mm, astartea and nanuzae combined SVL ranges of 39-42 mm. Hyla hylax further differs from carvalhoi 658 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Holotype of Hyla hylax, dorsal view. in that hylax has dark vertical stripes on the flanks, such distinct marks are absent in carvalhoi; hylax has a single, somewhat expanded vocal sac, carvalhoi has a bilobed subgular vocal sac. Hyla hylax is larger than H. izecksohni (SVL 34-51 mm), and the tympanum in hylax is noticeably smaller than in izecksohni. Hyla hylax is most similar to H. circumdata, from which it differs in having a moderate tympanum, its diameter just less than 2 the diameter of the eye (tympanum large VOLUME 98, NUMBER 3 659 Fig. 2. Head profiles of Hyla hylax (above) and Hyla circumdata (below). Drawings based on specimens from Boracéia. in circumdata, *,—*/, eye diameter) (Fig. 2), the head is narrower than long (broader than long in circumdata), and the male forearm is slightly hypertrophied (mod- erately to strongly hypertrophied in circumdata). Description of holotype.—Snout shape rounded in dorsal outline, vertically rounded in profile; canthus rostralis indistinct; tympanum moderate sized, di- ameter about equal to '2 diameter of eye; vomerine teeth in arched series, contigu- ous medially, arching posteriorly, between choanae anteriorly, extending behind posterior edges of choanae posteriorly; vocal slits present; vocal sac single, barely expanded externally; finger disks large, disk on finger III equal to tympanum; finger web formula I trace II 2~—3* III 2'4—2* IV; subarticular tubercles prominent, rounded, single; very faint brown nuptial pad between base of thurnb and pre- pollex; forearm noticeably hypertrophied but not extensively so; prepollex well developed, spines not exposed; distinct tuberculate ulnar ridge; short, distinct supratympanic fold, weak glandular transverse fold above anus; dorsal texture smooth; throat smooth; belly areolate; toe disks moderately large, slightly smaller than finger disks; toe web formula I 1'/—2 II 14%4-2% III 1*-2'2 IV 2+—1* V; inner metatarsal tubercle large, ovoid; outer metatarsal tubercle small, indistinct, round- ed; distinct inner tarsal fold extending length of tarsus; moderately large calcar on left heel, small calcar on right. Color in preservative: Face almost uniform brown with darker brown washes, no distinct pattern; top of head light brown with darker brown anastomosing stripes; dorsum of body light brown with darker brown transverse cross bars; upper limbs also brown with dark brown cross bands; anterior flanks with incom- plete vertical stripes, mid- and posterior flanks light cream and scattered brown dots (same as dorsal ground color) with distinct dark vertical stripes, light areas 2-4 times width of stripes; posterior face of thigh brown with distinct black vertical stripes, lighter areas 1—2 times width of stripes, about every other stripe continuous dorsally with stripes on front surface of thigh; throat cream and profused with brown melanophores; belly cream with a few scattered brown melanophores. 660 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ff | NRPS ca f° AR f nn xt | i { KILOHERTZ SECONDS Fig. 3. Advertisement call of Hyla hylax. Recorded from holotype, 2011 h, 4 Nov 1983, 17.5°C air temperature. Measurements in mm: SVL 58.5, head length 20.8, head width 20.4, femur 28.3, tibia 29.8, foot 25.1. Variation. —Males range from 55-62 mm SVL. The two females available measure 60.0 and 63.4 mm, respectively. The larger female has an obviously more flared and broader head. Other than this female, there is relatively little variation in most morphological features including extent of digital webbing. The dorsal pattern ranges from uniform to a crossbar pattern. A mid-dorsal dark stripe extending from the tip of the snout to at least the scapular region is present in some individuals. Some individuals have bold dark outlined enamel cream spots on the elbows, the heel calcars and the anal fold. There may be slight to distinct demarcation of dorsal and ventral color along the ulnar fold and a light tan stripe on the outside edge of the foot and tarsus may extend to the calcar. The dark stripes on the flanks may equai the width of the light areas, or be only %—% as broad as the light areas. In some individuals every other flank stripe is continuous with a dorsal cross band. The stripes on the posterior face of the thigh may be closely packed, such that the space between stripes is about equal to the stripe width, or spread out, such that the lighter areas between stripes are 2—3 times the stripe width. Color in life: (From Rand and Rand notes on specimen MZUSP 30904) “Dirty tan with brown middorsal stripe from snout, fading out at vent. Few small cream blotches lower back. Legs and arms same color as back, banded with wide orange tan except for femur which has very narrow tan band above and is flesh colored on front and back with dark grey bands. This continues onto concealed surface of tibia. Sides yellow gold, deeper at groin with pale transverse banding. Chin and throat cream with tiny black flecks. Belly and under femur anteriorly pale orange cream. Under femur behind this dirty tan. Under arms, hands, behind shoulder and under legs flesh color.” Advertisement call. —(Description based on calls from several specimens.) Call duration about 0.5-1.0 s; calls given sporadically; calls pulsed, pulses grouped together forming notes within call initially, more continuous but still not even pulse rate at end of call; pulse rate about 100 per s; calls not noticeably frequency modulated; call intensity moderately modulated, first half of call loudest; domi- VOLUME 98, NUMBER 3 661 nant frequency range 400-2300 Hz; no apparent harmonic structure; distinct sidebands due to pulsatile nature of call (Fig. 3). Etymology. —Latinized Greek for barker, in allusion to the distinctive dog bark- like advertisement call. Referred specimens. — Brazil: Parana; Prainha, MZUSP 36002; Sao Paulo; Bor- acéia; MZUSP 34578-80, 34588-91, 34599-601, 34603-04, 37795 (Guveniles); Sao Paulo; Ferraz de Vasconcelos, MZUSP 34474; Sao Paulo; Paranapiacaba, MZUSP 8870, 9629, 10621-23, 10978, 13873-76. There are four species of Eleutherodactylus from Boracéia to which names are readily applied: E. binotatus, guentheri, hoehnei, and parvus. In addition, there are other Eleutherodactylus specimens that are small in size as adults and their variation is considerable and difficult to interpret. Conservatively, these additional specimens represent a minimum of two species and possibly as many as six. I think at least four species are represented, one of which is represented only by a single juvenile specimen. For this specimen (USNM 243675), description is de- ferred until an adult is obtained. These small Eleutherodactylus all belong to the section of the genus that has at times been recognized as a distinct genus of its own, Basanitia. Several names have been proposed for members of this group; examination of the original descriptions, topotypic material, and most of the types indicates that only one name applies to the species from Boracéia. The names that could be or have been applied to the Boracéia frogs are Eupemphix bolboa- actyla, Basanitia gehrti, B. lactea, Hylaplesia nigriventris, Eleutherodactylus uni- strigatus holti, and E. venancioi. Each of these names is briefly discussed. Eupemphix bolbodactyla A. Lutz, 1925.—Two cotypes, USNM 96542-43, are in very poor condition, owing in part from incisions made through the angle of the jaw and body wall almost to the groin. Several digit tips are missing in both specimens. USNM 96542, an approximately 16 mm SVL male with vocal slits, has the snout region destroyed. USNM 96543, a 14.5 mm specimen, shares with 96542 the following character states: disks not greatly expanded, vomerine teeth present, and pattern completely non-existent (due to fading, as Lutz described a color pattern for this species in the original description). The disks are small relative to the Boracéia species, indicating that the name bolbodactyla does not apply to any of the Boracéia species. Further clarification of bolbodactyla will depend on examination of well preserved topotypic material and the MNRio cotype. Designation of a lectotype at this time is premature. Basanitia gehrti Miranda-Ribeiro, 1926.—The holotype and only known spec- imen of B. gehrti could not be located at either MNRio or MZUSP in May 1984. The MZUSP collection has been reorganized recently and it is unlikely that the type was missed but it is possible that it was missed during the search of the MNRio collection. Miranda-Ribeiro’s description is brief. The characters that pertain to association with the Boracéia species are: tympanum small, barely evident; vomerine teeth present; three outer fingers with very large disks, the first without a disk; skin smooth; dorsum uniform black with a white cephalic-coc- cygeal line, under sides of disks, belly, and thighs, lighter. This combination of characteristics does not match any of the species from Boracéia; the most similar is the juvenile specimen USNM 243675. Bokermann (1966a:15) placed gehrti, from Alto da Serra, Sao Paulo, in the synonymy of Basanitia nigriventris, another 662 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON species described in part from the same locality. The dorsal texture of nigriventris is tuberculate, but preservation differences could account for gehrti having been described as having a smooth skin. The identity of gehrti will not be resolved satisfactorily until the type specimen is located and examined. Nevertheless, the characters that Miranda-Ribeiro used to describe the animal differentiate it from the Boracéia species, with the exception of the above mentioned juvenile. Until evidence to the contrary is presented, I believe it best to allow gehrti to remain in the synonymy of nigriventris. Basanitia lactea Miranda-Ribeiro, 1923.— Miranda-Ribeiro described /actea on the basis of two specimens, a 32 mm specimen from Iguape, Sao Paulo (MZUSP 828), and a 19 mm individual from Campo Grande (MZUSP 504). MZUSP 828 is the specimen described and figured by Miranda-Ribeiro. MZUSP 504, the second syntype, is an adult female Eleutherodactylus parvus. MZUSP 828 is hereby selected as the lectotype of Basanitia lactea Miranda-Ribiero. The lec- totype is no longer in good condition— most of the finger disks have disintegrated and the feet are torn apart. A very faint pattern still persists, consistent with the pattern shown in Miranda-Ribeiro’s illustration. The back of the thigh in particular is boldly mottled, and the front of the thigh has a faint suggestion that it, too, was boldly mottled. The tympanum is hidden, the dorsum is smooth, and there is no indication that the belly was ever dark. The specimen is an adult female with the following measurements: SVL 32.2 mm, head length 11.7 mm, head width 11.3 mm, femur 15.1 mm, tibia 14.9 mm, and foot approximately 14.5 mm. The size and pattern distinguish /actea from the species at Boracéia. Hylaplesia nigriventris A. Lutz, 1925.—Lutz described nigriventris on the basis of more than one specimen (number not specified) from Itatiaia and Serra de Cubatao. Bokermann (1966a:65) stated that in the Adolpho Lutz collection he found two specimens (720, 721) labelled as types from Paranapiacaba (=Serra de Cubatao). Professor Antenor Leitao de Carvalho kindly located three specimens indicated as types for my examination: 719, 720, 721. All three specimens have the same locality data of Alto da Serra de Cubatao [=Paranapiacaba]. There is an additional cotype, USNM 96846, from the same locality. Thus, the Itatiaia specimen or specimens are probably lost. Based on zoogeography, it would not be surprising if the Itatiaia and Paranapiacaba specimens were different. The 21 mm SVL specimen Lutz described is apparently lost and may have been from Itatiaia. The remaining types are 17.8 mm SVL (AL 719), 8.2 mm (AL 720), 7.0 mm (AL 721) and 6.5 mm (USNM 96846). AL 720 is a juvenile Eleutherodactylus guentheri. The other three specimens appear to be conspecific and are used for application of the name nigriventris. The specimens are almost black, a coloration presumably artificially caused as the insides of the mouths are also black. As AL 719 is the largest specimen and easiest to compare with other specimens, it is hereby designated as the lectotype of Hylaplesia nigriventris A. Lutz. The lectotype has a series of warty tubercles on the dorsum and upper eyelid; a particularly prominent calcar, an indistinct tympanum, very large disks on the outer fingers and toes, and even though the bones have become decalcified, vomerine teeth are visible. As indicated above, the lectotype is dark and shows no indications of pattern. The lectotype shares all diagnostic character states with a 19.0 mm SVL male from Boracéia, MZUSP 37787; the two specimens are considered conspecific. The color pattern in preservative of MZUSP 37787 is: face with a dark brown VOLUME 98, NUMBER 3 663 mark in canthal region just behind nostril, rest of face brown with indistinct darker brown oblique bars below eye, dark supratympanic spot; dorsum a complex pat- tern of lighter and darker browns with large brown blocks; upper limbs brown with darker brown cross bands; anterior flank a meeting of dorsal and ventral patterns, in groin area a bold dark brown and cream mottle extending onto anterior surface of thigh; posterior surface of thigh with a light ovate cream area in mid- thigh to knee surrounded by more or less uniform brown, light area extending a bit more medially as a broken pattern; hidden and under surfaces of tibia with bold dark brown and cream mottle; throat uniform dark brown with very small, scattered light punctations; belly dark brown with light dots. Eleutherodactylus unistrigatus holti Cochran, 1948.—The holotype, from “Alto Itatiaya,’’ matches quite well with recently collected specimens of Eleutherodac- tylus from Brejo da Lapa, Itatiaia. At Brejo da Lapa, only one species of Eleu- therodactylus has been collected. The species is common, however. The holotype matches recent specimens in: (A) body proportions, including relatively long fingers with large truncated, dorsally notched disks on fingers III and IV; (B) small vomerine tooth patches; (C) tympanum indistinct but externally visible; (D) are- olate belly; (E) small calcar; (F) 2—3 distinct darkish tubercles on sole of foot. The holotype is in poor condition, and the pattern is no longer discernible. The figures Cochran drew showed some pattern details, which are matched in the recent specimens. The holotype is a male with vocal slits. Cochran gave the SVL as 19 mm; I measure no more than 17 mm, but the specimen is compacted. There are proportional differences between the Itatiaia form and E. unistrigatus, most strikingly in head shape and finger lengths. The structure of the finger and toe disks is very different—wunistrigatus with truncate or ovately truncate disks lacking a dorsal indentation; the Brejo da Lapa form has large, dorsally indented disks. The differences are of a magnitude more consistent with species level dif- ferentiation and are so considered. With the name nigriventris restricted to the lectotype from Paranapiacaba, Eleutherodactylus holti is the available name for the species from the higher elevations of Itatiaia. E/eutherodactylus holti and nigriventris are distinct species, differing in body shape, degree of dorsal wartiness, and particularly in calcar size. The calcar of nigriventris is large and prominent; the calcar of holti is small and indistinct, if expressed. Eleutherodactylus venancioi B. Lutz, 1959.—Recently collected topotypes agree well with Lutz’s description and illustration. This large disked species differs from all Boracéia large disked species in having a combination of a long slender snout, a distinct face mask, and small size, males about 17 mm and females 24 mm SVL. There remain three species of Eleutherodactylus from Boracéia lacking names. As indicated above, description ofa juvenile specimen (USNM 243675) is deferred until associated adults are collected. The two species described below differ from all other Eleutherodactylus from coastal Brazil except holti, lactea, nigriventris, parvus, pusillus, and venancioi in having a hidden or indistinct tympanum in contrast to a distinct, exposed tympanum with the annulus obvious externally at least on the sides and ventrally (also see Heyer 1984, Table 19). Of these, parvus and pusillus have small, pointed finger disks in contrast to the large triangularly ovate disks on the outer fingers of holti, lactea, nigriventris, venancioi, and the 664 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 4. Holotype of Eleutherodactylus randorum, dorsal and ventral views. new species. For comparative purposes, the new species are diagnosed only be- tween themselves and holti, lactea, nigriventris, and venancioi. Eleutherodactylus randorum, new species Fig. 4 Holotype. —MZUSP 59936, an adult male from Brazil; Sao Paulo; Boracéia, approximately 23°38’S, 45°50'W. Collected by W. Ronald Heyer on 2 Nov 1983. Paratopotypes. —MZUSP 23665-70, 23672-74, 23676-77, 34635, 36865, 37555, 49644, USNM 243475-78, 244635. Referred specimen. —-MZUSP 23675, a slightly morphologically distinct spec- imen from Boracéia. Diagnosis. — Eleutherodactylus randorum differs from holti, lactea, nigriventris, spanios, and venancioi in lacking vomerine teeth; vomerine teeth are either visible or can be felt by probe in the other species. Description of holotype. —Snout shape subovoid in dorsal outline, rounded- protruding in profile; canthus rostralis indistinct; tympanum hidden; no vomerine teeth; vocal slits present; vocal sac large, single, externally expanded from mid- throat to just past posterior level of arm insertion; thumb lacking disk, large disk on finger II, very large disks on fingers III and IV, disks ovately triangular, upper surfaces medially indented; fingers free of web; subarticular tubercles indistinct; no nuptial asperities or other secondary sexual characteristics; no prepollex; no ulnar ridge; no distinctive body glands or folds; dorsum smooth except for few low tubercles on upper eyelid; throat smooth; belly weakly granular; toe disks large, but smaller than largest finger disks; toes free of web; inner metatarsal tubercle oval, outer small and round; no tarsal fold or tubercle; heel with single modest tubercle. Color in preservative: Dark canthal stripe, dark outlined forwardly directed oblique bar under eye on upper lip, dark stripe from posterior corner of eye continuous with flank stripe; dorsum more or less uniform tan mid-dorsally with broad light dorsolateral stripes, light incomplete mid-dorsal pin stripe, light broad VOLUME 98, NUMBER 3 665 3,— ae 6\- oe. = Eph sae = a 3 7 = ea a) ran ae cer Site = eee See Sal Sets - - x a 4 WW Se _ z as ices “= ax - S x 2 — — —— —_——_ ——+— = ——— = By pe SECONDS Fig. 5. Advertisement call of Eleutherodactylus randorum. Recorded from holotype, 2045 h, 2 Nov 1983, about 19°C air temperature. dorsolateral stripe bordered above and below with dark brown stripe; upper limbs indistinctly cross banded; flanks with continuation of broad light dorsolateral stripe to groin; posterior surface of thigh mottled, faint dark seat patch, light transverse stripe under anus; more or less uniformly scattered melanophores on throat and belly, superimposed with scattered dark speckles. Measurements in mm: SVL 14.8, head length 5.9, head width 5.1, femur 6.9, tibia 7.3, foot 6.5. Variation. — There is considerable morphological and pattern variation among the specimens at hand. The upper eyelid varies from smooth to having several distinct tubercles. The tympanum is either hidden or indistinct, when indistinct of moderate size, about '2 diameter of eye. The male vocal sac begins either from under the nostril level or from a level under the posterior edge of the eye; the smallest male (11.8 mm) does not have an expanded sac. The dorsum is smooth to scattered with tubercles. The outer metatarsal tubercle is indistinct in some individuals. The heel tubercle ranges from indistinct to very distinct. The mid- dorsal area ranges from almost uniform brown or tan to mottled. Mid-dorsal and dorsolateral stripes are present or absent. The referred specimen, MZUSP 23675, has a broad light mid-dorsal area bordered laterally with dark pin stripes. The face pattern is either boldly demarcated or fuzzily so, but there is still a discernible pattern of dark oblique bars under the eye, except in referred specimen MZUSP 23675, which has a more or less uniform face pattern. One specimen has a light snout. In individuals lacking broad dorsolateral stripes, the posterior flank has a bold or subtle dark and light mottled pattern in the groin or a light uniform groin pattern. Referred specimen MZUSP 23675 has an almost uniform brown throat and chest and the belly is mottled but not quite distinctly speckled. The anterior and posterior surfaces of the thigh range from almost uniformly light to boldly mottled. The seat patch ranges from indistinct to noticeable. Males range from 11.8-15.0 mm SVL, females 15.0-18.2 mm. Color in life: (Based on composite color notes of W. R. Heyer and A. S. Rand for several individuals.) Iris with transverse brown stripe, above which lemon yellow, below which duller yellow; dorsolateral stripes cream; light areas under 666 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 6. Holotype of Eleutherodactylus spanios, dorsal view. eye cream; 2 light brown bars under eye; dorsum olive drab, pin stripes yellowish; venter dirty white with black markings; posterior surface of thigh mottled dirty white and brown/black. Life colors vary considerably. The dorsum ranges from light brown, orange- tan, red-brown, to deep rust. The groin is black and white, brass yellow, bright yellow, or orange. The tops of the legs are brown, orange-tan, or greenish-yellow. The belly is pale, white, cream, pearl, or bronze with dark brown or black markings. VOLUME 98, NUMBER 3 667 The anterior and posterior surfaces of the thighs are black and indistinct white, bright yellow, or faint orange. Advertisement call. —(Description based on several calls from one individual.) Call consisting of 3-8 notes and of about 2—5 s duration, intercall interval 4-8 s when actively calling; note duration 0.04—0.10 s; note rate about 2 per s; notes strongly partially pulsed, each note with 3-8 discernible pulses; note pulsatile rate about 280 per s; calls not noticeably frequency or intensity modulated; dominant frequency range 3800-5200 Hz; no apparent harmonics (Fig. 5). Etymology. —Named for Patricia and Stanley Rand, who through their field efforts added appreciably to our knowledge of the natural history of the frogs from Boracéia. Eleutherodactylus spanios, new species Fig. 6 Holotype. —MZUSP 23664, an adult male from Brazil; Sao Paulo; Boracéia, approximately 23°38'S, 45°50’W. Collected by Patricia and Stanley Rand, 13 Feb 1963. Paratopotype. —MZUSP 23671. Diagnosis. —Eleutherodactylus spanios is a small species, the known male 14.7 mm SVL and the known female 21.4 mm SVL; the known males and females of holti, lactea, nigriventris, and venancioi exceed 16 and 24 mm respectively. Eleutherodactylus spanios is most similar to E. randorum, from which it differs by having vomerine teeth and a non-expanded vocal sac in the males. Description of holotype. —Snout shape subovoid in dorsal outline, rounded in profile; canthus rostralis indistinct; tympanum indistinct, about '2 eye diameter; vomerine teeth discernible by probe, not visible under low magnification, posterior and medial to choanae; vocal slits present; vocal sac single, not expanded; tip of thumb swollen, disk on finger II moderately large, disks on fingers III and IV large, just smaller than tympanum, ovately triangular, slightly medially indented dorsally; fingers free of web; subarticular tubercles indistinct; no nuptial asperities or other secondary sexual characteristics; no prepollex; no ulnar ridge or fold; dorsum smooth; throat and belly smooth; toe disks large, just smaller than finger disks; toes free of web; inner metatarsal tubercle oval, outer small, round, distinct; no tarsal fold or tubercle, tarsus smooth; heel with moderate sized tubercle. Color in life: (Edited from Rand and Rand field notes.): Brown with lighter patterning and dark brown or black under eye, behind tympanum, over shoulders, at vent and heels; bright orange-red on concealed surfaces of legs and arms ex- tending up onto belly where edges of red area blotchy; fingers and toes red; darker and lighter brown banding on legs and arms. Color in preservative: Face with indistinct canthal stripe; indistinct, forwardly directed oblique bar under eye to upper lip; dark tympanic spot; dorsum with indistinct pattern of darker and lighter brown; limbs indistinctly cross banded; posterior flank light, patternless; posterior surface of thigh very light, not much pattern except for dark seat patch; throat almost uniform brown; belly mottled brown anteriorly, mostly plain posteriorly, not dark speckled. Measurements in mm: SVL 14.7, head length 5.9, head width 4.9, femur 6.5, tibia 7.6, foot 6.5. 668 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Variation. — The single, brittle female, 21.4 mm SVL, has one broken hind leg. The vomerine teeth are visible under low magnification. The tympanum is hidden. The belly is weakly granular. The groin is not much distinguished from the dorsal pattern. The posterior surface of the thigh is mottled, but with a lighter area toward the knee and the front of the thigh is patternless. Advertisement call. —Unknown. Etymology. — From the Greek spanios, rare, in allusion to its rarity of collection. Comparisons. — This species is most similar to randorum. It differs in having vomerine teeth (absent in randorum), somewhat larger size, male with non-ex- panded vocal sac (greatly expanded in randorum), non-speckled belly (speckled in randorum), and bright orange-red flash colors (no flash colors or yellow or dull orange in randorum). The combined differences are interpreted to mean species level differentiation rather than intraspecific variation. Four species of Physalaemus occur at Boracéia: cuvieri, maculiventris, olfersi, and a species that has been identified in collections and referred to in the literature as signiferus. The type locality of Rhinoderma signifera Girard, 1853, is Rio de Janeiro, Brazil. Comparison of topotypic specimens and calls of signiferus with the Boracéia specimens and calls indicates that the two are not conspecific. Physalaemus franciscae, new species Fig. 7 Holotype. -MZUSP 59935, an adult male from Brazil; Sao Paulo; Boracéia, approximately 23°38'S, 45°50’W. Collected by W. Ronald Heyer, 8 Dec 1976. Paratopotypes. —-MZUSP 3947, 4084, 4114, 25831, 25853, 25856, 25857-70, 25872-82, 26036-37, 37565-68, USNM 243507-45, 244636, taken by various collectors on various dates. Diagnosis. —Lynch (1970) delineated species-groups of Physalaemus, and he thought that all but the signiferus group were monophyletic. Members of the signiferus group were thought to be heterogeneous, but collectively could be dif- ferentially diagnosed from other Physalaemus by the following combination of characteristics: small to moderate size; slender build; smooth skin; finger I shorter than IJ; no inner tarsal tubercle; small, non-compressed metatarsal tubercles; small to large inguinal glands; lacking parotoid glands. The members or this group, which are diagnosed from franciscae, are maculiventris (A. Lutz, 1925), nanus (Boulenger, 1888), obtectus Bokermann, 1966, o/fersi (Lichtenstein and Martens, 1856), and signiferus (Girard, 1853). Physalaemus franciscae lacks the light stripe from eye to arm insertion ventrally bordering the dark flank stripe of olfersi, and is smaller than olfersi (adult fran- ciscae no larger than 29 mm SVL, adult o/fersi 30-35 mm SVL). Physalaemus franciscae has a few scattered light tubercles on the sole of the foot; the sole of the foot is smooth in maculiventris and nanus. The posterior belly is boldly mottled in franciscae, the posterior belly lacks pattern in obtectus and signiferus. Description of holotype. —Dorsal outline of canthus rostralis and tip of snout pointed, of upper lip rounded; snout acutely rounded in profile; canthus rostralis sharp; tympanum indistinct, almost hidden, just less than '2 eye diameter; max- illary teeth not visible, but discernible by probe; vomerine teeth absent; vocal VOLUME 98, NUMBER 3 669 yy Jay a oa Fig. 7. Holotype of Physalaemus franciscae, dorsal and ventral views. slits present; vocal sac single, slightly expanded externally, extending to border of chest with belly; finger tips barely swollen, not disked; fingers free, lacking fringe, ridge, or web; subarticular tubercles moderately large, rounded, not pun- gent; pair of brown nuptial pads on each thumb, narrowly separated from each other, ovoid pad at base of thumb associated with inner metacarpal tubercle, elongate pad covering rest of upper and inner thumb surface; no prepollex; no ulnar ridge; weak supratympanic fold; dorsolateral fold from eye to groin along demarcation of dorsal and flank color, large tear-shaped lumbar gland in groin, narrow portion directed anteriorly; dorsum smooth with scattered low glandular warts, tubercles warty, tubercles most noticeable on upper eyelids, throat and belly smooth; toe tips very slightly swollen, not disked; toes free, lacking web, sides of toes weakly ridged (left fourth toe missing); inner and outer metatarsal tubercles ovoid, outer about */, size of inner, outer slightly compressed; short, weak tarsal fold on distal % of tarsus, approaching but not contacting inner metatarsal tubercle, rest of tarsus smooth; left heel smooth, right heel with low glandular tubercle; foot smooth with few smallish light fleshy tubercles. Color in preservative: Black canthal stripe extending to upper lip sharply de- lineated from dorsal color above, indistinctly set off from loreal pattern below, loreal region gray, upper lip cream with dark bars; dorsal ground color cream-tan on which a somewhat darker, continuous, almost symmetrical pattern consisting of an interorbital bar, a broad forwardly pointing arrowhead just past the shoulder region and another broken triangular pattern in the sacral region with the bases of the triangles extending to and joining the black groin spots, all outlined with cream pin stripes, faint indications of other darker markings paralleling the tri- angular marks; forelimb with a dark cross band, rest of upper surfaces of arms mottled light brown-tan; when leg folded, a single dark cross band continuous across femur, adjacent tibia and adjacent tarsus, rest of upper leg surfaces indis- tinctly mottled tan and brown; forearm with sharply demarcated line separating light dorsal and dark ventral pattern; inner and outer surfaces of tibia tan with 670 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON vise eyo che ey reat sae 3S Seen WEY Gestaey es Chanraee cy | (ithe casei . by dats th athafleadhy Math ei bit ail in | | f (eee YH Lh 4th yee ial th 1) a td KIt seamed! ttt | sme 4 Hialpeall i {pave | lean My Wat 1 Vw Wad td AW 1 Wl | Wi a | fat | | hi fl : er toh fae |e | : } HE ah | i yt Wu 1 | | { | Ru Wt al ty ete j La ei fale fetal iat a) cme a aie KILOHERT Z SECONDS Fig. 8. Advertisement call of Physalaemus franciscae. Wave form signal length 0.214 s showing portion of call on left of audiospectrogram. Recorded from holotype, 8 Dec 1976, 21.5°C air tem- perature. bold black spots; black flank stripe sharply demarcated from dorsal color above from posterior edge of eye to lower groin, not quite extending to femur, dark flank stripe fading into ventral color rather than being sharply defined from it; very distinct white outlined black spot in groin on lumbar gland, spot about 4 size of gland; dark seat patch bordered by filigree pattern; light dorsal pattern extending deeply onto posterior surface of thigh, lower distal portion of posterior surface of thigh with darker mottle, lower mid-portion with 1—2 distinct dark blotchy spots; throat dark gray-brown mottle extending onto chest; belly with bold dark gray and cream mottle, especially on posterior 4 of belly. Measurements in mm: SVL 25.4, head length 8.5, head width 8.8, femur 11.5, tibia 11.9, foot 11.9. Variation. —The tympanum ranges from hidden to almost distinct. The male vocal sac is either expanded from the tip of the chin through the chest or not expanded externally. The first finger is about the same length as the second, but may be either just shorter or just longer than second. The outer metatarsal tubercle ranges from about 4 to almost as large as the inner. The short tarsal fold contacts the inner metatarsal tubercle in a few specimens; the fold is usually distinct but sometimes indistinct. There is relatively little variation in color patterns, mostly involving intensity of patterns. The males range from 25.0—27.0 mm, females 24.8-28.2 mm. Color in life: (From USNM 243520.) Iris bronze; throat with salmon cast; belly cream and black with brown cast; groin with salmon cast. (Rand and Rand color notes from MZUSP 25863-65.) Light khaki with patterning on back, edged with dark. Dark spots at end of belly and on tibia. Back of thighs and heels dark brown. Orange on groin. Belly and legs speckled. Advertisement call. —Call duration 0.50-0.70 s; calls given frequently when calling; call of complex structure including harmonics and pulses, beginning and most of call with 22—28 pulses, call ending with short non-pulsed portion, pulse rate about 50 per s; calls with slight but noticeable frequency modulation (warble) throughout call; calls quiet, starting and ending with lower intensity; dominant frequency range 600-1600 Hz; several distinct harmonics (sidebands?) at a fun- damental frequency of about 250 Hz (Fig. 8). VOLUME 98, NUMBER 3 671 Etymology. —Named in honor of Francisca Carolina do Val, who for the last several years has been the administrator of Boracéia and has greatly facilitated research at the field station. Referred specimens. —Brazil: Sao Paulo; Casa Grande, MZUSP 37327; Sao Paulo; Paranapiacaba, MZUSP 6477, 8851. Comparisons. —Physalaemus franciscae is most similar to P. obtectus and P. signiferus. In addition to the belly pattern differences given in the diagnosis, the calls of the three differ from each other in terms of length and pulse structure. The call of P. obtectus is composed of 3 or 4 notes, each note comparable to the entire call of franciscae and signiferus, with a note duration of about 0.08 s. The calls of signiferus are longer in duration, about 0.30 s, and the calls of franciscae longest, about 0.60 s. Further, only the final portion of the call of P. franciscae is non-pulsed, whereas the entire calls of obtectus and signiferus seem to lack the pulsation of franciscae. The entire call of signiferus appears very similar to the final portion of the call of franciscae. (Comparative data for P. obtectus and signiferus from Bokermann 1966b.) Acknowledgments Carlos Alberto Goncalves da Cruz and Oswaldo L. Peixoto pointed out the distinctiveness of Hyla hylax to me and kindly allowed me to describe it. Charles W. Myers and George R. Foley, American Museum of Natural History (AMNH) allowed me to borrow the holotype of Eleutherodactylus unistrigatus holti. An- tenor Leitao de Carvalho once again provided access to the collections in the Museu Nacional, Rio de Janeiro (MNRio). P. E. Vanzolini characteristically gave me complete access to the Museu de Zoologia da Universidade de Sao Paulo (MZUSP). Reginald B. Cocroft produced the audiospectrograms used for call analysis and entered and revised the text on the word processor. Esta Johnson drew the original illustrations for Figs. 2, 4, and 7. Stanley H. Weitzman kindly prepared the photograph for Fig. 6. George R. Zug, Smithsonian Institution (USNM), critically read the manuscript. Research leading to this paper was sup- ported, in part, by the Museu de Zoologia da Universidade de Sao Paulo; a Scholarly Studies Award from the Smithsonian Institution; and the I.E.S.P. Neo- tropical Lowland Research Program, Smithsonian Institution. Literature Cited Bokermann, W.C. A. 1966a. Lista anotada das localidades tipo de anfibios brasileiros.—Servico de Documentag¢ao, Sao Paulo, 183 pp. ———. 1966b. Dos nuevas especies de Physalaemus de Espiritu Santo, Brasil (Amphibia, Lepto- dactylidae).— Physis 26:193-202. Heyer, W. R. 1984. Variation, systematics, and zoogeography of Eleutherodactylus guentheri and closely related species (Amphibia: Anura: Leptodactylidae).—Smithsonian Contributions to Zoology 402:1—42. Lynch, J. D. 1970. Systematic status of the American leptodactylid frog genera Engystomops, Eu- pemphix, and Physalaemus. —Copeia 1970:488—496. Miranda-Ribeiro, A. 1926. Notas para servirem au estudo dos Gymnobatrachios (Anura) Brasileiros. Tomo primeiro.— Archivos do Museu Nacional, Rio de Janeiro 27:1—227, 22 plates. Department of Vertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, U.S.A. PROC. BIOL. SOC. WASH. 98(3), 1985, pp. 672-677 LOISETTEA AMPHICTENA, NEW GENUS, NEW SPECIES, FROM THE SUBLITTORAL OF NORTHWESTERN AUSTRALIA (ECHINODERMATA: HOLOTHUROIDEA) F. W. E. Rowe and David L. Pawson Abstract. —Loisettea, new genus, differs from others in the dendrochirotid subfamily Colochirinae in possessing body wall ossicles in the form of deep com- plex cups, buttons, and large scales. Loisettea amphictena, new species, is dis- tinctive in possessing conspicuous ventrolateral papillae; it occurs off the coast of northwestern Australia. Colochirus gazellae Lampert, from Sumbawa in the Sun- da Islands and from northwestern Australia, is also referred to the new genus. Over the past 25 years, largely through the dedicated collecting of Raymond W. George, Barry R. Wilson, and Loisette M. Marsh, the echinoderms of the Western Australian region, from the shore to upper bathyal depths, have become increasingly well known. The regional collections of the Western Australian Mu- seum are now extensive, and include numerous echinoderms which represent new taxa, or are new records for the region. During our separate and joint investigations of Australian holothurians, we found that both the Australian Museum (AM) and the Western Australian Museum (WAM) collections contained specimens of a unique and distinctive holothurian—this new genus and species is described be- low. We thank Loisette M. Marsh of the Western Australian Museum for making specimens and a photograph of this interesting animal available to us. John Miller, Harbor Branch Foundation, Inc., Florida, kindly and patiently led David L. Paw- son through SEM preparation and photography of the ossicles, and reviewed the draft manuscript of this paper. Partial support for David L. Pawson was provided by Smithsonian Fluid Research Funds. Sequence of authors of the paper was decided by the toss of a coin. Order Dendrochirotida Family Cucumariidae Ludwig, 1894 Subfamily Colochirinae Panning, 1949 Loisettea, new genus Diagnosis. —Tentacles 10, 2 ventral tentacles smaller than others. Calcareous ring with undulating posterior margin, lacking posterior projections. Body wall thick, brittle, packed with ossicles. Ventral radii with tube feet in zigzag rows, dorsal radii with single rows of widely spaced feet; smaller feet scattered in dorsal and ventral interradii. Ventrolateral and dorsal papillae present or absent. Ossicles lenticular plates or scales, knobbed buttons and deep complex cups. Type-species. — Loisettea amphictena, new species. Other species included. — Colochirus gazellae Lampert, 1889. VOLUME 98, NUMBER 3 673 Distribution. —Both species occur off the northwestern coast of Australia; C. gazellae is also known from Sumbawa, Sunda Islands. Etymology. —The genus-name is of feminine gender. We take great pleasure in naming the genus in honor of Loisette M. Marsh of the Western Australian Museum, Perth, who has contributed so much to our knowledge of Australian echinoderms, and who is a very good friend and colleague to both of us. The species-name is derived from the Greek amphi—both sides or double, and ktenos—a comb, in reference to the comb-like pattern of ventrolateral papillae in this species. Remarks. — Within the Subfamily Colochirinae, some species in the genera As/ia Rowe, 1970, Pentacta Goldfuss, 1820, and Trachythyone Studer, 1876, possess deep baskets or cups in the body wall. The new genus diagnosed here differs from Aslia species in the disposition of the tube feet and in ossicle form, ossicles of Aslia comprising buttons and baskets with no evidence of lenticular plates. Body wall cups of Pentacta are generally shallow and simple; in P. anceps (Selenka) the cups are deep, but the body wall is thinner and softer than in Loisettea. Also, the ossicles of P. anceps include delicate hollow spheres, but lack lenticular plates. Trachythyone species possess smooth plates, a characteristic that immediately separates that genus from Loisettea. Panning (1971) included Colochirus gazellae Lampert and C. dispar Lampert in the predominantly western Atlantic genus Thyonella; both species occur in northwest Australia, and the latter species is also known from the Somali Coast (Clark and Rowe 1971). These species differ from others in Thyonella in the form of the ossicles and in distribution of the tube feet. We believe that C. gazellae is congeneric with the new species described below, and we transfer it to the genus Loisettea herein. Colochirus dispar is provisionally referred to the genus Pentacta since 1n our opinion it is at least congeneric with P. australis (Ludwig). Key to species of Loisettea, new genus 1. Conspicuous ventrolateral papillae present ....... amphictena, new species — No-conspicuous ventrolateral papillae ................. gazellae Lampert Loisettea amphictena, new species Figs. 1-3 Diagnosis. —Conspicuous ventrolateral papillae present. Material examined.—HOLOTYPE: Australian Museum J1427, 17°38’S, 121°27’E, 78 m, 16 Jun 1980, collected by R. B. Martin on CSIRO Soela. PARA- TYPES: Australian Museum: J14283, 19°36’S, 118°37’E, 36 m, 4 Jun 1980, R. B. Martin on CSIRO Soela, 1 specimen; J14288, Timor Sea, 13°11'S, 129°43’E to 13°09'S, 129°43’E, 34-38 m, 7 Jul 1979, 1 specimen. Western Australian Museum: Catalogue No. 54-63, 36 miles SW of Adele Island, northwest Australia, 72 m, 17 Oct 1962, dredged by R. W. George on Davena, 1 specimen; 70-63, 3— 4 miles off East Delambre Island, Dampier Archipelago, 18 m, 5 Jun 1960, rocky, B. R. Wilson on Davena, 1 specimen; 96-63, same data as 70-63, 1 specimen; 330-71, off Carnavon, Western Australia, Jul-Sep 1967, W. and W. Poole on Bluefin, 1 specimen; 332-71, Blocks 2-6, Shark Bay, Western Australia, Aug 1965, 674 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON Fig. 1. Loisettea amphictena, live specimen, total length 70 mm, dorsal view. FRV Peron, 1 specimen; 676-82, 38 nautical miles north of Port Walcott, Western Australia, 19°59’S, 117°16’E to 20°01'S, 117°16’E, 52-50 m, 15 Apr 1982, silty sand and bryozoa, triangle dredge towed for 15 minutes, L. Marsh on Soela, 1 specimen; 677-82, 36 nautical miles north of Port Walcott, Western Australia, 20°01’S, 117°08’E to 20°00’S, 117°10’E, 50 m, 16 Apr 1982, silty sand and large Table 1.—Loisettea amphictena: Disposition and numbers of papillae and radial tube feet. Lateral papillae Specimen Length (mm) Left Right Dorsal papillae Other AM J14287 94 9 8 4 per radius jl AM J14283 80 7 8 6-8 per radius — AM J14288 92 5 6 absent — WAM 54-63 56 7 6 6—8 per radius — WAM 70-63 77 7 9 4 left, 5 right — WAM 96-63 40.5 9 7 4 per radius 2 WAM 330-71(1) 91 7 7 absent 3* WAM 332-71 96.5 7 6 — — WAM 676-82(1) 63 6 7 3 left, 4 right 4* WAM 677-82(1) 72 8 9 5 left, 7 right — BM(NH) 1892.1.16.99 53 — 5 BM(NH) 1892.1.14.269-71 56 BM(NH) 1892.1.14.269-71 36 No | nn | | | 1* tube feet in single row in lateral ventral radii, in double row in midventral radius. 2* ventral tube feet in single rows in ventral radii. 3* tube feet in irregular single row in each lateral ventral radius, in 2—3 rows in midventral radius. 4* tube feet in single row in each lateral ventral radius, 15 feet per row; in midventral radius 29 feet in zigzag single row (or 2 scattered rows). 5* tube feet in single row in each lateral ventral radius, 13-14 feet per row; in midventral radius approximately 20 feet scattered in single row. VOLUME 98, NUMBER 3 675 A 2D oan ——————————— 2mm Fig. 2. Loisettea amphictena. A, Outline drawings of preserved specimens, ventral view. Holotype is at top left. B, Radial (right) and interradial (left) pieces of calcareous ring. sponges, Engel trawl towed for 30 minutes, L. Marsh on Soela, 1 specimen. Additional material examined (not type-material): British Museum (Natural His- tory): 1892.1.16.99, no locality, presented by The Admiralty, 1 specimen; 1892.1.14.269-71, Holothuria Bank, off Cape Londonderry, Western Australia, presented by The Admiralty, 2 specimens. Description. —Body 36—96.5 mm long, approximately 4—5 times as long as broad, often quadrangular in cross-section, slightly flattened, tapering slightly anteriorly and posteriorly. Ventral margins with conspicuous concial lateral pa- pillae (Figs. 1, 2A), 5—9 in each radius (see Table 1), longest papillae approximately 15 mm. Dorsal radii either unadorned or with up to 8 inconspicuous to prominent dorsal papillae. Body wall rigid, coriaceous, in larger specimens approaching 2 mm in thickness in midventral interradius. Oral area covered by 5 radially placed more or less conspicuous triangular oral valves. Ventral radii with scattered con- spicuous tube feet (see notes to Table 1). Feet not extending to anterior and posterior extremities of radii, ending approximately | cm short of extremities; ventral and dorsal interradii with scattered smaller tube feet. Color in life more or less uniform orange-red, bases of dorsal papillae brick red. In alcohol, color fading to uniform dirty white. Radial longitudinal muscles poorly developed broad thin straps. Two Polian vesicles. Gonad consisting of tufts of light orange unbranched tubules. Internal surface of body wall on ventral side having large holes for tube feet in radii and smaller holes for tube feet in interradii. Calcareous ring of 10 simple pieces (Fig. PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON VOLUME 98, NUMBER 3 677 2B) lacking posterior projections. Radials elongate, anterior projection with shal- low notch. Interradials with bluntly pointed anterior projections. Ossicles in body wall large multi-layered scales (lenticular plates) overlain by complex cups and by buttons of varying complexity. In larger specimens (WAM 330-71, total length 91 mm), oval to circular scales can reach diameter of 5 mm in dorsal and ventral body wall. Dorsal buttons 80-220 um in length (Fig. 3A-B), simply knobbed, but often with more or less well developed secondary network of calcite, making them double-layered. Ventral ossicles tend to be smaller than dorsal. Ventral cups (Fig. 3H) average 69 wm in width and 56 um in height, ventral buttons (Fig. 31) approximately 90-150 um in length. In smaller specimen (WAM 96-63, total length 40.5 mm), ossicles generally smaller and less complex; dorsal and ventral ossicles identical. Cups (Fig. 3D, 3K—L) average 63 wm in width and 47 um in height. Buttons (Fig. 3C, 3J) greatly variable in size. Lateral papillae of larger specimen contain cups (Fig. 30, 3S) and buttons (Fig. 3M-—N) similar to those occurring elsewhere in the body wall. In smaller specimen, the cups (Fig. 3R, 3T) and buttons (Fig. 3Q) also similar to those from elsewhere in body wall, and simpler than their counterparts from larger specimens. Distribution. —Currently known only from the continental shelf of northwestern Australia, where it ranges from Shark Bay to the vicinity of Darwin, in depths of 18-78 meters. The species has usually been collected from a silty sand bottom, where large sponges and other echinoderms are common. Literature Cited Clark, A. M., and F. W. E. Rowe. 1971. Monograph of shallow-water Indo-west Pacific echinoderms. British Museum, London. 238 pp., 100 figs., 31 pls. Panning, A. 1971. Bermerkungen tiber die Holothurien-familie Cucumariidae (Ordnung Dendro- chirota). 6. Teil (Schluss). Die Gattungen um Ocnus Forbes 1841 und um Pentacta Goldfuss 1820.— Mitteilungen aus dem Zoologischen Staatsinstitut und Zoologischen Museum in Ham- burg 67:29-51. (FWER) Division of Invertebrate Zoology, Australian Museum, College St., Sydney, N.S.W., Australia. (DLP) Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Insitution, Washington, D.C. 20560, U.S.A. —_— Fig. 3. Loisettea amphictena, body wall ossicles from large (91 mm total length) and small (40.5 mm total length) specimens: A, B, Dorsal buttons from large specimen, < 280; C, Dorsal button from small specimen,