ino Di fanttsAtetia ss wom Ered Eatin Se HPAL Les bsorte etna E eS TSE Dar ECCN Ree eet ce can innoanie feat anh hind eSCatheM CIDA RSA Toe9 OTRAS MAR TNE TAR te tet Ne oO WiSATSREL MLDS SPARTAN Ran Penh Eh eAha Melse Lieto h «Gr EN AS PPR AA CL Ahh te weST ge fone ae SEP tw Sen RI Bet ee eatin ate erbnite 2 Finis f wale eon SLE on ‘ aera Dime re ma NTA NGA Cet ORR APT LLNG nener reer rrr yrenn etn i Une c nnn tis a HAAG NINN Palate wy NO dita ay aD AVE NAN AE MoI Es AN AMART SPA DEALS ATE oe epee dco ee Paseo hitate Sut Ten eesite’ eB TN AEE petd gratia Aone tatty wanton Mat CE ENR te MS A denen phe wnt Chega mean AM.NA ARMA TH TNA A AIT ENA TM AST ALN RICES NN Ma fey ep OAS fon Drea nan eee ene: Pasar Eas ssieed: een ahah caadelNeaaut AAD sian OM 9A Meuen Pasa rednd Aro TRS Yn Shan Sena AyD S IAI RAN NY DAIS A ts MF a Md gehen asta ONY NaS Dante ina NEAT peereca wren Meena i ery te rie ners Pn Bn sete Deana Seminar eres rece Cone Rho a ors plone me PEE wet ne 2 he Peete ahs Bawe We nte Mah aie RUSE UNE REA Ee teh tre ntele Sogn A EUAN Doge be Bind tae EAA ENE TORT HAGROL EL eaten Cnt LOM tonretdis ee ee gti ie fie Snatesh cyte ADNAN) AEE oto Seats ey spine tere, woo nay REPS Ga Behan er iere Orie honk a OF Panty Ae teotee 7 » Beir he Meech sie Rene MODELS Pe eae Lear nce Anse Pete nina of ansiet Pop teehricha on onmenms Onin a aay etme AARCata nose Gna eck in tog. 29 2 nF BAM NAA AUD AU REN CPR WMIPSINAY RM ABD a TAS STea te hitotina rah Sete ne AOL AD LIAS aaa ceee nasi shatieseonchet7aik 9 NGG aa iy MR SSAA wc eo Te a RAALAER STRIATE? Qa te Rane LO ama eet tN easy Nee ys week eni earh teen tna teAVia at gi sete bmn ran anne PRAT SAN ST ap spese ter nas et en Weary Ree are Bi pate tae oe Whee Mn bam mete TD rent iet Aronia fea Aaa hinted here Ara TANEN Potted aur insnls Sen tT tea Mabe ay ec aarih gee aMaeataetoa ga bas ATL AAS Saab ata bens woke A ATH eae ites oN ayy heat ata Sie tne Ag tel Aud NR OMAN siten Dt arena siya fo maybe DARA ca ge rc NSE TANT NG bacene Sporn Tyo Dono enne ray Peer en nT ea ety ee Pepe nah aon gh pees PA Me OP sith; Fart nen Bien ter 2 hes ade ie ae Dea Pt he BE MN AI CG TO Ny etm ethno eMbett pane tye ene her tte eh Aa Petia Rea E inte ne™ ee Ja Pet eme nas ee wind wean tied 4 nO ite ersten Ve pae P Om Eee seed Ei TBM AS dovdcerinae Mees sn We RARE AES eat ne NE EM FO yal ere ea series ren tet fae ety Boel ene van miicnse nent dei Bos 385 Of eee satan tt Hirst Eate xt tie MN Sten ERC AT tin cant oy eere tee ttn re eee cong emesatn s co enna rotates eds ree rte eens Re rene wadap steels TOY aa wakaiiva peanuts ey bein ae what ANS AAA sit synimange ae Seanarit nen a aATSD He Ds Bode Fea etenmqe nee ant NN See tenner ay ath itn in MO padre Paha SIREN ODN SAHA EAE AND TON Aap thay tog aed 9 ENNEN AHA Promeyie Net, ee eT eee et ELSIE OED Lomisetes We ser te eRe reat S Sehr e erate a mrenraper sere) ie Cadeasictie i Bor wrote ia ret : : 2 . “ chy et Wis Aininan en are Hh it e , m a SO eT ADD SPAN eb TEEN NEA HD Nertnerareeure | vetak eesescomgnen Soret aMaingsnt ree hTR AEN Seek eaaeeNeTe see cAI Seaberaa! eee Shane so pees arene ree ano saath ete Resto ltee' nen eyo Pie amine a Feta t ed oon hot ase eee c= the Peet nbatncee ene ait Gt alle fe Cal Rise He ak Ler Hel: vereetet nee fae EMER ACh ce miatiotene FAP ne ALI ed SHSM ehrar obra 9 AA 8AM NaN He NTRS Tam PTS Bom fe Darth ent tem 9 a MAN Pain Nate yay cut pm vente crams ERS AS > AFL RASNY dan sali Ge scam saat a Rw Fa caasaaitcyayit Rese D9 Soe were St ee isan nes et Pc eA Eo ED E CH MERE IGE Eee nen Te serra ts ska a Not tn Shoda TR ENTE TS ere eter a Pee Letty LETHE Qos PMA UTES ME oh nN OEE tag menaiyene eS nea MT ates cpiemasks ncn T= ANTS nim Denke Brepeeer errr Pawcien Sat p ide imeiy ATMA RM Hehe tge ed ares Sheen pcre ony cites RA NLP Re yeaa MTS ENT Seren a Peateineon et EEE NEY MATa wba dew feona tt ea The eh Rtn DT ne an BATE aN MUTA ANP deren essen ANnet mote Mae AD INN Ze TENTS TE 2 ANTM ANY hema te ait teat tee SPAN en gta Hea awten 7s er senegthan ee AIRS Te Rossel Ce wes nied eet fee wet pe retin cocio ADC Ee EDDIE aes cawetrees me ots eA EET tll Ee A ero n eee ee aa Cope ncmn Aerseiei si tH et etpas morris ie cla cuyt VR) S ASE oe oaks pacmuuemeraubatngsoremn «et esd acheinees CAND MATES EAT IM AN ITAA peqweren tent neraro roar oe eM nase kN eT 8 Ot pase ig a tA eh AEE OE beens MEAN EN eee bas prices MRM ne tanieretal tatu ae en AT Meee 6 AUNT Na pee eet At Oot pee anol peer nee ees paterintend co ip e e nt POON i pe ene ae ae era rere ne dwn ea eae g sie mee van TEY Oe Dhan Shoat ints pk ain PEctmsvatnsleh annem tna a Foren ome ASSMRERA AU LE Piacente esi Tene S marke ° _ 5S Ey, \—-} (eo) Bo = w . = a Oo > > Ly, = a = ‘Ly, = 5 WWE a ie Uf ly m NA es YVUGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHIIW Lae ae 9 ee 0 Z z = = = wy, = Zz = z = Wy = 2 as Yi: e 3 : i HL? = 2 E 2,%f fC OF > — = a “y 2 = » = = See eee LL cia elena Mia abet glee SMITHSONIAI z S tn uJ n J ta = EON = 4 = Yn = “EN NS co = a er fe) BS ae 5 ca. IMVUdIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIW sa1uvugin INSTITUTION INSTITUTION We NS Saiuvddi saiuvudl i » TITUTION NOILOALILSNI NVINOSHLINS S3ZIYVYEIT_LIBRARIES, SMITHSON! ae = = < aS KE = = Satay uf bj z XS at BES BE (@) x= SW ty (@) Qe. G Se SW 8 = bd 2 XX” 8 ‘ Z = = Wy & 2 . > Z = > = ae é, Zz (7p) ees (ep) we ee iyvUug ae LIBRARI ES SMITHSONIAN INSTITUTION NOl Nl ee 2 af : - a A ae Ho BE yg EP Vey a zB + = ae. UY = ac a é we z : a 4 x Oy ar & 5 a “YW & = = ze aay Fa = = Ne STITUTION ES NOILNLILSNI_NVINOSHLIWS oS {yvVuYd!i1 LIBRARI ES = rs 2 w ° = Oo ey 2 > 2 > > of fen EY a Ma /, i : : 2 A z i z is z 1yuvuald a | B RARI ES SMITHSONIAN INSTITUTION NOILALILSNI Bh moe < ae = ee = = = 2 8 a 5 = . 8 recep} yen) w ame af ep} == O 5a wie} eS 5 Ie z Fe a Ee = SS > = >’ = n i. 2 a > 2 7) STITUTION NOILALILSNI NVINOSHLIWS S3JIYVYUEIT LIBRARIES SMITHSONI ef = uw 2 a ~ —_ [aa = oe < a 4 < [oe = [om S lo’ eo iS a 3S an oy 2 ay z a ! JIYVUGIT LIBRARIES SMITHSONIAN INSTITUTION NOILOLILSNI NVINOSHLIV (> = fer: ae et Zz [ am ‘a 1g!11 LIBRARIES SMITHSONIAN S3Iu¥VvVual it J yay INSTITUTION INSTITUTION saruvuaty INSTITUTION saiuvugin INSTITUTION NOILALILSNI NVINOSHLIWS S: Z ~~. 5) Zz: 2) Ze, n Uy. 5 a z = Wy nee Sj We 3 NN G 9 & GBS : Ne fe) ae We) Uf fe. So fe) Se. 2 = ZH ike Zy ae 3 = S 3 eo NVINOS REINS _LIBRARIES SMITHSONIAN IN in ES w & WW & os rn oc = ia ae fad ha re + ae — a S oO Oo oO = co. = = Bs: re) a fe) a Zz ay Zz am z 4911 LIBRARIES SMITHSONIAN INSTITUTION NOILOLILSNI NVINOSHLINS S$ i = ~ z rc = = oO eas oO == O a = ee = roa 5 = Ey z S z E a a a - 2 - rm ee m Zz m 2 — — Ww » — TION NOILALILSNI NVINOSHLINS S3IYVYSIT LIBRARI ES SMITHSONIAN if wee = =i = (oz) e) = > 2 wn z . Es 2 = Q = As < x = Zz 4 Ky yn = B 2 9 5 GY 3 S S = SGEE = WW E s 2 = = Gy = = R. > Z = > = = - Fa 7p) Fa Ww) 5 Tp HT LIBRARI ES SMITHSONIAN INSTITUTION NOILALILSNI_ NVINOSHLIWS S > & 2 1a ES oe = oc. = ~ a oe e Soe c = =a s ar oc = o 5 oc o = o 23 o = Z ay ‘ Zz as Zz ee, TION NOILALILSNI NVINOSHLINS S3!1YVUGIT LIBRARIES SMITHSONIAN II Zz ile z Tes z oc ° = S = S) ye & iE = = = 5 GHG >» Ez > = > = tg “, Zz m N 2 m oad es m z ® as 2 (ap) = n HG IT _LIBRAR! ES SMITHSONIAN INSTITUTION NOILALILSNI NWINOSHLIWS |S a = 4 = 3 oO Ele ag KO 2p) Se Y oO ae ro) oO = = 2 zy > s >’ >" we Pd ” oe Zz Zz NOILNLILSNI NVINOSHLINS SSIYVYEIT LIBRARIES it SMITHSONIAN Vi LIBRARIES SMITHSONIAN LIBRARIES SMITHSONIAN LIBRARIES NOILNLILSNI \ WN SS NOILALILSNI NOILNLILSNI di? LIBRARIES SMITHSONIAN _{NSTITUTION NOILNLILSNI NVINOSHLINS S Hy pe —_ —_ | ont Se PROCEEDINGS of the Biological Society of Washington VOLUME 88 1975-1976 WASHINGTON PRINTED FOR THE SOCIETY EDITOR AusTIN B. WILLIAMS ASSOCIATE EDITORS Classical Languages Invertebrates GErEorGE C. STEYSKAL THomas E. BOWMAN Insects Vertebrates RoBERT D. GORDON WILLIAM R. TAYLOR 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 1975-1976 OFFICERS President CLYDE J. JONES Vice President CLYDE F. E. ROPER Secretary W. DUANE HOPE Treasurer OLIVER S. FLINT Custodian of Publications JOHN H. MILES COUNCIL Elected Members BRUCE B. COLLETTE MAUREEN E. DOWNEY RICHARD S. COWAN HORTON H. HOBBS, JR. PAUL J. SPANGLER tit TABLE OF CONTENTS Volume 88 Abele, Lawrence G. A new species of freshwater shrimp (genus Atya) from the Pacific drainages of Panama ____- 51-58 Allen, Gerald R., and Roger Lubbock. Glyphidodontops springeri, a new species of damselfish (Pomacentridae) from the Philippine and Mollucca Islands __._.-__»_»_______ _ 345-350 Barnard, J. Laurens, and Margaret M. Drummond. Clarifi- cation of five genera of Phoxocephalidae (marine Amphip- OLA) fea 2h LS A eS Se lc cee A ch) 515-548 Baumann, Richard W. A report on the Fifth International Sym- posium on Plecoptera Bogan, Michael, and Clyde Jones. Observations on Lepus callotisin, New Mexico: 2 eee eee 45-50 Bouchard, Raymond W., and Judith W. Bouchard. Orco- nectes saxatilis, a new species of crayfish from eastern Okla- homays (sate sh So Le ed en hos tae 439-446 Bouchard, Raymond W., and Judith W. Bouchard. Orco- nectes etnieri, a new species of crayfish from western Ten- nessee and northern Mississippi with notes on Procambarus ablusus and Orconectes wrightt 459-468 Bowman, Thomas E., and Charlotte Holmquist. Asellus (Asellus) alaskensis, n. sp., the first Alaskan Asellus, with re- marks on its Asian affinities (Crustacea: Isopoda: Asellidae) 59-72 Bowman, Thomas E., and Glenn Longley. Redescription and assignment to the new genus Lirceolus of the Texas trog- lobitic water slater, Asellus smithii (Ulrich) (Crustacea: Tsopoda: Asellidae))) 222-3 eee oe ee ei 489-496 Bradley, J. Chester. Dates of publication of Westwood’s Arcana’ Eintomolovica 2222... ae eee eee ee 91-94 Chace, Fenner A., Jr. Cave shrimps (Decapoda: Caridea) from) the Dominican’ Republic = = = eae _ 29-44 Child, C. Allan. The Pycnogonida types of William A. Hilton. f WBhoxichilidiidae: 0. ee ee 189-210 Crabill, R. E., Jr. A new Watophilus from Utah, including a list of all known species (Chilopoda: Geophilomorpha: Ghilenophilidae))\})o:..0 25 12 ov) See 395-398 Cressey, Roger F. Shiinoa elagata, a new species of parasitic copepod (Cyclopoida) from Elagatus (Carangidae) __.. 433-438 Dawson, C. E., and J. E. Randall. Notes on Indo-Pacific pipe- fishes (Pisces: Syngnathidae) with description of two new SPECIES!) Re uo gia EE ee te 263-280 Deevey, Georgiana B. Two new species of Bathyconchoecia (Myodocopa Halocyprididae) from the Caribbean Sea ___ 141-158 Etnier, David A. Percina (Imostoma) tanasi, a new percid fish from the Little Tennessee River, Tennessee _________--__.------- 469-488 Ferrari, Frank D. Taxonomic notes of the genus Oncaea (Copepoda: Cyclopoida) from the Gulf of Mexico and north- ErnnC@arin beanqoca =e mean eR I oth Oh fol 217-232 Fink, William L. A new genus and species of characid fish from the Bayano River basin, Panama (Pisces: Cyprini- HCO ERT ES) ) ae enema Nas el ta en! EIEN SINE IPE EI) 331-344 Fraser, Thomas H., and John E. Randall. Two new Indo-West Pacific cardinalfishes of the genus Apogon __....--..--___. 503-508 Gore, Robert H., and Linda J. Becker. Studies on stomatopod Crustacea of the Indian River region of Florida. I. Redis- covery and extension of range of Heterosquilla armata CES rine ees Ss EMR dete a GI Cul ea oe ES ar el et ei 21-28 Harman, Walter J. A new species of Stephensoniana (Oligo- chaeta: Naididae) from North America —_------___. 1-4 Harman, Walter J., and Michael L. McMahan. A reevaluation of Pristina longiseta (Oligochaeta: Naididae) in North ENOTES EGE po cag a Aga ep a te NS 2 TUR oe 167-178 Heyer, W. Ronald. Adenomera lutzi (Amphibia: Leptodacty]- idae), a new species of frog from Guyana 315-318 Hobbs, Horton H., Jr., and Margaret Walton. New ento- cytherid ostracods from Tennessee with a key to the species rate GENUS PASCELOCULNETE 8 5-20 Hoffman, Richard L. A new genus and species of euryurid millipeds from Chiapas (Polydesmida: Platyrhacidae) ___. 211-216 Kenk, Roman. Freshwater triclads (Turbellaria) of North America. VIII. Dugesia arizonensis, new species __..- 113-120 Knutson, Lloyd. Key to subfamilies of robber flies based on pupal cases, with a description of the pupal case of Doryclus distendens (Asilidae: Megapodinae) —.- 509-514 Kornicker, Louis S., and Marcia Bowen. Sarsiella ozotothrix, a new species of marine Ostracoda (Myodocopina) from the At- lantic and Gulf coasts of North America 497-502 Lynch, John D. A new high Andean slope species of Eleuthero- dactylus (Amphibia: Leptodactylidae) from Colombia and ISSA Cl Grp reco Nc a eh aL Ans eR tol a eA 351-354 Manning, Raymond B. Eurysquilla pacifica, a new stomatopod chustacean) som News Britain ees le ee 249-252 Manning, Raymond B. Two new species of the Indo-West Pacific genus Chorisquilla (Crustacea, Stomatopoda), with moteston Cz-excavata: (Miers) es ee ee 253-262 Oliver, Michael K. Labidochromis textilis, a new cichlid fish (Teleostei: Cichlidae) from Lake Malawi 319-330 Olson, Storrs L. A new species of Milvago from Hispaniola, with notes on other fossil caracaras from the West Indies (Aves: "Halconidae)) 2.20 al Eh Nae ee ea 355-366 Pettibone, Marian H. Review of the genus Hermenia, with a description of a new species (Polychaeta: Polynoidae: Tepidonotinae,) (2s Ni AON I U0 Oe 233-248 Ray, Clayton E. The relationships of Hemicaulodon effodiens Cope 1869 (Mammalia: Odobenidae) —-.-- 281-304 Rickner, Jack A. Notes on members of the family Porcellanidae (Crustacea: Anomura) collected on the east coast of Mexico 159-166 St. John, Harold. More variants of Scaevola taccada (Gooden- iaceae) Hawaiian plant studies 43 ___..________ 73-16 Schwartz, Albert, and Ronald I. Crombie. A new species of the _genus Aristelliger (Sauria: Gekkonidae) from the Caicos MssT cpa AS? Nese C0 C6 SN ec 305-314 Schwartz, Albert, and Richard Franz. A new species of Sphaerodactylus (Sauria: Gekkonidae) from Hispanola ___ 367-372 Schwartz, Albert, and Orlando H. Garrido. A reconsideration of some Cuban Tropidophis (Serpentes, Boidae) 77-90 Segun, A. O. Two new genera of eudrilid earthworms from INVES reLe ole SDR) OO OU UN Ma ee 383-394 Shelley, Rowland M. The identity of Nannaria conservata Chamberlin, with notes on an abnormal male and descriptions _of two new species of Nannaria from North Carolina ( Diplop- oda: Polydesmida: Xystodesmidae) __....._________ 179-188 Shelley, Rowland M. Two new diplopods of the genus Poly- zonium from North Carolina, with records of established spe- cies (Polyzoniida: Polyzoniidae) __.______-____ 373-382 Thomas, Lowell P. The ophiacanthid genus Amphilimna (@phturcideat) Echinodermata), 22 ee eae 127-140 Voss, Gilbert L. Two new species of octopods of the genus Graneledone (Mollusca: Cephalopoda) from the Southern OGeam Fe) eRer es AMVAEE Mala IG ENE EE SMC AER, See 447-458 Watling, Les. Parametopella inquilinus, new species from Dela- ware Bay oyster beds (Amphipoda: Stenothoidae) ___-_-. 429-432 Watling, Les, and Don Maurer. Chiridotea stenops Menzies and Frankenberg, a juvenile of C. arenicola Wigley (Crusta- CEA SIO) Meee eae No NN an a NE a eRe Oa 121-126 Wetzel, Ralph M. The species of Tamandua Gray (Edentata, Mymmecophagidae)) Uviuiiit stan aii en vee 2 aCe 95-112 v1 BIOLOGICAL SOCIETY OF WASHINGTON PROCEEDINGS 1049th Meeting—12 March 1975 NINETY-SIXTH ANNUAL MEETING The meeting was called to order by President Thomas E. Bowman at 2:05 p.m. in the Baird Auditorium of the National Museum Natural History. Twenty-six members attended. Reading of the minutes of the previous annual meeting was dispensed with, since they were printed in the Biological Society of Washington Proceedings. Treasurer Oliver Flint read the Financial Report and Budget for 1975 (attached copy). He pointed out that only $1500 of the original $7000 borrowed from the endowment fund remains to be returned, and that some of the endowment fund has been used to purchase high-interest certificates. The report was accepted by motion. Mr. C. A. Child, member of the Financial Committee, reported that the Financial Report had been audited and found to be correct. President Bowman reported that the Council for the Society had held two meetings during the year. In one of the meetings it had been neces- sary to raise the dues for membership in the Society from $6.00 to $7.00, and to increase slightly the charges for reprints of papers published in the Proceedings. Austin Williams, Chairman of the Editorial Committee, presented a report stating that Volume 87 of the Proceedings for 1974 consisted of 43 papers totaling 500 pages. Twenty-eight of these were paid, four partially paid and three were unpaid. The first issue of Volume 88, con- sisting of 11 papers totaling 112 pages, is now in galley proof. Seven of these papers are paid, one partially paid, and three unpaid. Sixteen manuscripts have been received and are under review. Of these, three are paid, seven unpaid, and six have an uncertain status still to be nego- tiated. In Volume 87 of 1974, biological groupings of papers published were: botany 1; insects 7; invertebrates 18 (crustaceans 11, other 7); vertebrates 17 (fishes 6, amphibians 4, reptiles 3, fossil birds 1, mam- mals 2). Secretary Louis S. Kornicker gave results of the election for Officers and Council members for 1975: President, Clyde J. Jones; Vice President, Clyde F. E. Roper; Secretary, W. Duane Hope; Treasurer, Oliver S. Flint; Members of the Council: Bruce B. Collette, Richard S. Cowan, Maureen E. Downey, Horton H. Hobbs, Jr., Paul J. Spangler. President Bowman then turned over the meeting to the incoming vii Secretary, W. Duane Hope. The incoming President and Vice-President were unable to attend the meeting. Meredith L. Jones made a motion thanking the Officers and Council members of the Society for their efforts on the behalf of the Society for the preceding year. The motion was seconded and passed. A motion te adjourn at 2:50 p.m. was made, seconded and passed. Louis S. Kornicker Secretary vii eh . 1, pp. 1-4 23 April 1975 NH PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW SPECIES OF STEPHENSONIANA (OLIGOCHAETA: NAIDIDAE) FROM NORTH AMERICA By WALTER J. HARMAN Department of Zoology, Louisiana State University, Baton Rouge, Louisiana 70803 Aiyer (1926:139) described Naidium (?) trivandranum as a new species from soft mud in an old tank in Trivandrum, India. He was able to culture the species and to get some of the specimens to reach sexual maturity. As a result, in 1929, he published an account of its reproductive biology and recog- nized the generic distinctiveness of the species. In that paper (1929:27), he described Stephensonia as a new genus and placed S. trivandrana in it as the only species. Cernosvitov (1938:536), in his report on the oligochaetes of Palestine, found S. trivandrana in the waters of Lake Huleh. He correctly pointed out (1938:539) that Stephensonia had previously been used in the Aphidae and had priority. He therefore proposed the name Stephensoniana. Brinkhurst (1966:142) recorded S. trivandrana from small coastal rivers in the Republic of South Africa. A generic description, slightly modified in view of the new species, is presented. Stephensoniana Cernosvitov, 1938 No eyes. No proboscis. Dorsal setae beginning in II. Hairs and needles, either simple or bifid. Ventral setae all of one type. No coelomocytes. Foreign matter adhering to cuticle. Stephensoniana tandyi n. sp. Distribution: Louisiana; Catahoula Parish. 2.4 mi. NE of the parish line on Louisiana Highway 28. 26 March 1966. Known only from the type locality. Holotype: U.S. National Museum Number 45481 (1 specimen). I—Proc. Biot. Soc. WasH., Vou. 88, 1975 (1) 2 Proceedings of the Biological Society of Washington A B @Q Fic. 1. Stephensoniana tandyi n. sp.: A, Ventral seta of VI; B, Dorsal needle seta of X. Paratypes: LSU Number 746 (8 specimens). Description: Dorsal setae beginning in II, containing 1 or 2 hairs and 1 bifid needle (Fig. 1). Ventral setae 1 or occasionally 2 per bundle (Fig. 1). Posterior one third of body without setae. Clitellum in VIT-IX. Foreign matter adhering to cuticle. Discussion: This species is considered as a member of Stephensoniana and not Pristina because of the adhering foreign matter, the single ventral seta that makes up most of the ventral bundles, and the little differentia- tion between anterior and posterior setal morphology. It is distinct from S. trivandrana in its possession of bifid needles, fewer setae in all bundles, and in the presence of so great an asetigerous zone at the posterior of the body. Often, the coat of foreign matter made it difficult or impossible to see the proximal ends of the setae in freshly prepared material. The points of insertion of setae in these specimens, mounted in synthetic resin, did not become apparent until some months after preparation. Hair setae, 1 or occasionally 2 per bundle, are shorter anteriorly and posteriorly than in midbody. In II, hair setae range from 107 to 124 yp; in VII, 115-186 »; in XVI, 44-100 u. The hairs do not appear to be serrated and are fine in structure, about 1 uw in diameter. Needle setae are nearly straight and have a very inconspicuous nodulus. They are only slightly more than 1 yu in diameter. The bifid condition is new to the genus. The distal tooth is about 4,4 in length, twice as long as the proximal, and is slightly thicker. Needles are 35 to 74u long, longer in midbody than at either end. New oligochaete from North America 3 Ventral setae, 50 to 81, resemble the needles. A slightly distal nodulus is inconspicuous but visible in II and III. If a nodulus is present in more posterior setae, it is not visible because of the coat of foreign matter. One can be certain that there is not a distal nodulus because one-half of each seta is extended from the body wall. There is no con- spicuous difference in shape of seta or tooth structure from anterior to posterior. The distal tooth is over 4 long and twice as long as the proximal, A clitellum is judged to be present in one specimen because of a great change in body diameter. This enlargement covers part or all of VII, VIII, and IX. The diameter of the body in IV is 112 uw, changing in VII to 1874. The coat of foreign matter keeps one from seeing all cell structure to confirm the clitellum. Neither could internal anatomy be observed. A characteristic of S. tandyi is the asetigerous posterior third of the body. Four specimens are available for study. Segments VII, XI, XVI, and XVII contained the last setae. Asetigerous portions of the body were 663 u, 649 u, 687 uw, and 374 u in length, respectively. The greatest dis- tance between setal bundles in midbody was 604. Assuming the same segmental size for the posterior of the body, there would be from 6 to 11 asetigerous segments. This would be a minimal number, because the pre-periproct segments frequently are small. Therefore, the range of number of segments per worm conservatively is estimated to be 17-28. _ The habitat of S. tandyi is a low and swampy portion of Catahoula Parish, Louisiana. It is likely that the species is more widely distributed than the collecting data would indicate. There are many intermittent streams present, and surface water may cover much land area following rains, allowing easy distribution between drainage systems, but resulting in discontinuous distribution during the dry season. S. tandyi adds a second species to the previously monotypic genus and establishes it in the fauna of North America. It is a pleasure to name this species for Dr. Richard E. Tandy who collected it during the study of other forms of Oligochaeta. Dr. John P. O’Neill drew the setal illustrations. LITERATURE CITED Arver, K. S. P. 1926. Notes on the aquatic Oligochaeta of Tranvan- core. II. Ann. Mag. Nat. Hist. Ser. 9. 18:131-142. —. 1929. An account of the Oligochaeta of Travancore. Rec. Ind. Mus. 31:13—76. Brinkuurst, R. O. 1966. A contribution towards a revision of the aquatic Oligochaeta of Africa. Zool. Afr. 2:131-166. CeRnosvitov, L. 1938. The Oligochaeta, p. 585-550. In R. Wash- bourn and R. F. Jones, Report of the Percy Sladen Expedition to Lake Huleh; a contribution for the study of the fresh waters of Palestine. Ann. Mag. Nat. Hist. Ser. 11. 2:517-560. 4 Proceedings of the Biological Society of Washington MUN Ls j GH i re toate a ae ah A | fee 2, pp. 5-20 23 April 1975 BAY NH PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON NEW ENTOCYTHERID OSTRACODS FROM TENNESSEE WITH A KEY TO THE SPECIES OF THE GENUS ASCETOCYTHERE By Horton H. Hosss, JR., AND MARGARET WALTON Smithsonian Institution, Washington, D.C. 20560 and Mountain Lake Biological Station, Pembroke, Virginia 24136 During a survey of the crayfishes of Tennessee, Raymond W. Bouchard (at present Postdoctoral Fellow, Smithsonian Institution) saved the sediments from the containers that he Was using to preserve the specimens caught, and these were donated to us. From these sediments, the branchiobdellids (now in the collection of Perry C. Holt, Virginia Polytechnic Institute and State University), entocytherids, and copepods were retrieved, and the descriptions of new entocytherids in- cluded here are largely based on specimens obtained by Dr. Bouchard. Additional material was contributed by Dr. Holt. Included among the descriptions herein are four new mem- bers of the genus Ascetocythere (together with a key to the 15 members of the genus), two species of the genus Dactylo- cythere, and one species assigned to a new genus, Psittocythere. We are most grateful to Drs. Bouchard and Holt for donat- ing these specimens to us, and we wish to thank Fenner A. Chace, Jr., Margaret A. Daniel, and C. W. Hart, Jr., for their criticisms of the manuscript. Ascetocythere Hart, 1962 In order to encompass the new species, Ascetocythere veruta, the diagnosis of the genus offered by Hobbs and Hobbs (1970:3) must be emended as follows (The altered portion is in italics.): “Terminal tooth of mandible pectinate. Male copulatory complex without finger guard; peniferum extending ventrally much beyond clasping apparatus, elongate, comparatively slender, with subterminal bulbous enlargement bearing one to several projections; penis with gap between dorsal spermatic and 2—Proc. Biot. Soc. WasH., Vou. 88, 1975 (5) 6 Proceedings of the Biological Society of Washington ventral prostatic elements along part of its length except in A. veruta, and distal portion directed anteriorly, anteroventrally, or posteroventrally; both elements always shorter than clasping apparatus; accessory groove lacking. Clasping apparatus well developed and may or may not be clearly divisible into vertical and horizontal rami; external border of horizontal ramus entire, internal border with 2, 3, or no teeth along apical half, if present, often grouped far distally with 3 apical denticles. Triunguis female lacking pectinate process on distal podomere of second antenna; genital complex consisting of genital papilla but lacking J-shaped rod and amiculum.” Key TO THE SPECIES OF THE GENUS ASCETOCYTHERE (Modified from Hobbs and Hobbs, 1970:3) Penis with contiguous spermatic and prostatic elements ___ Pe SME ING NWR aia ROLE as ea Oe A. veruta, new species Penis with gap between spermatic and prostatic elements along “part oF ats leneth <2 222 ee ee 2 Preaxial (internal) border of distal half of clasping apparatus with one or more teeth situated proximal to apical denticles, or preaxial border unarmed (Asceta Group) —_______-._ 3 Preaxial (internal) border of clasping apparatus with teeth and apical denticles grouped distally (Coryphodes Group) 14 Bulbous portion of peniferum with well-developed anterior DEO COS atte RC PU 8 4 Bulbous portion of peniferum lacking anterior process _____ 12 Ventral portion of peniferum with angular flange ____ 5 Ventral portion of peniferum lacking angular flange ______ 7 Flange on ventral portion of peniferum subtriangular _______ pe cee ah is sb mean Conti, deepal A. bouchardi, new species Flange on ventral portion of peniferum subrectangular ____ 6 Ventral extremity of peniferum with fingerlike projection extending ventrally beyond angular flange Diket SELF a ene A. asceta (Hobbs and Walton, 1962:39) Ventral extremity of peniferum lacking fingerlike projection eM ee i ee A. ozalea Hobbs and Hart, 1966:40 Veniral portion of peniferum with posterior process _________ 8 Ventral portion of peniferum lacking posterior process 10 Anterior process triangular, broadly fused basally with ventral DIOGESS = 22 ke ae ee A. triangulata, new species Anterior process slender and elongate, distinctly set off from digitiform» ventral.2process 22 40 eae i ee 9 Anterior process of peniferum much shorter than ventral (digitiform) process and situated on level above base of DCTS Reet ile. Aba steele Sere A. sclera Hobbs and Hart, 1966:42 Anterior process of peniferum as long as ventral (digitiform ) process and situated on level below base of penis ________ eines Ae ree ee A. holti Hobbs and Walton, 1970:853 New ostracods from Tennessee 7 10(7’) Peniferum with tablike penis guide extending anterodorsally fomemesialuside of ventral process 222 ee cM CRE NO oes A. pseudolita, new species 10’ Peniferum lacking tablike penis guide —. Il 11(10’) Postaxial (external) border of clasping apparatus with prox- imal bend at end of proximal fifth of apparatus; penis reaching posterior surface of digitiform process —_- POG AON iste MESS es UE A. lita Hobbs and Hobbs, 1970:4 ie Postaxial (external) border of clasping appartus with proxi- mal bend at end of proximal third of apparatus; penis extending along anterior side of digitiform process 12(3’) Weasel portion of peniferum lacking angular flange 12% Ventral portion of peniferum with angular flange _______ 13 13(12’) Ventral portion of peniferum with distinct curved digitiform (ventral) process extending posteroventrally from flange jit Sie Merete PEARED Maen ne eee A. hoffmani Hobbs and Hart, 1966:40 13’ Ventral portion of peniferum lacking distinct digitiform (ven- tral!) process) A. hyperoche Hobbs and Hart, 1966:41 14(2’) Anterior surface of subterminal or terminal enlargement of peniferum with anteroventrally projecting, clublike prom- inence and two additional prominences —_____-_______- htc eet DE BA eae A. cosmeta Hobbs and Hart, 1966:46 14’ Anterior surface of terminal enlargement of peniferum with single prominent snoutlike projection _..--___»___ 15 15(14’) Snoutlike prominence less than half as long as longitudinal diameter of peniferum at level of prominence LUN eee ates Br te A. myxoides Hobbs and Hart, 1966:45 UB Snoutlike prominence more than two-thirds as long as longi- tudinal diameter of peniferum at level of prominence __ AE ee COON eae Oe A. coryphodes Hobbs and Hart, 1966:44 Ascetocythere bouchardi, new species (Figure la—d) Male: Eye pigmented (not clearly defined in holotype) and located approximately one-fourth shell length from anterior margin. Shell (Fig. 1b) ovate with greatest height one-third length from posterior end where about 1.3 times height at level of eye; margin entire but with slight broad impression ventrally just posterior to level of eye. Submar- ginal setae present except dorsally, somewhat evenly spaced ventrally but closer together anterodorsally and posteroventrally. Copulatory complex (Fig. la) with bulbous portion of peniferum bearing following processes: anterior process short (length about one- sixth diameter of peniferum at base), thick, rounded distally, and directed anteriorly; ventral process very complex, consisting of broad winglike 8 Proceedings of the Biological Society of Washington Fic. la—d, Ascetocythere bouchardi, new species; le-g, Ascetocythere pseudolita, new species; 1h—k, Ascetocythere triangulata, new species; ll-n, Ascetocythere veruta, new species. a, f, j, 1, coplatory complex of male; b, e, h, m, dextral view of shell of male; c, g, k, n, clasping ap- paratus of male; d, i, dextral view of shell of triunguis female. Scales In mm. New ostracods from Tennessee 9 lobe supporting terminal anteroventral digitiform, ventromesially directed projection and small flange projecting anteromesially from cephalic base of process; and posterior process consisting of short subrectangular lobe on posteroventral angle of bulbous area. Penis very prominent, C-shaped, with apex reaching posterodorsal base of flange of ventral process. Clasping apparatus (Fig. lc) not reaching level of midlength of bulge of peniferum, L-shaped, with margins of very short, broad, tapering vertical ramus entire; horizontal ramus with almost uniform diameter along proximal third, tapering in middle third, and with subparallel dorsal and ventral margins in distal third; internal border of distal half with 8 equally spaced teeth, and apex with 3 terminal denticles. Dorsal finger moderately heavy and about one-half as long as slender ventral finger; latter slender and evenly curved posteroventrally almost from base. Triunguis Female: Eye pigmented and situated approximately one- fifth shell length from anterior margin. Shell (Fig. 1d) subpyriform, highest slightly posterior to midlength, greatest height approximately 1.5 times height at level of eye. Submarginal setae disposed as in male. Genital complex consisting of short posteroventrally directed papilla with heterogeneous mass adhering to apical part. Measurements (in mm): 5 males and 5 females. Holotype Males Allotype Females Length (range) 0.43 0.438—-0.46 0.42 0.42-0.47 Average 0.44 0.44 Height (range) 0.25 0.25-0.27 0.27 0.25-0.28 Average 0.26 0.26 Type-locality: Soddy Creek off County Road 5306, southwest of Hendon, approximately 0.3 mile north of the Sequatchie County line, Bledsoe County, Tennessee, where it was infesting the stream-dwelling Cambarus (Jugicambarus) parvoculus Hobbs and Shoup, 1947, and Cambarus ( Jugicambarus) sp. nov. Disposition of Types: The holotypic male and allotypic female are deposited in the National Museum of Natural History (Smithsonian Institution), numbers 150640 and 150641, respectively. Paratypes are in the collection of H. H. Hobbs III and the Smithsonian Institution. Range and Specimens Examined: Tweny-two specimens from the fol- lowing localities in Tennessee: Type-locality; Renfro Creek at USS. Highway 70, northwest of Ozone, Cumberland County; Duskin Creek at County Road 5881, northeast of Milo, Bledsoe County; and Cain Creek, tributary of North Chickamauga Creek, off road between County Road 4264 and U.S. Highway 27, southeast of Dunlap, Sequatchie County. Hosts: Cambarus (Jugicambarus) parvoculus Hobbs and Shoup, 1947, Cambarus (Hiaticambarus) longirostris Faxon, 1885, and Cambarus (Jugicambarus) sp. nov. 10 Proceedings of the Biological Society of Washington Entocytherid Associates: Dactylocythere pachysphyrata Hobbs and Walton, 1966, Donnaldsoncythere hiwasseensis (Hobbs and Walton, 1961). Relationships: Ascetocythere bouchardi has its closest affinities with Ascetocythere asceta (Hobbs and Walton, 1962) and, more distantly, is related to A. hoffmani Hobbs and Hart, 1966, and A. ozalea Hobbs and Hart, 1966. The flattened flange on the ventral end of the peniferum, the caudally curved penis, and 8 teeth on the preaxial border of the horizontal ramus of the clasping apparatus will distinguish A. bouchardi from all of its congeners. Etymology: This ostracod is named in honor of its discoverer, Ray- mond W. Bouchard, who also collected and donated to us most of the specimens on which this report is based. Ascetocythere pseudolita, new species (Figure le-g) Male: Eye pigmented and located one-fourth shell length from an- terior margin. Shell (Fig. le) egg-shaped, with greatest height slightly posterior to midlength and 1.4 times height at level of eye. Margin entire but with faint broad ventral excavation short distance anterior to midlength. Submarginal setae present except posterodorsally; setae slightly closer together anteriorly and posteriorly than ventrally. Copulatory complex (Fig. 1f) with usual ventral bulbous area bearing anteriorly directed, slender, slightly curved anterior process; ventral sur- face of bulbous area produced in conical, gently procurved ventral process bearing tablike penis guide mesially. Penis also curved and following contour of mesial longitudinal groove almost to apex of process. Clasping apparatus (Fig. 1g) not nearly reaching ventral margin of peniferum, not clearly divisible into horizontal and vertical rami but with subangular bend proximally and gently tapering distally; preaxial margin bearing 1 small tooth at base of distal third and 2 additional ones evenly spaced between it and 8 apical denticles. Dorsal finger slightly less than half as long as gently curved ventral finger, latter with bend similar although not subparallel to contour of clasping apparatus. Female: Unknown. Measurements: Holotypic male (only known specimen) 0.41 mm long and 0.24 mm high. Type-locality: Waurel Branch off U.S. Highway 25W, northwest of La Follette, Campbell County, Tennessee, where it was found in com- pany with Donnaldsoncythere hiwasseensis (Hobbs and Walton, 1961) and Dactylocythere spinata Hobbs and Walton, 1970. These were re- trieved from a collection of crayfishes containing Cambarus (Jugi- cambarus) distans Rhoades, 1944, and Cambarus (Puncticambarus) buntingi Bouchard, 1978. Disposition of Types: The holotypic male is deposited in the National Museum of Natural History (Smithsonian Institution), number 150642. New ostracods from Tennessee Il Range and Specimens Examined: The holotype is the only known specimen. Hosts and Entocytherid Associates: See “Type-locality.” Relationships: Ascetocythere pseudolita, as its name suggests, shares more in common with Ascetocythere lita Hobbs and Hobbs, 1970, than with any other species. The most striking similarities exist in the terminal elements of the peniferum and the armature of the clasping apparatus. The most conspicuous differences are in the curvature of the clasping apparatus which approaches that of A. asceta (Hobbs and Walton, 1962) and in the presence of the tablike penis guide directing the penis toward the distal extremity of the ventral process rather than posteroventrally. Ascetocythere triangulata, new species (Figure 1h-k) Male: Eye pigmented and located about one-fourth shell length from anterior margin. Shell (Fig. 1h) subovate with greatest height about one-third shell length from posterior margin, there about 1.8 times height at level of eye; ventral margin with shallow excavation anterior to mid- length. Submarginal setae present except dorsally; setae somewhat closer together anteriorly and posteriorly than ventrally. Copulatory complex (Fig. 1j) with ventral subbulbous area bearing prominent broadly based subtriangular cephalic process (whence the specific name), posteroventral margin of which poorly delimited from greatly reduced, slightly recurved ventral process; posterior process small, thumblike, procurved, and opposing posterior margin of ventral process. Penis U-shaped and emerging from body of peniferum between ventral and posterior processes. Clasping apparatus (Fig. 1k) not nearly reaching ventral extremity of peniferum and not divisible into horizontal and vertical rami; apparatus curved in proximal half, almost straight in distal half, and gently tapering almost from base to 3 terminal denticles; pre- axial margin bearing 3 teeth on distal third, with penultimate tooth closer to ultimate than to antepenultimate. Dorsal finger stouter than ventral and approximately half as long; ventral finger gently curved caudoventrally from base. Triunguis Female: Eyes situated as in male. Shell (Fig. 11) more highly vaulted than in male but with greatest height also about 1.3 times that at level of eye; concavity on ventral margin more strongly pro- nounced than in male. Submarginal setae disposed as in male. Genital complex consisting of single or bilobed prominence surrounded by heterogeneous, amorphous, slightly refractile mass situated postero- dorsally. Measurements (in mm): 2 males and 2 females. Holotype Male Allotype Female Length 0.41 0.42 0.41 0.41 Height 0.22 0.23 0.25 0.25 12 Proceedings of the Biological Society of Washington Type-locality: Pole Bridge Branch (Caney Fork River System), tributary to Lost Creek at County Road 4385, south of De Rossett, White County, Tennessee. There it was associated with Donnaldsoncythere hiwasseensis (Hobbs and Walton, 1961), Dactylocythere sp., and Entocythere illinoisensis Hoff, 1942. These ostracods were obtained from a collection of crayfishes containing Cambarus ( Depressicambarus ) sphenoides Hobbs, 1968, and C. (Jugicambarus) parvoculus Hobbs and Shoup, 1947. Disposition of Types: The holotypic male and allotypic female are deposited in the National Museum of Natural History (Smithsonian Institution), number 150643 and 150644, respectively, as are the para- typic male and female. Range and Specimens Examined: Known only from the type-locality. The four specimens in the type-series are the only known specimens. Hosts and Entocytherid Associates: See “Type-locality.” Relationships: Ascetocythere triangulata has its closest affinities with members of the Asceta Group of the genus, and shares more in common with A. holti Hobbs and Walton, 1970, than with any other species. It may be distinguished from all members of the genus by the anterior process of the peniferum which is subtriangular with its broad base situated on the anterodistal margin of the peniferum. Three teeth on the preaxial border of the clasping apparatus will also distinguish this species from A. holti. Ascetocythere veruta, new species (Figure 1/-n) Male: Eye pigmented and located slightly anterior to anterior fourth of shell length from anterior margin. Shell (Fig. lm) almost egg- shaped with greatest height short distance posterior to midlength where about 1.4 times higher than shell at level of eye; ventral margin with shallow concavity anterior to midlength. Submarginal setae present ex- cept dorsally and posterodorsally; setae only little closer together anterodorsally and posteroventrally than elsewhere. Copulatory complex (Fig. 11) with ventral subbulbous area lacking cephalic process but bearing thin shelflike obliquely vertical flange on cephalic margin, ventral part extending onto prominent posteroventrally curved ventral process; latter bifid distally with more posterodistal lobe supporting penis; posterior process consisting of short, strongly sclerotized, clawlike projection situated posterodorsal to ventral process and with apex directed anteriorly. Penis unique in genus, its prostatic and spermatic elements being contiguous rather than separated for some dis- tance; both elements emerging free from distal extremity of posterior lobe of bifid ventral process. Clasping apparatus (Fig. 1n) L-shaped with horizontal ramus noticeably slenderer (slightly tapering) than vertical ramus and bearing single tooth on preaxial surface slightly distal New ostracods from Tennessee 13 to midlength; ramus terminating in 4 apical denticles. Both dorsal and ventral fingers rather slender, former little more than half length of latter; ventral finger straight along proximal third and gently curved caudally to base of apical seta. Female: Unknown. Measurements: Holotypic male (only known specimen) 0.41 mm long and 0.24 mm high. Type-locality: Poplar Creek (Tennessee River System) on road be- tween Tennessee routes 115 and 61, in Laurel Grove, Anderson County, Tennessee, where it was found associated with Dactylocythere falcata (Hobbs and Walton, 1961), and Donnaldsoncythere hiwasseensis (Hobbs and Walton, 1961). It was infesting either Cambarus ( Hiaticambarus) longirostris Faxon, 1885, or C. (Jugicambarus) distans Rhoades, 1944. Disposition of Type: The holotypic male is deposited in the National Museum of Natural History (Smithsonian Institution), number 150645. Range and Specimens Examined: Known only from the type-locality; the holotype is the only specimen available. Host and Entocytherid Associates: See “Type-locality.” Relationships: Ascetocythere veruta is not closely allied to any known entocytherine. The general conformation of the peniferum resembles that of members of the Asceta Group of the genus more closely than those of other members of the subfamily, and the clasping apparatus is strikingly similar to that of Ascetocythere lita Hobbs and Hobbs, 1970, but differs in that only 1 tooth is present on the preaxial margin. The contiguous condition of the spermatic and prostatic elements of the penis, as stated above, is unique in the genus, as are the posteroventrally di- rected ventral process and the heavily sclerotized clawlike posterior process. The latter is suggestive of the posteroventral, sometimes movable, clawlike projection in the Microsyssitrinae and some Notocytherinae. (See Hart, Nair, and Hart, 1967, and Hart and Hart, 1967.) Dactylocythere Hart, 1962 The two new members of the genus Dactylocythere described below may be included in the key to the members of that genus provided by Hobbs and Hobbs (1970:6) by inserting the following: 12(7’) Dorsal margin of accessory groove not reaching beyond midlength of spermatic loop, 222 13a ox Dorsal margin of accessory groove reaching almost to, or slightly beyond, dorsal extremity of spermatic loop __.. 16 13a(12) Preaxial (internal) border of horizontal ramus of clasping apparatus with 2 or more teeth or emarginations; tip of penis not nearly reaching level of posteroventral ex- tremity) (Of peniterumi ee 2. 2 Os A ee 13 13a’ Preaxial (internal) border of horizontal ramus of clasping apparatus never with more than 1 (sometimes without) 14 Proceedings of the Biological Society of Washington tooth and it adjacent to apical denticles; tip of penis reaching level of posteroventral extremity of peniferum 30 30(13a’) Ventral surface of peniferum almost straight; peniferal groove conspicuous; anteroventral protuberance of peniterum lacking, = aes Dt. scissura, new species 30’ Ventral surface of peniferum curved; peniferal groove lack- ing; peniferum bearing anterovental, slightly tapering, truncate protuberance _________________- Dt. crena, new species Dactylocythere crena, new species (Figure 2a-f) Male: Eye pigmented, located approximately one-third to one-fourth shell length from anterior margin. Shell (Fig. 2d) oblong ovate, pro- duced posteroventrally in broad rounded lobe. Submarginal setae around entire perimeter, somewhat evenly dispersed except anterodorsally and posteroventrally where closer together. Sternal spine lacking. Copulatory complex (Fig. 2a) with rather short, somewhat tapering, distally bifid finger guard; peniferum robust with small, rounded postero- ventral lobe bearing delicate anteroventral and somewhat mesial sheath- like protuberance, latter broader at base than distally; peniferal groove not evident. Penis L-shaped with distal arm subequal in length to proximal arm, conspicuously longer than that of all other congeners ex- cept Dactylothere scissura (see below), and extending well into antero- ventral protuberance of peniferum. Accessory groove slender, extending posterodorsally from base of penis, variable in length but rarely reaching level of ventral extremity of spermatic loop. Clasping apparatus (Fig. 2c) extending ventrally beyond ventral extremity of peniferum, somewhat L-shaped with rami joined in rather abrupt curve; vertical ramus almost straight with both pre- and postaxial margins entire; horizontal ramus subequal in length to vertical ramus, very gently curved dorsally from base to apex, its margins entire, and terminating in 3 or 4 small denticles. Dorsal and ventral fingers slender, latter about twice length of former and with proximal third to half straight, distal portion curved gently caudally. Triunguis Female: Eye pigmented, situated slightly posterior to anterior fourth of shell. Shell (Fig. 2e) tapering anteriorly, 1.5 times as high posteriorly as at level of eye. Posterior margin of shell with conspicuous, often subrectangular, notch (from whence the name crena, L.=notch) height of which approximately one-fourth to one-fifth greatest height of shell. Genital complex (Fig. 2b) situated at fundus of notch consisting of vertically disposed sinus encompassing slightly curved hyalin rod (probably corresponding to “J-shaped rod” of most congeners ) in dorsal half and tubular cavity with posteroventral angle ventrally. No trace of amiculum present. New ostracods from Tennessee 15 Measurements (in mm): 10 males and 10 females. Holotype Males Allotype Females Length (range) 0.68 0.61—0.69 0.69 0.61-0.70 Average 0.67 0.65 Height (range) 0.35 0.33-0.37 0.34 0.34-0.42 Average 0.35 0.39 Type-locality: Field off County Road 2428, (Tennessee River Basin), southeast of Martel near Cedar Hills Golf Course, Loudon County, Tennessee, where this ostracod was infesting an undescribed burrowing crayfish, Cambarus (Depressicambarus) sp. Disposition of Types: The holotypic male and allotypic female are deposited in the National Museum of Natural History (Smithsonian Institution), number 150646 and 150647. Paratypes are in the collection of H. H. Hobbs III and the Smithsonian Institution. Range and Specimens Examined: Approximately 175 specimens from the type-locality and immediate vicinity (Tennessee River Basin). Hosts: Cambarus (Lacunicambarus) diogenes subsp. and Cambarus (Depressicambarus) sp. Entocytherid Associates: Uncinocythere simondsi Hobbs and Walton, 1960, Cymocythere gonia Hobbs and Hart, 1966, and Dactylocythere sp. Relationships: Dactylocythere crena has its closest affinities with Dt. scissura (described below). They share a longer ventral arm of the penis than other species of the genus, and the females of both possess a deep notch in the caudal margin of the shell. This species may be distinguished from Dt. scissura by its larger size, in possessing a rounded, as opposed to almost straight, anteroventral margin of the peniferum, and the penis opens into an anteroventral sheathlike projec- tion from the lobe. Dactylocythere scissura, new species (Figure 2g-k) Male: Eye pigmented, located slightly posterior to anterior fifth of shell. Shell (Fig. 2g) oblong ovate with slight broadly rounded postero- ventral bulge. Submarginal setae around entire perimeter of shell, some- what evenly distributed except anterodorsally and posteroventrally where closer together. Sternal spine absent. Copulatory complex (Fig. 2i) with short to moderately long trifid finger guard; peniferum moderately robust with posteroventral extremity subangular and with almost straight ventral margin forming, in optical section, acute anteroventral angle. Peniferal groove distinct, opening anteroventrally, its diameter about half that of least diameter of vertical ramus of clasping apparatus. Penis L-shaped with distal arm distinctly 16 Proceedings of the Biological Society of Washington Fic. 2a-f, Dactylocythere crena, new species; 2g—k, Dactylocythere scissura, new species; 2l—p, Psittocythere psitta, new species. a, i, l, copulatory complex of male; b, k, n, genital apparatus of female; c, 7, m, clasping apparatus of male; d, g, 0, dextral view of shell of male; e, h, p, dextral view of shell of triunguis female; f, dextral view of shell of biunguis female. Scales in mm. New ostracods from Tennessee 17 longer than proximal arm; distal arm proportionately longer than that of any other of its congeners and sometimes reaching almost to distal extremity of peniferal groove. Accessory groove slender, extending posterodorsally from base of penis, never reaching dorsally beyond level of midlength of spermatic loop. Clasping apparatus (Fig. 2j) C-shaped without hint of angle at junction of horizontal and vertical rami; rami subequal in length; vertical ramus entire, and horizontal ramus with 3 apical denticles and 1 small tooth immediately proximal to latter. Dorsal and ventral fingers moderately slender, former slightly greater than half length of latter; ventral finger weakly sinuous and directed antero- ventrally. Triunguis Female: Eye pigmented, situated approximately one-fourth shell length from anterior margin. Shell (Fig. 2h) very high posteriorly and tapering anteriorly, greatest height about 1.4 times that at level of eye. Posterior margin of shell with broad subrectangular notch (sug- gesting the name scissura, L. = cleft), height of which about one-third greatest height of shell. Genital complex (Fig. 2k) consisting of highly refractive J-shaped rod with tapering, dorsally disposed extension from body of rod; heterogeneous frothy-appearing mass suspended from ventral extremity of refractive body opposite cleft in posterior margin of shell, this body perhaps corresponding, at least in position, to amiculum of most members of genus. Measurements (in mm): 10 males and 10 females. Holotype Males Allotype Females Length (range) 0.50 0.48-0.53 0.53 0.49-0.55 Average 0.50 0.52 Height (range ) 0.28 0.27-0.29 0.31 0.29-0.35 Average 0.28 0.32 Type-locality: Field at Ten Mile Creek (Tennessee River Basin) at U.S. Highway 11, Knox County, Tennessee, where it was infesting an undescribed burrowing crayfish, Cambarus (Depressicambarus) sp. Disposition of Types: The holotypic male and allotypic female are deposited in the National Museum of Natural History (Smithsonian In- stitution), numbers 150648 and 150649, respectively. Paratypes are in the collection of H. H. Hobbs III and the Smithsonian Institution. Range and Specimens Examined: Approximately 30 specimens from the following localities in the Tennessee River Basin, Knox County, Tennessee: Field at County Road 2404, Middlebrook Pike, opposite Wesley Road, Knoxville. Field at Interstate 75 N, near Emory Road eastern exit, Knoxville. Field at Hardin Valley Road, approximately 2 miles southwest of Lovell Road off Interstate 40, west of Ball Camp. Field southwest of Byrd Chapel, just west of Hickory Creek. Hosts: Cambarus (Depressicambarus) sp. novy., Cambarus (Cam- 18 Proceedings of the Biological Society of Washington barus) bartonii bartonii (Fabricius, 1798), and Cambarus (Jugicam- barus) dubius Faxon, 1884. Entocytherid Associates: Donnaldsoncythere hiwasseensis (Hobbs and Walton, 1961), Uncinocythere simondsi (Hobbs and Walton, 1960), Dactylocythere sp., and Entocythere sp. Relationships: Dactylocythere scissura is more closely related to Dt. crena than to any other ostracod. The unique characteristics of the two are pointed out in the discussion of relationships of the latter. Dacty- locythere scissura may be distinguished from Dt. crena by its smaller size, the presence of a well defined peniferal groove instead of a tapering anteroventral sheath on the peniferum, and the former possesses a ven- trally flattened peniferum as opposed to a ventrally rounded one in Dt. crena. Psittocythere, new genus Diagnosis: Terminal tooth of mandible pectinate. Male copulatory complex (Fig. 21) without finger guard. Peniferum extending ventrally only slightly beyond clasping apparatus, elongate, slender, with sub- terminal enlargement bearing single acute hooklike projection from posteroventral extremity; hook curved anterodorsally; penis consisting of contiguous spermatic and prostatic elements directed anteroventrally, neither element more than one-fourth length of clasping apparatus; accessory groove lacking. Clasping apparatus (Fig. 2m) not reaching ventral margin of peniferum but well developed, clublike, and not clearly divisible into horizontal and vertical rami; postaxial border entire; pre- axial border of swollen distal third with 6 rounded teeth. Triunguis female with second antenna lacking accessory pectinate process on distal podomere; genital complex (Fig. 2n) consisting of short bilobed prominence, suspended in posterodorsal part of body, surrounded by amorphous irregularly shaped refractive mass. Type-species: Psittocythere psitta, described below. Gender: Feminine. Psittocyithere psitta, new species (Figure 2l-p) Male: Eye pigmented and located about one-sixth shell length from anterior margin. Shell (Fig. 20) subovate with greatest height distinctly posterior to midlength, about 1.5 times height at level of eye; ventral margin with faint concavity anterior to midlength. Submarginal setae present except dorsally and somewhat closer together anteriorly and posteriorly than ventrally. Copulatory complex (Fig. 21) as described in generic diagnosis. In addition, dorsal finger distinctly angular and slightly more than half length of slender, gently and evenly curved ventral process. Triunguis Female: Eye as in male. Shell (Fig. 2p) more highly vaulted than in male, with greatest height distinctly posterior to mid- New ostracods from Tennessee 19 length, about 1.7 times that of height at level of eye; ventral margin with concavity as in male. Submarginal setae present except dorsally between level of eye and genital complex, distinctly closer together anteriorly and posteriorly than ventrally. Measurements (in mm): 7 males and 5 females. Holotype Males Allotype Females Length (range) 0.39 0.38—0.40 0.37 0.37-0.42 Average 0.39 0.40 Height (range) 0.23 0.20-0.23 0.24 0.22-0.25 Average 0.22 0.23 Type-locality: Laurel Creek, 9.7 miles northeast of Jamestown on State Route 154, Fentress County, Tennessee. There this ostracod was associated with Dactylocythere spinata Hobbs and Walton, 1970, and Donnaldsoncythere hiwasseensis (Hobbs and Walton, 1961). These ostracods were infesting Cambarus (Jugicambarus) distans Rhoades, 1944, Disposition of Types: The holotypic male and allotypic female are deposited in the National Museum of Natural History (Smithsonian Institution), number 150650. Paratypes are in the collection of H. H. Hobbs III and the Smithsonian Institution. Range and Specimens Examined: Known from the type-locality and the following two localities in Pickett County, Tennessee: Rock Creek at State Route 154, and Thompsons Creek at State Route 154. Only 12 specimens are available from the three localities, all of which are in the Big South Fork of the Cumberland River system. Host: Cambarus (Jugicambarus) distans Rhoades, 1944. Entocytherid Associates: Dactylocythere spinata Hobbs and Walton, 1970, Donnaldsoncythere hiwasseensis (Hobbs and Walton, 1961), Dactylocythere sp., and Entocythere sp. Relationships: Psittocythere psitta is perhaps more closely allied to Cymocythere clavata Crawford, 1965, than to any other species, the greatest similarity occurring in the clasping apparatus. In this respect, however, it is markedly dissimilar to the other two species assigned to the latter genus. The arched hooklike projection anterodorsally from the posteroventral side of the subterminal bulbous area is unique in this ostracod, and it is this unique characteristic that lends a parrotlike aspect to the inverted peniferum that, in turn, suggested the generic and specific names proposed for this ostracod. LITERATURE CITED (References to descriptions of crayfishes are omitted; these may be found in Hobbs, Horton H., Jr. 1974. A Checklist of the North and 20 Proceedings of the Biological Society of Washington Middle American Crayfishes (Decapoda: Astacidae and Cambaridae). Smithsonian Contrib. Zool. 166:1-161, 294 figs.) CRAWFORD, Epwarp A., Jr. 1965. Three new species of epizoic ostracods (Ostracoda, Entocytheridae) from North and South Carolina. Amer. Midl. Nat. 74(1):148-154, 16 figs. Hart, C. W., Jr. 1962. A revision of the ostracods of the family Entocytheridae. Proc. Acad. Nat. Sci. Philad. 114(3):121-147, 18 figs. , AND DasneEy G. Harr. 1967. The entocytherid ostracods of Australia. Proc. Acad. Nat. Sci. Philad. 119(1):1-51, 95 figs. , A. BALAKRISHNAN Nair, AND DaBNEy G. Harr. 1967. A new ostracod (Ostracoda: Entocytheridae) commensal on a wood- boring marine isopod from India. Notulae Naturae, Acad. Nat. Sci. Philad. 409:1-11, 17 figs. Hosss, Horton H., Jn., anD C. W. Hart, Jr. 1966. On the ento- cytherid ostracod genera Ascetocythere, Plectocythere, Phy- mocythere (gen. nov.), and Cymocythere, with descriptions of new species. Proc. Acad. Nat. Sci. Philad. 118(2):35-61, 37 figs. ——. AND H. H. Hosss III. 1970. New entocytherid ostracods with a key to the genera of the subfamily Entocytherinae. Smith- sonian Contrib. Zool. 47:1-19, 9 figs. ——, AND Marcaret Watton. 1960. Three new ostracods of the genus Entocythere from the Hiwassee Drainage System in Georgia and Tennessee. Journ. Tenn. Acad. Sci. 35(1):17— 23, 20 figs. ——. 1961. Additional new ostracods from the Hiwassee Drainage System in Georgia, North Carolina, and Tennessee. Trans. Amer. Micros. Soc. 80(4):379-384, 8 figs. ——. 1962. New ostracods of the genus Entocythere from the Mountain Lake region, Virginia (Ostracoda, Entocytheridae). Virginia Journ. Sci., new series 13(2):42-48, 12 figs. —. 1966. A new genus and six new species of entocytherid ostracods (Ostracoda, Entocytheridae). Proc. U.S. Nat. Mus. 119( 3542) :1-12, 2 figs. —. 1968. New entocytherid ostracods from the southern United States. Proc. Acad. Nat. Sci. Philad. 120(6) :237—-252, 3 figs. —. 1970. New entocytherid ostracods from Tennessee and Vir- ginia. Proc. Biol. Soc. Wash. 82(68):851—-863, 3 figs. Horr, C. CLayton. 1944. New American species of the ostracod genus Entocythere. Amer. Midl. Nat. 32(2):327-357, 33 figs. (y Ley Ne we ICT\AT 7: AANA ANLIINGA A GH / Pe arian IES i 3, pp. 21428 ArK 3) W/o 23 April 1975 BAX \ NH PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON STUDIES ON STOMATOPOD CRUSTACEA OF THE INDIAN RIVER REGION OF FLORIDA. I. REDISCOVERY AND EXTENSION OF RANGE OF HETEROSQUILLA ARMATA (SMITH, 1881)? By Rosert H. Gore ANp LinpA J. BECKER Smithsonian Institution, Ft. Pierce, Florida 33450 Heterosquilla armata (Smith) is an uncommon species of mantis shrimp in the family Lysiosquillidae which was orig- inally described in 1881 from material obtained primarily by the U.S. Fish Commission Steamer “Fis Hawk’ off the coast of New England (see Manning, 1969: 52). In spite of numer- ous collections by many agencies in the waters off the north- eastern United States and elsewhere along the eastern seaboard, apparently only one additional specimen has been taken. This specimen, a female, was captured in 1963 by the National Marine Fisheries Service from 128 m on the margin of the continental slope east of New Jersey. The range of the species thus seemed to be confined to waters off the New England area and vicinity in depths from 96-218 m ( Manning, 1974: 4). It was of considerable interest, then, to find a single fe- male specimen of H. armata in collections which we had made on the continental shelf off the central eastern Florida coast. This rediscovery was deemed of sufficient interest to warrant a short report at this time. We thank Dr. Raymond B. Manning and Thomas E. Bowman, National Museum of Natural History, for critically reading the manuscript. Dr. Manning also confirmed our identification of the species, and provided the data on the one other recently collected specimen. Dr. Roland Wigley, National Marine Fisheries Service Laboratory (NMFS), Woods Hole, Massa- 1 Scientific Contribution No. 36 from the Smithsonian Institution-Harbor Branch Foundation Consortium, Science Laboratories. 3—Proc. Biou. Soc. WAsH., Vou. 88, 1975 (21) 22 Proceedings of the Biological Society of Washington chusetts, graciously allowed one of us (LJB) to examine and measure this specimen, and provided additional information on its collection. Photographs of the Floridan and New En- gland specimen are by Mr. William Davenport, Harbor Branch Foundation Laboratory, Ft. Pierce, Florida. Materials and Methods: Our system of measurements follows that of Manning (1969) except for total length (TL) which was measured along the dorsal midline from rostral tip to tip of the spines on the boss of the telson. All measurements were made with a Wild M-5 stereomicroscope using an ocular reticle and a stage micrometer, and were rounded off to the nearest tenth of a millimeter. Research vessel cruise numbers are followed by station number. For complete synonymies the reader is referred to Manning’s (1969) monograph on western Atlantic Stomatopoda. We in- clude in this report only the citations of the original author, and subsequent changes and additions in the generic and subge- neric nomenclature. Heterosquilla (Heterosquilloides) armata (Smith, 1881) Figures 1-3 Lysiosquilla armata Smith, 1881: 446—Manning, 1963: 321 [listed]. Heterosquilla ( Heterosquilloides) armata.—Manning, 1963: 321 [listed]; 1966: 119, 124 [listed]; 1969: 44, 45, 52, fig. lla, b; 1974: 1, 3, 4, fig. 7, 9 [listed, key]. Material examined: 1 9, cl 12.0 mm, TL 72.4 mm, R/V ALBATROss IV, 63-7/166, 39°26’N, 72°25’W, 128 m, 36’ otter trawl, 11 December 1963. 1 9, cl 7.4 mm, TL 42.8 mm, R/V Gosnoip 237/507, 27°08.5’N, 79°54’W, 210 m, 5’ Blake trawl, 11 June 1974. Remarks: The New England specimen had been examined previously and its identification confirmed by Dr. R. B. Manning. The single Floridan specimen available for examination agreed well with the descrip- tion and illustration provided by Manning (1969: 52) and with the male lectotype deposited in the National Museum of Natural History. In our specimen, the median projection of the telson was much more noticeably raised than would be inferred from R. P. Bigelow’s illustration > Fic. 1. Heterosquilla armata (Smith), Florida specimen, dorsal views: A, Whole animal. B, Carapace. C, Posterior abdominal somites and telson. Southern record—Heterosquilla armata 24 Proceedings of the Biological Society of Washington as used by Manning. In addition, the lectotype and other material in the National Museum exhibited a more distinct telsonal boss than the recently collected New England specimen. We also noted minor varia- tion in the number and shape of the dorsal and marginal spines on the telson in the Floridan specimen (i.e. more sharply pointed) when compared with the above mentioned material but consider this variation to be nonsignificant (see Figs. lc, 2c). We compared our specimen with all available material (much of which was in poor shape) deposited in the National Museum of Natural History (USNM 7199, 21491), and found no other notable differences. Color: The color of the Floridan specimen after two months in alcohol was as follows. The animal is grayish-white overall with a fine network of brown chromatophores scattered over the body. The rostrum is brown anteriorly, becoming lighter at the suture with the carapace. The carapace has 2 brown spots anterolaterally, plus 4 spots on the gastric grooves, 2 of which occur on the posterior parts of the grooves, followed by 2 medially near the posterior margin of the carpace. The thoracic somites have brown chromatophores as follows: fifth somite, 2 small single “snowflake” dots on the anterolateral margin; somites 6-8, scattered brown specks posteriorly, becoming less numerous anteriorad, so that a clear margin appears to be present in unmagnified view. Abdominal somites 1-3 have alternating transverse bands of grayish-white anteriorly, and brown posteriorly, on each somite; somite 4 dorsolaterally with 2 subparallel transverse bands of brown, becoming off-white medially; somites 5 and 6 similar to somites 1-3. The telson has a row of scattered brown chromatophores laterally along the longitudinal margin of the raised median boss as far as the first row of transverse spinules. The carpus and merus of the raptorial claws, the antennules, the antennal peduncles, and the eyestalks all have numerous scattered brown chromatophores. Measurements: New England female, TL 72.4, carapace length by width (cl x cw) 12.0 x 11.1, cornea width 3.6, rostral carapace length (Rel) 16.2, rostral length by width (Rl x Rw) 3.3 x 3.6, width of abdominal somite 5 (A-5w) 14.5, telson length by width (tl x tw) 7.7 < 12.2 mm.—Florida female, TL 42.8, cl & cw 7.4 xX 7.5, cornea width 2.4, Rel 9.5, Rl & Rw 2.1 x 2.0, A-5w 7.9, tl x tw 4.0 X 6.8 mm. Distribution: Off New England in waters 96-218 m depth. The Floridan specimen reported herein, collected in 210 m, extends the known range approximately 1400 km southward to a point about 19 km east of St. Lucie Inlet, Martin County, Florida. > Fic. 2. Heterosquilla armata (Smith): A, New England specimen, whole animal, dorsal view. B, Same, posterior abdominal somites and telson, dorsal view. C, Florida specimen, posterior abdominal somite and telson, ventral view. SUG oO LS Ze Lip 2 SS aS < aS SY Ze 26 Proceedings of the Biological Society of Washington AUS ooomot Fic. 3. Heterosquilla armata (Smith): A, Florida specimen, right raptoral claw, mesial view. B, New England specimen, anterior part, lateral view. Southern record—Heterosquilla armata oT Discussion: Although extensive sampling has been carried out on the continental shelf of central eastern Florida between 27° and 28° N latitude over the past 10 months (November 1973 to August 1974) by the R/V Gosnoxp, with over 500 biological collection stations occupied, only the single female specimen of H. armata has thus far been taken. The Florida record, therefore, may be extralimital. Depth records, in- cluding the data on the NMFS specimen and the Floridan specimen, indicate that Heterosquilla armata is a species inhabiting deeper waters on the continental shelf. Station records for the U.S.F.C.S. Fish Hawk and A.spatTross, those from the R/V Arsatross IV provided by Dr. Roland Wigley of the NMFS, and the R/V Gosnotp data all indicate that Heterosquilla armata should be looked for on the distal edge of the continental shelf and upper margins of the continental slope. The tilefish record noted in Manning (1969: 52, 343) was a FisH Hawk station on the continental slope due east of New Jersey. The presence of H. armata as far south as central eastern Florida may be either a reflection of its preferred habitat, i.e. the proximity of the continental slope, or a preference for the cooler oceanic water found along the 200 m line in this vicinity, where temperatures of 10-12°C prevail. Whatever the reason, the rediscovery so far southward of this distinctively ornamented lysiosquillid after an interval of over 90 years simply reinforces the fact that our knowledge of many benthic species and their distribution along the continental shelf still remains incomplete. The New England specimen has been returned to Dr. Roland Wigley, National Marine Fisheries Service Laboratory, Woods Hole, Massa- chusetts. The specimen from Florida has been deposited in the National Museum of Natural History, Washington, D.C., USNM 150273. LITERATURE CITED Manninc, R. B. 1963. Preliminary revision of the genera Pseudosquilla and Lysiosquilla with descriptions of six new genera (Crustacea; Stomatopoda). Bull. Mar. Sci. Gulf & Carib. 13(2):308-328. —. 1966. Notes on some Australian and New Zealand Stomatopod Crustacea, with an account of the species collected by the Fisheries Investigation Ship Endeavour. Rec. Australian Mus. ) 97 (4): 79-137, figs. 1-10. ——. 1969. Stomatopod Crustacea of the Western Atlantic. Stud. Trop. Oceanogr. Miami 8: viii + 380 pp., figs. 1-91. ———. 1974. Marine flora and fauna of the northeastern United States. Crustacea: Stomatopoda. NOAA Tech. Rpt. NMFS Circ. 387: iii + 6 pp., figs. 1-10. SmitH, S.I. 1881. Preliminary notice of the Crustacea dredged, in 64 to 325 fathoms, off the south coast of New England, by the U.S. Fish Commission in 1880. Proc. U.S. Nat. Mus. 3:413-452. 28 Proceedings of the Biological Society of Washington 0. 4, pp. 29-44 23 April 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON CAVE SHRIMPS (DECAPODA: CARIDEA) FROM THE DOMINICAN REPUBLIC By Fenner A, CHAcE, Jr. Smithsonian Institution, Washington, D.C. 20560 A small but excellently preserved collection of shrimps re- cently received from Renato O. Rimoli of Santo Domingo, Dominican Republic, contained three specimens of particular interest from subterranean habitats. The first is a species of the atyid genus Typhlatya that had been recorded previously from Isla Mona off Puerto Rico and from Barbuda in the Lee- ward Islands. The second is an apparently undescribed species of the palaemonid genus Macrobrachium that shows no obvious adaptation for troglobitic existence. The third is a second species of the recently proposed hippolytid genus Calliasmata, the type-species of which is known from anchialine pools (Holthuis, 1973:3) in the Sinai Peninsula, Ellice Islands, and Hawaiian Islands. I am very grateful to Mr. Rimoli for making this material available to me and for his assistance during the preparation of the paper, as well as to my colleagues, Thomas E. Bowman, Horton H. Hobbs, Jr., Raymond B. Manning, and Austin B. Williams, for their critical review of the manuscript. ATYIDAE Typhlatya monae Chace Typhlatya monae Chace, 1954:318, fig. 1—Chace and Hobbs, 1969:80, fig. 16. Typhlatya nana Vandel, 1964:178 (incorrect spelling of monae); 1965: 139. Material (carapace lengths in mm in brackets): La Furnia de Los Corrales, Villas del Mar, Provincia de San Pedro de Macoris, Dominican Republic; 16 August 1973; Renato O. Rimoli: 21 2 [3.3-4.4]. 4—Proc. Brow. Soc. WasH., Vou. 88, 1975 (29) 30 Proceedings of the Biological Society of Washington Color: Yellow-orange in life. Habitat: Wa Furnia de Los Corrales is a subterranean cavity sur- rounded by abundant vegetation about 400 meters from the sea. The entrance has a maximum height of 1.85 meters, with a cornice one meter thick. The distance to the innermost end of the cavity is 9.20 meters. There is sufficient phreatic water to form a pool about two feet deep, the bottom of which is covered with a large amount of sediment of organic origin. Light penetrates into the interior, providing some illumination, but the presence of the shrimps, which were quite numer- ous, was detected by flashlight. Distribution: The species has been recorded previously from subter- ranean fresh water on Isla Mona (Puerto Rico) and Barbuda (Leeward Islands). Remarks: Not even a vestige of an exopod can be found on the fifth pereopods of these specimens. They agree so well, however, with speci- mens of T. monae from Isla Mona and Barbuda in all other characters that there is little justification for regarding them as taxonomically distinct. Like all previously recorded specimens of the species, none of the series from the Dominican Republic has an appendix masculina on the second pleopods, and all have consequently been presumed to be fe- males. Inasmuch as males of most of the other species of the genus are known—even of the closely related T. garciai Chace, 1943, from Cuba —the failure thus far to find males of T. monae suggests two possible explanations: either they are unrecognizable from external characters or they are restricted to a habitat niche that has not yet been investigated. The mysterious presence of part of the type-series of T. monae in a concrete water catchment basin on the high central mesa of Isla Mona (Chace, 1954:319) has finally been explained. Thomas A. Wiewandt, who has been conducting an ecological survey of Mona under the auspices of the Puerto Rico Department of Natural Resources, informs me that, during dry periods on the island, water from the wells in which the shrimp normally occurs is sometimes pumped into the catchment basins, thereby accounting for the presence of Typhlatya in such an apparently inimical habitat. PALAEMONIDAE Macrobrachium crybelum new species Figures 1—4 ? Macrobrachium olfersii—Rathbun, 1919:324.—Schmitt, 1936:372. [Not Palaemon olfersii Wiegmann, 1836.] ? Macrobrachium sp. (near M. faustinum) Chace and Holthuis, 1948:23. ? Macrobrachium aff. faustinum Holthuis, 1952:95. Description: Rostrum with dorsal margin usually regularly concave (Fig. 3a), rarely straight (Fig. la), overreaching antennal scale, 0.7 to Cave shrimps from Dominican Republic 31 Fic. 1. Macrobrachium crybelum, holotype ¢. a, Carapace and an- terior appendages in lateral aspect; b, Anterior region in dorsal aspect; c, Abdomen; d, Posterior end of telson; e, Right mandible; f, Right first maxilla; g, Right second maxilla; h, Right first maxilliped; i, Right second maxilliped; j, Right third maxilliped. (a—c, j7, x 2; g-i, x 4; Q> ji She GE S< UN) 0.9 as long as carapace in larger specimens, as long as or slightly longer than carapace in juveniles; armed dorsally with 11 to 16 teeth, including 4 to 6 on carapace posterior to orbit, spaced farther apart on rostrum than on carapace, except anteriormost 2 or 3 occasionally set close to- - gether, all but those on anterior half of rostrum with indistinct basal suture; armed ventrally with 3 to 6, most frequently 4, teeth; lateral ridge sharp in posterior % of length, becoming blunt anteriorly, equi- distant from dorsal and ventral margins over much of length. Carapace smooth, antennal spine arising submarginally from sub- orbital lobe, somewhat longer than hepatic spine; rather distinct suture extending from base of hepatic spine to anterior margin. Abdomen smooth, pleuron of 4th somite with posteroventral angle varying from subrectangular (Fig. 3c) to rounded (Fig. 1c); pleuron of 5th somite acute posteroventrally; 6th somite about 1.5 times as long as 5th, acute posteroventrally, posterolateral lobe drawn out to sharp tooth. Telson about 1.5 times as long as 6th somite, normally 32 Proceedings of the Biological Society of Washington with 2 pairs of distinct dorsal spines, 1 near midlength, 1 about midway between first pair and posterior end of telson; posterior margin (Figs. 1d, 3d) acutely triangular (obviously damaged in holotype), apex falling far short of tips of larger of 2 pairs of posterior spines. Eyes (Figs. 1b, 3b) rather short, cornea inflated and fully pigmented, with ocellus on dorsomesial surface. Antennular peduncle (Figs. 1b, 3b) with strong distolateral spine on proximal segment. Lateral flagellum fused for 5 to 10 short articles, mesial branch nearly as long as postorbital length of carapace, lateral branch nearly 3 times as long as carapace; mesial flagellum about twice as long as carapace. Antennal scale (Figs. 1b, 3b) about 3 times as long as wide, disto- lateral spine nearly or quite reaching level of truncate distal margin of blade. Antennal peduncle not reaching midlength of scale, basal seg- ment with strong lateral spine. Flagellum nearly 6 times as long as carapace. Mouthparts as illustrated (Figs. le-j, 3e-j). Third maxilliped falling slightly to considerably short of distal end of antennal scale. First pereopod (Figs. 2a, 4a) overreaching antennal scale by length of chela and about % of carpus in holotype, but slightly less than length of chela in some smaller specimens; carpus 1.5 to 2.0 times as long as chela. Fully developed second pereopods distinctly dissimilar and un- equal. Major 2nd pereopod (Fig. 2b) overreaching antennal scale by length of chela and nearly or more than entire length of carpus; chela (Fig. 2c) with fingers slightly shorter than palm; opposable margins of both dactyl and fixed finger with single row of 12 to 17 rounded teeth, one larger than rest near bases of both fingers, distal ones small and indistinct; surface of palm concealed by closely spaced tufts of long fine hairs covering margin continuous with fixed finger and extending far onto both flattened surfaces; long dark setae becoming especially numer- ous on fingers and forming dense fringe in gape; most of surface covered with distally directed spines increasing in length on fingers and forming distinct row along margin continuous with fixed finger, those centrally located on palm shorter than those on proximal and distal portions; carpus slightly longer than palm and about 1.4 times as long as merus. Minor 2nd pereopod (Fig. 2d) more slender, overreaching antennal scale by % or nearly entire length of carpus; chela (Fig. 2e) with fingers as long as palm; opposable margins of both dactyl and fixed finger with large tooth near bases and 3 minute tubercles proximal thereto, re- mainder of margins entire and rather sharp; surface of chela with numerous long dark setae becoming crowded on fingers and forming dense fringe in gape; carpus distinctly longer than palm and longer than merus. Third pereopod (Fig. 2f) overreaching antennal scale by length of dactyl and about 1% of propodus. Fourth pereopod (Fig. 2g) over- reaching antennal scale by length of dactyl and about %4 of propodus. Fifth pereopod (Fig. 2h) overreaching antennal scale by length of dactyl and about 4 of propodus. Cave shrimps from Dominican Republic 33 Fic. 2. Macrobrachium crybelum, holotype ¢. a, Right first pereo- pod; b, Right second pereopod; c, Same, extensor aspect of chela (denuded); d, Left second pereopod; e, Same, extensor aspect of chela (denuded); f, Right third pereopod; g, Right fourth pereopod; h, Right fifth pereopod; i, Right first pleopod; 7, Same, endopod; k, Right second pleopod; /, Same, appendices interna and masculina. (a-i, k, xX 2; pipes aL XaliZ.) Endopod of first pleopod (Figs. 2i, 7), appendix masculina (Figs. 2k, 1), and uropods characteristic of genus. Variation: In the largest male in the collection—a specimen with a carapace length of 17.8 mm from Cueva de Valiente—the second pair of pereopods (Figs. 4b, c) are much less dissimilar and less unequal than they are in both the holotype and a similar male with a carapace length of 13.0 mm from the type-locality. In the Cueva de Valiente male, the major cheliped is almost identical with those in two females with carapace lengths of 18.4 and 19.2 mm from La Caleta. The chela is less swollen and less densely furred than in the holotype and the 34 Proceedings of the Biological Society of Washington ae gs SINE, 7 | | Sa Fic. 3. Macrobrachium crybelum, gynecomorphous paratype ¢, cara- pace length 17.8 mm, from Cueva de Valiente (except d). a, Carapace and anterior appendages in lateral aspect; b, Anterior region in dorsal aspect; c, Abdomen; d, Posterior end of telson of paratype ¢, carapace length 14.3 mm, from Cueva de Valiente; e, Right mandible; f, Right first maxilla; g, Right second maxilla; h, Right first maxilliped; i, Right second maxilliped; j, Right third maxilliped. (a—c, j, x 2; e, g-i, x 4; iy SSH CD. S< Ms) fingers are proportionately longer and less gaping, with the opposable margins (Fig. 4e) armed as in the minor cheliped, lacking teeth on the distal two-thirds of their lengths. Also, the rows of spines on the merus and ischium are much less prominent than the comparable series in the holotype. In all of the specimens except these five (two males from Ciudad del Caribe, one male from Cueva de Valiente, and two females from La Caleta), the second pereopods are more slender, not very dis- Cave shrimps from Dominican Republic 30 Fic. 4. Macrobrachium crybelum, gynecomorphous paratype é, cara- pace length 17.8 mm, from Cueva de Valiente. a, Right first pereopod; b, Right second pereopod; c, Left second pereopod; d, Same, chela; e, Same, basal teeth of fingers; f, Right third pereopod; g, Right fourth pereopod; h, Right fifth pereopod; i, Right first pleopod; 7, Same, endo- pod; k, Right second pleopod; J, Same, appendices interna and masculina. (Gaon jt ks x 2d, <4; 6) 7}, x18; 1 Sc IT) similar, equal or only slightly unequal, and the major chelae are not noticeably furred. Size: Carapace length of holotype male, 14.9 mm; of paratype males, 5.9 to 17.8 mm; of paratype females, 5.6 to 19.2 mm. Type-locality: Cave at Ciudad del Caribe (18°58’N, 70°23’W), Santo Domingo, D.N., Dominican Republic. Material (carapace lengths in mm in brackets): Cave at Ciudad del Caribe, Santo Domingo, D.N., Dominican Republic; 14 July 1973; Renato O. Rimoli: 10 ¢ [7.8—-14.9] (largest is holotype, USNM 151199) 5 @ [9.6-11.7].—El Caimito (Cave No. 2), Dominican Republic; 17 February 1974; Renato O. Rimoli: 1 2 [14.2].—La Caleta, Dominican 36 Proceedings of the Biological Society of Washington Republic; 15 March 1974; Renato O. Rimoli: 2 ¢ [5.9, 11.3] 6 @ [5.6- 19,.2]—“Cueva de Valiente,’ km 21, Autopista “Las Americas,” Santo Domingo, D.N., Dominican Republic; 27 May 1974; Alberto Otten- walder: 4 6 [9.0-17.8] 149 [8.2—-15.5]. Habitat: The cave at Ciudad del Caribe is in Pleistocene limestone about 50 meters from the sea and at approximate sea level. The pool within is brackish and varies in depth from one to five feet, with con- siderable sediment of vegetable origin. The shrimps occurred throughout the pool but seemed to prefer hiding among rocks near the edge. A fish, Lophogobius cyprinoides, occupied the same pool and preyed on the shrimps, at least in captivity. At El Caimito, the pool is situated at the end of a large cave where nearly absolute darkness prevailed. The single specimen taken from this pool was found in shallow water, but the depth varied from a foot to several meters. The bottom is entirely rocky and devoid of organic sediment. The water was almost fresh or very slightly brackish and completely clear—the deeper part blue by flashlight—and with only a slight current. A fish was observed in the pool but it avoided capture. The fauna near the entrance of the cave was scanty, consisting principally of arachnids and lizards (Anolis, sp.). The cave at La Caleta was not completely dark. The water was little brackish and varied in depth from one to 15 feet. An unidentified fish was also found there. Name: The specific name—from krybelos, G., = hidden, secret-—was suggested by the habitat. Remarks: Typical mature males of Macrobrachium crybelum super- ficially resemble comparable specimens of M. faustinum (De Saussure, 1857), which occurs in exposed freshwater habitats throughout the West Indian islands, but they may be readily distinguished by the following characters. The rostrum is longer, overreaching the antennal scales rather than reaching no farther than the end of the antennular peduncle. The spines on the major second pereopod are more numerous and less erect, especially those on the margins corresponding to that of the fixed finger of the chela. The fixed finger of the minor second pereopod is less bowed, so that the fingers gape less noticeably. The three posterior pairs of pereopods are longer, considerably overreaching the antennal scale, rather than reaching at most only to the distal end of the scale. Specimens of both sexes of M. crybelum differ from all other American species of the genus known thus far in the combination of a long rostrum, overreaching the antennal scales, and more than three teeth of the dorsal rostral series situated on the carapace posterior to the orbit. There is a possibility that the large male from Cueva de Valiente (Figs. 3, 4) and the two large females from La Caleta represent a distinct species, because of the very different form of the major second pereopod. It seems more likely, however, that the females display the normal development of the chelipeds at this size and that the male is gynecomorphous. There are no other distinguishing characters to support Cave shrimps from Dominican Republic 37 the recognition of a second species. Holthuis (1952:12) noted that in this genus: “Sometimes males may be found, which are as large as males with fully developed second legs, but in which these legs are still shaped like in the females and juveniles.” Perhaps regeneration is the cause of this apparent dimorphism. It is quite conceivable that the specimens from wells on Bonaire, identified by Holthuis (1952:95) as “Macrobrachium aff. faustinum,” belong to this species. The specimens of that material available to me are compatible with that conclusion, but the lack of a male with fully developed second pereopods precludes a positive determination. HIPPOLYTIDAE Calliasmata rimolii, new species Figures 5-7 Description: Rostrum (Figs. 6b, c) short and simple, directed anteroventrad and largely concealed by eyes from lateral view; ventral margin sinuous; apex sharply pointed; dorsal aspect narrowly acute, lateral margin passing into orbital margin with broadly rounded obtuse angle. Carapace (Fig. 5) noticeably inflated dorsally and laterally; surface largely smooth, not closely pitted, without apparent scales but with widely scattered, very short, microscopic setae set perpendicular to surface. Antennal spine (Fig. 6b) triangular, curving slightly ventrad, almost imperceptibly separated from orbital margin, distinctly over- reaching level of tip of rostrum. Anteroventral margin of carapace broadly rounded. Without cardiac notch posteriorly. Abdomen (Fig. 5) broadly rounded dorsally; surface largely smooth as on carapace. Pleura of 4 anterior somites broadly rounded, of 5th somite acutely produced posteroventrally. Sixth somite nearly 1% again as long as 5th, posteroventral angle bluntly triangular, posterolateral lobe drawn out into narrowly acute, dorsally straight spine (Fig. 6d). Telson (Fig. 6e) about 1% again as long as 6th somite, slightly less than 214 times as long as wide, tapering rather regularly to broadly triangular posterior margin armed with distinct median spine; dorsal surface bearing 2 pairs of rather inconspicuous spines, anterior pair placed slightly anterior to midlength, posterior pair about midway between anterior pair and pos- terior margin; posterior margin (Fig. 6f) armed with 3 pairs of movable spines, intermediate pair much the longest and lateral pair much the shortest. Eyes (Fig. 6g) apparently immovably fused together basally, variably divergent or approximate distally, usually slightly overreaching level of tips of antennal spines; somewhat variable in shape, depending on degree of development of subtruncate anteromesial projection; anterolateral margin consisting of rather sharp, broadly rounded flange; rudimentary cornea lateral to base of anteromesial projection; persistent black pig- 38 Proceedings of the Biological Society of Washington Zz oe S " Ss Se cae Fic. 5. Calliasmata rimolii, holotype 9. x 2.8. ment variably situated elsewhere beneath both dorsal and ventral sur- faces of eyestalk. Antennular peduncle (Figs. 6a, c) with Ist segment subequal in length to combined lengths of 2 distal segments. Stylocerite suboval or subtriangular in lateral view, distal end blunt, with or without minute denticle, falling far short of distal margin of first segment. Second seg- ment longer than 8rd. Dorsolateral flagellum nearly 3 times as long as carapace, setiferous (fused) portion consisting of 19-21 articles, free part of shorter branch reduced to single short article (Fig. 6h). Ventro- mesial flagellum slightly shorter than dorsolateral. Antennal scale (Fig. 6c) about twice as long as wide, barely over- reaching 2nd segment of antennular peduncle, lateral margin nearly straight, terminating in strong distal tooth reaching nearly as far as distal margin of blade. Antennal peduncle reaching little beyond mid- length of scale, basal segment with stout lateral tooth below base of scale and bluntly triangular lobe dorsally that appears toothlike in lateral view. Flagellum at least 434 times as long as carapace, about % again as long as entire animal. Mandibles (Fig. 61) without palp or incisor process; opposable surface of molar process dissimilar on right and left mandibles. First maxilla (Fig. 67) with palp bilobed. Second maxilla (Fig. 6k) with proximal endite much reduced, distal endite rather deeply divided into broadly rounded distal portion and bluntly triangular proximal lobe. All maxilli- peds with well-developed exopods. First maxilliped (Fig. 61) with two Cave shrimps from Dominican Republic 39 Fic. 6. Calliasmata rimolii, holotype @ (except f). a, Anterior region in lateral aspect; b, Rostrum and orbit in lateral aspect; c, An- terior region in dorsal aspect; d, Base of telson and uropod; e, Telson and uropods; f, Posterior end of telson of paratype 2, carapace length 10.8 mm; g, Eyes in dorsal aspect; h, Distal end of setiferous portion of dorsolateral flagellum; i, Left and right mandibles in dorsal (oral) aspect; 7, Right first maxilla; k, Right second maxilla; J, Right first maxil- liped; m, Right second maxilliped. (a, c, e, X 4; b, d, g, i-m, X 8; jis Ves S< Alta) endites separated by shallow notch, palp well-developed and 8-seg- mented, exopod with lash and narrow caridean lobe, epipod prominent. Second maxilliped (Fig. 6m) with terminal segment attached lengthwise to penultimate; exopod, epipod, and podobranch well-developed. Third maxilliped (Fig. 7a) overreaching antennal scale by length of distal and nearly 4% of penultimate segment, distal segment considerably less than twice as long as penultimate, terminating in 2 small curved spines and armed with about 6 movable spines on distal half or less of extensor margin; exopod reaching beyond midlength of antepenultimate segment; epipod and arthrobranch present. 40 Proceedings of the Biological Society of Washington Fic. 7. Calliasmata rimolii, holotype 9. a, Right third maxilliped; b, Right first pereopod; c, Same, chela; d, Right second pereopod; e, Same, chela; f, Left second pereopod; g, Right third pereopod; h, Same, dactyl; i, Right fourth pereopod; j, Same, dactyl; k, Right fifth pereopod; I, Same, dactyl; m, Right first pleopod; n, Same, endopod. (a, b, d, f, Ge A. [Es thls S< Ae CSC OR Gy ioe fel Mp SK Ue) First pereopod (Fig. 7b) overreaching antennal scale by more than length of chela; chela (Fig. 7c) with fingers fully % as long as palm, dactyl terminating in 2 unequal, dark-colored spines embracing single dark tip of fixed finger; carpus about % as long as chela; merus distinctly Cave shrimps from Dominican Republic 4] longer than chela; coxa bearing both straplike epipod and tuft of long filaments. Second pereopods (Figs. 7d-f) similar but unequal, right slightly to considerably longer than left; right overreaching antennal scale by combined lengths of chela, carpus, and 1% to % of merus, left by combined lengths of chela and %o of carpus to combined lengths of chela carpus, and 4 of merus. Carpus subdivided into 29 to 32 articles, merus into 12 to 19, and ischium into 3 to 5; ischium bearing series of stout curved setae along proximal half of flexor margin; coxa bearing both straplike epipod and tuft of long filaments. Third pereopod (Fig. 7g) overreaching antennal scale by combined lengths of dactyl, propodus, and about % of carpus; dactyl (Fig. 7h) armed with 3 or 4 movable spines on flexor margin, in addition to terminal spine; propodus more than 4 times as long as dactyl, bearing about 11 spinules distributed over entire length of flexor margin; carpus about 114 times as long as propodus, bearing about 10 spinules on flexor margin; merus about 11% times as long as carpus, bearing 3 appressed curved spines near and proximal to midlength; ischium no more than ¥% as long as merus, bear- ing single submarginal spine in distal half of flexor margin; coxa with both straplike epipod and tuft of long filaments. Fourth pereopod (Fig. 7i) overreaching antennal scale by combined lengths of dactyl, propodus, and about % of merus; dactyl (Fig. 7j) armed with 3 or 4 movable spines on flexor margin, in addition to terminal spine; propodus about 4 times as long as dactyl, bearing about 12 spinules on flexor margin; carpus nearly 114 times as long as propodus, bearing 6 or more spinules on flexor margin; merus more than 1% times as long as propodus, bear- ing single appressed spine near proximal end of flexor margin; ischium less than 1% as long as merus, bearing single submarginal spine near midlength of flexor margin; coxa with both straplike epipod and tuft of long filaments. Fifth pereopod (Fig. 7k) overreaching antennal scale by length of dactyl and about % of propodus; dactyl (Fig. 7/1) armed with 5 movable spines on flexor margin, in addition to terminal spine; propodus about 6 times as long as dactyl, bearing nearly 20 spinules on flexor margin, those near distal end closer together and partially con- cealed by tufts of long setae; carpus slightly more than % as long as propodus, unarmed; merus nearly twice as long as carpus, unarmed; ischium about 24 as long as merus, unarmed; coxa with tuft of long filaments but no straplike epipod. Endopod of Ist pleopod of female (Figs. 7m, n) divided into 2 lobes distally, mesial lobe bearing retinacular hooks. Uropods (Fig. 6e) usually slightly overreaching telson; lateral branch with prominent mov- able spine mesial to strong tooth at distal end of lateral margin and with nearly complete diaeresis. Color: Pale red in life. Size: Carapace length of female holotype, 11.1 mm; of female para- types, 10.0 to 12.0 mm. Type-locality: Cave 4 km from town of Estero Hondo (19°51’N, 71°11’W), Provincia de Puerto Plata, northern Dominican Republic. 42 Proceedings of the Biological Society of Washington Material: Five females (holotype, USNM 151205) collected at the type-locality, April 20, 1978, by Renato O. Rimoli. Habitat: The cave in which C. rimolii was found is situated in a Pleistocene escarpment about one kilometer long and about 500 meters from the sea. The cave, the entrance to which was shaded by dense vegetation, was completely full of water, forming an underground pool with a maximum depth of two meters. Light penetrated to the bottom of the pool and there was little organic sediment. The water was clear and barely brackish with a temperature of 25°C, compared with an out- side air temperature of 26°C at the time of collection. Calliasmata rimolit was numerous and was the only animal found in the cave. Probably this cave qualifies as an anchialine habitat, as defined by Holthuis (1973:3), even though the water in the lake was “barely brackish” and no tidal influence was apparent. Name: The species is named for the collector, Renato O. Rimoli. Remarks: This species is very similar in general appearance to the previously unique type-species, C. pholidota Holthuis, 1978, from com- parable habitats in the Sinai Peninsula, Ellice Islands, and Hawaiian Islands. Comparison of the five females from the Dominican Republic with part of the original series of C. pholidota (one female from the Sinai Peninsula and one male, two females from the Ellice Islands) de- posited in the Smithsonian Institution reveals, however, that C. rimolii differs from the Indo-Pacific species in several characters, some of which might be considered to be of supraspecific importance. The in- tegument of C. rimolii is much less firm and seems to be devoid of the scales and corresponding pits that are characteristic of C. pholidota. The carapace is more strongly inflated and the rostrum is consequently directed anteroventrad so that it is largely concealed by the eyes from lateral view; in dorsal view, the rostrum is more slender subapically, but there is considerable variation in this regard. Perhaps the most obvious differences between the two species are the rounded, rather than rectangular posteroventral margin of the first abdominal pleuron, the absence of a ventral spine on each of the third, fourth, and fifth abdominal pleura, and the acute, rather than rounded, posteroventral angle of the fifth pleuron in C. rimolii. The eyes are variable in shape in both species, but the anteromesial angle is usually more strongly produced into a subtruncate projection in C. rimolii. The stylocerite tends to be less produced anteriorly and to be blunter apically in C. rimolii, but it, too, is variable in both species. A final striking difference lies in the form of the endopod of the first pleopod of the female; in C. pholidota, it tapers to a simple blunt tip with terminal retinacular hooks, whereas in C. rimolii it is split distally into two divergent lobes, only the mesial one of which bears retinacular hooks. Three other differences suggested by comparing the descriptions of the two species may not be significant. The dorsal lobe of the basal segment of the antennal peduncle is blunt or rounded in dorsal aspect in the specimens of C. pholidota examined, little more acute than it is in Cave shrimps from Dominican Republic 43 C. rimolii; in lateral aspect, this lobe appears toothlike in both species. The third maxilliped may be a little longer and the spines on the distal segment less numerous in C. rimolii, but there is probably sufficient variation to minimize the importance of these differences when longer series of both species are compared. Only two of the available specimens of C. pholidota have both second pereopods; in each specimen, the right member of the pair is slightly or distinctly longer than the left, as in C. rimolii, but this character may prove to be of little importance in both species when additional material is examined. Two of the characters mentioned above—the integumental scales and the spines on the abdominal pleura—were considered by Holthuis (1973:387) to be of generic significance, and a third—the form of the endopod of the first pleopod—might have been so treated if C. rimolii had been discovered first. Even the relatively few specimens thus far known of each species, however, suggests a variability that may be reflected in genetic instability at the generic level; there would therefore seem to be little justification for treating these two species as other than congeneric at the present time. LITERATURE CITED CHACE, FENNER A., JR. 1948. Two new blind prawns from Cuba with a synopsis of the subterranean Caridea of America. Proc. New England Zool. Club 22:25—40, 3 pls. —. 1954. Two new subterranean shrimps (Decapoda: Caridea) from Florida and the West Indies, with a revised key to the American species. Journ. Washington Acad. Sci. 44(10):818—- 324, 2 figs. , AND Horton H. Hosss, Jr. 1969. The freshwater and terres- trial decapod crustaceans of the West Indies with special refer- ence to Dominica. U.S. Nat. Mus. Bull. 292:i-vi, 1-258 pp., 76 figs., 5 pls. , AND LipxE B. Hottuuts. 1948. Land and fresh water decapod Crustacea from the Leeward Group and northern South Amer- ica. In Hummelinck, Studies on the fauna of Curacao, Aruba, Bonaire and the Venezuelan islands 3:21-28. HoxttHuis, Lirke B. 1952. The subfamily Palaemoninae. Part II in A general revision of the Palaemonidae (Crustacea Decapoda Natantia) of the Americas. Allan Hancock Found. Occ. Pap. 12:396 pp., 55 pls. —. 1978. Caridean shrimps found in land-locked saltwater pools at four Indo-West Pacific localities (Sinai Peninsula, Funafuti Atoll, Maui and Hawaii Islands), with the description of one new genus and four new species. Zool. Verh. Leiden 128:1-48, 13 figs., 7 pls. RaTHBUN, Mary J. 1919. Stalk-eyed crustaceans of the Dutch West Indies. In Boeke, Rapport betreffende een voorloopig onderzoek 44 Proceedings of the Biological Society of Washington naar den toestand van de Visscherij en de Industrie van Zeeproducten in de Kolonie Curacao ingevolge het Ministerieel Besluit van 22 November 1904 2:317-348, 5 figs. DE SaussuRE, H. 1857. Diagnoses de quelques crustacés nouveaux de lAmérique tropicale. Rev. Mag. Zool. (2)9:501-505. ScumitrtT, Watpo L. 1936. Macruran and anomuran Crustacea from Bonaire, Curacao and Aruba. No. 16 in Zoologische Ergebnisse einer Reise nach Bonaire, Curacao und Aruba im Jahre 1930. Zool. Jahrb., Syst. 67:363-378, 3 pls. VANDEL, A. 1964. Biospéologie. La biologie des animaux cavernicoles. Paris xviii + 619 pp. 1965. Biospeleology. The biology of cavernicolous animals. New York (Pergamon Press) xxiv + 524 pp., 80 figs. lo. 5, pp. 45-50 23 April 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON OBSERVATIONS ON LEPUS CALLOTIS IN NEW MEXICO By MicHart A. BoGAN AND CLYDE JONES National Fish and Wildlife Laboratory, National Museum of Natural History, Washington, D.C. 20560 The occurrence of Lepus callotis gaillardi Mearns in the United States has been unclear since the original description (Mearns 1896:560). During the resurvey of the Mexican- United States boundary in 1892 and 1893, E. A. Mearns and F. X. Holzner obtained six of these unique jack rabbits. The skin labels on three specimens taken 16 and 29 June 1892, state the animals were taken on the “Mexican Boundary Line, near White Water, Chihuahua.” Two others, including the type of L. gaillardi, taken on 17 June 1892, read “near White Water, Chihuahua, on the Mex. Bound. Line.” The skull tags of these five specimens read variously: “Plyas (sic) Valley, near White Water, Chihuahua, Mex.; West fork of Plyas (sic) Valley, near White Water, Chihuahua, Mexico; Mex. Bound. Line; Plyas (sic) Valley near White Water, Chihuahua.” The holotype (USNM 20525/35714, skin and skull) also possesses red type labels, presumably added at the time of description, stating: “Playas Valley, near Mon. No. 63, Mex. Bound. Line.” The original description (Mearns 1896:562) gives the type- locality as: “west fork of the Playas Valley near monument No. 63, Mexican boundary line.” The remaining specimen, taken 15 September 1893, came from “Plyas (sic) Valley, Mex. Bd. Line” according to the skin tag, or “East Fork of Plyas (sic) Valley” according to the skull label. The skin tag of this specimen also bears the notation “Topotype.” All six of the above specimens were taken while the Bound- ary Commission was encamped at White Water, 1 mi S Monument No. 61. During the first visit there (13-29 June 5—Proc. Biot. Soc. WasH., Vou. 88, 1975 (45) 46 Proceedings of the Biological Society of Washington 1892) Mearns explored the east and west Playas Valleys (Mearns 1907:11), traveled south to San Francisco Canyon (5 mi SW Monument No. 63), and explored the San Luis Mountains in this area (Mearns 1907:12). Mearns’ original “Field Catalog” (files of USNM) for this first visit gives “Whitewater, Chihuahua, Mexico” as the locality for specimens taken 16-17 June 1892, and “Mexican boundary line, near Whitewater” as the locality for the 29 June 1892 specimen. We can find no original Field Catalog listing the specimen taken in September 1893. During his travels with the Boundary Commission, Mearns was in the company of competent surveyors. The main missions of the party were to resurvey the border and erect new monuments. It thus seems likely that he would have been cognizant of the actual provenance of these specimens. Had the animals been taken in New Mexico, we presume he would have labeled them as from Grant County (now Hidalgo County), New Mexico, as he did with other specimens taken before and after his stay at Whitewater. Although no definite answer as to the exact locality of these animals can be given now, our conclusion is that all the animals came from the Mexican side of the boundary. The plethora of localities for these jack rabbits has resulted in considerable confusion by later workers. Lyon and Osgood (1909:28), apparently relying on the type description (Mearns 1896:562), gave the type-locality as “West fork of the Playas Valley, near Monument No. 63, Mexican boundary line.” Bailey (1932:53) ascribed the holotype to the “Mexican bound- line, near Monument 63, west arm of Playas Valley, south- western New Mexico.” Furthermore, he stated (Bailey 1932) that “besides the type-specimen that he (Mearns) collected, he secured a series of five from White Water just below the international boundary line.” Bailey (1932) also reported, on the basis of field work in New Mexico, that “ranchmen in 1908 reported white-sided jack rabbits, which they called antelope rabbits” in the south- ern end of both the Playas and Animas Valleys. E. A. Goldman (files of Biological Survey 1908) also reported having seen Jack rabbit rare in New Mexico 47 this species in these areas, although no specimens were known from the Animas Valley. Poole and Schantz (1942:211), in their catalog of type- specimens, added to the confusion by giving the type-locality for L. gaillardi as: “West fork of the Playas Valley, near Monu- ment No. 63, Mexican boundary line, Grant County, N. Mex.” Hall (1951:188) listed all localities known to him for L. gaillardi, including Poole and Schantz’s Grant County “record” and Bailey’s reported sight records for the Animas Valley. The map drawn from these localities artifically extended the range of this species far into New Mexico because present day Grant County is well to the north of the Mexican boundary; the County having since been divided to form Luna and Hidalgo Counties. Hall and Kelson (1959:286) gave the type-locality as in Hidalgo County, although their map (p. 288) does not reflect this change. Anderson and Gaunt (1962) and Anderson (1972) did much to correct the confusion regarding Mearns’ localities, and Anderson and Gaunt (1962:8) also recorded two specimens taken in the “south end of west side of Playas Valley, 4600 feet” on 6 September 1931, and housed in the Museum of Vertebrate Zoology, University of California, Berkeley. These two specimens appear to be the only unquestioned records from New Mexico, and are reported as such by Findley et al. (1975). Anderson and Gaunt (1962), however, allocate the type and the 1893 specimen to New Mexico, an allocation that seems erroneous to us. Regardless of the actual provenance of these early specimens, L. callotis gaillardi must be regarded as an extremely rare mammal in this area of New Mexico and Chihuahua, and the recent acquisition of a specimen seems noteworthy in this respect. We obtained an adult female L. callotis about 0.5 km N Cloverdale, Hidalgo County, New Mexico (labeled “Clover- dale”), on 19 July 1974. The animal, the first taken in 43 years in New Mexico, also substantiates the 1908 reports of this species in the Animas Valley. Our specimen (USNM 506267) was lactating and also con- tained two embryos measuring 9 mm (crown-rump). Of the 48 Proceedings of the Biological Society of Washington five females taken by Mearns and Holzner, one was noted to contain “three small fetuses.” Anderson (1972) has noted the tendency for these hares to occur in pairs, an observation also made by E. A. Goldman and V. Bailey (reports in files of Biological Survey). Our own observations confirm this, and during our field work around Cloverdale on 18 and 19 July 1974, we saw at least two and perhaps as many as four pairs. On 20 July we returned to this area and made an intensive search for L. callotis, but found none. The jack rabbits, resting within 5 m of one another, flushed in front of us at distances from 5 to 25 m and ran swiftly to- gether for distances up to 0.5 km. Their nests, which concealed them perfectly, consisted of bare oval patches (approximately 30 cm long) in clumps of tabosa grass (Hilaria mutica). All jack rabbits seen by us were flushed within stands of this grass, which occurred in large patches in this part of the Animas Valley. Once seen (as a jack rabbit vacated it), the nests were easy to spot; one such nest contained a partially chewed flower of Cirsium sp. The stomach of our specimen contained finely-chewed green plant material. Our own observations agree closely with those made by Vernon Bailey in Chihuahua in 1932 and contained in un- published files of the Biological Survey: “My greatest thrill of the trip was the first view of this wonderful rabbit in life, and my first specimen collected. Riding across a high grassy mesa, about 10 miles southwest of Ramos, at 5,500 feet in (the) Upper Sonoran Zone on October 30, two of these rabbits jumped out of the brown grass just in front of my horse, and after running a few rods, squatted in the grass which matched their color exactly. The curly buffy brown hair of the backs was all that could be seen except the eyes of the nearest one which I quickly auxed from the saddle. The other, a little farther off, I tried to stalk on foot but did not get near enough for the shotgun and missed it with the rifle as it sped over the smooth prairie with white sides glistening and great ears erect, as striking a mark as an antelope, and almost as swift. “They are said to be common on these high prairie mesas Jack rabbit rare in New Mexico 49 and always two together. The local name is Snow Sides. The specimen was an adult male, perfectly healthy and as fat as a rabbit ever gets. Well stewed it was delicious meat and made two meals for four of us. Its stomach contained tender grasses and other prairie plants of this high prairie flora.” ACKNOWLEDGEMENTS We thank B. Harvey and P. Mehlhop for assistance in the field. Henry Setzer and Don Wilson read the manuscript. LITERATURE CITED ANDERSON, S. 1972. Mammals of Chihuahua, taxonomy and distribu- tion. Bull. Amer. Mus. Nat. Hist. 148:149-410. ——,, AND A. S. Gaunt. 1962. A classification of the white-sided jack rabbits of Mexico. Amer. Mus. Novitates 2088:1—-16. Batty, V. 1932. Mammals of New Mexico. North Amer. Fauna, No. 53: 1-412. FinDLey, J. S., A. H. Harris, D. E. Witson AND C. Jones. 1975. The Mammals of New Mexico. Univ. New Mex. Press. In press. Hau, E. R. 1951. A synopsis of the North American Lagomorpha. Univ. Kans. Publ., Mus. Nat. Hist. 5:119-202. , AND K. R. Ketson. 1959. The Mammals of North America. Ronald Press, New York. Vol. 1. xxx + 546 + 79 pp. Lyon, M. W., Jn. anp W. H. Oscoop. 1909. Catalogue of the type- specimens of mammals in the United States National Museum, including the Biological Survey Collection. U.S. Nat. Mus. Bull. 62: x + 325 pp. Mearns, E. A. 1896. Preliminary description of a new subgenus and six new species and subspecies of hares, from the Mexican border of the United States. Proc. U.S. Nat. Mus. 17:551-565. 1907. Mammals of the Mexican boundary of the United States. U.S. Nat. Mus. Bull. 56:xv + 530 pp. Poo.E, A. J. AnD V. S. ScHantz. 1942. Catalog of the type specimens of mammals in the United States National Museum, including the Biological Surveys Collection. U.S. Nat. Mus. Bull. 178:xiii + 705 pp. 50 Proceedings of the Biological Society of Washington baie oo HL Es yet iy ey, L vy ; ats Pants) Ny tole ss ree a ip if ', pp. 51-58 23 April 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW SPECIES OF FRESHWATER SHRIMP (GENUS ATYA) FROM THE PACIFIC DRAINAGES OF PANAMA By Lawrence G. ABELE Department of Biological Science, Florida State University, Tallahassee, Florida 32306 This new species of Atya was collected during a survey of the decapod crustacean fauna of Panama (Abele, 1972) and is described here as a contribution towards a knowledge of that fauna. The abbreviation cl refers to carapace length measured from the orbital margin to the posterior margin of the carapace; tl to total length measured from the apex of the rostrum to the distal margin of the telson; UPRC to the University of Panama Reference Collection, Panama City, Republic of Panama; USNM to the National Museum of Natural History, Wash- ington, D.C. Atya dressleri new species Figures 1-2 Material: Panama, Pacific drainage, Veraguas Province, Rio Santa Maria drainage, small stream N of Santa Fe, fast flowing, riffles present, 600 m elevation, 9 February 1962, H. L. Loftin, E. W. Tyson, 6 specimens, USNM.—Panama, Pacific drainage, Veraguas Province, Rio Santa Maria drainage, headwaters of Rio San Juan about 15 km above Calobre, fast flowing, riffles present, 566 m elevation, 20 February 1973, L. G. Abele, M. H. Robinson, LGA 73-16, 1 male holotype, 2 female paratypes, USNM, 1 male paratype UPRC.—Panama, Pacific drainage, Panama Province, Rio Pacora drainage, Rio Cabra above Cerro Azul, fast flowing, riffles present, 650 m elevation, 1968, 2 males (molts of aquarium specimens ), R. Dressler, USNM. Measurements: Male holotype cl 20.2 mm, tl 52.4 mm; male paratype cl 9.5 mm, tl 31.4 mm; female paratypes cl 18.5, 20.2 mm, #l 59.4, 60.2 mm. 6—Proc. Biot. Soc. WasuH., Vou. 88, 1975 (51) 52 Proceedings of the Biological Society of Washington Diagnosis: Rostrum triangular in dorsal view, no lateral teeth; third pereiopod with small, depressed tubercles; podobranch present on second maxilliped. Description: The rostrum (Fig. 1A) is acute, triangular in dorsal view, laterally compressed with a distinct dorsal carina. It just reaches to the distal margin of the basal antennular segment. The lateral margins lack lobes or extensions, being almost straight from the orbits to the apex. There are 1—2 small dorsal notches, which may be absent, just pos- terior to the apex and 1-2 distinct ventral teeth (Fig. 1B). The antennal spine arises at the lower margin of the orbit and a bluntly acute pterygostomial spine is present. There are no other spines on the cara- pace, which lacks pubescence. The first three pleura are broadly rounded; the fourth is slightly angled and the fifth and sixth are distinctly angled but blunt (Fig. 2). There is a subtriangular lobe with a dorsal notch above the sixth pleuron. There are no spines on the ventral margins of any pleura in either sex. The fourth abdominal segment is about 1.25 times the length of the fifth, 1.1 times the sixth and 1.06 times the length of the telson. The sixth segment is about 1.15 times the length of the fifth segment and only slightly shorter than the telson. The length of the telson is slightly less than twice the width (Fig. 1E). It is armed dorsally with two rows of 10-12 spines (one row on each side of the midline). The rows begin about four-sevenths of the distance from the posterior margin with each row diverging distally and ending with 1 or 2 small spines at the lateral angles adjacent to a longer medial spine on the distal margin of the telson. There is a distinct depression between the rows of spines which terminates in a large spine just proximal to the distal margin of the telson. There are long, strong setae on the distal margin of the telson. The anal tubercle (Fig. 1D) is long and acute. The eyes (Fig. 1B) are well-developed and pigmented. The basal antennular segment (Fig. 1C) is armed with a well- developed stylocerite which does not extend to the distal margin. The distal margin is armed with 7-9 dark colored spines which decrease in size laterally. The second segment is about twice as long as wide and has a row of 7-10 small spines on the mid-dorsal surface. The distal margin is armed with 5-6 spines. The ultimate segment is armed with 5-8 small spines at the insertion of the lateral flagellum. The ultimate segment is about 0.56 times the length of the second; the second is slightly shorter than the total length of the basal segment but is about 1.77 times the visible portion of the basal segment. The basicerite has a blunt lobe dorsally and a small lateral spine which does not extend to the tip of the stylocerite. The lamella of the scaphocerite extends beyond the antennular peduncle; a lateral tooth arises about two-thirds of the distance from the base. The carpocerite is shorter than the antennular peduncle and is armed with about 12 small spines at the distal margin. The antenna is slightly longer than the body. New freshwater shrimp from Panama 53 The branchial formula is: Maxillipeds Pereiopods 1 2} 3 1 2 3 4 5 pleurobranch - - - a 1 1 1 1 arthrobranch = = 9} 1 _ = = = podobranch - iL = a ») se epipod 1 1 1 1 1 1 1 — exopod. il 1 1 = = ni a as The third maxilliped extends to the distal end of the carpocerite or slightly beyond it; it does not extend to the distal end of the antennular peduncle. The exopod extends to the distal margin of the penultimate segment. The antepenultimate segment is about 1.3 times the length of the penultimate which is about 1.2 times the length of the ultimate segment. There are rows of setae on the penultimate and ultimate segments, the latter having large serrate, spoon-shaped setae distally. The first two pereiopods are typical of the genus and similar. The third pereiopod (Fig. 1F) is unarmed and is only slightly longer than the fourth and fifth; the propodus extends beyond the antennular pe- duncle. There is a row of pubescence from the proximal lateral part of the merus to the distal part of the propodus. Depressed squamose tuber- cles are present on the lateral surfaces of the propodus, carpus and distal three-fourths of the merus. The merus is slightly more than 5 times as long as wide; it is about 3 times the length of the carpus, more than twice the length of the propodus and about 7 times the length of the dactylus. The dactylus (Fig. 1G) has a circular depression laterally with setae pres- ent; the ventral margin is armed with about 7 strong spines and the distal portion is narrow and acute. The tip of the dactylus of the fourth pereiopod extends to the distal margin of the antennular peduncle. The fourth pereiopod has many fewer tubercles than the third and is armed with 4 spines: 1 on the distal lateral surface and 2 long movable ones on the inner surface of the merus; the distal one is very long being about one-sixth the length of the merus; the fourth spine is on the proximal part of the carpus. The carpus is slightly less than half the length of the merus and shorter than the propodus. The dactylus is similar to that of the third pereiopod. The fifth pereiopod extends just to the basicerite; it is armed on the inner surface of the merus with 3 evenly spaced spines and 1 spine on the distal outer surface of the merus; the propodus is armed with rows of small dark-colored spines on the inner surface. The merus is longer than the carpus but both are shorter than the propodus. The dactylus is similar in form to the others. In terms of the relative lengths of segments the merus becomes shorter from the third to the fifth pereiopod while the propodus becomes longer. The endopod of the first male pleopod (Fig. 11) is subrectangular in 6 54 Proceedings of the Biological Society of Washington Fic. 1. Atya dressleri new species. A, Dorsal view of anterior region of carapace. B, Lateral view of anterior portion of carapace. C, An- tennal peduncle. D, Preanal carina. E, Telson. F, Third pereiopod. G, Dactylus of F. H, Appendix masculina of male. I, Endopod of first male pleopod. D, H, I, Holotype male, others paratype female. Scale = 15 mm for A, F; 7 mm for B, C, E; 4 mm for D, G, H, I. shape, it is covered on one side with very small backwardly directed spines; the other side is unarmed. It is about subequal in length to the exopod. The appendix masculina (Fig. 1H) is slightly more than twice as long as wide; it has strong curved setae present on the margins and a few scattered setae anterior to the appendix interna. The uropods New freshwater shrimp from Panama 5o Fic. 2. Atya dressleri new species. Lateral view of male showing one of the more common color patterns. Scale = 15 mm. both extend well beyond the telson; the diaeresis of the exopod is armed with about 18 to 21 spinules. Color: The color pattern of a live male is shown in Figure 2. The ground color is light brown with both yellow and black specks present. The carapace is light brown with a dark brown oblong area on the anterior lateral surface just above an elongated black spot. Posterior to and above these is a short yellow rectangular area with a black posterior border. On the lower posterior part of the carapace is a long yellow stripe distinctly bordered by black. The first abdominal segment has a sinuous black area covering the anterior part of the segment; the black is bordered by a distinct narrow yellow band. The second abdominal segment has the anterior and posterior pleural angles bordered in black with an oblique black stripe medial to each. There is a distinct yellow area bordered by a black band on the dorsal posterior part of the seg- ment. The third segment has the pleural angle black with an oblique stripe anterior to it. There is an hourglass shaped yellow area bordered in black on the posterior dorsal surface of the fourth segment; the pleural angle is black with an oblique black stripe anterior to it. The pleural angle of the fifth segment is black with an oblique black stripe anterior to it. The sixth segment has almost the entire dorsal surface yellow with a black border.. There is a distinct black spot at the base of the uropods. The antennular peduncle has yellow markings on the dorsal surface of the antepenultimate and penultimate segments. The flagella are brown. 56 Proceedings of the Biological Society of Washington The third through the fifth pereiopods have 7 yellow bands: 1 at the coxa-ischium, 1 about the middle and 1 towards the end of the merus, 1 at the merus carpus joint, 1 distally on the carpus and 1 each at the proximal and distal ends of the propodus. The remainder of the shrimp is an even brown color. The color pattern varied among individuals with some individuals being an even brown, others having the dorsal markings fused into a dorsal stripe and others having variations on the color pattern described. There was no obvious sexual dimorphism although small specimens were usually an even light brown without any pattern. Etymology: This species is named for Robert L. Dressler who intro- duced me to the natural history of Panama. Type locality: Panama, Veraguas Province, approximately 15 km above the town of Calobre, 566 meters elevation, Pacific drainage, Rio San Juan emptying out of Laguna Yeguada eventually entering the Rio Santa Maria drainage system. The stream was from 0.2 to 4 m wide with the depth to 1 m; riffles were present in narrow sections and there was overhanging vegetation. Habitat: Atya dressleri occurs in small fast running freshwater streams from about 560 to 650 m elevation. All of the specimens were collected from vegetation or overhanging roots in areas where the current was swift. Distribution: The species has been taken at three localities in Panama; the type locality, above Santa Fe in the same drainage and on Cerro Jefe in the Pacora drainage, Panama Province. Remarks: Atya dressleri is most closely related to A. lanipes Holthuis, 1963 which occurs on St. Thomas, Virgin Islands and on Puerto Rico. The two species can be immediately distinguished in that A. dressleri has squamose tubercles on the third pereiopods while A. lanipes lacks tubercles. Other features which appear to separate the two species are the following: the merus and carpus of the third pereiopod are unarmed in A. dressleri while the merus is armed with 3 and the carpus with 1 spine in A. lanipes; the telson is less than twice as long as wide in A. dressleri while it is more than twice as long as wide in A. lanipes; the dorsal surface of the telson is armed with 10-12 spines in A. dressleri while it is armed with 6-7 spines in A. lanipes; a podobranch is present on the second maxilliped in A. dressleri while it is lacking in A. lanipes. Atya dressleri can be distinguished from all other American species of Atya in that the carapace is unarmed in the hepatic and postorbital regions and the rostrum lacks lateral lobes. The species is fairly long lived. Robert Dressler has kept adult indi- viduals alive in an aquarium for more than 5 years. Discussion: Six species of the genus Atya occur in Panama. Atya scabra (Leach, 1815) and A. innocous (Herbst, 1792) occur in Atlantic drainage streams while A. dressleri new species, A. tenella Smith (1871), A. rivalis Smith (1871) and A. crassa Smith (1871) occur in Pacific drainage streams. Doflein (1900) reported A. crassa from the Atlantic New freshwater shrimp from Panama 57 coast of Panama but this seems to be the only record. No other locality data were given so it was not possible to recollect the area. Atya crassa was collected in the Pacific drainage of Rio Bayano and from no other areas in Panama, and A. dressleri is known from only a few localities, but all other species are widely distributed in Panama. Some question has been raised concerning the specific distinction of the Pacific A. tenella from the Atlantic A. innocous and of the Pacific A. rivalis from the Atlantic A. scabra (see Bouvier, 1925; Chace and Hobbs, 1969). Holthuis (1966) considered A. scabra and A. rivalis to be distinct and pointed out some differences between them. While a review of the genus is not the objective of the present report, it seems worthwhile to point out that A. tenella and A. rivalis appear to be valid species. In addition, a comparison of the figures and description of A. ortmannoides Villalobas (1956) with A. tenella reveals that they are very similar. A detailed comparison between these species is necessary to resolve their status. In A. innocous the ventral borders of the third through the fifth pleura are armed with short strong denticles (Chace and Hobbs, 1969:65, fig. 10c), while in A. tenella they are unarmed. In A. scabra the ventral borders of the second through the fifth pleura are armed with short strong denticles (Chace and Hobbs, 1969:65, fig. 10f), while in A. rivalis only the third through fifth pleura are armed; the second pleuron is unarmed. These differences were found to be constant on more than 25 specimens of each of the 4 species. There appeared to be other dif- ferences in the spination and form of the antennal peduncles, in the shape of the preanal carinae and in the robustness of the legs but these were not examined in detail. LITERATURE CITED ABELE, L. G. 1972. Comparative habitat diversity and faunal rela- tionships between the Pacific and Caribbean Panamanian deca- pod Crustacea: A preliminary report with some remarks on the crustacean fauna of Panama. pp. 125-138. In M. L. Jones (ed.). The Panamic biota: some observations prior to a sea level canal. Bull. Biol. Soc. Washington, No. 2, 1972. Bouvier, E. L. 1925. Recherches sur la morphologie, les variations, la distribution geographique des Crevettes de la famille des Atyides. Encyl. Entom. Ser A, 4:1-370, figs. 1-716. Cuace, F. A., JR. anD H. H. Hosss, Jr. 1969. The freshwater and terrestrial decapod crustaceans of the West Indies with special reference to Dominica. Bull. U.S. Nat. Mus. 292:v, 1-258, figs. 1-76. Dor tein, F. 1900. Weitere Mitteilungen tiber dekapode Crustaceen der k. bayerischen Staatssammlungen. Sitzung. Math.-Phys. Classe Kgl. Bayer. Akad. Wissen. Munchen 1900, 1:125—145. Hersst, J. F. W. 1791-1796. Versuch einer Naturgeschichte der 58 Proceedings of the Biological Society of Washington Krabben und Krebse nebst einer systematischen Beschreibung ihrer verschiedenen Arten, 2:i-viii, 1-226, pls. 22-46. Hottnuis, L. B. 1963. Two new species of fresh-water shrimp (Crustacea Decapoda) from the West Indies. Proc. Kon. Nederlandsche Akad. Wetensch. 66:61—-69, 3 figs. 1966. The freshwater shrimps of the Island of Annobon, West Africa. In The R/V Pillsbury deep-sea biological expedition to the Gulf of Guinea, 1964-65. II. Stud. Trop. Oceanogr., No. 4(1):224-239, figs. 1-5. Leacu, W. E. 1815. A tabular view of the external characters of four classes of animals which Linné arranged under Insecta; with the distribution of the genera composing three of these classes into orders, etc. and descriptions of several new genera and species. Trans. Linn. Soc. London 11:306—400. SmitH, S.I. 1871. List of the Crustacea collected by J. A. McNiel in Central America. Rep. Peabody Acad. Sci. 1869:87-98. VitLALogos, A. F. 1956. Contribucion al conocimiento de las Atyidae de Mexico I. Una nueva especie de Atya de la vertiente del Pacifico del estado de Michoacan. An. Inst. Biol., Univ. México 26(2[for 1955] ) :459—475, figs. 1-6. i 7, pp. 59-72 23 April 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ASELLUS (ASELLUS) ALASKENSIS, N. SP., THE FIRST ALASKAN ASELLUS, WITH REMARKS ON ITS ASIAN AFFINITIES (CRUSTACEA: ISOPODA: ASELLIDAE) By THomas E. BOWMAN AND CHARLOTTE HOLMOUIST Department of Invertebrate Zoology Smithsonian Institution, Washington, D.C. 20560, and Sektionen for Evertebratzoologi Naturhistoriska Riksmuseet, S-10405 Stockholm North American freshwater isopods of the family Asellidae are widespread east of the Great Plains in the United States and southern Canada (to about 46°N). West of the Great Plains asellids apparently do not occur naturally until they reappear in the Pacific Coast states and the province of British Columbia ( Williams, 1970; Bowman, 1974). Asellus communis Say, a species that is distributed mainly in the northeastern United States and southern Canada, occurs in Boulder and Larimer counties, Colorado, but was probably introduced there by man (Williams, 1970). Winger, Peters, Donahoo, Barnes, and White (1972) reported Asellus tomalensis Harford from the Provo River, Utah. Dr. Winger kindly sent some of the Utah specimens, and they proved to be A. racovitzai racovitzai Williams, a form with a distribution similar to that of A. communis. It appears likely that the Provo River populations were also introduced. Assuming artificial transference of the Colorado and Utah asellids, a wide geographic gap separates the eastern and west coast populations of Asellus. It has been suggested that the western species, A. occidentalis Williams and A. tomalensis Harford, are more closely related to Asian than to eastern North American asellids (Henry and Magniez, 1970), but this seems unlikely (Bowman, 1974). 7—Proc. Biot. Soc. WAsH., Vou. 88, 1975 (59) 60 Proceedings of the Biological Society of Washington The northernmost record for a west coast Asellus is Double Bay, Hansen Island, Queen Charlotte Strait (50°35’N, 126° 46’W ) [Williams (1970) erroneously located Double Bay on Vancouver Island; the correct location is given in Bousfield’s (1963) station list (Bousfield, in litt.)]. The 4 samples re- ported below of a new species from Alaska, from which asellids were previously unknown, extend the record slightly north of the Arctic Circle. The new species is noteworthy not only because it is the first known Alaskan asellid, but also because it belongs to Asellus (Asellus), restricted by Henry and Magniez (1970) to include only 5 Eurasian species and 8 troglobitic species from Japan. The Alaskan asellid thus has affinities with Eurasian rather than with North American species of Asellus, most of which have been assigned to Conasellus Stammer (1932) by Henry and Magniez. The re- sulting zoogeographical implications are discussed below. Asellus (Asellus) alaskensis new species Figures 1-10, 12-24, 27-32 [?] Asellus sp.—Watson, Davis, and Hanson, 1966, p. 576 [in list]. Material examined: All from the Noatak and Selawik lowlands, Alaska. Niglaktak Lake (66°35.5’N, 159°56’W), 22 August 1968, 1 6, 2 9; 19 September 1970, 59 ¢ 9; collected by Charlotte Holmquist. Inland Lake (66°28’N, 159°50’W), 22 August 1968, 1 @, collected by Charlotte Holmquist. (For details of the 2 lakes see Holmquist, 1975). “Lake No. 1,” Noatak River Drainage Basin, 14.5 km east of Kiligmak Inlet, collected by Donald G. Huggins: 15 August 1973, 43 69; 18 August 1973, open water on Lemna, 40 6 @ Types: Holotype, 7 mm ¢, from “Lake No. 1,” August 1973, USNM 151224. The remaining specimens listed above are paratypes. The paratypes from Niglaktak Lake have been deposited in the Naturhistoriska Riksmuseet, Stockholm; those from “Lake No. 1” have been deposited in the National Museum of Natural History, Smithsonian Institution, Washington, except for 10 specimens from the 18 August collection which were returned to Dr. Huggins. > Fics. 1-10. Asellus alaskensis: 1, 2, dorsal; 2, 2 head and pereonite 1, dorsal; 3, @ antenna 1, dorsal; 4, 2 antenna 2, ventral; 5, ¢ right mandible; 6, Incisor of ¢ right mandible; 7, Incisor and lacinia of ¢ Remarks on first Alaskan Asellus 61 left mandible; 8, 6 mandibular palp; 9, 2 right uropod, dorsal; 10, Protopod of 9 left uropod, ventral. Fic. 11, Asellus latifrons, 6 urosome (copied from Birstein, 1951). 62 Proceedings of the Biological Society of Washington Diagnosis: With the characters of the subgenus Asellus as defined by Henry and Magniez (1970) and closely resembling A. latifrons Birstein (1947), but distinguished by the absence of a rostrum, the shorter and blunter postmandibular lobes, the less prominent caudomedial lobe of the telson, the relatively shorter uropodal rami, and the narrower 9 pleopod 2. Description: Length of largest ¢ (excluding antennae and uropods) 7.4mm. Other ¢ ¢ with differentiated pleopods 1 and 2, 5.3-6.2 mm. 2 @ with small oostegites, 6.2 mm. Body gradually increasing in width to pereonite 7; greatest width about % length. Head about twice as wide as long; anterior margin shallowly concave; postmandibular lobes separated by deep fissure, latter broader and more rounded medially than in A. latifrons; head more rounded anterior to fissure and post- mandibular lobes shorter and more obtuse than in A. latifrons. Pereonites broadly expanded laterally; insertions of pereopods set well in from lateral margins of pereonites; coxae much reduced, not visible in dorsal view. Pereonite 7 overlapping anterolateral parts of pleotelson. Telson about 44 wider than long; caudomedial lobe relatively broader and less sharply delimited than in A. latifrons. Margins of telson densely setose. Antenna 1 reaching distal third of last segment of antenna 2 peduncle; flagellum of 6-7 segments, last 3 segments each bearing esthete. An- tenna 2 about 0.6 length of body; last segment of peduncle about 1.5 times as long as preceding segment; flagellum of 25-35 segments. Mandible with 4-cuspidate incisors and lacinia mobilis; spine row of right mandible with 9 dentate and plumose spines, of left mandible with 7 plumose spines; palp as in Fig. 8. Maxilla 1, apex of outer lobe with 11 dentate spines and 1 plumose seta, body of outer lobe with 2 additional plumose setae; inner lobe with 4 apical plumose setae. Maxilla 2, outer lobe with 12 and 10 setae on outer and inner lamina respectively; inner lobe with oblique row of 23 setae. Maxilliped with 4—5 retinaculae. Pereopod 1, propus nearly twice as long as wide, palm defined by 3 spines borne on smal] boss in ¢, but flush with palm in @; dactyl with 7-8 spines on posterior margin. Dactyls of pereopods 3-7 with 2-5 spines, fewer on more posterior dactyls. Pereopod 4 more robust in ¢ than in 9; merus and carpus broader and more spinose in ¢; propus of ¢ with cluster of slender spines distally. Male pleopod 1 quadrate, slightly longer than pleopod 2; protopod half as long as exopod, with 3-4 retinaculae; exopod subrectangular, nearly twice as long as broad, lateral and distal margins armed with long plumose setae, those on distal margin longer, shorter naked setae inserted submarginally. Male pleopod 2 protopod about as long as wide, with 3-5 distomedial setae. Exopod % as long as protopod; proximal segment with 1-4 setae on lateral margin; distal segment oval, with 8-10 lateral setae and 1—2 larger medial setae, proximomedial part of posterior surface with well developed catch lobe. Endopod (see discussion below for terms used) Remarks on first Alaskan Asellus 63 Fics. 12-18. Asellus alaskensis: 12, 6 maxilla 1; 13, ¢ maxilliped; 14, $ right pereopod 1; 15, 2 left pereopod 1; 16, 2 pereopod 4; 17, ¢ pereopod 4; 18, @ left pereopod 7. 64 Proceedings of the Biological Society of Washington Fics. 19-24. Asellus alaskensis: 19, ¢ pleopod 2; anterior; 20, Tip of endopod of ¢ pleopod 2, anterior; 21-22, Same, posterior; 23, Same, lateral; 24, Exopod of ¢ pleopod 2, posterior. Fics. 25-26, Asellus aquaticus, tip of endopod of ¢ pleopod 2: 25, Anterior; 26, Lateral. Abbreviations: c, capitulum; a.l., anterior lobe; f, fissure; l.s., labial spur. Remarks on first Alaskan Asellus 65 Fics. 27-31. Asellus alaskensis, anterior views of pleopods: 27, 2 pleopod 2; 28, $ pleopod 1; 29, ¢ pleopod 3; 30, ¢ pleopod 4; 31, ¢ pleopod 5. slightly longer than exopod, about twice as long as wide (width not including spur); curving gently laterally. Fissure bounded proximally by transverse sclerotized bar; medial lip of fissure with U-shaped labial spur connecting proximally with transverse bar. Anterior lobe low, broadly rounded, without medial spine present in A. aquaticus. Ca- pitulum with central cannula, produced posteriorly into ledge and cau- dally into spinose lobe. Female pleopod 2 oval, about 1% times as long as wide, with about 15 plumose setae on distal and lateral margins. Pleopod 3 exopod about 1% times as long as wide, proximal segment about % length of distal segment. Exopods of pleopods 4 and 5 with about 15 and 8 setae re- spectively on proximolateral margins; exopod of pleopod 4 with a partial suture; both pleopods with double false suture distally in form of crescent. Uropods about %4 as long as pleotelson, not so slender and relatively shorter than in A. latifrons; armature of long and short setae as in Figure 9. _ Etymology: The specific name, alaskensis, refers to the occurrence of the new species in the state of Alaska. Relationships: The presence of 4 setae on the inner lobe of maxilla 1, the boss proximal to the palm on the male pereopod 1, and the rounded female pleopod 2, taken together, place the Alaskan species in Asellus (Asellus) as restricted by Henry and Magniez (1970). Within the sub- genus, A. alaskensis and the west Siberian A. latifrons are set apart from the other species by having lateral head incisions and reduced coxae not visible in dorsal view. The male pleopod 2 endopod tip of A. latifrons is not illustrated in sufficient detail to permit a comparison of this taxonomically important structure in the 2 species, but it seems evident 66 Proceedings of the Biological Society of Washington from the overall similarity that A. alaskensis and A. latifrons are descendents of a common ancestor and have no close relatives among other species of Asellus (Asellus). They, as well as A. epimeralis Birstein from Lake Bount, exhibit certain characters that Birstein con- sidered primitive because they are found in juveniles but not in adults of A. aquaticus: the large size of the rami in proportion to the protopod of the male pleopod 2; the elongate female pleopod 2; and the straight lateral margins of the exopod of the male pleopod 1, lacking the excava- tion of the adult A. aquaticus. The wide taxonomic gap between Asellus alaskensis and other North American Asellidae has been referred to already. Other examples of Alaskan freshwater invertebrates with Eurasian rather than North Amer- ican affinities can be cited, for example among the Turbellaria (Kenk, 1953) and the diaptomid Copepoda (Wilson, 1953). Terminology for endopod of 8 pleopod 2: American zoologists, be- ginning with Steeves (1963) and followed by Williams (1970, 1972) and Fleming (1972, 1973) have applied standard terms to the 4 or fewer elements forming the tip of the ¢ pleopod 2 endopod. The num- ber and form of these elements have been given great weight in recent works on North American asellids and have formed the basis for pro- posed phylogenies, either wholly (Steeves, 1966) or in part (Williams, 1970). We have not attempted herein to apply Steeves’ terms to A. alas- kensis because homologies cannot be assumed between the Alaskan asellid and the species of Conasellus. In fact, the assumed homologies of the endopodal elements among species of Conasellus cannot be taken to have been proven beyond doubt. Perhaps because this component species can be distinguished readily by other characters, consistent terms have not been applied to the struc- tures of the endopod of the ¢ 2nd pleopod in Asellus ( Asellus) species. We found it necessary, therefore, to devise the terms used in the above description of A. alaskensis. Our terms are listed below, together with those used by Maercks (1930) and Needham (1938) for the same structures in A. aquaticus. This paper Maercks Needham basal spur processus calcariformis spur or “Sporn” fissure Spalt opening of seminal cavity labial spur processus cylindriformis conical spur, or head of + Hakchen hammer anterior lobe = —--—-----— mobile lappet capitulum Spiralhaken recurved distal tooth of lateral edge Zoogeography and ecology: Birstein (1951) interpreted the known present distribution of Asellus (Asellus) spp. in Siberia (Fig. 32) as Remarks on first Alaskan Asellus 67 A.(Asellus) Approximate border of any glaciation a . . last glaciation Fic. 32. Known distribution of epigean species of the subgenus Asellus in Asia and Alaska. Asian data based on Birstein (1951). remnants of a vast continuous distribution in preglacial times. He be- lieved that severe conditions associated with Pleistocene glaciation wiped out all but a few populations, namely A. latifrons in the Ob-Irtysh basin, A. epimeralis in Lake Bount, and A. hilgendorfi martynovi in the Lena delta. With the discovery of A. alaskensis, Birstein’s hypothesis would re- quire the preglacial distribution to stretch nearly the entire distance across northern Siberia and across the Bering Strait into an area of Alaska on the north side of Kotzebue Sound. The Bering Strait presents no obstacle to this distribution, since the Bering Land Bridge was present during the early Tertiary until the late Miocene and again during most 68 Proceedings of the Biological Society of Washington of the Pleistocene (Hopkins, 1967). In preglacial times A. latifrons and A. alaskensis may have still been a single species, with the now remaining marginal populations beginning to diverge or having already differentiated as extremes of a variable species or as subspecies. Birstein would then have the Siberian glaciers of the Pleistocene exterminate the entire latifrons-alaskensis populations between the existing “relicts,” i.e., over the vast extent of about 6100 km (3800 miles). Birstein’s hypothesis may prove to be correct—only time will tell—but it suffers from 2 major deficiencies. 1. It requires more intolerable conditions resulting from glaciation in northern Siberia than can be supported by the available evidence; 2. It assumes that the freshwater fauna of northern Siberia is well known, and that further collecting will not change the overall distribution patterns of Siberian A. (Asellus) spp. As in other areas the exact extent of glaciation in Siberia is still some- what obscure, but because there was less precipitation the ice sheets were less extensive and thinner than those in Europe and North America (Fig. 32; Charlesworth, 1957; Woldstedt, 1954-1965). Areas uncovered by snow and therefore dark would have absorbed much radiant heat during summers, and the many bodies of water acted as heat reservoirs during winters. Even so, conditions during glacial periods probably would have been severe enough to have eliminated some asellid popu- lations, but other asellid populations might have survived. It is known that Alaskan lakes deeper than about 2 m generally do not freeze to the bottom (Brewer, 1958; Holmquist, 1973), and there must be many such lakes in Siberia. These lakes also have a layer of unfrozen ground beneath them; the thickness of this layer is correlated with the depth of the lake and the volume of unfrozen water in it. If only a little oxygen is present asellids should be able to survive in this layer of unfrozen water. A. aquaticus can survive up to 48 h without oxygen at a tem- perature of 16-20°C and normally requires only 0.4-0.5 mg/I1 ( Levanidov, 1949). Even if a lake should freeze to the bottom it is possible that asellids might survive. Some invertebrates, including certain insect larvae and crustaceans, can withstand being frozen in ice (Scholander, Flagg, Hock and Irving, 1953), and although there is no evidence that asellids could survive such treatment, the possibility cannot be ruled out. Although Birstein (1951) gave the impression that the distribution of asellids in northern Siberia is well known, there appear to be a number of unexplored areas where asellids might be expected to occur. A most useful summary of limnological investigations in the U.S.S.R. was given by Zhadin and Gerd (1963), and from this work one can get some idea of the extent of the areas not investigated for Asellus. Moving eastward from the Ob basin, inhabited by Asellus latifrons, we find that nothing is said about the Pur and Taz Rivers. The authors listed 1 isopod in the Yenisei fauna, which must be A. ( Baicalasellus) angarensis Dybowsky, which inhabits the upper reaches of the tributary Angara River as well as Lake Baikal. On the Taimyr Peninsula, Lake Taimyr has been studied and no asellids found, but the Pyasina and Remarks on first Alaskan Asellus 69 Taimyr Rivers have not been investigated. No information is given about the bottom fauna of the Khatanga, Anabar, and Olenek Rivers. Asellus hilgendorfi martynovi Birstein (not mentioned by Zhadin and Gerd) inhabits lakes in the delta of the Lena River (Birstein, 1951). Zhadin and Gerd listed A. aquaticus from the middle reaches of the Lena, but this must be a misidentification, since A. aquaticus does not occur in Siberia. The bottom fauna of the Yana, Indigirka, and Kolyma Rivers have not been studied, nor has that of the Anadyr. In addition to the foregoing major river systems, there are numerous uninvestigated shallow lakes in Siberia, and these may offer more favorable habitats for asellids than the deeper lakes and rivers that have received more attention from hydrobiologists. The reason why Asellus alaskensis was taken only in 2 lakes of the western part of the Holmquist survey area is obscure. This survey was undertaken during 6 summers of the period 1961-1970. About 100 lakes were investigated over a large area, from about 129°33’ to 165° Al’W, and from about 65°07’ to 71°23’N (Holmquist, 1975). Huggins’s collection (see above) comes from an area not very far from the 2 lakes mentioned, and the report by Watson, Davis and Hanson (1966) of an Asellus in the Cape Thompson area concerns the same north- western part of Alaska. Other invertebrates showing such a western occurrence are the oligochaetes Alexandrovia onegensis and Styloscolex opisthothecus (Holmquist, 1974a, 1974b). A. onegensis was taken in approximately the same area as Asellus, but it is known otherwise only from its type locality, Lake Onega in Europe. S. opisthothecus, a species originally described from localities on Kamchatka, was found in the survey area in 2 lakes slightly further north than Asellus. It belongs to a genus other- wise known from Japan, Manchukuo and Lake Baikal. These facts are apt to strengthen the significance of the Bering Land Bridge as a means of dispersal for freshwater invertebrates. They also emphasize the lack of knowledge of the freshwater fauna of vast northern areas. Niglaktak Lake is fairly weedy. It is rather shallow (maximum depth noted, about 1.8 m) with the vascular plants consisting mainly of Potamogeton and Myriophyllum. On the first visit to this lake a great quantity of the blue-green alga Gloeotrichia was obtained in the net. The invertebrate fauna of the lake also seemed luxuriant. Among others, the above-mentioned oligochaete Alexandrovia was taken here. Inland Lake appeared equally shallow (maximum depth noted about 1.9 m) but less weedy. Only small fragments of Potamogeton were secured. The plankton seemed rich and varied, and the benthic fauna was varied. Lakes of the Noatak valley are generally about as shallow. Such lakes may be presumed to warm up fairly quickly in this area, once the ice is gone from the surface. However, there is no protection from the frequent winds, which are apt to effectively stir up the water of these rather large lakes and thus moderate the heating rate. In the fall the winds are cool and presumably cause the whole body of water to cool 70 Proceedings of the Biological Society of Washington down close to freezing temperature before the surface is iced over, protecting the deeper parts from freezing. Thus, during the long winter a temperature between 1° and 4°C, but closer to the lower limit, is presumed for the benthic environment. At the end of the summer (22 August 1967) a temperature of about 14°C was measured in both lakes. Summer temperatures may be higher than those recorded during this survey. On September 19, 1970, water in Niglaktak Lake measured only 4°C. These facts may give some idea of the conditions under which A. alaskensis is living. In summary, Asellus alaskensis and A. latifrons are descendants of a common ancestor that once ranged across the entire breadth of northern Siberia and into Alaska. The slight morphological differences between the 2 species may already have existed before Pleistocene glaciation or may have developed only after the Ob-Irtysh and Alaskan populations were isolated by extermination of intervening populations during Pleisto- cene glaciations. It is possible that at least some of the intervening populations were not wiped out during the Pleistocene but still exist in unstudied lakes and rivers, and even if further investigations should demonstrate the absence of intervening populations, this would not prove that they were eliminated during the Pleistocene. Isolation and specia- tion might have occurred much earlier. ACKNOWLEDGMENTS We are grateful to Dr. Huggins for the samples from the Noatak River drainage basin and to Dr. Winger for specimens from the Provo River. The samples from Niglaktak Lake and Inland Lake were col- lected during a survey sponsored by The Arctic Institute of North America with the approval and financial support of the Office of Naval Research, under contract N 00014-70-A-0219-0001 (subcontract ONR- 412), and supported by the Naval Arctic Research Laboratory, Barrow, Alaska. Dr. Roman Kenk kindly reviewed the manuscript. LITERATURE CITED Binstein, J. A. 1947. Water slaters of Siberia. Doklady Akademii Nauk SSSR 57(4):409-412 [in Russian]. 1951. Freshwater isopods (Asellota). Fauna SSR, Crustacea 7(5):1-140 [in Russian; English translation by Israel Program for Scientific Translation, 148 pp. 1964]. BousFieLp, E. L. 1963. Investigations on sea-shore invertebrates of the Pacific coast of Canada, 1957 and 1959. I. Station list. Nat. Mus. Canada Bull. 185:72-89. Bowman, THomas E. 1974. The California freshwater isopod, Asellus tomalensis, rediscovered and compared with Asellus occi- dentalis. Hydrobiologia 44:431—441. Remarks on first Alaskan Asellus 71 Brewer, MAx C. 1958. Some results of geothermal investigations of permafrost in northern Alaska. Trans. American Geophys. Union 39;19-26. CHARLESWORTH, JoHN K. 1957. The Quaternary Era, with special references to its glaciation. 2 volumes, 1700 pp. London. FLEMING, LAURENCE FE. 1972. The evolution of the eastern North American isopods of the genus Asellus (Crustacea: Asellidae). Part I. Intern. Jour. Speleol. 4:221—256. ——. 1973. The evolution of the eastern North American isopods of the genus Asellus (Crustacea: 100 pp.) In press. Horxins, Davi M. 1967. The Cenozoic history of Beringia—a syn- thesis. In: Hopkins, D. M., editor, The Bering Land Bridge. 495 pp. Stanford University Press. Kencx, Roman. 1953. The fresh-water triclads (Turbellaria) of Alaska. Proc. U.S. Nat. Mus. 103(3322):163-186. LevaAnipov, V. J. 1949. The importance of allochthonous material in the water as a food supply for the water slater (Asellus aquaticus L.). Trudy Vesesoyuznogo Gidrobiologicheskogo Obshchestva [Trans. All-Union Hydrobiological Society] 1:100- 117 [in Russian]. Maercks, HANS HetMutT. 1930. Sexualbiologische Studien an Asellus aquaticus 1.. Zool. Jahrb. Abt. Allg. Zool. Physiol. 48:399-— 508, pl. 13. NEEDHAM, A. E. 1938. Abdominal appendages in the female and copulatory appendages in the male Asellus. Quart. Jour. Microscop. Sci. 81:127—150. SCHOLANDER, P. F., WALTER FLAcc, R. J. Hock, AND LAURENCE IRVING. 1953. Studies on the physiology of frozen plants and animals in the arctic. Jour. Cell. Comp. Physiol. 42, Suppl. 1:1-56. STEEVES, Harrison R., III. 1963. The troglobitic asellids of the United States: the Hobbsi group. American Midl. Nat. 71:445—451. 72 Proceedings of the Biological Society of Washington 1966. Evolutionary aspects of the troglobitic asellids of the United States: the Hobbsi, Stygius, and Cannulus groups. American Midl. Nat. 75(2):392—403. SramMeER, H. J. 1932. Zur Kenntnis der Verbreitung und Systematik der Gattung Asellus, inbesondere der mitteleuropaischen Arten (Isopoda). Zool. Anz. 99:113-131. Watson, D. G., J. J. Davis, AnD W. C. Hanson. 1966. Terrestrial invertebrates. Pp. 565-584 in Norman J. Wilimovsky and John N. Wolf, eds. Environment of the Cape Thompson Region, Alaska, xvi + 1250 pp. U.S. Atomic Energy Commission, Di- vision of Technical Information, Oak Ridge. WitiiaMs, W. D. 1970. A revision of North American epigean spe- cies of Asellus (Crustacea: Isopoda). Smithsonian Contrib. Zool. 49:1-80. 1972. Freshwater isopods (Asellidae) of North America. Biota of Freshwater Ecosystems Identification Manual No. 7: ix + 45 pp. Water Pollution Control Research Series, Environ- mental Protection Agency, Washington, D.C. Wincer, PARLEY V., EDWARD J. PETERS, MicHAEL J. DONAHOO, JAMES A. BARNES, AND Davin A. Wurre. 1972. A checklist of the macroinvertebrates of the Provo River, Utah. Great Basin Naturalist 32(4):211-219. Witson, Mitprep S. 1953. Some significant points in the distribution of Alaskan fresh-water copepod Crustacea. Proc. Second Alaskan Sci. Conf. (1951) :315-318. Wo.pstepT, Paun. 1954-1965. Das Eiszeitalter. Grundlinien einer Geologie des Quartars. 1-3. Stuttgart. ZHADIN, P. I., anp S. V. Gerp. 1961. Fauna and flora of the rivers, lakes and reservoirs of the U.S.S.R. State Training and Pedagogical Literature Publishing House, Moscow, 599 pp. [in Russian; English translation by Israel Program for Scientific Translations, 626 pp., 1963]. hy bape ek 0 peaintien ne nt ee | pled tipaaades =the: PRA, ae aaa | ‘Wire ke ee Nok! Ney Pte hie Nighy ant 65) peat NR ES IN hae, ance i sodnineaine gin an fie, pare ‘al a ei yy biel : i '* W by ae : % i pul hl v ip Ma tas iw + A XN > 5 =e wh “ i 4 “OK j y re 7 i) aayey N A ( li as ioe 8, pp. 78-76 23 April 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON MORE VARIANTS OF SCAEVOLA TACCADA (GOODENIACEAE) HAWAIIAN PLANT STUDIES 43 By Harotp St. JOHN B. P. Bishop Museum, Honolulu, Hawaii 96818 The commonest marine beach shrub in the Hawaiian Islands is the “naupaka kahakai” or Scaevola sp. It thrives on rock, or soil of gravel, sand, clay, or loam, and it extends inland where the salt spray and marine influence does. As on many other tropical shores, there are two recognized variations: S. Taccada (Gaertn.) Roxb., var. Taccada which has its leaves glabrous, and its calyx lobes 2-3 (—4) mm long, 0.3-0.5 (-—1) mm wide, lanceolate to linear. This is on all of the islands of the Leeward Islands, Hawaiian Islands, and it is also present on Kauai. The var. sericea (Vahl) St. John differs by having its leaves sericeous. This variety is infrequent on the Leeward Islands, but it has been collected on Ocean, Midway, Laysan, and French Frigates. It is also present and abundant on all of the main islands of the Hawaiian group, and it is the only one on Niihau, Oahu, and the Hawaiian Islands to the south. In the uplands of the Hawaiian Islands, there are recognized about 11 species and numerous varieties and hybrids of the genus, all of them endemic. It is assumed that they have evolved from the pantropic marine beach species. These up- land ones are quite different in appearance, having the fruit smaller and black, and the corollas white, blue, or yellow. Nevertheless, S. Taccada is looked upon as the logical an- cestor of these local endemics. On Kauai, the maritime species extends inland and upland to a conspicuous degree. This was noted, and upland specimens of it were collected in 1909 by Urbain Faurie, in 1909 and 8—Proc. Bro. Soc. WAsH., Vou. 88, 1975 (73) 74 Proceedings of the Biological Society of Washington 1916 by C. N. Forbes, in 1927 by L. H. MacDaniels, in 1933 by H. St. John, and in 1961 by O. & I. Degener. These collections were mostly made in southem Kauai, near Lihue, on the Haupu (or Hoary Head) Range, a line of jagged mountains running directly inland. Most of the collections were made at Kahoaea, or the Knudsen Gap, a pass at 500 ft altitude, 5% mi inland. Here, at the lower edge of the forest, among trees of Metrosiders collina, and with Dicranopteris linearis, the species abounds and makes vigorous bushes 1-3 m tall. It follows along the range at least to Laaukahi and to an elevation of 1,300 ft. Its leaves are thinner and of a brighter green than that of the usual coastal varieties. Because of their distinctive differences these plants are here made a new va- riety, though they have had other placements. Scaevola Taccada ( Gaertn.) Roxb., var. Fauriei (Lévl.) comb. nov. S. Fauriei Lévl., Fedde Repert. 10:150, 1911. S. sericea Vahl, var. Fauriei (Lévl.) Deg. & Deg., Phytologia 6:321 1958. S. sericea Vahl, var. Fauriei (Lévl.) Deg. & Deg., forma saintjohnii Deg. & Deg., Taxon 10:227, 1961. Original diagnosis: “Rami arcuati, teres, striati, brunei; folia oblongo- spathulata, obtusissima, glabra, flaccida, ciliata, in petiolum alatum longe attenuata, 10-16 x 4-5 cm; inflorescentiae cymosae et axillares 2-3 cm ad basin plumoso-villosae; pedunculi et pedicelli pubescentes; dentibus calycinis anguste lanceolatis, acuminatis persistentibus; drupa glabosa (sic).” Diagnostic differences: Leaves glabrous as in the species, but the calyx lobes 3.5-6 mm long, linear, glabrous. Holotype: Hawaiian Islands, Kauai, Koloa, dec. 1909, Faurie 651 (P). Type examined. Isotype (A). Specimens examined: Hawaiian Islands, all from Kauai Island, Omoe, in woods on both sides of terminal ridge, Mar. 16, 1961, O. & I. Degener 27,313 (NY, BISH), (holotype of S. sericea, var. Fauriei f. saintjohnii) ; Honopu near Lihue, on very tip of ridge, July 9, 1909, C. N. Forbes 81.K (BISH); Moloaa Gulch, 4% mi from the sea, 750 ft elev., Oct. 9, 1916; Forbes 554.K (BISH); Haupu, Lihue, top of ridge, July 9, 1909, Forbes 31.K (BISH); Waioli Valley, in uluhi fern (Dicran- opteris linearis), 2% mi from the sea, 50 m alt., Feb. 27, 1929, L. H. MacDaniels 904 (BISH); Haupu, Kipu, reforested area, 500 ft alt., Dec. 25, 1933, H. St. John & F. R. Fosberg 13,634 (BISH); Laaukahi, Koloa, grassy slope and in woods with Metrosideros and Pandanus, 900-— 1,300 ft alt., Dec. 22, 1933, St. John & Fosberg 13,464 (BISH); Variants of Scaevola Taccada 75 Hanakapiai, Napali Coast, in Pandanus forest, Dec. 31, 1930, St. John et al. 10,906 (BISH). Discussion: This Scaevola has invaded the uplands in several places on the south side (leeward) along the Haupu Range, and on the north side (windward) in Waioli Valley, and at Hanakapiai. This upland, inland variety may well represent the connecting link between the maritime, halophytic species and the upland endemic ones. Scaevola Taccada (Gaertn.) Roxb., var. Bryanii var. nov. Diagnosis holotypi: Folia glabra sunt, lobis calycis 4-5 mm longis 2-2.5 mm latis ellipticis glabris. Diagnosis of holotype: Leaves glabrous; calyx lobes 4-5 mm long, 2-2.5 mm wide, elliptic, glabrous. Holotypus: Hawaiian Leeward Islands, Midway Island, Sand Island, 4-8 ft high, on the sand mounds, Aug. 21, 1902, W. A. Bryan (BISH). Discussion: This new variety is closely related to S. Taccada (Gaertn. ) Roxb., var. Taccada, but that typical variety has the calyx lobes 2-3 mm long, and 0.5-1 mm wide. The variety Bryanii is named for its collector, William Alanson Bryan (1875-1942), professor of zoology and geology in the College of Hawaii, Honolulu; curator of ornithology, B. P. Bishop Museum; author of “Natural History of Hawaii,” 1915. 76 Proceedings of the Biological Society of Washington 9, pp. 77-90 23 April 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A RECONSIDERATION OF SOME CUBAN TROPIDOPHIS (SERPENTES, BOIDAE) By ALBERT SCHWARTZ AND ORLANDO H. Garrmo Miami-Dade Community College, Miami, Florida 33167 and Academia de Ciencias de Cuba, Instituto de Zoologia, La Habana, Cuba The most recent treatment of the small Cuban members of the boid genus Tropidophis is that of Schwartz and Marsh (1960). They recognized four species of these multispotted boas: pardalis Gundlach, 1840; maculatus Bibron, 1840; pilsbryi Bailey, 1937; nigriventris Bailey, 1937. In addition to these species, Cuba also is inhabited by other, either larger or quite differently patterned, Tropidophis: melanurus Schlegel, 1837; semicinctus Gundlach and Peters, 1865; wrighti Stull, 1938, feicki Schwartz, 1957. Schwartz and Marsh (1960: 72-74) also discussed four questionable Cuban specimens in American collections; one of these they assigned to Tropido- phis canus Cope (a species that is otherwise Bahamian; the presumed Cuban specimen may be mislabeled as to prove- nance). The three remaining snakes Schwartz and Marsh considered representative of Tropidophis haetianus Cope, a widespread Hispaniolan species which had not been previ- ously reported from Cuba. Thus, Cuba has at least nine spe- cies of Tropidophis, whereas no other Antillean island with these small boas (Hispaniola, Jamaica, Cayman Islands, Navassa Island, Bahama Islands, Caicos Islands) has more than one species. Through the efforts of the junior author, specimens from various collections in Cuba have gradually been assembled in the Instituto de Zoologia (IZ), Academia de Ciencias de Cuba. As this material accumulated, Garrido felt that addi- 9—Proc. Bio. Soc. WasH., Vou. 88, 1975 CO) 78 Proceedings of the Biological Society of Washington tional specimens of one of the less well-known smaller boids (T. pilsbryi) was included among material previously identi- fied as the widespread T. pardalis. In addition, there was a specimen which could not be easily identified as any known taxon. These specimens were sent to Schwartz, who borrowed pertinent material of this complex from the following collec- tions: American Museum of Natural History (AMNH), Acad- emy of Natural Sciences of Philadelphia (ANSP), Cleveland Museum of Natural History (CMNH), Museum of Compara- tive Zoology (MCZ), Museum of Zoology, University of Michi- gan (UMMZ), and National Museum of Natural History (USNM). These specimens, the IZ material, and specimens given to Schwartz by George R. Zug, now in the Albert Schwartz Field Series, comprise the suite of snakes examined. We wish to thank Richard G. Zweifel, George W. Foley, Edmond V. Malnate, William E. Scheele, Patricia Helwig, Ernest E. Williams, Arnold G. Kluge, George R. Zug, and Ronald I. Crombie for the loans of specimens, Fred G. Thompson for pointing out to the senior author the existence of the CMNH specimen, and Pastor Alayo for collecting speci- mens for the junior author. Our concepts of T. pardalis have been greatly influenced by the series of 44 snakes from Marianao, Habana Province, taken by Bruce B. Collette. This is the longest series of any of these small boids from a single locality. To summarize distributional data (Fig. 1) on the three species considered in the present paper: T. pardalis occurs throughout the Isla de Pinos and Cuba (but is known from only one locality—San German in Oriente Province); T. pilsbryi is known from a series of localities in and near the Sierra de Trinidad in Las Villas Province, and from Oriente Province (Cayo del Rey, Miranda; Guantanamo); T. nigri- ventris occurs in the lowlands near the Sierra de Trinidad and in eastern Camagiiey Province (Marti; 24 km SW Camagiiey). Theoretically, all three species should occur syntopically, since their ranges are apparently to some extent sympatric. How- ever, T. pardalis and T. pilsbryi are known sympatrically only from Soledad, Las Villas Province, and T. pardalis and T. 719 Some Cuban boas—Tropidophis ‘[, ‘sepsuely, “ugrounsy ey WorF (74 “hpioy ‘U *[, ‘Sa[BURITY PI[OS ‘spquaawiau “U * T, ‘sojsueiy uedo ‘sijuaaiioi snpyaovjps “d *7,,, ‘arenbs prjos-tutes isnpyaovjps ‘d “Jy, “sexenbs pros ‘whagspd -d “J, ‘satenbs uado ‘ifugspd *y, :soxenbg ‘syppsnd * 7, “saporto pos ‘syydopidou J, JO setoeds sar} JO UOTNALYsIp 94} surmoys “eqny ‘TT ‘Iq Sema | 7 Tes 08 aa 80 Proceedings of the Biological Society of Washington nigriventris also occur together at Soledad. In both these cases, actual syntopy requires confirmation. Schwartz and Marsh (1960) discussed the variation in these small boas; we will not repeat their detailed discussion. Briefly, the three species may be distinguished as follows: T. pardalis usually has 6 rows of blotches around the body, is small in size (maximum total length 315 mm in males, 303 mm in females), lacks prominent pale occipital spots, has dorsal scales usually smooth, modally 2/2 postoculars, and modally 23 scale rows at midbody; T. nigriventris usually has 8 rows of blotches around the body, is large in size (maximum total length in males about 351 mm, 367 mm in females), lacks prominent pale occipital spots but may have an extensively dark venter, has dorsal scales smooth, modally 2/2 postoculars, modally 23 scales at midbody, and has the head notably smaller when compared to that of T. pardalis; T. pilsbryi has either 8 (Oriente) or 10 (Las Villas) rows of blotches around the body, is large (judging from females) in size (maximum total length in males 213 mm, in females 374 mm), has a prominent pair of pale occipital spots at times fused into a complete pale occipital collar, has keeled dorsal scales, modally 3/3 postoculars, and modally 25 scales at midbody. From these brief diagnoses, one can readily observe that T. pilsbryi differs conspicuously in having pale occipital spots or a pale collar, a feature which is absent in T. nigriventris but which occurs as a variant in Isla de Pinos and western Cuban T. pardalis, where the spots may be present but are never so prominent and contrasting as are those of T. pilsbryi. We are convinced that the three nominal species recognized by Schwartz and Marsh are distinct. Recently gathered speci- mens of T. pilsbryi in the IZ from Las Villas and Oriente Provinces and specimens of T. nigriventris previously un- studied in the USNM from Las Villas Province, as well as reconsideration of one of the questionable specimens (USNM 137084) and other material formerly available to Schwartz and Marsh, all help to amplify the variation in these two species. In addition, a single snake (IZ 1246) from El Babiney north of Monte Libano, Oriente Province, is of interest. In the following descriptions, we admit that we may have over- Some Cuban boas—Tropidophis 81 simplified a complex problem by electing to name new taxa as subspecies rather than species, since in both cases the ranges as presently known seem broadly disjunct, and the taxa involved are completely separable from each other. How- ever, it seems presently preferable to follow this philosophy rather than name these taxa as new species. Still, the hiatuses and other complexities in distribution might suggest specific, rather than subspecific, status. Bailey (1937:42) named Tropidophis maculatus pilsbryi from three specimens from widely separated localities: Cayo del Rey near Miranda (type-locality) and Guantanamo, both in Oriente Province, and Mina Carlota in the Sierra de Trini- dad in Las Villas Province. The holotype (ANSP 20822) has 164 ventral scales (as counted by the method proposed by Dowling, 1951) and the Guantanamo specimen (MCZ 9884) has 166 ventrals; the Mina Carlota snake (UMMZ 65043), on the other hand, has 178 ventrals. In addition, the two Oriente snakes have 8 rows of blotches around the body, whereas the Las Villas snake has 10 rows. Four recently taken specimens from these two general areas plus two previously secured snakes confirm these differences, and accordingly we name the Las Villas populations as follows: Tropidophis pilsbryi galacelidus, new subspecies Holotype: 1Z 4052, from near Cafetal de Gavifia, Sierra de Trinidad, Las Villas Province, Cuba, taken 9 July 1973 by Pastor Alayo. Paratypes: YZ 1075, Pico de Potrerillo, Trinidad, Las Villas Province, Cuba, June 1955, P. Alayo; UMMZ 65043, Mina Carlota, Las Villas Province, Cuba, 19 March 1925, E. R. Dunn; USNM 137084, Soledad, Las Villas Province, Cuba, December 1954, J. D. Hardy; CMNH 1021, Guabairo, Las Villas Province, Cuba, 23 March 1941, P. N. Moulthrop. Definition: A subspecies of T. pilsbryi characterized by a greater number of ventral scales (177-183 versus 160-166 in T. p. pilsbryi), 10 rows of blotches around body at midbody, more blotches (44-50; x = 47.8) in the paramedian series than in the nominate subspecies (36-47; x = 39.5), and 25 or 27 scale rows at midbody (in contrast to 23, 24, or 25 in T. p. pilsbryi). Description of holotype: A male with snout-vent length of 187 mm, tail length 26 mm (total length 213 mm), 177 ventral scales, 35 unpaired subcaudal scales, 10/10 supralabials, 12/12 infralabials, 1/1 preoculars, 3/3 postoculars, dorsal scales keeled and in 25-25-19 rows, 48/44 blotches 82 Proceedings of the Biological Society of Washington in paramedian dorsal blotch rows, arranged in 10 longitudinal series at midbody, parietal scales not in contact. As preserved, dorsal ground color medium gray, blotches black, blotches in paramedian series largest (each encompassing about 4 to 6 scales), ventralmost series on each side confined to ventral scales and first dorsal scale row; a prominent white collar representing fused pair of occipital spots, collar constricted medially and about 8 dorsal scales long at its maximum length; upper surface of head with more or less rectangular darker gray figure, its sides slightly concave, diffuse on snout and abutting against white collar posteriorly; supralabials pale gray; infralabials, chin and throat pale gray, grading to white venter invaded about midbody by lateralmost blotch series so that in some instances blotches slightly join each other; tail tannish yellow above, clear yellow below, with black tip. Variation: The series of T. p. galacelidus consists of the holotypic male and four females ranging in total length from 183 mm to 459 mm, in tail length from 22 mm to 54 mm. One specimen (IZ 1075) is de- capitated, but its collector stated that there was a white nuchal pattern, and the ventral count of the specimen (178+) as well as the size of the dorsal blotches and the presence of 10 rows at midbody assures us that this individual is indeed assigned to this taxon. In the entire series, ventrals range from 177 to 183, subcaudals from 29 to 35; supralabials either 10 or 11, and infralabials either 11 or 12; parietal scales not in contact in any specimen; preoculars 1/1 and postoculars 3/3 in all snakes; dorsal scales regularly keeled above vent, less prominently so elsewhere on body; scale rows behind head usually 25 (one with 27), at midbody 25 (three specimens) or 27 (two), and anterior to vent 19 (one with 20). The blotches are arranged in 10 longitudinal series in all indi- viduals, and the paramedian blotch series vary between 44 and 50 (x = 47.9); the dorsal surface of the tail has from 4 to 9 blotches and is pale ventrally in all specimens except USNM 137084, whose tail is incomplete. In color and pattern the paratypes resemble the holotype except that in IZ 1075 the dorsal pattern is less obvious, and in USNM 137084 and CMNH 1021 the dorsal color is more brown and the blotches darker brown but not especially conspicuous. Large individuals have scattered pale scales associated with at least the paramedian blotch rows. Comparisons: The two subspecies of T. pilsbryi are easily distin- guished on the basis of number of ventral scales (177-183 in galacelidus, 160-166 in pilsbryi), presence of 8 rows of blotches in pilsbryi and 10 in galacelidus, and a higher number and mean of paramedian blotches in galacelidus (44-50; x = 47.9) than in pilsbryi (36-47; x= 39.5). In addition, galacelidus has either 25 or 27 scale rows at midbody, whereas in pilsbryi midbody scales are 23, 24, or 25. Both subspecies have a prominent pair of pale occipital blotches or a collar. The blotches are fused in two galacelidus (UMMZ 65058, IZ 4052) and in two pilsbryi (MCZ 9884, USNM 12361). Like galacelidus, specimens of pilsbryi lack Some Cuban boas—Tropidophis 83 parietal contact. Two pilsbryi (IZ 1076-77) have 3/2 postoculars, a condition not observed in galacelidus. Remarks: USNM 137084 is one of the four specimens which Schwartz and Marsh were unable to locate taxonomically. They stated (1960:73) that this snake had occipital spots, and thus might well have been consid- ered T. pilsbryi. However, the spots are less well defined than in all other T. pilsbryi, and the size of the snake (total length 374 mm) in com- parison with other T. pilsbryi available to them (maximum 183 mm), combined to obscure its affinities. The even larger CMNH specimen (total length 459 mm) also has obscure occipital spots. We imagine that these spots become more obscure with age (and increasing size), but they are nevertheless a constant feature (albeit faint in large adults ) of T. pilsbryi. The distribution of T. p. galacelidus centers in the Sierra de Trinidad in Las Villas Province. Three specimens (from Mina Carlota, Pico de Potrerillo, and Cafetal de Gavifia) are all from within this range, and the specimen from Soledad is from near the foot of the Sierra de Trinidad and may well have been taken within the mountains themselves. Guabairo also is near the base of this range. Comparisons of T. p. galacelidus with T. pardalis and T. nigriventris will be made below in the present paper. The name galacelidus is from the Greek for “milk” and “spot,” in allusion to the white occipital markings. Tropidophis p. pilsbryi is now known from four localities in Oriente Province, of which one is vaguely “eastern Cuba.” The closest approxi- mation of the two subspecies is about 440 kilometers, across all of Camagiiey Province and well into Oriente Province. Whether the species occurs in this hiatus remains unknown. In Oriente, T. p. pilsbryi seems to be widely distributed, from Miranda east to Guantanamo, but the localities whence this subspecies is known may also be forested uplands as appears to be the case with T. p. galacelidus. We have deliberately not designated as a paratype a T. p. galacelidus (IZ) from La Asuncién, Baracoa, Oriente Province. This specimen is a young female (snout-vent length 144 mm) with a pair of pale occipital spots, 177 ventrals, 29 subcaudals, 10/10 supralabials, 11/12 infralabials, lack of parietal contact, 1/1 preoculars, 3/3 postoculars, smooth dorsal scales in 25-27-19 rows, and 8 rows of dark blotches with 38/35 blotches in the paramedian series. In every way, this specimen agrees with our concepts of T. p. galacelidus and not with adjacent T. p. pilsbryi. The geographic hiatus between this specimen and the nearest T. p. galacelidus is about 575 km, yet T. p. pilsbryi is known from a locality only about 70 km distant. There is no question that the locality of this specimen is correct. We would have no reluctance in assigning this snake to T. p. galacelidus were it not for the extensive geographic gap between it and that subspecies and the proximity of the range of T. p. pilsbryi. The gap between La Asuncién and Trinidad encompasses much of Las Villas Province, all of Camagiiey Province, and almost all of Oriente Province; 84 Proceedings of the Biological Society of Washington this of itself is in no way prohibitive, but the absence of specimens from this entire region makes the subspecific status of the La Asuncién snake problematical. The absence of T. pilsbryi from much of the inter- vening region is by no means affirmed, and the species may persist there in suitably forested situations. On the other hand, it is possible that there is another isolated population, differing in some details from both T. p. pilsbryi and T. p. galacelidus, in extreme northeastern Oriente Province. Still another possibility is that T. p. pilsbryi and T. p. galacelidus are distinct species, a possibility that we are presently unwilling to accept. Specimens of T. p. pilsbryi examined: Cups, ORIENTE PROVINCE, Cayo del Rey, near Miranda (ANSP 20822, holotype); Santa Faz, near San Vicente (IZ 1076-77); Guantanamo (MCZ 9884); “eastern Cuba” (USNM 12361). In the introduction to the present paper we mentioned a single ques- tionable specimen from El Babiney, Hoyo del Rio Guaso, to the north of Monte Libano, hydroelectric plant, road to Las Ninfas, Guantanamo, Oriente Province. This snake (IZ 1246) is from the Ramsden collection and was collected 17 February 1921. It is indeed a puzzling snake. It is a male with a snout-vent length of 310 mm, tail 40 mm. There are 165 ventral scales and 26 subcaudals, 1/1 preoculars, 3/3 postoculars, parietals not in contact, the dorsals are smooth, and the scale row formula is 25-23-19. The dorsal blotches are in 8 rows and there are 41/40 paramedian blotches on the body and 5 on the tail. The number of ventral scales (167) eliminates pardalis (ventrals 136-155) and macu- latus (189-208) from consideration. The ventral number falls within the range of nigriventris from Las Villas Province (153-172) but not from adjacent Camagiiey Province (144-150). The ventral count does fall within the parameters of T. p. pilsbryi and the snake agrees with our concept of that taxon except that it now lacks any clear-cut indication of pale occipital blotches. The occipital area is no paler than the balance of the dorsal ground color; but this region is clear and unmarked, sug- gesting that in life the occipital area may well have been pale. A label in Ramsden’s writing identifies the snake as T. maculatus, but it surely is not that species; perhaps a factor which caused Ramsden’s assignment of this snake to T. maculatus is that that species does not occur further east in Cuba than Habana Province. It is possible that Ramsden, who lived in Oriente, had never seen specimens of T. maculatus and, realizing that this specimen was different from any other local Tropidophis, con- sidered it maculatus. We had ourselves at first considered that IZ 1246 was another Cuban specimen of T. haetianus Cope, but the scale counts and other charac- teristics eliminate that species from consideration. It seems most likely that this is a peculiarly patterned T. pilsbryi or one whose pale occipital blotches have faded during preservation. Another possibility which should not be dismissed is that it represents still another species of Some Cuban boas—Tropidophis 85 dwarf boa from eastern Cuba, but militating against this is the fact that adult specimens of T. pilsbryi galacelidus have the occipital blotches obscured, but much more obvious than in this individual; all specimens of the nominate subspecies have the occipital blotches very prominent, and a maximally sized specimen has a snout-vent length of 181 mm, thus is much smaller than IZ 1246. Tropidophis pardalis is a widespread Cuban species that occurs throughout Cuba (with the exception of most of Oriente Province, where it is known only from San German) and the Isla de Pinos. It presents a uniformity of pattern, size and scale counts throughout much of its range, except that some Pinar del Rio and Isla de Pinos specimens show indications of pale occipital spots. We now have counts on 106 T. pardalis. The ventral scales vary between 186 (Habana Province) and 155 (Isla de Pinos); the largest male has a total length of 315 mm (Isla de Pinos) and the largest female 303 mm (Habana Province). The rows of dark body blotches are modally 6, with 8 rows occurring occasionally in extreme western snakes. Dorsal scale rows at midbody are either 23 (mode) or 25. In general aspect, then, T. pardalis is a small snake with 6 rows of large and conspicuous blotches and usually 23 scale rows at midbody. There remain eight snakes in the USNM and the AMNH which are from the area between Soledad and Trinidad in extreme southern Las Villas Province. Of them, Schwartz and Marsh examined two (AMNH 77784, UMMZ 76109—both from Soledad) and considered them T. pardalis. These two specimens, however, have relatively high ventral counts (153, 155) for that taxon. The additional six specimens agree with these two snakes in large size and in having even higher scale counts (161-172), and we group them together. We feel confident that these eight snakes represent a new taxon, distinct at some level from T. pardalis. The options are these: 1) the extreme southern coast of Las Villas Province is inhabited by a local subspecies of T. pardalis which is large and has high ventral counts, 2) these snakes represent a species distinct from not only T. pardalis but all other Cuban Tropidophis, or 3) they are allied to another named species other than T. pardalis. We cannot refute or confirm any of these possibilities; specimens of T. pardalis from elsewhere in Las Villas Province are very few and from only two localities (La Sierra, N of Vega Alta; Cumanayagua), and these snakes seem typical of T. pardalis in size, ventral counts, and pig- mental and pattern characteristics. One feature which is dis- 86 Proceedings of the Biological Society of Washington tinctive is the fact that the eight snakes, at both comparable and larger sizes than T. pardalis, have relatively smaller heads. Schwartz and Marsh (1960:79) assigned two of these snakes (USNM 188512, USNM 138510) to T. nigriventris; however, they commented (p. 70) that the two Las Villas snakes dif- fered from Camagitiey material, including the holotype descrip- tion and the paratype, in not having the venter almost entirely dark, a distinguishing feature of Camagtiey T. nigriventris and indeed of the species. They stated (loc. cit.) that the major differentiating feature of the two species (T. pardalis and T. nigriventris) “is the much smaller head of nigriventris; if specimens of the two taxa are compared, the smaller head of nigriventris is immediately apparent.” For this reason, we feel most inclined to regard these southern Las Villas snakes as a subspecies of T. nigriventris, despite the apparent hiatus between them and the nominate subspecies. Accordingly, in honor of Jerry D. Hardy, who collected the majority of the specimens, we propose the following combination: Trepidophis nigriventris hardyi, new subspecies Holotype: USNM 138510, an adult male, from 10 mi. (16 km) W Trinidad, Las Villas Province, Cuba, taken 5-10 September 1956 by J. D. Hardy. Paratypes: USNM 188511-12, same data as holotype; USNM 140471-72, near Trinidad, Las Villas Province, Cuba, 23-27 June 1957; USNM 137085, Trinidad, Las Villas Province, Cuba, 23 December 1954, J. D. Hardy; UMMZ 76109, Soledad, Las Villas Province, Cuba, 18 July 1983, N. A. Weber; AMNH 77784, Soledad, Las Villas Province, Cuba, 18 July 1957, W. H. Gehrmann. Definition: A subspecies of T. nigriventris characterized by high number of ventral scales (153-172 versus 144-150 in nigriventris), 6 or 8 (modally 6) rows of blotches around body at midbody, and venter never extensively dark brown to blackish because of greatly enlarged ventrolateral rows of blotches across venter as in nominate subspecies. Description of holotype: A male with snout-vent length of about 303 mm, tail about 48 mm (total length about 351 mm), 170 ventral scales, 33 unpaired subcaudal scales (tail incomplete), 10/? supralabials, 11/? infralabials, 1/1 preoculars, 2/2 postoculars, dorsal scales smooth and in 21-23-18 rows, 40/41 blotches in paramedian dorsal blotch rows, ar- ranged in 8 (incomplete) rows at midbody, parietal scales not in contact. As preserved, dorsum medium brown with dark blotches darker brown, ventrolateral rows of blotches on each side not expanded midventrally to give basically all-dark venter; top of head dark and patternless; venter Some Cuban boas—Tropidophis 87 white, chin and throat dark brown; upper labials dark brown, separated by an almost black loreal stripe from brown top of head; tail dark brown above with about 6 dorsal blotches, ventrally with about 13 partially paired or staggered dark blotches on white ground; no indication of any pale occipital markings. Variation: The series of T. n. hardyi consists of three males and five females; the largest male has a snout-vent length of 303 mm (USNM 138510), the largest female 8334 mm (USNM 140472). The incomplete tail of the male measures 33 mm, that of the female 21 mm (also incom- plete). The ventrals in the entire series range from 153 to 172, and specimens of both sexes with complete tails have subcaudals varying between 31 (male) and 46 (female); one female with an incomplete tail has 48 subcaudals. The supralabials are 9/9 (two snakes), 9/10 (three) and 10/? (one). Infralabials are 9/9 (one snake), 9/10 (one), 10/10 (one), 10/11 (one), 11/11 (one), and 11/? (one). The parietal scales are not in contact except in one snake (USNM 140471). Pre- oculars are 1/1 in all specimens which can be counted and postoculars are either 2/2 (five specimens), 2/3 (one), or 3/3 (one). Dorsal scales are smooth in all snakes except UMMZ 76109 where they are slightly keeled. Scale rows behind the head are 20, 21, 23 or 24 (mode 21), at midbody 23 or 25 (mode 23—seven snakes), and anterior to vent 18, 19, or 20 (mode 19). The modal dorsal scale row formula is 21-23-19. The blotches are arranged in 6 rows in six specimens and in 8 rows in two; in these latter snakes, the lowermost rows are obsolete, and one of the lateral rows on each side is weakly indicated; paramedian blotches range from 32 to 44 (x = 87.8); the dorsal surface of the tail has from 4 to 6 dark blotches (the high count is from a snake with an incomplete tail). The dorsal ground color is medium brown to dark brown, and the venter white to pale tan, in all specimens. The rows of blotches are dark brown and conspicuous, although the ventrolateral- most rows may be obsolescent and the median lateral row may also be present (although it is absent completely in all but two snakes). In large snakes the head is conspicuously small, and the same condition is obvious but somewhat less so in smaller individuals. The upper surface of the head is unpatterned, except that at times there is a vague, longi- tudinal, dark rectangular figure; the supralabials are tan and set off sharply from the dorsal ground color of the head. Comparisons: Tropidophis n. hardyi differs from T. n. nigriventris in two obvious ways: 1) the number of ventrals in hardyi is greater (152-172) than in nigriventris (144-150), and 2) nigriventris has the ventralmost blotch rows large and expanded, so much so as to render the venter almost totally dark due to random blotch fusions across it. One hardyi (AMNE 77784) tends slightly toward this latter condition, but all other specimens do not demonstrate it at all; in fact, the ventral- most rows are often obsolete. The dark venter in nominate nigriventris is not ontogenetic, since the condition occurs in snakes with snout-vent 88 Proceedings of the Biological Society of Washington lengths of 162 mm and 184 mm, much smaller than any hardyi, in which subspecies the venter never has this hypermelanic condition. There is no problem differentiating hardyi from nigriventris; what is problematical is whether hardyi should be associated with it nomenclaturally. One feature is suggestive: both populations have distinctly smaller heads than, for instance, T. pardalis. The larger size of hardyi suggests that this subspecies is indeed a larger snake than nigriventris, but the sam- ples in each case are small. It is intriguing, however, that all nigriventris are small (maximum snout-vent length 227 mm) whereas all hardyi are large (minimum snout-vent length 184 mm); the latter subspecies reaches a maximum snout-vent length of 334 mm. More pertinent than comparisons of T. n. hardyi with T. n. nigriventris are comparisons of T. n. hardyi with T. pardalis and T. pilsbryi. Of these two species, T. pardalis has been taken sympatrically with T. n. hardyi at Soledad (USNM 134358), and T. pilsbryi galacelidus is also known from Soledad, as well as Guabairo which is within a few kilo- meters of Soledad. In the Sierra de Trinidad, T. p. galacelidus’ may approach T. n. hardyi closely, since the latter seems to be (Trinidad) a lowland snake, whereas the former seems to occur in the more mesic uplands. Tropidophis n. hardyi differs from T. p. galacelidus in lacking pale occipital blotches, in having fewer ventral scales (153-172 versus 177— 183), in having 6 (rarely 8) rows of blotches rather than 10, and in having fewer paramedian blotches (32—44 versus 44-50). Both taxa reach about the same size, but there is no difficulty in separating them. Differentiating between T. n. hardyi and T. pardalis is more difficult. Ventrals in the long series of T. pardalis vary between 186 and 155, whereas this count in T. n. hardyi varies between 153 and 172. The high counts of ventrals in T. pardalis, however, are not from Las Villas Province, but rather from the Isla de Pinos and Habana Province. The largest male T. pardalis has a total length of 315 mm and the largest female 303 mm, in contrast to T. n. hardyi males which reach a maxi- mum snout-vent length of about 851 mm and females which reach a total of 367 mm. Tropidophis pardalis regularly has 6 longitudinal rows of blotches, whereas T. n. hardyi may have 8 but modally 6 rows of blotches. Both T. pardalis and T. n. hardyi often have 2/2 postoculars. The smaller head of T. n. hardyi also tends to distinguish these two taxa. We can do no better than to reiterate our previous comments: we feel certain T. n. hardyi is a distinct entity, but whether we correctly asso- ciate it with T. nigriventris or whether it is indeed a peculiarly local and circumscribed population of T. pardalis remains to be determined. So much of central and western Cuba remains unknown as far as these small multispotted boas are concerned that speculation on their relation- ships is indeed futile. Finally, comparisons of T. p. galacelidus and T. pardalis are easily made. The former has a pair of pale occipital spots, whereas the latter Some Cuban boas—Tropidophis 89 rarely does; and, if present, the spots are much smaller, less contrasting, and less conspicuous than in T. p. galacelidus. Tropidophis p. galacelidus has 10 rows of small blotches (6 rows of large blotches in T. pardalis), a greater number of ventral scales (177-183 versus 136-155 in T. pardalis), and (at least females) reaches a much larger size (874 mm) than T. pardalis (303 mm). If T. pardalis and T. p. galacelidus occur syntopically, they should be easily distinguishable on the basis of the above characters. Specimens of T. n. nigriventris examined: Cupa, CamMactry Prov- INCE, 6 mi. (about 9.6 km) E Marti (UMMZ 70889-holotype; UMMZ 70887); Finca El Porvenir, 24 km SW Camagiiey, Loma de Yagua (AMNH 81182-83). LITERATURE CITED Batty, J. R. 19387. A review of some recent Tropidophis material. Proc. New England Zool. Club 16:41-52. Dowuinc, H. G. 1951. A proposed standard system of counting ven- trals in snakes. Brit. J. Herpetology 1(5):97-99. ScHwartz, A., AND R. J. Marsu. 1960. A review of the pardalis- maculatus complex of the boid genus Tropidophis in the West Indies. Bull. Mus. Comp. Zool. 123(2):49-84, 10 figs. 90 Proceedings of the Biological Society of Washington ne 4 (thie (hag QH i B4k NH A) 10, pp. 91-94 23 April 1975 PROCEEDINGS OF THE ~.~LOGICAL SOCIETY OF WASHINGTON DATES OF PUBLICATION OF WESTWOOD’S ARCANA ENTOMOLOGICA By J. Cuester BRADLEY! Comstock Hall, Cornell University, Ithaca, N.Y. 14850 John Obadiah Westwood published Arcana entomologica; or illustrations of new, rare, and interesting insects, in London in two volumes, the title page of each of which bore the date MDCCCLV. Each volume was a republication of pages or groups of pages which had originally been independently published on the precise date printed at the foot of the page, or at the foot of the first page of a group of pages, and were being republished without any change in these originally indicated dates of publication. Dealing with a world-wide fauna, a great many new taxa, now well known, were described or commented upon in the Arcana. Some cataloguers and authors referring to the taxa included in it ascribed to them the date of the republished volume, i.e. December 31, 1845. That is false. Each dates from when the part in which it appeared was actually pub- lished. This will change the publication date of a great many taxa from that which is ordinarily accorded them. In the analysis that follows the pagination of each part and its date of publication is given, so that anyone knowing the page on which a taxon was described can instantly discover its actual date. For example, in Kirby’s Catalogue of the Orthoptera, Phylocrania insignis is dated from 1845, the date of the volume in which it was republished. It was actually published in Part E2 on November 1, 1843. The purpose of this paper is to make it easy for any biologist to discover the correct date of publication of every taxon in- cluded in the Arcana. 1 Deceased. 10—Proc. Brot. Soc. Wasu., Vou. 88, 1975 (91) 92 Proceedings of the Biological Society of Washington These changes in the dates accorded to Westwood’s taxa in many catalogues are not capricious but are compelled by the International Code of Zoological Nomenclature. The style of the Arcana is complicated, but after the frontis- piece of volume one, Westwood has applied it consistently through both volumes without so much as a single variation. The arrangement may be comprehended from an examination of the following excerpts: “No. 8.—Ist July, 1842 (I 113-114, pls. 29,30) (12 115-128, pls. 31,32) No. 9.—Ist September, 1842 (K 129-130, pls. 33,34,35,36)(K2 131-144, pls. 37,38, 39,40)” The division of each lettered section into two parts, such as K and K2, is maintained throughout the entire work. The numbered sections were issued bimonthly, each pub- lished on the first day of the month. Tue DATE oF PUBLICATION OF THE NUMBERED SECTIONS OF THE ARCANA VoLuME 1, 1845 Pl. 15 (Frontispiece) Ist January, 1842 (B 1-2, pl. 1)(B2 3-16, pls. 2,3,4) (C 17-18, pl. 5)(C2 19-32, pls. 6,7,8) (D 33-34, pl. 9)(D2 35-48, pls. 10,11,12) (E 49-50, pls. 13,14) (E2 51-64, pls. 15,16) No. 5.—I1st January, 1842 (F 65-66, pl. 17) (F2 67-80, pls. 18,19,20) No. 6.—I1st March, 1842 (G 81-82, pls. 21,22,23) (G2 83-96, pl. 24) No. 7.—Ist May, 1842 (H 97-98, pl. 25 Frontispiece) (H2 99-112, pls. 26,27,28) No. 8.—Ist July, 1842 (I 113-114, pls. 29,30) (12 115-128, pls. 31,32) No. 9.—Ist September, 1842 (K 129-130, pls. 33,34,35,36) (K2 131-144, pls. 37,38,39,40) Dates of Westwood’s Arcana 93 No. 10.—lst November, 1842 (L 145-146) (L2 147-160) No. 11.—lIst January, 1843 (M 161-162, pl. 41)(M2 163-176, pls. 42,43,44) No. 12.—1st March, 1843 (N 177-178, pls. 45,46) (N2 179-190, pls. 47,48) VoLuME 2, 1845 No. 13.—Ilst May, 1843 (B 1-2, pls. 49,50) (B2 3-16, pls. 51,52) No. 14.—Ist July, 1843 (C 17-18, pls. 53,54) (C2 19-32, pls. 55,56) No. 15.—Ist September, 1843 (D 33-34, pl. 57) (D2 35-48, pls. 57,58,59,60) No. 16.—1st November, 1843 (E 49-50, pl. 61) (E2 51-64, pls. 62,63,64) No. 17.—Ist January, 1844 (F 65-66, pl. 65)(F2 67-80, pls. 66,67,68) No. 18.—Ilst March, 1844 (G 81-82, pl. 68) (G2 83-96, pls. 69,70,71) No. 19.—Ist May, 1844 (H 97-98, pl. 72) (H2 99-112, pls. 73,74,75) No. 20.—Ist July, 1844 (I 113-114, pls. 76,77) (12 115-128, pls. 78,79) No. 21.—Ist September, 1844 (K 129-130, pl. 80) (K2 131-144, pls. 81,82,83) No. 22.—1st November, 1844 (L 145-146) (L2 147-160, pls. 84,85,86,87 ) No. 23.—Ist February, 1845 (M 161-162, pls. 88-95) (M2 163-176) No. 24.—Ist June, 1845 (N 177-178) (N2 179-191) On page 191 the “Addenda et Corrigenda” apply to both volumes and include some corrections in synonymy. There are also “Errata” preceding page 1 in Volume One. The Arcana is a work of great rarity. In some libraries the work is kept under lock and key. 94 Proceedings of the Biological Society of Washington No. 11, pp. 95-112 23 April 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON THE SPECIES OF TAMANDUA GRAY (EDENTATA, MYRMECOPHAGIDAE) By Rates M. WETZEL Biological Sciences Group, University of Connecticut Storrs, Connecticut 06268 Over an extensive range from southern México to northern Argentina and Uruguay, the anteaters of the genus Tamandua display great variation in colors of coats and in proportions of skulls. Central American populations are black-vested, but in South America the pelage varies from a monochromatic black, brown, tan or white through a gradient of bichro- matic patterns to a black-vested condition. After a taxonomic history in which writers gave not less than ten specific names to the tamanduas, treatment of the genus during this century has been chiefly either (1) as containing two species—T. tetradactyla L. for the black-vested forms and T. longicaudata (Wagner) for those lacking a complete vest, or (2) as mono- typic, the species T. tetradactyla containing different sub- species of which one was the nonvested or partially vested T. t. longicaudata. One melanistic form, nigra Geoffroy, has been assigned to either of the two species, while a second melanistic form, quichua Thomas, has been more uniformly retained as a subspecies of T. tetradactyla. This study was initially concerned with the identification of the extensive series of Tamandua collected by the Smith- sonian Venezuelan Project under the direction of Dr. Charles O. Handley, Jr. The black-vested tamanduas of northwestern Venezuela proved to be distinctly separable from the geograph- ically adjoining samples of nonvested forms from a variety of habitats. In turn, the nonvested forms were bound through a clinal chain to black-vested forms south of Venezuela. 11—Proc. Biot. Soc. WasH., Vou. 88, 1975 (95) 96 Proceedings of the Biological Society of Washington Correlating these two differences—a disjunction line in north- western Venezuela and a cline to the south—with taxonomic levels required samples from areas well beyond the original focus of the study. To this end, I examined specimens which were available in other collections from throughout the range of the genus. PROCEDURES Color and Color Patterns: 1 recorded patterns of pelage (Fig. 1) as: A) Vested. A vivid black area on the trunk, continuous from black shoulder stripes posteriorly to the rump and widening behind the shoulders to encircle the body. This vest is surrounded by a uniformly pale color of white, tan or gold on the rest of the body, the legs, and the heavily furred base of the tail. The pale color also divides the vest as a middorsal stripe from the neck to the middle or posterior portions of the back. Such pelage is found in all specimens from the northwestern portion of the range of the genus (AMNH 17272, México, Vera Cruz, Pasa Nueva; USNM 443242, Venezuela, Zulia, El] Rosario) and from the lower Amazon basin south to Uruguay (NRMS 508, Brazil, Para, Cameta; Uruguay, Cerro Largo, Estancia La Formosa, see Ximénez, 1972: plate). B) Nonvested. Specimens have completely uniform pelage, light tan, buffy white, or gold in color, with no obvious indication of vest pattern (USNM 441973, Venezuela, Falcén, Capatarida; ROM 31883, Guyana, Rupununi, Dadanawa Ranch). In some specimens (USNM 441971, Venezuela, Capatarida) a vague vest pattern could be seen only by parting the hairs, as the bases of hairs were slightly darker than the tips. C) Partially Vested. Obvious appearances of the vest first occur usually in the shoulder area as smudges or even vivid stripes, though the rest of the body is uniformly pale (ANSP 4635, probably Chapada, Mato Grosso, Brazil, holotype of M. bivitatta straminea Cope). At the darker end of the gradient, the vest appears grizzled or gray because of dark hairs with tan tips; black shoulder stripes usually accompany this coloration (FMNH 25261, Bolivia, Santa Cruz; USNM 361029, Guyana, Dadanawa Ranch). In searching for correlations of variation in pelage with cranial and external measurements and proportions, I rated the gradient from completely pale and uniform pelage to the nearly complete pattern of the vest as stages of one to nine. D) Melanistic. Specimens in this category have overall, completely black (BMNH 27.1.1.156, Peri, San Martin, Yurac Yacu, holotype of M. t. quichua Thomas) or dark brown pelage. Variants, not recorded as melanistic, are individuals with pale foreheads and legs but with uniformly dark brown pelage from ears to rump. This color form, as well as those with uniformly dark brown pelage, strongly suggests a gradient from complete melanism to the partially vested condition. Variation in Pelage by Age, Sex, and Season: I found no major Species of Tamandua @ vested @ melanistic © partially vested © non vested Localities of Tamandua specimens examined. Three peripheral records from the literature are added: I, Dalquest, 1958; II, Carvalho, 1962; III, Ximénez, 1972. Fic. 1. © 98 Proceedings of the Biological Society of Washington changes in pelage between immaturity and adulthood; but specimens of young tamanduas have longer dorsal hair, in proportion to size of the body, which gives them a fluffy appearance. In addition, the immature vested specimens tend to have a greater proportion of pale hairs visible in the vest areas. Aside from this I found no basis for statements by Azara (1801:107) and Liais (1872:360) that color variation is correlated with age. No pronounced sexual difference in pelage was found except that in localities presenting a gradient of partially vested forms, the darker specimens are usually males (USNM 406686, female, Capibara vs. USNM 406688, male, San Juan de Rio Manapiare, both from Territorio Amazonas, Venezuela). One sample (Guyana, Dadanawa Ranch, in USNM and ROM) is large enough to demonstrate that there is little or no seasonal variation in pelage. However, more data are needed on this point and on rates of growth of hair and hair replacement. Cranial Characters: Each skull was assigned to one of five arbitrary age classes on the basis of sequence of closure of sutures. Measurements of skulls in age classes 0 and 1 (no closures except for sutures of the occipital in class 1) were not included in geographical comparisons be- cause of the magnitude of differences in proportions and measurements, particularly shorter and narrower rostra. Age class 1 was considered a stage of transition to sexual maturity as the Smithsonian Venezuelan Project collected a few females of this age with embryos. Comparison of adult males and females (age classes 2-5) from similar geographical areas revealed no significant sexual differences in cranial measurements or proportions. The following cranial dimensions and their abbreviations are included in this paper (Fig. 2): ARW-—anterior rostral width; taken with the anterior flat face of the calipers at the level of the most posterior lateral indentation of the anterior part of the maxilla. FMW—maximal width of foramen magnum. IFL—infraorbital foramina, mean of minimal lengths of pair. NL-a, NL-b—nasal bones: a, minimal length at midline and b, maximal length. OL—occipito-nasal length, distance from maximal posterior extension of occipital, dorsal to the formen magnum, to anterior tip of nasal bone. RL—rostral length, derived as in the altitude of a triangle where the distance between the lacrimal foramen to the anterior tip of the nasal bone (probably the RL of Reeve, 1942 and Davis, 1955) is considered the hypotenuse and half the antorbital width is considered the base. PRL—post-rostral length, derived as the difference between OL and RL. AOB—antorbital breadth, minimal width of skull at the level of the lacrimal foramina. ORB—interorbital breadth, minimal width between the middorsal borders of the orbits. Species of Tamandua 99 Qs anterior borders of lacrimal foramen and bone Fic. 2. Illustration of measurements of skulls of Tamandua. BW—breadth of braincase, maximal breadth of skull posterior to zygomatic processes. Two cranial characters that were scored, rather than measured, proved to have taxonomic value (Fig. 3): 1) The presence of four pairs of foramina (optic, orbital fissure, foramen rotundum, foramen ovale) in the orbit or only three pairs. The latter condition is caused by the absence of a bony septum between the orbital fissure (= anterior lacerate foramen) and the foramen rotundum. Bilateral asymmetry in some of the specimens required an averaging of this condition between left and right sides as well as partial scoring for incomplete septa. Since these septa are incompletely ossified in the youngest specimens, as well as more easily destroyed in the cleaning of immature skulls, no count of septa for immature specimens is included in the geographical comparisons. 2) The shape of the posterior (palatal) border of the infraorbital foramen. The maxilla either (a) extends as far, or farther, posteriorly on the medial side of the foramen as it does on the lateral side, resulting 100 Proceedings of the Biological Society of Washington A. _-anterior limit of palatine-~ >> -anterior limit of - ~ lacrimal jugal __ jugal slender broad T. mexicana B. T. mexicana Ts tetradactyla Fic. 8. Comparison of selected cranial characters in Tamandua. T. mexicana, CONN 16875, Chepo, Panama; T. tetradactyla, CONN 16831, Km 285, Trans-Chaco highway, Paraguay. Both are black-vested specimens. Species of Tamandua 101 in a crescent-shaped border of the foramen, or (b) terminates on the medial side of the foramen distinctly anteriorly to the lateral side, result- ing in a subequal or incomplete crescent-shaped border of the foramen. When compared with the distances between the anterior border of the lacrimal foramen and bone (Fig. 2), the distances between the anterior borders of the palatine and lacrimal bones (parallel to the long axis of the skull, Fig. 3) are proportionally different between the two species. Count of Vertebral Regions: A total of 39 vertebral columns was ex- amined, but only individual vertebral regions that were without missing vertebrae are included in the diagnoses of the species. My count of the sacral region began with the first vertebra completely articulating with the ilia and ended with the last vertebra completely articulating with the ischia. Sources of Specimens: Although tamanduas from over 40 collections have been examined, only the following collections and their curators at the time of study are cited here: AMNH—American Museum of Natural History, New York; Richard G. Van Gelder. ANSP—Academy of Natural Sciences of Philadelphia, Philadelphia; H. Radcliffe Roberts. BMNH— British Museum (Natural History), London; G. B. Corbet. CONN— University of Connecticut Museum of Natural History, Storrs; Robert E. Dubos. FMNH—Field Museum of Natural History, Chicago; Joseph Curtis Moore. MCZ—Museum of Comparative Zoology, Harvard Uni- versity, Cambridge; Barbara Lawrence. MNHN—Muséum National d'Histoire Naturelle, Paris; F. de Beaufort and Jean Dorst. MZUV— Musée Zoologique de l’Université et de la Ville, Strasbourg; F. Gouin and Jean-Pierre Rieb. NRMS—Naturhistoriska Riksmuseet, Stockholm; Ulf Bergstrom (deceased) and Greta Vestergren. PARA—Museu Paraense “Emilio Goeldi,’ Belém; Fernando C. Novaes. ROM—Royal Ontario Museum, Toronto; Randolph L. Peterson. SMF—Natur-Museum und Forschungsinstitut Senckenberg, Frankfurt am Main; Heinz Felten. TCWC—Texas Cooperative Wildlife Collection, Texas A & M Univer- sity, College Station; Dilford C. Carter. USNM—National Museum of Natural History, Smithsonian Institution, Washington, D.C.; Charles O. A. Views of right palate. Infraorbital foramen is marked by an arrow and the medial and lateral margins of the posterior opening, by bars. Compare: posterior limits of maxilla on either side of foramen, anterior limits of the palatine and lacrimal bones, and dimensions of jugal bone. B. Right side of skulls and detailed view of orbits, tilted to show foramina. 1, optic foramen; 2, orbital fissure; 3, foramen rotundum; 4, foramen ovale. Foramina 2 and 8 are separate in T. mexicana but combined in T. tetradactyla or separated by a deeply set, slender septum which is visible from lateral view only at its base. 102 Proceedings of the Biological Society of Washington Handley, Jr. ZSM—Zoologische Sammlung des Bayerischen Staates, Munich; Theodor Haltenorth. Genus Tamandua Gray Tamandua Gray, 1825:348. Tamanduas F. Cuvier, 1829:501. Uroleptes Wagler, 1830:36; Palmer, 1899:73. Dryoryx Gloger, [1841]:112. The black-vested populations of Tamandua are separated by 2,000— 3,000 kilometers (1,200-2,000 miles) containing the intervening non- vested, partially vested, and melanistic forms (Fig. 1). The vested group occupying the northwestern part of the range of the genus (México to northwestern Venezuela) has been traditionally placed in the same species, T. tetradactyla, with the vested group in the south- eastern portion of the generic range (Amazon basin to southeastern Brazil and northern Argentina). The few comparisons that have been made over much of the range of the genus, such as Allen (1904), Krumbiegel (1940), and Reeve (1942), have assumed a specific rela- tionship of these two vested populations. Krumbiegel and Reeve, as well as Schréder (1937), then considered the question of the uniqueness of the nonvested and partially vested forms. Upon finding these T. longicaudata clinally integrating with T. tetradactyla in Brazil, and lacking specimens for adequate comparison of the two forms at their juncture in Venezuela, these writers concluded that Tamandua was a monotypic genus. Because of the new material collected by the Smithsonian Venezuelan Project, as well as the opportunity to study many other specimens not seen by the previous authors, I could compare longicaudata at all of its boundaries with the vested tamanduas. Geographical gradients in both pelage and skulls were found between the southern populations of vested tamanduas and the longicaudata pelage forms. This was particu- larly evident in the Amazon basin where a gradual cline occurred from nonvested and partially vested longicaudata in eastern Ecuador, Pert, and western and central Brazil to vested tetradactyla in eastern Amazonas to eastern Para, Brazil. A south-to-north gradient was found between the vested forms from the left or north bank of the Rio Amazonas to longicaudata of the Guyana highlands. To the north, the Rio Orinoco proved to be only a partial barrier for longicaudata. The second question was the taxonomic level of distinction between the black-vested tamanduas west of the Andes and those lacking com- plete vests and occurring immediately east of the Andes. This question involved comparing specimens from (a) either side of the Andes in Ecuador, (b) the Rio Magdalena valley with those from the eastern side of the Cordillera Oriental in Colombia, and (c) northwestern Venezuela and Colombia with those from adjacent Venezuela and Co- lombia east of the Cordillera Oriental. All of these comparisons indicated Species of Tamandua 103 TABLE 1. Comparison of the species of Tamandua. mexicana tetradactyla Black vest always present Ear shorter (means 40-46 mm) Skull slender (means of AOB, ORB, and BW less than 34, 25, and 42 mm) 4 pairs of orbital foramina in most skulls Posterior border of infraorbital fo- ramen distinctly crescent shaped Jugal bone slender (height usually no more than 31% of length) Distance between anterior borders of palatine and lacrimal bones pro- portionally less Caudal vertebrae 40-42 Color of pelage geographically variable; black vest always present only in portions of range most dis- tant from T. mexicana Ear longer (means 50-54 nearest to range of T. mexicana) Skull broad (means of AOB, ORB, and BW greater than 34.5, 25, and 41.5) 8 pairs of orbital foramina in most skulls Posterior border of infraorbital foramen an incomplete crescent; maxilla forming subequal borders about foramen Jugal bone broad (height more than 31% of length) Distance between anterior borders of palatine and lacrimal bones pro- portionally greater Caudal vertebrae 31-39 major lines of interruption coinciding with the Andean chain. This separation within the genus is abrupt, certainly not clinal as are the variations that extend from this point to the peripheries of the range, and is of such a major nature as to require consideration of two species: (1) a black-vested species, T. mexicana, occurring west of the eastern Andean divide from northwestern Venezuela and the Pacific slope of northern Peri to México, and (2) a black-vested to nonvested species confined to South America east of the Andes and including the nominate form, T. tetradactyla, as well as geographical color variants such as longicaudata, quichua, and nigra. The comparison of these two species is presented in Table 1. All measurements are in millimeters. The lists of synonyms, although incomplete, include most of the original names and new combinations, the names used in the more comprehensive studies of the genus, and the names used by the standard references for the two continents (Cabrera, 1958; and Hall and Kelson, 1959). Holotypes and certain other critical specimens that I have examined are cited. 104 Proceedings of the Biological Society of Washington Tamandua mexicana (Saussure ) Myrmecophaga tamandua (?), Desm. (var. Mexicana, Sauss.) Saussure, 1860:9. Tamandua tetradactyla—Tomes, 1861:287; Sclater, 1871:546. Myrmecophaga tetradactyla—Frantzius, 1869:307. Tamandua bivittata Var. 3. Opistholeuca Gray, 1873a:27, only the speci- mens from Central America and Colombia. Lectotype (herewith selected): BMNH_ 50.7.8.39, skin and skull, “Colombia,” coll. Par- zudaki; labeled as “type”; listed by Gray as the second syntype. ? Tamandua tetradactyla, var. leucopygia Gray, 1873b:469. Colombia, Antioquia, Concordia or Medellin. Nomen nudum; although BMNH 73.4.23.8, same locality, coll. J. K. Salmon, was probably the basis for this name, Gray did not support his name with description, ref- erence, or indication. ? Myrmecophaga quadridactyla True, 1884:588, in part. Nomen nudum. Myrmecophaga sellata Cope, 1889:133. Type-locality: Honduras. Tamandua tetradactyla instabilis Allen, 1904:392; Krumbiegel, 1940:174; Reeve, 1942:300; Cabrera, 1958:204. Holotype: AMNH 23420, Co- lombia, Magdalena, Bonda. Tamandua tetradactyla tenuirostris Allen, 1904:394. Holotype: AMNH 17272, México, Vera Cruz, Pasa Nueva. Tamandua tetradactyla chiriquensis Allen, 1904:395; Krumbiegel, 1940: 172; Hall and Kelson, 1959:238. Holotype: AMNH 18883, Panama, Chiriqui, Boqueron. Tamandua tetradactyla mexicana.—Allen, 1906:200; Reeve, 1942:300; Hall and Kelson, 1959:239. Tamandua tetradactyla punensis Allen, 1916:83; Cabrera, 1958:204. Holotype: AMNH 36452, Ecuador, Isla Puna. Tamandua tetradactyla sellata—Goldman, 1920:63, footnote; Lonnberg, 1937:27; Krumbiegel, 1940:172. Tamandua tetradactyla tambensis Lénnberg, 1937:25. Holotype: NRMS 18, Colombia, Cauca, El Tambo. Tamandua tetradactyla tetradactyla—Reeve, 1942:300, in part in which T. t. chiriquensis, tambensis, and punensis are considered as probable synonyms of subspecies tetradactyla. Tamandua tetradactyla hesperia Davis, 1955:558; Hall and Kelson, 1959: 238. Holotype: TCWC 5322, México, Guerrero, near Acahuizotla. Type-locality: México, Tabasco (Saussure, 1860:9). Range: From the lower forested slopes of the southern edge of the Mexican plateau, in southeastern San Luis Potosi, Guerrero, and north- western Oaxaca, through Central America to northwestern Venezuela, the valleys and montane forests of Colombia west of the divide of the Cordillera Oriental, and south in the forested coastal and Pacific Andean slopes of Colombia, Ecuador, and northwestern Pert (Fig. 1). Diagnosis: Pelage—always black-vested on a paler background, rather Species of Tamandua 105 than uniformly black, brown or tan. Ear—range of means of length from notch, 40-46. Skull—slender, with means of AOB, ORB and BW for all samples falling below 34, 25 and 42, respectively, except for a gradient to larger skulls in the upper valleys of Rio Cauca and Rio Magdalena in Colombia; four orbital foramina (Fig. 3B) present in at least one side of all skulls from southeastern part of range of species and present in both sides of skulls of 80-100% of all specimens except for those from Guerrero, México (70%), and coastal Ecuador (60%); posterior (palatal aspect) border of infraorbital foramen (Fig. 8A) distinctly crescent-shaped and formed entirely by maxilla, with medial border of foramen extending as far posteriorly as (or farther than) lateral border; distance (measured perpendicular to midline of skull) between most anterior extension of palatine and lacrimal bones (Fig. 3A) less than half the distance between anterior borders of lacrimal foramen and lacrimal bone (Fig. 2); jugal (malar) slender (Fig. 3A), its greatest height usually 31% or less of its greatest length. Vertebral count—14 specimens from México (1, FMNH), Guatemala (3, USNM), Honduras (1, FMNH), Panama (1, CONN), and northeastern Colombia (7, AMNH; 1, USNM): cervical 7 (in 10); thoracic 17.0 (15-18, in 11); lumbar 2.4 (2-3, in 14); sacral 5 (in 14); caudal 40.8 (40-42, in 8). Comparison: See Table 1. Tamandua tetradactyla (Linné) Myrmecophaga tetradactyla Linné, 1758:35. Myrmecophaga myosura Pallas, 1766:64. Type-locality: Brazil. Myrmecophaga tamandua G. Cuvier, 1798:143. No locality; pelage described as “jaunatre.” Myrmecophaga nigra Geoffroy St.-Hilaire, 1803:217. “La Guyane?” In part: Azara, 1809:plate 7; Desmarest, 1817:107. Myrmecophaga tamandua Geoffroy St.-Hilaire, 1803:217, in part [based upon both M. tridactyla L. and T. tetradactyla (L.)]. No locality. Myrmecophaga bivittata Desmarest, 1817:107. Brazil. Uroleptes tetradactyla—Wagler, 1830:36. Myrmecophaga tridactyla (not Linné, 1758:35).—McMurtrie, 1831:166, as synonym of M. tamandua Cuv. Tamandua crispa Riippell, 1842:179. “Guiana.” Holotype not listed in Mertens, 1925, nor found in SMF in 1970. Tamandua tetradactyla— Gray, 1843:191. Brazil. Myrmecophaga longicaudata Wagner, 1844:211. Type-locality restricted to Surinam by Cabrera, 1958:203. ? Myrmecophaga longicaudata Turner, 1851:218. No locality or refer- ence, nomen nudum. Uroleptes bivittatus—Fitzinger, 1860:395. Brazil. Tamandua biwvittata—Gray, 1865:384. Brazil, Paraguay. Tamandua ursina.—Gray, 1865:384, as synonym for T. bivittata. 106 Proceedings of the Biological Society of Washington Tamandua longicaudata.—Gray, 1865:384; Allen, 1904:385. Tamandua brasiliensis Liais, 1872:360. Brazil. Tamandua bivattata. Var. 1. Opisthomelas Gray, 1873a:27. Lectotype (herewith selected): BMNH 40a, skin, Brazil, coll. Lt. Mawe; labeled as type; first in Gray’s list of syntypes. ? Myrmecophaga quadridactyla True, 1884:588, in part. Nomen nudum. Tamandua tamandua.—Jentink, 1888:215. Surinam. Myrmecophaga bivattata straminea Cope, 1889:132. Holotype: ANSP 4635, skin, Brazil, probably Mato Grosso, SAo Joao or Chapada. Tamandua tetradactyla nigran—Menegaux, 1902:494,. French Guiana, Cayenne, Ouanary River (MNHN 1902-62, melanistic skin). Tamandua_ tetradactyla tetradactyla—Allen, 1904, by implication; Schroder, 1937:135; Krumbiegel, 1940:175, in part; Reeve, 1942:300, in part; Cabrera, 1958:205, in part. Tamandua tetradactyla chapadensis Allen, 1904:392; Krumbiegel, 1940: 174; Reeve, 1942:300; Cabrera, 1958:203. Holotype: AMNH 869, Brazil, Mato Grosso, Chapada. Tamandua longicaudata longicaudata.—Beaux, 1908, by implication; Cabrera, 1958:2038. Tamandua longicaudata nigra—Beaux, 1908:417. “Brasilien (?).” Tamandua tetradactyla quichua Thomas, 1927:371; Reeve, 1942:300; Cabrera, 1958:205. Holotype: BMNH 27.1.1.156, Peri, San Martin, Yurac Yacu. Tamandua_ tetradactyla longicaudata—Pittier and Tate, 1932:255; Schroder, 1937:186; Krumbiegel, 1940:175; Reeve, 1942:300. Tamandua tetradactyla kriegi Krumbiegel, 1940:171. Holotype: ZSM 1931-284, Paraguay, Estancia Zanja Moroti. Tamandua tetradactyla straminea.—Krumbiegel, 1940:174. ? Tamandua longicaudata mexianae.—Cabrera, 1958:203. Type-locality: Brazil, Para, Ilha Mexiana. Nomen nudum, not in Hagmann, 1908:29, as cited by Cabrera; no specimens from I]ha Mexiana found at PARA, MZUV, or elsewhere. Type-locality: Brazil, Pernambuco (Thomas, 1911:133). Range: South America east of the Andes, from Venezuela and Trinidad south to the state of Rio Grande do Sul in Brazil, the department of Cerro Largo in Uruguay (Ximénez, 1972), and the provinces of Santa Fe, Chaco, Salta, and Jujuy in Argentina. Diagnosis: Pelage—color geographically variable, black vest always present only in specimens from southeastern portion of range (Fig. 1), gradating to partially vested or nonvested blond, tan or brown individuals elsewhere in range; melanistic forms from eastern foothills of Andes in Pertti and Ecuador east along Amazon River to Amapaé and coastal French Guiana. Ear—range of means of length from notch, 50-54 in all samples except 43-48 in southeastern portion of range. Skull—broad, with means of AOB, ORB and BW greater than 34.5, 25 and 41.5, respectively, except in samples from portions of Amazon basin and thorn forests of Species of Tamandua 107 northwestern Venezuela; only three orbital foramina present in at least one side of all skulls from a northwestern portion of range and in both sides of 80-100% of the skulls from eastern edge of Andes, decreasing to 60% in Amazon basin and 50% in southeastern Brazil (Fig. 3B); posterior (palatal aspect) border of infraorbital foramen an incomplete crescent, its medial border not extending as far posteriorly as its lateral border, and medial border formed either entirely by palatine or by a combination of maxilla and palatine (Fig. 3A); distance (measured perpendicular to midline of skull) between anterior borders of palatine and lacrimal bones (Fig. 3A) at least one-half distance between anterior borders of lacrimal foramen and lacrimal bone (Fig. 2); jugal relatively wide, its greatest height 31% or more of its greatest length (Fig. 3A). Vertebral count—25 specimens from Guyana (2, AMNH); lower Ama- zon basin (1, AMNH), Minas Gerais (1, MCZ), and Mato Grosso (1, USNM), Brazil; northeastern Bolivia (10, AMNH); the Chaco of Paraguay (10, CONN): cervical 7 (in 19); thoracic 17.3 (16-18, in 22); lumbar 2.5 (2-3, in 25); sacral 5 (in 25); caudal 34.5 (31-39, in 19). Comparison: See Table 1. Comments: Except as noted above, the extensive geographical vari- ation within the genus prohibits a satisfactory delineation of the two species based only on external measurements and size and proportions of skulls. Both species contain large and small subspecies, clinally arranged, some undescribed and others with geographical limits incorrectly defined in the literature. The geographical span of samples compared by Reeve (1942) embraced, in the main, too wide a spectrum of subspecific variation to be useful in resolving the question of specific separation. However, my comparison of smaller geographical units, made possible by larger samples available, permits the separation of any one subspecies of T. mexicana from any one of T. tetradactyla. This is particularly striking where the geographic ranges of the two species are contiguous in northwestern Venezuela. There all measurements of body and skull, as well as pelage color and the discrete skull characters of Fig. 3, are significantly different between the two species. The features of pelage color, discrete skull characters, and certain measurements and proportions also show pronounced character displacement; examples of this displace- ment are shown in Fig. 4. . The isolation leading to separation into the present two species on either side of the Andes has been relatively complete as compared to events occurring in South America east of the Andes. Even so, my data indicate two centers of isolation and later dispersal there, one in the highlands of Brazil and the other in the Guyana highlands and eastern foothills of the Andes. Along the boundary between the two centers, marked roughly by the Amazon and its western tributaries, occur the greatest variation in color of pelage and a clinal decrease in size of speci- mens. For example, melanistic specimens have been collected from coastal French Guiana and the state of Amapa, Brazil, west through the immediate Amazon basin to the eastern foothills of Ecuador and Pert. 108 Proceedings of the Biological Society of Washington = mexicana | T. tetradactyla Sy SSyriFen ve Nin aneiyh\ San oa Ss WIE Tye By p25 FO) tty, {ie aaeaercealaee Length, ear 2 I a) &) || FMW 4 BS fl 2 + ARW/PRL 1,4,5 23 - RL/PRL 5,1,3,.4 2 = PRL/OL 5,4] 3 1 2 + IFL/OL 5. 2 I = NL-a/NL-b 5 23,1 4 - T mexicana, west of Andes T.tetradactyla , Venezuela east of Andes Y—mean, Venezuela: Mérida and Zulia W of Lake Maracaibo (n 20) 1-W of Rio Yaracuy, thorn shrub: Falcon, Lara, and Yaracuy (n 23) 2-E of Rio Yaracuy, montane, moist evergreen: Carabobo, Aragua, and Miranda (n[5) C— Colombia: César and NW Magdalena (1 52) 3-N of Rio Orinoco, non montane (n 1!) 4-Valley of Rio Orinoco, south bank (n 7) 5- Guyana Highlands of Bolivar and T; F Amazonas (n 19) Fic. 4. Ratio diagram for measurements of Tamandua from juncture of the two species in northern South America. S¢ is estimate of standard error of mean. Samples of T. tetradactyla are numbered 1-5 in increasing order of geographical distance from nearest population of T. mexicana. See text (pages 98-99) for definitions of abbreviations used for cranial measurements. This distribution coincides with the areas of major variation between nonvested and vested forms. The inclusion of a figure entitled “Le Tamandua noir, variété du Tamandua ordinaire” in Azara (1809:plate 7) must be disregarded as evidence that melanistic tamanduas have been found in Paraguay. Azara (1801 and 1809) made no textual reference to melanistic tamanduas. The plate in question is a composite of a giant anteater, Myrmecophaga tridactyla L., and a black Tamandua tetradactyla (L.). The lateral head stripe and contours of the head, ears, and apparently the Ist and 2nd claws (despite depicting digits 1, 2 and 4 as medial to the longest claw on digit 3) are those of Myrmecophaga tridactyla, while the remaining body, legs, and tail are those of Tamandua. The textual description (Azara, 1809:255) cited for this plate is for M. tridactyla. Desmarest (1817:107) referred this figure to the black tamandua, M. nigra (presumedly that of Geoffroy St.-Hilaire, 1803:217, which in turn is based upon MNHN 481 from “La Guyane?”) without reference to geographical origin. I have found no melanistic specimens from Paraguay, Mato Grosso or Bolivia, an area containing both vested tamanduas and those having only black shoulder stripes. Species of Tamandua 109 The melanistic tamanduas are, perhaps, examples of the saturate stage before progressive subtraction of melanin to paler, more uniformly colored indivuals as discussed by Hershkovitz (1968). This may have been triggered by cross-mating between populations formerly isolated in late Pleistocene along the Andean foothills and the Guyana highlands with those from the highlands of Brazil (see Vanzolini, 1973). An alternate explanation of the distribution of melanistic forms could begin with a population having a high incidence of melanism that spread from the foothills of the Andes easterly along the Amazon drainage to the Atlantic coast. This does not explain the evidence of clinal variation in size, the highly variable pelage, and differences in orbital foramina as does the concept of two dispersal centers. ACKNOWLEDGMENTS I am grateful to the curators of the collections cited for allowing me to examine specimens in their care. Additional appreciation is due Dr. Charles O. Handley, Jr., National Museum of Natural History, for his support and encouragement of this study and for critically reading the manuscript. I also thank my wife, Drew, who assisted in many ways, Mary Hubbard who prepared the illustrations, and D. Wilcox who read the manuscript. This paper is a contribution of the Smithsonian Venezuelan Project, supported by a contract (DA—49-193-MD-2788 ) of the Medical Research and Development Command, Office of the Surgeon General, United States Army. Travel funds from the University of Connecticut Research Foundation, the National Geographic Society, and the American Philosophical Society also aided this study. LITERATURE CITED ALLEN, J. A. 1904. The tamandua anteaters. Bull. Amer. Mus. Nat. Hist. 20:385-398. 1906. The proper name of the Mexican tamandua. Proc. Biol. Soc. Wash. 19:200. —. 1916. New South American mammals. Bull. Amer. Mus. Nat. Hist. 35:83-87. AzarA, F. bE. 1801. Essais sur lhistoire naturelle des quadrupédes de la province du Paraguay. Vol. 1. Paris. —. 1809. Voyages dans 1’Amérique méridionale. Collection de planches, pl. 7. Paris. BEaux, O. pE. 1908. Uber die schwarze Varietiit der Tamandua long- icaudata Gray. Zool. Anz. 33:417-418. CasrerA, A. 1958. Catalogo de los mamiferos de America del Sur. I. Rev. Mus. Argentino Cienc. Nat. “Bernardino Rivadavia,” 1957, Zool. 4(1):1—807. CarvaLtHo, C. T. 1962. Lista preliminar dos mamiferos do Amapa. Papéis Avulsos Dept. Zool., S. Paulo 15:283—297. 110 Proceedings of the Biological Society of Washington Corr, E. D. 1889. On the Mammalia obtained by the naturalist ex- ploring expedition to southern Brazil. Amer. Nat. 23:128—-150. Cuvier, F. 1829. Mammalogie, list of genera. In Dictionnaire des sciences naturelles. Vol. 59. LeNormant, Paris. Cuvier, G. 1798. Tableau élémentaire de histoire naturelle des animaux. Baudoin, Paris. Dataquest, W. W. 1953. Mammals of the Mexican state of San Luis Potosi. Louisiana State Univ. Stud., Biol. Sci. Ser. 1:1—229. Davis, W. B. 1955. A new four-toed anteater from Mexico. Jour. Mamm. 36:557—559. Desmarest, A. 1817. Nouveau dictionnaire d’histoire naturelle. Vol. 12. Chez Deterville, Paris. Frrzincer, L. J. 1860. Die Ausbeute der osterreichischen Naturforscher .. . Fregatte Novara. Sitz. Akad. Wiss. 42:385-416. Frantzius, A. 1869. Die Saugethiere Costaricas, ein Beitrag zur Ken- ntniss der geographischen Verbreitung der Saiugethiere Amerikas. Arch. Naturg. 1:247-333. Grorrroy Sr.-Hmame, E. 1803. Catalogue des mammiféres du Muséum National d’Histoire Naturelle. Paris. GiLocer, C. W. L. [1841] 1842. Hand-und Hilfsbuch der Nat- urgeschichte. Aug. Schulz, Breslau. GoLpMAN, E. A. 1920. Mammals of Panama. Smithsonian Misc. Coll. 69: 1-307. Gray, J. E. 1825. An outline of an attempt at the disposition of Mammalia into tribes and families, with a list of the genera. ... Ann. Philos., New Ser. 10:337-344. ——. 18438. List of the specimens of Mammalia in the collection of the British Museum. London. ——. 1865. Revision of the genera and species of entomophagous Edentata, founded on the examination of the specimens in the British Museum. Proc. Zool. Soc. London 1865:359-886. —. 1873a. Hand-list of the edentate, thick-skinned and ruminant mammals in the British Museum. London. ——. 1873b. On Mammalia from the neighborhood of Concordia, in New Granada. Ann. Mag. Nat. Hist., Ser. 4, 11:468—469. Hacmann, G. 1908. Die Landsiugetiere der Insel Mexiana. Arch. Rassen-GesBiol. 5:1-31. Hau, E. R. anp K. R. Ketson. 1959. The mammals of North Amer- ica. Vol. 1. Ronald Press Co., New York. HersuKxovitz, P. 1968. Metachromism or the principle of evolution- ary change in mammalian tegumentary colors. Evolution 22: 556-575. Jentinx, F. A. 1888. Catalogue systématique des mammiféres ... . Mus. Hist. Nat. Pays-Bas. Vol. 12. E. J. Brill, Leiden. KrumpigceL, I. 1940. Die Saugetiere der Siidamerika-Expeditionen Prof. Dr. Kriegs: 2, Ameisenbiren. Zool. Anz. 131:161-188. Species of Tamandua 111 Liats, E. 1872. Climats, géologie, faune et géographie botanique du Brésil. Part 2, Faune du Brésil, pp. 301-556. Garnier fréres, Paris. LinnE, C. von. 1758. Systema naturae, 10th ed. Vol. 1. LOnNBERG, E. 1937. Notes on some South American mammals. Ark. Zool. 29A:1-29. McMourtniz, H. 1831. The animal kingdom arranged in conformity with its organization by the Baron Cuvier . . . translated from the French with notes and additions. Vol. 1. New York. Menecaux, A. 1902. Catalogue des mammiféres rapportés par M. Geay de la Guyane Francaise en 1898 et 1900. Bull. Mus. Natl. Hist. Nat. 8:490—496. MerTENS, R. 1925. Verzeichnis der Sdugetier-Typen des Sencken- bergischen Museums. Senckenbergiana 7:18—37. Patxas, P. S. 1766. Miscellanea zoologica, quibus novae imprimis atque obscurae animalium species describuntur et observa- tionibus iconibusque illustrantur. Hagae Comitum. Parmer, T. S. 1899. Notes on Tatoua and other genera of edentates. Proc. Biol. Soc. Wash. 18:71-73. Pirtier, H. anp G. H. H. Tare. 1932. Sobre fauna venezolana. Bol. Soc. Venez. Cienc. Nat. 1:249-278. REEVE, FE. C. R. 1942. A statistical analysis of taxonomic differences within the genus Tamandua Gray (Xenarthra). Proc. Zool. Soc. London 111A:279-802. RuUprPett, E. 1842. Verzeichniss der in dem Museum der Sencken- bergischen naturforschenden Gesellschaft aufgestellten Samm- lungen. Mus. Senckenb. 3:144—-196. SaussurE, H. pe. 1860. Note sur quelques mammiféres du Mexique. Rev. Mag. Zool., Ser. 2, 12:3—11. Scur6perR, W. 1937. Uber Tamandua tetradactyla longicaudata (Wagn.). Zool. Anz. 119:124—138. Sciater, P. L. 1871. Report on additions to the Society’s menagerie in May 1871. Proc. Zool. Soc. London 1871:5438-546. Tuomas, O. 1911. The mammals of the Tenth Edition of Linnaeus; an attempt to fix the types of the genera and the exact bases and localities of the species. Proc. Zool. Soc. London 1911: 120-158. ———. 1927. The Godman-Thomas expedition to Peru.—V. Ann. Mag. Nat. Hist., Ser. 9, 19:361-875. Tomes, R. F. 1861. Report of a collection of mammals made by Osbert Salvin .. . at Duefas, Guatemala .... Proc. Zool. Soc. London 1861:278-288. TruE, F. W. 1884. A provisional list of the mammals of North and Central America, and the West Indian Islands. Proc. U.S. Nat. Mus. 7:587-611. Turner, H. N., Jr. 1851. On the arrangement of the edentate Mam- malia. Proc. Zool. Soc. London 1851:205—221. 112 Proceedings of the Biological Society of Washington VANZOLINI, P. E. 1973. Paleoclimates, relief, and species multiplica- tion in equatorial forests. Pp. 255-258 in B. J. Meggers, E. S. Ayensu, and W. D. Duckworth, eds. Tropical forest ecosystems in Africa and South America: a comparative review. Smith- sonian Institution Press, Washington. WacterR, J. 1830. Natiirliches System der Amphibien, mit vorange- hender Classification der Saugthiere und Vogel. J. G. Cotta’ schen, Munich. Wacner, J. A. 1844. Die Saugthiere in Abbildungen nach der Natur mit Beschreibungen von Dr. Johann Christian Daniel von Schreber. Suppl. 4:211. Erlangen. XIMENEZ, A. 1972. Hallazgo de Tamandua tetradactyla (Linné, 1758) en el Uruguay (Edentata-Myrmecophagidae). Neotropica 18: 134-136, plate. BAX ». 12, pp. 113-120 15 August 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON FRESHWATER TRICLADS (TURBELLARIA ) OF NORTH AMERICA. VIII. DUGESIA ARIZONENSIS, NEW SPECIES By RoMaAn KENK Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560 Eight species of the genus Dugesia Girard are known at present to inhabit the continental Americas north of the Isthmus of Panama. Two of them are widely distributed over the entire North American continent, Dugesia tigrina (Girard ) and D. dorotocephala (Woodworth). Dugesia microbursalis (Hyman), which may be a form of D. tigrina, has been re- ported from the northeastern United States. Four species of the genus have recently been described from Mexico by Mitchell and Kawakatsu, i.e. D. typhlomexicana, D. barbarae, D. guatemalensis, and D. mckenziei. A European immigrant, D. polychroa (O. Schmidt) has been discovered in the waters of the St. Lawrence River system. Five additional species of Dugesia occur on the Caribbean islands: D. antillana Kenk in Puerto Rico, D. cubana Codreanu and Balcesco in Cuba, D. aurita (Kennel) (a somewhat prob- lematic species) and D. arimana Hyman on the island of Trini- dad, and D. festai (Borelli), a South American species, re- ported from Curagao. The new species described in this paper is an additional member of the planarian fauna of the United States. Dugesia arizonensis n. sp. Figures 1-3, 5 Type material: Holotype from Bog Springs, Arizona, on four slides of sagittal sections, deposited in the collections of the National Museum 12—Proc. Biot. Soc. WasH., Vou. 88, 1975 (113) oe SO Adwiar yOu \ Wo \ 114 Proceedings of the Biological Society of Washington 1 y) Fic. 1 and 2. Dugesia arizonensis. 1. Photograph of living speci- men, X 4.6. 2. Outline drawing. of Natural History, Smithsonian Institution, Washington, D.C. (USNM 51980). Paratypes in the author's collection. External features: (Fig. 1 and 2): At first glance, the species re- sembles Dugesia dorotocephala (Woodworth) by its general appearance. Mature specimens from the type-locality, when gliding quietly, are up to 30 mm long and 4 mm wide. Animals from Rucker Creek appear more slender. The anterior end is triangular and sharply pointed, with a rather acute anterior angle. The two sides of the triangle may appear straight (Fig. 2) or slightly curved (Fig. 1). At the base of the triangle is a pair of long, slender, pointed auricles which in gliding are held elevated. Behind the auricles the body is slightly constricted, then the lateral margins diverge gradually and run almost parallel to the level of the copulatory apparatus, where they converge again to meet at the bluntly pointed posterior end. The two eyes are situated anterior to the level New freshwater planarian from U.S.A. 115 of the auricles. Their distance from each other amounts to about one- third, or a little more, of the width of the head at eye level. The color of the animals from Bog Springs is brown, almost black, dorsally and somewhat lighter ventrally. Under magnification, the pigment appears almost uniformly distributed, without lighter specks (such as are usually seen in the related D. dorotocephala). The mouth and the genital aperture are marked with small white spots. Specimens from Rucker Creek are lighter in color, grayish, with an almost white ventral surface which, however, shows scattered pigment spots under magnification. Digestive system: The intestinal branching is difficult to analyze in the living specimens, particularly in the darkly pigmented individuals, because of the profuse ramification of the intestinal diverticula. The gut area reaches far into the head to a level anterior to the eyes. The pharynx is rather long, measuring about one-fifth the body length, and is inserted behind the middle of the body. Its surface is uniformly pigmented, light gray, leaving only a short area of the tip free of pigment or white. The anatomy of the pharynx corresponds to that of related species of the same genus. Its external surface is covered by a rather thin, ciliated, infranucleate epithelium (about 3 um thick) underlain by a narrow layer of longitudinal muscle fibers (4 um), followed by a thicker sheet of circular fibers (19 um). There is no internal longi- tudinal layer developed, such as is seen in some species of Dugesia of the Old World. Then follows a wide (130-170 um) intermediate or mesenchymal zone traversed by numerous pharyngeal gland ducts, the cell bodies of which lie in the mesenchyme anterior to the pharyngeal root. The internal muscle zone consists of two muscle layers, a thinner layer of longitudinal fibers (8 um) and a wider layer of circular muscles (60 um). The internal pharyngeal canal is lined with an epithelium (5 um) which is infranucleate in the posterior and nucleate in the an- terior part of the pharynx. Reproductive system: The testes (Fig. 3) are subdorsal, arranged in a pair of longitudinal rows beginning a certain distance behind the Ovaries and extending posteriorly to the level of the copulatory ap- paratus. The region behind the gonopore generally is devoid of testic- ular follicles. Moderate numbers of testes are located in the dorsal parts of the mesenchyme and in the spaces between the intestinal branches. In the zone of the testes, the yolk glands or vitellaria are developed chiefly in the ventral regions, some of them also reaching to the dorsal side. The thin vasa efferentia descend from the individual testicular follicles ventrally and connect, on either side, with the vas deferens which runs along the ventral nerve cord medial to the ovi- duct. In the region of the pharynx, the vasa deferentia expand to form the usual contorted spermiductal vesicles filled with masses of sperm. De Beauchamp (1939: 66) reported that in a South American species of Dugesia, “Euplanaria aurita’ [probably Dugesia festai (Borelli)], the testes are usually ventral, particularly in specimens which are not 116 Proceedings of the Biological Society of Washington Fic. 3. Dugesia arizonensis, parasagittal section of prepharyngeal region, < 74. ed, efferent duct; i, intestine; n, ventral nerve cord; f, testes; v, vitellaria. Fic. 4. Dugesia dorotocephala, parasagittal section of prepharyn- geal region, X 134. Abbreviations as in Fig. 3. fully mature sexually and have the vitellaria incompletely differentiated. Other specimens, however, with fully developed vitellaria, have the testes located exclusively dorsally. He assumes that there is a migration of the testes at full maturity or possibly after copulation. Ernesto Marcus (1960: 46), on the other hand, observed the testes of D. festai to be ventral, occasionally extending throughout the dorsoventral diameter of New freshwater planarian from U.S.A. 117 0.5mm vd bd sg [ (} ee mm” \ \ i b v de am gp ac od Fic. 5. Dugesia arizonensis, semidiagrammatic view of copulatory ap- paratus in sagittal section. ac, common atrium; am, male atrium; b, copulatory bursa; bd, bursal duct; de, ejaculatory duct; gp, gonopore; od, oviduct; sg, shell glands; vd, vas deferens; vs, seminal vesicle. the body; in only one young individual they were dorsal for a certain distance. Other authors who have examined D. festai (Borelli, 1898: 4; Hyman, 1939: 420 [D. titicacana]; Eveline Marcus, 1953: 66) indicate that the testes are ventral. Similarly, Ball (1971: 11) reports that he found the testes of D. arimana to be predominantly dorsal, except in two specimens, one of which had the testes predominantly ventral on the right side and dorsal elsewhere. The other individual had ventral testes throughout. Ball considers these conditions to be anomalies. Ermesto Marcus (1946: 152) states that in D. nonatoi Marcus the testes originate dorsally but, in the fully mature state, are positioned ven- trally, with only a few dorsal testes in the postcopulatory region. In D. hypoglauca Marcus (1948: 169) the testes are lateral to the vitel- laria, originate centrally, but during growth extend to both the dorsal and ventral sides. I have examined sections of 14 mature as well as semi- mature individuals of D. arizonensis and find the testes in the semi- dorsal position at all stages. The paired ovaries, lacking parovaria, are situated at a considerable distance behind the brain, at the level of the fourth to sixth lateral branches of the intestinal trunk. The vitellaria, beginning anteriorly to the ovaries, occupy chiefly the ventral region of the mesenchyme, but also extend dorsally through the spaces between the _ intestinal diverticula. The copulatory apparatus (Fig. 5) is somewhat removed posterior to the pharyngeal chamber. The genital aperture (gp) leads into the 118 Proceedings of the Biological Society of Washington © common atrium (ac) which receives the mouth of the bursal canal (bd) dorsally and connects with the male atrium (am) anteriorly. Between the two atria is a conspicuous constriction formed by folds of the atrial wall which leave only a small round opening connecting the two chambers. The epithelium of the common atrium is cuboidal and, at least in part, infranucleate. That of the male atrium is rather thick, cylindrical, and nucleate. The atrial walls have the usual two muscular layers, a circular one and a longitudinal one. The penis consists of a moderately developed, round bulb and an elongate, very flexible, easily twisted, pointed papilla, covered by a rather flattened epithelium. The bulb contains a lobed cavity, the seminal vesicle (vs), lined with a tall, apparently secretory epithelium. Toward the papilla, with a funnel-shaped transition, the cavity ex- tends into a narrow canal, the ejaculatory duct (de), which runs through the center of the papilla, opening at the tip. The lining of the duct is a cuboidal to flattened epithelium. Numerous faintly cyanophilic gland ducts enter the penis bulb from the surrounding mesenchyme. The vasa deferentia (vd), which approach the copulatory complex as widened spermiductal vesicles, run posteriorly to the level of the male atrium, then curve upward, bend forward, and enter the penis bulb dorsolaterally. This recurving of the sperm ducts is very characteristic and is seen in all mature specimens. Within the tissues of the bulb, the ducts narrow, acquire a coat of circular muscle fibers, and open into the seminal vesicle separately. The copulatory bursa (b) is a rather large, ovoid sac situated at a considerable distance posterior to the pharyngeal pouch. Its outlet, the bursal duct or bursal stalk (bd) runs posteriorly above the penis, widening gradually. At the level of the gonopore, or somewhat more posteriorly, it curves ventrally and proceeds anteroventrally to open into the common atrium. It is surrounded by a dense muscular coat, often rather thick, which is somewhat difficult to analyze. Apparently is consists mainly of circular fibers surrounded by longitudinal muscles. Below the downward bend, the bursal duct receives the two oviducts (od) entering separately from the sides. In the adjoining section, many eosinophilic gland ducts, the shell glands (sg) open into the bursal stalk. All epithelia of the copulatory complex are nucleate, with the ex- ception of the lining of the common atrium which is, at least partially, infranucleate. Distribution and ecology: Dugesia arizonensis was collected by Dr. Peter D’Eliscu in two localities in Arizona: Bog Springs in Madera Canyon, Santa Rita Mountains, Santa Cruz County (type-locality). 12 May 1972: 8 specimens, some of them semimature, collected on rocks and organic sediment, water temperature 60° F (16°C), pH 7.2. August 1973: Water flow very slow, tempera- ture 20° C, pH 7.4, several specimens collected. 3 November 1973: New freshwater planarian from U.S.A. 119 Water temperature 14° C, pH 7.4, about 35 specimens, some of them mature. Rucker Creek (or West Turkey Creek) in Chiricahua Mountains, Cochise County. 22 September 1973: 7 specimens, all immature (some matured later in the laboratory culture), collected about one-half mile upstream from Rucker Lake, water temperature 18° C, pH 6.7. Hyman (1931: 324) reported that a species of Euplanaria (= Dugesia) occurred in mountain springs near Tucson, Arizona, in size and shape similar to D. dorotocephala, but with a more pointed head and narrower and longer auricles. As no sexual specimens were obtained, she re- frained from describing and naming the species (which had been used by Fraps [1930] in his studies on the respiratory and glycolytic metab- olism in planarians ). It is hightly probable that this species was Dugesia arizonensis. Behavior in laboratory cultures: Specimens of the various collections were shipped to me alive and could be maintained in laboratory cultures of dechlorinated (aged) tap water, kept at a temperature of 14° C, and were fed beef liver once a week. Reproduction by fission was observed occasionally. Some of the worms became mature, but no egg capsules were laid in the cultures. The animals have considerable regenerative ability, as is known for all species capable of asexual reproduction by fis- sion. Several individuals were transmitted to Dr. Nicole Gourbault (of the Muséum National d’Histoire Naturelle, Paris) for an analysis of their chromosome morphology. Taxonomic relations: Dugesia arizonensis belongs to Ball’s (1974: 376) subgenus Girardia, which comprises species without penial dia- phragm, with the musculature of the bursal stalk consisting of inner circular muscles surrounded by longitudinal fibers, and the testes not confined to the prepharyngeal region. This group has numerous repre- sentatives in the Americas and a few in Australia and an Indian Ocean island. The distinctive characters of D. arizonensis are the position and number of the testes, the anatomy of the penis, and the configura- tion of the genital atria. Contrary to the opinion of de Beauchamp (1939: 66), I believe that the location of the testes, whether ventral, dorsal, or intermediate, has some taxonomic significance. A comparison of Figures 3 and 4, showing the arrangement of the testes in D. arizonensis and D. dorotocephala, illustrates differences which certainly are of systematic value. Cases of secondary displacement of the testes in the course of sexual maturation, such as observed by de Beauchamp (1939: 66) in D. festai(?), by Ernesto Marcus (1946: 152) in D. nonatoi, and by the latter author (1948: 169) in D. hypoglauca, may be exceptional and must be checked by examining individuals in various phases of sexual maturity. Dorsal or semidorsal testes have been re- ported in 12 of the about 30 species of the subgenus Girardia. The pro- nounced recurving of the vasa deferentia before entering the penis also appears to be a usable characteristic and is known to occur also in 120 Proceedings of the Biological Society of Washington several other species of Dugesia, e.g. D. dorotocephala. In the anatomy of the penis, the presence of a large seminal vesicle separates D. ari- zonensis from the group of species with a bifid penial lumen. The slender, pointed, and flexible penis papilla is a unique character within the subgenus. Another very characteristic feature is the separation of the male atrium from the common atrium by the striking constriction or atrial diaphragm, a condition which it shares only with D. arimana Hyman (see Ball, 1971: 6). Acknowledgments: I am sincerely thankful to Dr. Peter D’Eliscu for his kind collaboration in procuring ample materials of the new species, and to Dr. John C. Harshbarger of the Smithsonian Institution for his aid in preparing the photomicrographs used in this paper. LITERATURE CITED Batz, I. R. 1971. Systematic and biogeographical relationships of some Dugesia species (Tricladida, Paludicola) from Central and South America. Amer. Mus. Novitates No. 2472:1-25. —. 1974. A contribution to the phylogeny and biogeography of the freshwater triclads (Platyhelminthes: Turbellaria). In Bi- ology of the Turbellaria (N. W. Riser and M. P. Morse, ed.): 339-401. McGraw-Hill Book Co., New York. BEAUCHAMP, P. pE. 1939. The Percy Sladen Trust Expedition to Lake Titicaca in 1937 under the leadership of Mr. H. Cary Gilson, M.A. V. Rotiféres et turbellariés. Trans. Linn. Soc. London, ser. 3, 1:51—79, pl. 4. Boretur, A. 1898. Viaggio del Dr. Enrico Festa nell’Ecuador e regioni vicine. IX. Planarie d’acqua dolce. Boll. Mus. Zool. Anat. Comp. Univ. Torino 13(322):1-6. Fraps, M. 1930. Studies on respiration and glycolysis in Planaria. I. Methods and certain basic factors in respiration. Physiol. Zool. 3:242-270. Hyman, L. H. 1931. Studies on the morphology, taxonomy, and dis- tribution of North American triclad Turbellaria. IV. Recent European revisions of the triclads, and their application to the American forms, with a key to the latter and new notes on dis- tribution. Trans. Amer. Microsc. Soc. 50:316—335. —. 1939. New species of flatworms from North, Central, and South America. Proc. U.S. Natl. Mus. 86:419—-439. Marcus, Ernesto. 1946. Sobre Turbellaria brasileiros. Univ. Sao Paulo, Bol. Fac. Filos. Cién. Letras, Zool. 11:5-187, pls. 1-31. ——. 1948. Turbellaria do Brasil. Univ. Sao Paulo, Bol. Fac. Filos. Cién. Letras, Zool. 13:111—-202, pls. 1-20. —. 1960. Turbellaria from Curacao. Stud. Fauna Curacao other Caribb. Isl. 10:41-51. Marcus, EvELInE. 1953. Some South American triclads. Anais Acad. Brasil. Cien. 25:65—78. No. 13, pp. 121-126 15 August 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON CHIRIDOTEA STENOPS MENZIES AND FRANKENBERG, A JUVENILE OF C. ARENICOLA WIGLEY (CRUSTACEA: ISOPODA) By Les WaTLING AND Don MAURER Marine Studies Center, University of Delaware as : Lewes, Delaware 19958 To date, six species of the western Atlantic isopod genus Chiridotea Harger have been described (Bowman, 1955; Wigley, 1960, 1961; Menzies and Frankenberg, 1966). Of these, two (C. almyra Bowman and C., nigrescens Wigley ) oc- cur in brackish water while the remaining species [C. coeca (Say), C. tuftsi (Stimpson), C. arenicola Wigley, and C. stenops Menzies and Frankenberg] occur in marine waters. Until the description of C. stenops by Menzies and Franken- berg (1966), a common feature of the genus was the presence of deep clefts in the lateral cephalic margins. Menzies and Frankenberg (1966) commented that C. stenops was unique in possessing lateral cephalic margins that were not incised and thus it did not appear to be very closely related to any other species in the genus. It is the purpose of this paper to show that the lack of incised lateral cephalic margins is char- acteristic of immature Chiridotea and that C. stenops is a juvenile of C. arenicola. MATERIAL EXAMINED Specimens examined were obtained from the following sources: a) the paratypes of C. stenops (USNM 111072) and C. arenicola (USNM 104281) were obtained from the US. National Museum; b) several individuals were obtained from a series of samples taken on the Delaware-Maryland conti- nental shelf by the Environmental Protection Agency; c) a 13—Proc. Brot. Soc. WAsH., Vou. 88, 1975 (121) AN 4] 7 ¥ oT “ Go aD a a Cr . 122 Proceedings of the Biological Society of Washington Fic. 1. Lateral cephalic margin of several Chiridotea specimens. a, C. stenops paratype, 2.5-mm body length; b, C. stenops paratype, 3.0 mm; c, specimen from continental shelf off Delaware, 4.0 mm; d, C. arenicola paratype, male, 5.0 mm; e, C. arenicola paratype, female, 6.5 mm. a, b, d, e, drawn to scale different from c. series of specimens of C. tuftsi from Massachusetts Bay were obtained from the U.S. National Museum. RESULTS AND Discussion Menzies and Frankenberg (1966) listed the following fea- tures as being diagnostic of C. stenops: (1) lateral margin of cephalon not deeply incised; (2) preocular region of cephalon narrow as compared to the produced postocular region; (3) apex of pleotelson with only three to five lateral setae; and (4) eyes reduced, almost obscure. To establish which char- acters varied with increasing size, the paratypes of C. stenops and C. arenicola and locally obtained specimens from the Delaware-Maryland shelf were examined for the following characteristics: size of the cleft in the lateral cephalic mar- gin; number of spines on antenna 2 peduncle segment 4; num- Reevaluation of Chiridotea species 123 Taste 1. Change in value of selected features with increasing body length. No. spines No. antenna No. spines, 25 flagellar lateral peduncle segments, margin Length segment antenna telson Source of Specimen (mm ) 4 2 apex 1. C. stenops paratype 2.5 3 2; 3 2. C. stenops paratype 3.0 3 2(3)* 2-3 3. Off Bethany Beach, De. 3.0 5 2(3) 4 4, Mid-Atlantic shelf, 38°11.9’N, 74°32.9’W 3.0 5 2 5. Off Bethany Beach, De. 3.5 5 3 3 6. Off Bethany Beach, De. 4.0 5 2(3) 5 7. Mid-Atlantic shelf, 38°11.9'N, 74°32.9’W 4.0 8. C. arenicola paratype é 5.0 9. Mid-Atlantic shelf, 38°23.8’N, 74°15.3’W 5.0 0 3(4) 4 10. Mid-Atlantic shelf, 38°11.9’N, 74°32.9’W 5.5 9 4 8 11. C. arenicola paratype 9 6.5 10 4 9 12. Off Delaware Bay, 38°47.5'N, 74°45.5’W 6.5 11 4 W 13. Off Bethany Beach, De. 6.5 9 4 8 2 The numbers in parentheses indicate that within the exoskeleton of one of the segments, two new segments were visible. ber of antenna 2 flagellar segments; and number of spines on the lateral margins of the telson apex. The changes in the lateral cephalic margins are documented in Figure 1 and the values for the other characters examined are given in Table 1. While neither of Menzies and Franken- berg’s (1966) paratypes possessed a distinct cleft to divide the lateral cephalic margin into pre- and post- ocular lobes, there was a slight ventral indentation just forward of the eye. As larger specimens were examined, the cleft became more dis- tinct and the preocular lobe more quadrate as in the C. 124 Proceedings of the Biological Society of Washington Fic. 2. Chiridotea tuftsi, from Massachusetts Bay, USNM 35258, 2.0 mm body length: a, lateral cephalic margin; b, second antenna, peduncle segment 5 and 2 flagellar segments. arenicola paratypes (Fig. 1). Similarly, the number of spines on the postocular lobe increases from 1 on the Menzies and Frankenberg specimens to 4 on Wigley’s specimens of C. arenicola. The values for the other characters listed in Table 1 also changed with increasing body length. To substantiate that the changes in the above features were related to the age of the individuals, several specimens of the very distinct C. tuftsi were also examined. Individuals of this species, regardless of their size, can be recognized by the presence of several strong spines on the inner margin of the dactyl of pereopod 1. It was observed (Fig. 2) that the lateral cephalic margins of a 2.0-mm specimen (USNM 35258) showed a striking resemblance to the paratypes of Menzies and Frankenberg (1966). As larger individuals were examined, the cleft in the lateral cephalic margin became more distinct and the preocular lobe much more pronounced. Similarly, the number of spines on both the pre- and post- ocular lobes increased with increasing body length. Other features, such as the number of antenna 2 flagellar segments, also increased. On the basis of the evidence presented, it is concluded that Chiridotea stenops Menzies and Frankenberg (1966) is a synonym of Chiridotea arenicola Wigley (1960). It is also suggested that C. nigrescens and C. coeca be re-examined to determine if there is a similar gradational change in characters between those two species. ACKNOWLEDGMENTS We would like to thank Dr. Thomas E. Bowman for the loan of the U.S. National Museum specimens and for his en- Reevaluation of Chiridotea species 125 couragement of this study. We are also indebted to Dr. Donald Lear of the Environmental Protection Agency, Regional Field Office, Annapolis, Maryland, for obtaining specimens from the Delaware-Maryland continental shelf. LITERATURE CITED Bowman, T. E. 1955. The isopod genus Chiridotea Harger, with a description of a new species from brackish waters. Jour. Wash- ington Acad. Sci. 45(7):224—229. Menzies, R. J., AND D. FRANKENBERG. 1966. Handbook on the com- mon marine isopod Crustacea of Georgia. University of Georgia Press, Athens, Ga., 93p. Wictey, R. L. 1960. A new species of Chiridotea (Crustacea: Isop- oda) from New England waters. Biol. Bull. 119(1):153- 160. 1961. A new isopod, Chiridotea nigrescens, from Cape Cod, Massachusetts. Crustaceana 2(4):286-292. 126 Proceedings of the Biological Society of Washington Vo. 14, pp. 127-140 15 August 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON THE OPHIACANTHID GENUS AMPHILIMNA (OPHIUROIDEA, ECHINODERMATA) By Lowe. P. THOMAS Rosenstiel School of Marine and Atmospheric Science, Division of Biology and Living Resources, University of Miami, Miami, Florida 33149 Until recently Amphilimna Verrill, 1899 was considered an amphiurid genus and considerable confusion existed regard- ing the systematic characters of the type-species, Amphilimna olivacea (Lyman, 1869). Some of the mystery surrounding this species was cleared up when it was discovered (Thomas, 1967) that A. olivacea is not an amphiurid, and that a second nominal species, A. caribea (Ljungman, 1871), belongs in the ophiocomid genus Ophiopsila. Unfortunately, I considered Amphilimna reduced to monotypic status by my studies, over- looking a species described from the Philippines 45 years earlier (A. multispina Koehler, 1922). Consequently, as of 1967 Amphilimna actually included two nominal species, A. olivacea and A. multispina. Thomas and Schoener (1972) suggested that Amphitarsus H. L. Clark, 1941, with two species, A. mirabilis H. L. Clark, 1941, and A. nike Schoener, 1967 should probably be con- sidered a junior synonym of Amphilimna. Devaney (1974), describing a new species of Amphilimna from Pitcairn Island, acted upon this suggestion by making Amphitarsus a junior synonym of Amphilimna. Unfortunately, Devaney also over- looked Koehler’s Pacific species, considering his own new one, A. tanyodes, the first known from the Pacific. Another species of Amphilimna, A. cribriformis was described from southern Africa by A. M. Clark in 1974 in a paper published about the same time as Devaney’s. Finally, during a visit to the British 14—-Proc. Biot. Soc. WasH., VoL. 88, 1975 (127) Se 128 Proceedings of the Biological Society of Washington Museum (Nat. Hist.) in the summer of 1974, I discovered, as A. M. Clark (1974) had suggested, that Anamphiura valida H. L. Clark is in fact an Amphilimna; thus, including the latter species as a new combination, there are now seven species in the genus Amphilimna: A. olivacea (Lyman, 1869), A. multispina Koehler, 1922, A. nike (Schoener, 1967), A. mirabilis (H. L. Clark, 1941), A. tanyodes Devaney, 1974, A. cribriformis A. M. Clark, 1974, A. valida (H. L. Clark, 1939) new combination. In the following account I have included only selected references in the synonymies. These references should be con- sulted for more complete synonymies. The diagnoses have been deliberately kept short. As more material of the poorer known species becomes available, expanded diagnoses will un- doubtedly appear. Similarly the keys to the species must be considered provisional, subject to future elaboration and revision. Through the years, while working on Amphilimna and other genera, I have been assisted by Maureen Downey and David Pawson of the United States National Museum of Natural History (hereafter abbreviated USNM); Elizabeth Deich- mann, H. Barraclough Fell, Charlene Long, and Amy Schoener of the Museum of Comparative Zoology at Harvard (MCZ); and Ailsa Clark of the British Museum (Natural History) (BM). Mary Anna Bogle assisted with most of the technical work for this paper, and Claire Ulanoff made the illustrations of A. nike. I am grateful to all for their long and valued as- sistance and friendship, and particularly for their patience when it appeared academic and other responsibilities would preclude my completing some of the research projects I had begun. I am also grateful to Linda Pequegnat of Texas A & M University who arranged for me to look at the collection in which a fine large specimen of A. nike was discovered and to Dennis Devaney who kindly edited my diagnosis of his new species. This study has been conducted with the support of the following grants in aid for research from the National Science Foundation which I thankfully acknowledge: NSF GB 16556 and NSF GB 39871. Brittle star genus Amphilimna 129 OPHIACANTHIDAE Amphilimna Verrill, 1899 Amphilimna Verrill, 1899: 30.—Thomas, 1967: 123.—Devaney, 1974: 122. Amphitarsus H. L. Clark, 1941: 83. Diagnosis: The following is based upon consideration of the 7 nomi- nal species listed above which presently constitute the genus Amphilimna. The differences between this diagnosis and Devaney’s (1974) are dis- cussed below: Disc notched at radial shields; radial shields variable in size and shape; disc scales covered to varying degrees by spines, spinelets, and/or granules. Two to 6 oral papillae at apex of jaw, 2 or 3 at or near distal end of jaw around second oral tentacle pore, 1 or more papillae (1 an oral tentacle scale?) between distal and apical clumps of papillae. Some arm spines under disc flattened and fused together forming peculiar “flanges” in the genital slits. Ventral arm plates abruptly widen distally. Proximal arm segments usually with 2, oc- casionally to 5, tentacle scales, 1 to 3 on lateral arm plate, 1 or 2 usually arising from middle of side of ventral arm plate; tentacle scales spiniform, round and flat, or some intermediate condition. Discussion: Devaney’s (1974) treatment and diagnosis of Amphilimna requires some modification and comment, partly because of the addition of more species to the genus. The following range of variation may be found if one considers all 7 species: The distal oral papillae, bordering the outermost tentacle pores, are not necessarily “squamiform,” but may be either round and very flat, spiniform, or of some intermediate con- dition. There is, varying in expression from species to species, a tendency for the proximal oral papillae to occur in a distinct cluster at the jaw apex. The distal and proximal groups of papillae are usually separated from one another by 1 or more papillae (including what seem to be oral tentacle scales). The ventral arm plates of all species widen abruptly at their distal ends, a condition I consider of generic importance. With the addition of A. valida to Amphilimna the arm length to disc diameter ratio deviates considerably from Devaney’s figure of “six or more times disc diameter.” The arm length may be only 2 times the disc diameter in A. valida, and the ratio for ophiuroids in general is probably useful only within specific size classes of individual species. Earlier (Thomas, 1967: 125) I discussed the dental plates of Amphi- limna olivacea and pointed out their similarities with those of other ophiacanthid genera, but I also suggested that Ophioprium cervicorne (Lyman, 1883) was closely related to A. olivacea, After examining the holotype of O. cervicorne I still believe the 2 genera are quite closely related. Devaney has suggested that Amphilimna shows affinities with a group of 6 genera, with large tentacle pores, indicated by Matsumoto (1917) and (possibly?) with 2 genera included in that group by Fell (1960). Two of the genera Devaney lists seem to be synonyms of other 130 Proceedings of the Biological Society of Washington genera and the affinities of the 8 nominal genera with Amphilimna vary, in my opinion, from none whatsoever to a definite affinity in the case of Ophioprium. Considering the genera in question, Ophiopora Verrill, 1899 is treated by Fell (1960) as a synonym of Ophiotoma Lyman, 1883, whose type-species, Ophiotoma coriacea Lyman, 1883, except for lacking tentacle scales, looks vaguely like an Amphilimna. Ophiopristis Verrill, 1899 is considered a synonym of Ophiacantha Miiller and Troschel, 1842 by Fell (1960) but it seems to be mis- placed here, as the type-species of Ophiopristis, O. bartletti Lyman, 1883, has large tentacle pores and looks very much like an Ophiotoma. The remaining genera in Devaney’s list are Ophiambix Lyman 1880, Ophiomedea Koehler, 1906, Ophiotrema Koehler, 1896, Ophiodelos Koehler, 1931, Glaciacantha Fell, 1961 and Ophioprium, Clark, 1915. The type-species of Glaciacantha, Ophiodelos, and Ophiambix, in my opinion, cannot be considered as having any “affinity” with Aphilimna. The type-species of Ophiomedea, O. duplicata Koehler, 1906, has a similar tentacle scale arrangement and in other ways looks vaguely similar to A. olivacea, as does the type-species of Ophiotrema, O. alberti Koehler, 1896, which has naked radial shields, spines on the disc, large tentacle pores with tentacle scales on both the lateral arm plates and ventral arm plates. The major problem in determining relationships of other genera with Amphilimna appears to me to be in determining the generic char- acteristics of Amphilimna itself. Aside from the flattened fused arm spines under the disc, the features which seem to be most characteristic of this genus are the shape of the ventral arm plates and the arrange- ment of the tentacle scales around the large tentacle pores, the presence of the oral papillae, notches in the disc at each radial shield pair, and the presence of spines or granules on the disc. Characters which vary from species to species include the relative size and shape of the radial shields (these vary also with the size of the individual), the disc cover- ing, the shape and numbers of oral papillae, the shape and numbers of tentacle scales, and the number of arm spines. Amphilimna olivacea (Lyman, 1869) Ophiocnida olivacea Lyman, 1869: 340. Amphitarsus spinifer Schoener, 1967: 269. Amphilimna olivacea: Thomas, 1967: 123, figs. 1-6.—Madsen, 1970: 163, fig. 5.—Tommasi, 1970: 32, figs. 30, 31—A. M. Clark, 1974: 444 [in discussion] —Devaney, 1974: 121 [in discussion]. Material Studied: 14 spec.; GosNouip Sta. 1074; northeast of Hudson Canyon (39°43.0’N, 71°53.2’,W), 225 m; 12 May 1963.—6 spec.; Gosnotp Sta. 1261; Block Canyon (40°00.0’N, 71°16.8’W), 235 m; 4 October 1963.—4 spec.; GosNop Sta. 1571; off Florida Keys (24°32.0’N, 81°16.5’°W), 89 m; 31 May 1964.—6 spec.; Gosnoxp Sta. 1572; off Florida Keys (24°29.3’N, 81°27.7’W), 97 m; 31 May 1964. Brittle star genus Amphilimna 131 Diagnosis: Jaws with 2 or 3 papillae at apex, 2 or 3 slender, pointed oral papillae bordering each oral tentacle pore; 1 papilla (probably an oral tentacle scale) separating infradental papillae and papillae arising from oral tentacle pore. Oral shield rhomboidal, wider than long; adorals meeting or nearly meeting proximally; arms slender, 6 times disc diameter in large specimen, 8 to 10 proximal arm plates with 2 tentacle scales at each tentacle pore, innermost attenuated, arising from middle of ventral arm plate; beyond tenth ventral arm plate innermost absent; 7 or 8 slender arm spines, ventralmost longest; all but ventral 2 or 3 arm spines of arm segments under disc greatly flattened, fused with others to form peculiar flanges occupying genital slits; dorsal arm plates slightly wider than long. Disc scales studded with slender spines. Radial shields narrow, about 3 times longer than wide, joined proximally, slightly separated distally; disc deeply notched at each pair of radial shields. Two slender spines at distal ends of genital plates near junction with radial shields. Color variable, disc gray, tan or brown, arms pink or orange. Discussion: With its slender disc spines, attenuated inner tentacle scales, and slender oral papillae, O. olivacea appears to be most similar to A. cribriformis. It differs in having a rhomboidal oral shield, con- tiguous or nearly contiguous adorals, longer arm spines, and in a va- riety of other details discussed by A. M. Clark (1974: 444). Distribution: Widespread in the Atlantic Ocean from 60 to 350 m. Perhaps extending to nearly 500 m (Verrill, 1899b: 377). Liberia to Angola in the eastern Atlantic, Massachusetts to Uruguay in the western Atlantic. Amphilimna cribriformis A. M. Clark, 1974 Amphilimna cribriformis A. M. Clark, 1974: 442, Fig. 1. Material Studied: None. Diagnosis (modified after A. M. Clark, 1974): Jaws with 2 to 4 infradental papillae at apex, 3 slender, pointed oral papillae bordering each oral tentacle pore; one papilla (probably an oral tentacle scale) separating apical and distal oral papillae. Oral shields triangular, broadest distally; adoral plates separated at proximal angle of oral shield. First 8-10 arm segments bearing 2 tentacle scales, innermost arising from middle of ventral arm plate, initially long and slender; beyond tenth or twelfth ventral arm plate innermost absent; outermost arising from lateral arm plate, short, becoming flattened toward distal end of arm. Six or fewer arm spines each arm segment, under disc all but ventral spine flattened and fused to form peculiar flanges occupying genital slits; dorsal arm plates beyond first few ovate, broader than long, thin, semitransparent. Disc scales studded with slender spines. Radial shields long, narrow, joined proximally, slightly separated dis- tally; disc somewhat notched at each pair of radial shields. Two spines 132 Proceedings of the Biological Society of Washington stouter than disc spines at distal ends of genital plates near junction with radial shields. Discussion: A. M. Clark (1974) described this species in detail. It differs from O. olivacea, with which it seems most similar, in having triangular oral shields, slightly separated adoral shields, shorter arm spines which are never as long as an arm segment, 2-4 infradental papillae rather than 2-3, and in the oval shape and transparent, delicate nature of the dorsal arm plates. Distribution: Known only from along the continental shelf of eastern South Africa and southern Mozambique, 86 to 200 m. Amphilimna multispina Koehler, 1922 Figure 1 Amphilimna multispina Koehler, 1922: 183, p. 61, figs. 4-9, pl. 96, fig. 7 Material Studied: 1 spec. (paratype); ALBATROss Sta. 5220; between Marinduque and Luzon, Philippines (13°38’00”N, 121°58’00”E), 91 m; 24 April 1908; USNM Cat. No. E 23.—13 spec. (paratypes); ALBA- rross Sta. 5221; between Marinduque and Luzon, Philippines (13°38’15”N, 121°48715”E), 335 m; 24 April 1908; USNM Cat. No. E 26, E 27.—4 spec. (paratypes); ALBATROss Sta. 5222; between Marinduque and Luzon, Philippines (13°38’30"N, 121°42’45”E), 357 m; 24 April 1908; USNM Cat. No. E 25.—1 spec. (paratype); ALBA- TRoss Sta. 5223; between Marinduque and Luzon, Philippines (13°36’00”N, 121°25’30"E), 357 m; 24 April 1908; USNM Cat. No. E 24.—4 spec. (paratypes); ALBATROss Sta. 5375; Marinduque Island and vicinity, Philippines (13°42’15”N, 121°50’15”E), 196 m; 2 March 1909; USNM Cat. No. E 4. Diagnosis: Jaws with one or 2-6 short blunt papillae at apex, 2 large broadly flattened papillae longer than wide bordering each oral tentacle pore, oral papilla (probably an oral tentacle scale) separating infradental papillae and papillae arising from oral tentacle pore. Oral shield almost twice longer than wide, arrowhead shaped, pointed proximally, broadly rounded distally. Adoral shields joined at distal end of oral papillae except in very large specimens. Arms slender, probably at least 6 times as long as disc diameter. Tentacle scales flattened, oval, scale arising from lateral arm plate largest, completely covering tentacle pore, smaller similarly shaped scale arising from side of ventral arm plate for first 8-10 segments; 6-9 arm spines, slender, flattened, dorsal- most longest, longer than an arm segment. All but ventral 2 or 3 spines > Fic. 1. A, Amphilimna nike, Gosnoxp Sta. 1580, oral view, two jaws and base of arm. B, A. nike, Gosnoxp Sta. 1580, radial shields and base of Brittle star genus Amphilimna 133 D arm. C, A. multispina, Paratype, ALBATROss Sta. 5221, oral view, two jaws and base of arm. D, A. multispina, Paratype, ALBATROss Sta. 5221, radial shields and base of arm. Scale equals 1 mm. 134 Proceedings of the Biological Society of Washington of each side arm plate under disc flattened, fused with others to form peculiar flanges occupying genital slits. Disc scales studded with slender spines; radial shields narrow, 3-4 times longer than wide, usually slightly separated at their distal ends. Disc deeply notched at each pair of radial shields. One to 3 slender spines at distal ends of genital plates near junction with radial shields. Color pale tan, arms lightly banded with darker brown. Discussion: Amphilimna multispina bears large, round, flattened ten- tacle scales like those of A. valida, A. nike, A. mirabilis, and A. tanyodes. It differs from the last three, however, in possessing slender spines rather than granules or granulose spinules on the disc. The pos- sible relationship of A. multispina with A. valida or A. tanyodes is dis- cussed under those species. Distribution: Known only from Koehler’s (1922, p. 183) original material near Marinduque Island in the Philippines, 91 to 357 m. Amphilimna valida (H. L. Clark, 1939) new combination Anamphiura valida H. L. Clark, 1939: 70, figs. 26a, 27—A. M. Clark, 1974: 478, fig. 16. Material Studied: 1 spec. (holotype); John Murray Exped. Sta. 105; off Zanzibar, 238-293 m; 11 January 1934; BM 1948.5.26.87. Diagnosis: Jaws with 2-4 papillae at apex, 2 or 3 papillae around oral tentacle scale, one oral papilla (probably an oral tentacle scale) between the two groups. Oral shield rhomboidal, wider than long. Adorals meet at proximal end of oral shield. Arms short, only 2 or 3 times longer than disc diameter. First 3-6 arm segments bear 2 ten- tacle scales at each tentacle pore, 1 on lateral arm plate round, flattened, large, 1 on side of ventral arm plate smaller, round or conical. There- after only large outer scale. First 1 or 2 arm segments bear some arm spines flattened and coalesced into “flange,” 2 ventral spines free. Five or 6 arm spines on segments nearest disc; lowest longest, equalling or longer than arm segment, uppermost shortest, one-half length of arm segment. Disc scales include large primitive plates, scattered slender spines on disc. Radial shields short, up to 2 times longer than wide, joined proximally, widely separated distally, disc notched at distal ends of radial shields. Discussion: Although A. valida as presently known has all the char- acteristics of a young specimen and consequently may be a synonym of another species, there are characters that suggest it is distinct. The broad rhomboidal oral shield is conservative in A. olivacea and pre- sumably so in A. valida as well. This shape shield, combined with slender disc spines, occurs in no other nominal Pacific species of Am- philimna. Otherwise A. valida appears most similar to A. multispina which has similar tentacle scales and slender disc spines. Distribution: Known only from John Murray Sta. 105, Zanzibar area Brittle star genus Amphilimna 135 238-293 m (H. L. Clark, 1939), and near Tugela River mouth, 350 m (A. M. Clark, 1974). Amphilimna mirabilis (H. L. Clark, 1941) Amphitarsus mirabilis H. L. Clark, 1941: 83, pl. 8.—Schoener, 1967: 2, ro feaae Amphilimna mirabilis: Devaney, 1974: 22. Material Studied: 1 spec. (holotype); ATLANTis Sta. 3402; off Cayo Coco, Camaguey Province, 420 m; MCZ 6232.—1 spec. (paratype); ATLANTIS Sta. 3332; Bahia de Cochinos, Santa Clara Province, Cuba, 320-412 m; MCZ 6398.—1 spec. (paratype); ATLANTIS Sta. 3422; off Caibarien, Santa Clara Province, Cuba, 430 m; MCZ 6399. Diagnosis: Jaws with 3 or 4 oral papillae at apex of jaw, 2 or 3 flat, round papillae guarding oral tentacle pore; 2 groups of oral papillae separated by (probably oral tentacle scale) oral papilla. Oral shield thomboidal, slightly wider than long with rounded angles. Adoral shields meeting at distal side of oral shields. Arms slender, more than 6 times disc diameter, segments under disc with arm spines flattened and fused forming flanges, 1 or 2 ventral spines free. Eight or 9 arm spines near disc, ventralmost longest. Tentacle scales round, flat, largest arising from lateral arm plate, 1 or 2 smaller ones arising from side of ventral arm plate of basal arm segments. Disc scales thickly studded with tiny spinules and granules. Radial shields narrow, 4 times longer than wide, separated their distal third, disc deeply notched at each pair of shields, several slender spines arising from each genital plate within notch. Color, disc grayish white, arms yellowish white. Discussion: Amphilimna mirabilis seems to be most similar to A. nike and A. tanyodes, the only other species with granules or closely grouped spinules rather than slender spines on the disc. It differs from A. nike in having at least twice as many arm spines, a much more dense covering of granules, much longer radial shields and in the adoral shields meeting at the distal side of the oral shields. Amphilimna tanyodes ap- pears to be more like A. mirabilis in the above and other characters but seems to have somewhat longer spinules on the disc. In view of the distribution of the 2 species, A. tanyodes, though similar, must be dis- tinct from A. mirabilis. Distribution: Has been taken only off the north and south coasts of Cuba, 320 to 430 m. Amphilimna nike (Schoener, 1967) Figure 1 Amphitarsus nike Schoener, 1967: 2, fig. 2. Amphilimna niki: Devaney, 1974, p. 122 [lapsus]. Material Studied: 1 spec. (holotype); WHOI Cxain cruise 35, Dredge Sta. 33; 140 miles north of Surinam (7°53.5’N, 54°33.3’W), 136 Proceedings of the Biological Society of Washington 535 m; 25 April 1963; MCZ 6797.—1 spec.; disc diameter 10 mm; Gosno_p Sta. 1580; off Florida Keys (24°10.0’N, 81°22.0’'W), 681 m; 1 June 1964.—1 spec.; Texas A&M cruise 69A11, Sta. 58; off Veracruz, Mexico (19°02.6’N, 95°27.5’W), 476 m; no date (specimen returned to Texas A&M ). Diagnosis: Oral papillae varying from jaw to jaw, 2—4 small papillae at apex of each jaw, 1 or 2 large round flattened papillae distally at each oral tentacle pore, occasionally a smaller papilla also, apical and distal groups separated by 1 or 2 small papillae (probably oral ten- tacle scales). Oral shields slightly longer than wide, roughly pentagonal with proximolateral sides meeting at a rounded angle and _ disto- lateral sides meeting a distal side. Adoral shields widely separated by oral shields. Arms probably 6 or more times disc diameter; arm spines under disc flattened and fused to form flanges except ventralmost which is free. Large, round, flattened tentacle scale arising from lateral arm plate, 1 or 2 much smaller scales arising from sides of ventral arm plates of proximal arm segments for varying distances along arm. Three or 4 slender arm spines, longest about as long as arm segment. Dorsal arm plates wider than long, often slightly separated by lateral arm plates. Disc scales covered by scattered conical granules often as far apart as 2 or 3 times diameter of granule; granules giving way to spinules on ventral surface of disc. Radial shields separated, only one and one-half times longer than wide, broadest distally; disc notched at each pair of radial shields. Discussion: Amphitarsus nike and A. mirabilis both display con- siderable variation in their oral armature. Fortunately they seem easily separable by other characters, particularly the number of arm spines and proportions of the radial shields. In addition to the smaller speci- mens of the type material, A. nike is now known from a specimen with a disc diameter of 10 mm (see material studied). Distribution: Known from the type-material collected in the tropical Atlantic 140 miles north of Surinam at 535 m and other material from the Gulf of Mexico off Veracruz, Mexico at 476 m, and in the southern Straits of Florida at 681 m. Amphilimna tanyodes Devaney, 1974 Amphilimna tanyodes Devaney, 1974: 116, figs. 1-5. Material Studied: None. Diagnosis (modified after Devaney, 1974): Two to 4 small papillae extending around proximal end of jaw; 2 large round or oval papillae at each oral tentacle pore; papillae of proximal end of jaw separated from group at oral tentacle pore by 2 or 3 papillae, one nearest those at oral tentacle pore largest. Oral shields about as wide as long, bluntly pointed proximally, slightly lobed distally. Adoral shields meeting (or nearly so) at proximal end of oral shields. Arms slender, long, Brittle star genus Amphilimna 137 more than 10 times disc diameter. At size of 6-7 mm, 2 tentacle scales regularly to ninth arm segment, irregularly to 17th segment, single scale beyond; outer scale largest, oblong with rounded tip, flat, arising from side arm plate; inner scale smaller, more tapering, arising from ventral arm plate. Seven to 8 arm spines near disc; 1 to 3 lower arm spines on each lateral arm plate under disc unmodified, upper spines flattened and fused to form flanges; spines forming flanges di- vided beyond base. Disc scales studded with granules above, more widely spaced spinules below. Radial shields 3 times longer than wide, slightly separated proximally, more widely separated distally; disc notched at each pair of radial shields; 2 to 4 bluntly tipped spines at junction of radial shield and genital plate. Discussion: A. tanyodes, except for having granules and small spines (spinules) on the disc, appears rather similar to A. multispina Koehler. REMARKS ON THE GENUS AMPHILIMNA The seven nominal species of Amphilimna considered here seem to fall into three groups of similar species, one consisting of A. olivacea and A. cribriformis, one consisting of A. multispina and A. valida, and one consisting of A. nike, A. mirabilis and A. tanyodes. Unfortunately, so little is known about five of the seven species I feel it is impossible, at present, to construct a completely satisfactory key. The following provisional key may serve to assist in the identification of the species provided recourse to the diagnosis is made. PROVISIONAL KEY TO THE SPECIES OF AMPHILIMNA 1. On basal arm segment innermost tentacle scales (arising from ventral arm plates) long and slender, longer than outer scales Bnisingecromywaterallsarimn) platesjies soe le ee ee 2 On basal arm segments innermost tentacle scales round and flattened, or occasionally absent or granular, not long and slender, shorter than outer tentacle scale 3 2. Oral shield rhombic with angular or curved distal side _ A. olivacea Oral shield triangular with straight distal side ___ A. cribriformis 3. Disc scales of upper surface bear scattered spinelets Ea eA Disc scales of upper surface bear closely clustered tiny short spineletsh.and/Oneranvles) sa.8i ee ee ee 5 4, Oral shield longer than wide, 6-9 arm spines ___.._.___. A. multispina Oral shield wider than long, 6 or fewer arm spines _____. _ A. valida 5. Four arm spines or fewer, radial shields oval or round, as wide ASMlON Ga eeesete he eee tL ae ail a eee tee ae A. nike Six or more arm spines, radial shields at least twice as long CISMENV CLOW caniealmctatoee Ne SRY wes sR ie neared EES Te piety 1s: 6 GR AhlantiClispecles amie Se Putirs hy Sul fusion dte slant oily D A. mirabilis Raciticuspecies#usmts Meda eh Pears sah 31 shel) et AWN _ A. tanyodes 138 Proceedings of the Biological Society of Washington LITERATURE CITED Cuark, A. M. 1974. Notes on some echinoderms from southern Africa. Bull. Brit. Mus. Nat. Hist. (Zool.) 26(6):421—487, 16 text-figs., 3 pls. Ciark, H. L. 1915. Catalogue of recent ophiurans based on the col- lection of the Museum of Comparative Zoology. Mem. Mus. Comp. Zool. Harvard 25(4):165—-376, pls. 1-20. —. 1939. Ophiuroidea. Scient. Rep. John Murray Exped. 6:29- 136, 62 figs. —. 1941. Reports on the scientific results of the Atlantis expedi- tions to the West Indies, under the joint auspices of the Uni- versity of Havana and Harvard University. The echinoderms (other than holothurians). Mem. Soc. Cubana Hist. Nat. 15(1):1-154, pls. 1-10. DEVANEY, D. M. 1974. Shallow water asterozoans of southeastern Polynesia II]. Ophiuroidea. Micronesia 10(1):105-204. FELL, H. B. 1960. Synoptic keys to the genera of Ophiuroidea. Zool. Publs. Victoria Univ. Wellington (26):1—44, 6 figs. 1961. New genera and species of Ophiuroidea from Antarc- tica. Trans. Roy. Soc. N.Z. 88(4):839-841, 1 text-fig. KoErHLEerR, R. 1896. Note préliminaire sur les Ophiures recueillies pendant les campagnes de la “Princesse Alice.” Mem. Soc. Zool. France 9:241-253. —. 1906. Description des Ophiures nouvelles recueillies par le “Travailleur” et le “Talisman.” Mem. Soc. Zool. France 19: 5-35, pls. 1-3. ——. 1922. Ophiurans of the Philippine Seas and adjacent waters. Bull. U.S. Natl. Mus. 100, 5:1-186, pls. 1-103. ——. 1930. Ophiures recueillis par le docteur Th. Mortensen dans les mers d’Australie et dans l’Archipe] Malais. Vidensk. Medd. Naturhist. Foren. Kjobenhavn 89:1-295, 22 pls. Lyuneman, A. 1871. Forteckning 6fver uti Vestindien af Dr. A. Goes samt under Korvetten Josefinas expedition i Atlantiska Oceanen smalade Ophiurider. Ofversigt K. Vet. Akad. Forh. 28(6) :615-658. LyMAN, Tu. 1869. Preliminary report on the Ophiuridae and As- trophytidae dredged in deep water between Cuba and the Florida reef. . . Bull. Mus. Comp. Zool. Harvard 1( 10) :309-354. —. 1880. A structural feature, hitherto unknown among Echi- nodermata, found in deep-sea ophiurans. Anniv. Mem. Boston Soc. Nat. Hist.:1-12, pls. 1, 2. —. 1883. Reports on the results of dredging, under the super- vision of Alexander Agassiz, in the Caribbean Sea (1878-79), and on the east coast of the United States, during the summer of 1880, by the U.S. Coast Survey Steamer “BLAKE,” .. . Brittle star genus Amphilimna 139 XX.—Report on the Ophiuroidea. Bull. Mus. Comp. Zool. Harvard 10(6):227—-287, pls. 1-8. MapseEn, F. J. 1970. West African Ophiuroids. Atlantide Rpt. No. 11: 151-243, text-figs. 1-49. Matsumoto, H. 1917. A monograph of Japanese Ophiuroidea ar- ranged according to a new classification. Jour. Coll. Sci. Im- perial Univ. Tokyo 38(2):1—408, figs. 1-100, pls. 1-7. MULLER, J. AND TROSCHEL, F. H. 1842. System der Asteriden. Braun- schweig, Friedrich Viewig und Sohn. xviii + 125 pp., pls. 1-12. ScHOENER, A. 1967. Two new species of Amphitarsus (Ophiuroidea) from the western North Atlantic. Breviora No. 269: 1-9, figs. 1-3. Tuomas, L. P. 1967. The systematic position of Amphilimna (Echi- nodermata; Ophiuroidea). Proc. Biol. Soc. Washington 80:123- 130, figs. 1-9. THomas, L. P. anD SCHOENER, A. 1972. Growth changes in Amphi- limna olivacea (Lyman) and the systematic status of Amphi- tarsus spinifer Schoener. Breviora No. 387: 1-9, figs. 1, 2. Tommasi, L. 1970. Os ofiurdides recentes do Brasil e de regides Uizinhas. Contrcdes Inst. Oceanogr. Univ. Sao Paulo, Ser. Oceanogr. Biol. No. 20:1-146, pls. 1-46. VERRILL, A. E. 1899. North American Ophiuroidea. I. Revision of certain families and genera of West Indian ophiurans. II. A. faunal catalogue of the known species of West Indian ophiurans. Trans. Connecticut Acad. Arts. Sci. 10(7):301-385. 140 Proceedings of the Biological Society of Washington No. 15, pp. 141-158 15 August 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON TWO NEW SPECIES OF BATHYCONCHOECIA (MYODOCOPA HALOCYPRIDIDAE) FROM THE CARIBBEAN SEA = By GrorciANA B, DEEVEY \¢ Florida State Museum, University of Florida Gainesville, Florida 32611 The genus Bathyconchoecia Deevey 1968 was established to include 6 new species of epibenthic pelagic ostracods from the Gulf of Mexico and 2 species previously assigned to the genus Euconchoecia. Since then 7 more species have been described (Poulsen, 1969a, 1969b, 1972; Kornicker, 1969; Angel, 1970). Members of this genus apparently live just over the bottom, in regions difficult to sample with plankton nets, and have been caught over a known depth range of 130 to 3165 m. As noted by Angel (1973), some of the species seem to be more bathypelagic than epibenthic. Poulsen (1972) has divided the known species into four more or less natural groups. B. deeveyae Kornicker and B. septemspinosa Angel differ from all other species in having long dorsal, lat- eral, and rostral spines, as well as a long spine at the postero- dorsal corner of the right shell. B. sagittarius Deevey and B. latirostris Poulsen are distinguished by the high numbers of setae on the maxilla and the fifth and sixth limbs. A third group includes the small species less than 2 mm long, B. paulula Deevey, B. laqueata Deevey, B. kornickeri Deevey, B. lacunosa (Miller), B. galerita Deevey, and B. nodosa Poulsen, in which the shell is usually strikingly sculptured and the height of the shell is 60-70% of the length. Included in the fourth group are larger species 3.5-7 mm long: B. baskiae Poulsen, B. foveolata Deevey, B. darcythompsoni (Scott), B. crosnieri Poulsen, and B. subrufa Angel; in these species the height of the shell is 40-60% of the length and the 15—Proc. Brow. Soc. WAsH., Vou. 88, 1975 (141) 142 Proceedings of the Biological Society of Washington sculpturing on the shell is not as striking as in most of the smaller species. Three specimens of Bathyconchoecia were found in samples collected by Dr. Harding B. Owre Michel on PILusBury Cruises in the Caribbean Sea. Two of these, a mature male 4 mm long and a juvenile specimen 1.25 mm long, are new species; the third specimen is considered the immature male of a species described from the Gulf of Mexico. All three specimens belong in the fourth group of larger species. The station locations in the Caribbean Sea and Gulf of Mexico where 8 species of Bathyconchoecia have now been found are shown in Fig. 1. I am deeply indebted to Dr. Michel for permission to examine the Caribbean samples. This work was partially supported by grant GA-36512 from the National Science Foundation. Poulsen’s (1972) key to 15 species has been changed to in- clude the two new species. As Poulsen noted, the key is pro- visional since most of the species are known from only one or a few specimens, or from only one sex or a juvenile specimen. Sexual dimorphism is not marked in this genus. Aside from the male claspers on the endopodite of the second antenna and the male copulatory organ, the other differences noted may be in the relative height of the shell and in the length of the long dorsal seta on the first exopodite segment of the fifth and sixth limbs; in the known cases this is longer in the female than in the male. The present data indicate that the males are slightly larger than the females. Male and female specimens of the same species have been described by the same author for B. paulula, B. subrufa, B. crosnieri and B. darcythompsoni. In the latter species, ac- cording to Scott (1909), the male rostrum is bifid and the fe- male’s single, so the sexes must be separated in the key. Only mature males are known of B. laqueata, B. kornickeri, B. foveolata, B. baskiae and B. hardingae n. sp., one of the new species here described. In the case of B. deeveyae the im- mature female (Kornicker, 1969) and the male (George, 1971) are known. Only the females of B. galerita and B. latirostris, and the juvenile females of B. septemspinosa, B. lacunosa and B. nodosa have been described. The other new Two new ostracods from Caribbean 143 species now described, B. diacantha n. sp., is represented by a juvenile specimen. In the case of B. sagittarius, my male specimen from the intestine of Bassozetus normalis, an abyssal bottom fish, was crumpled into a ball and part of the posterior margin of the shell was damaged, but the posterodorsal corner of the right shell seemed to be produced in a blunt point. Both Poulsen (1972) and Angel (1973) have described the female of this species, but found no point at the posterodorsal corner. There are other discrepancies. The male was characterized by a chitinous knob on the coxa of the mandible; Angel did not mention this in his description of the female, but Poulsen noted it in his. On the first antenna Angel’s females had a ventral spinose seta on the segment proximal to the one bear- ing the cluster of sensory filaments; such a seta has not been reported for any other Bathyconchoecia species, although it occurs on the two new species here described. It was not noted on the male, and unfortunately the first antennae were missing from Poulsen’s specimen. Also, the male had 16 setae on the Ist segment of the endopodite of the maxilla, as ap- parently did Poulsen’s female specimen, but Angel reported 22 setae on this segment in his females. The three descriptions agree on the sculpturing on the shell and the extraordinary number of setae on the maxillae and fifth and sixth limbs, but the possibility exists that the specimens described as B. sagittarius may include more than one species. Key To SpreciES OF BATHYCONCHOECIA 1. Penultimate segment of 5th limb with 7-13 setae, of 6th limb Sth OSC LAG gael ceae stacy EN $0 MCRL rN RR MERE ed Ee AE 2 Penultimate segment of 5th limb with 3-5 setae, of 6th limb NWVACIN MEO eS SCL AC a air AM ies ed i Le 3) 2. Shell with dense vertical striations. Mature specimens over OREM IITINO LO TIS per se eee ae cae Be ae B. sagittarius 94+ 4 Shell with polygons or bands filled with pits; mature speci- mens less than 2.5 mm long __....... B. latirostris 9@ 3. Shell with long rostral, lateral, dorsal and posterior spines _. 4 Shell without dorsal, lateral or rostral spines — 5 4. Shell surface with short fringes of fine hairs ___________________-___- OEE, SUE Eo SOON. POE ea B. septemspinosa imm. @ Shell surface without fringes of fine hairs _.... SE: SS Se ee eee ee ee 2 1B, deenenae imm. 9+ 4 144 Proceedings of the Biological Society of Washington 5. Posterodorsal corners of both shells with 2 small sharp points, larger on nightyshell) 22s ae eee B. diacantha n. sp., juv. Right posterodorsal shell corner with a point or spine ____________ 6 Posterodorsal shell corners without points or spines ______________ 10 Gs) Hach® halt of rostrum: bifidy B. darcythompsoni ¢ Bach shalfvof rostrum: single). 32). 40 eee 7 7. Posterior shell glands open level with edge of shell —_____. Sts Shell glands open on tubercle below posterodorsal corner of bothiishells, 2).95 2) De ed ee 9 8. Posterodorsal corner of right shell with a wide strong point; strong chitinized triangular process on coxa of mandible; penultimate segment of 5th limb with 4 setae __.-_-- NCA A iia Mia ctv atte | tee «deals at sehen B. hardingae n. sp. ¢ Posterodorsal corner of right shell with a small point; no tri- angular process on coxa of mandible; penultimate segment Ove iday Waal oy Sido) Wetas B. darcythompsoni @ 9. First segment of lst antenna with large ventrodistal bulge __ EASE DS WS Ag) eM Sn Ra UE Na Woe Mean DO ae SENSE B. crosnieri 9+ ¢6 First segment of lst antenna without bulge ____. B. subrufa 2+¢6 10. Length of mature individuals over 4 mm __--.---- 11 Length of mature individuals less than 4 mm _...- 12 11. Posterior shell margin fairly straight with a single row of dentate structures extending anterior to the posteroventral CORTOE Merete orate ENR SNS ME O98 RS a B. foveolata 8 Posteroventral corner well rounded; no marginal dentate struc- UETGS) fread AT tae ee eh le acide erode B. baskiae 8 12. Length of mature individuals less than 1 mm; height 70% of Vn eee ec nn, Ka Na oan nt Ree B. paulula 9+ 6 Length of mature individuals 1.0-1.55 mm ___......._-_ 13 Length of mature individuals over 1.5 mm __. 14 13. Shell glands open on large rounded processes at posterodorsal comers obvedchi shell: ian eae ea ae B. nodosa imm. Posterodorsal shell corners sharply right-angled __ B. laqueata 3 14. Shell strikingly sculptured with reticulations forming po- Iyzonallicells) filled’ wath tiny. pits 222 ee 15 Shell not strikingly sculptured though punctate pattern may bebepresentt). 2 lkork hy. Pike: Whee ses ied) oe Aa B. galerita @ 15. Flange of sculpturing on posterior margin beneath shell glands Eo Le ae | A Bt ee UE aN, AR, eer B. kornickeri No flange of sculpturing on posterior margin __ B. lacunosa imm. @ Bathyconehoecia hardingae new species Figures 2-4 Holotype: Male, 4.0 mm long by 1.6 mm high. Three slides, de- posited in the National Museum of Natural History, Smithsonian Institu- tion (USNM 152437). Two new ostracods from Caribbean 145 85° 80° 75° 70° 65° Ww 1 GULF x5 OF MEXICO x4 | =) ATLANTIC OCEAN ees 20° JAMAICA SSHISPANIOLA) PUERTO RICO d SS C25 i CARIBBEAN SEA 15° 10° SOUTH AMERICA =i = 1 Fic. 1. Station locations for species of Bathyconchoecia described from the Caribbean Sea and Gulf of Mexico. X 1: B. hardingae, n. sp.; X 2: B. foveolata, imm. male; X 3: B. diacantha n. sp. juvenile; X 4: B. sagittarius male, B. foveolata male; X 5: B. paulula female + male; B. laqueata male; B. kornickeri male, B. galerita female. Type-Locality: Pituspury Cruise 6911, Sta. 6: 13°30’N, 62°50’W, 1324 m depth; bottom depth 1656 m. See Fig. 1: X 1. Etymology: This species is named for Dr. Harding B. Owre Michel. Description of male: Shell (Fig. 2a): The height of the shell is 40% of the length, anteroventral and posteroventral corners well rounded, dorsal margin nearly straight, large point at posterodorsal corner of right shell, posterodorsal corner of left shell broken off. Rostrum pointed on both shells, and projects about 20% of the shell length beyond the anterior margin. Shell gland opens beneath point on postero- dorsal corner of right shell. Posterior margin of shell with a single row of dentate structures running from the shell gland and decreasing in size to anterior of the posteroventral corner. Faint scalelike sculpturing on the rostrum and anterior dorsal region, with vertical and oblique lines more visible ventrally, as shown in Fig. 2a. First antenna (Fig. 3a): This consists of 5-6 visible segments, al- though Poulsen (1969a) suggested that 7 joints are represented but not clearly defined. The most distal joint bears 4 setae, the principal seta exceptionally long with fine spinules dorsally and distally, broken off as figured. The 6th segment bears a stout plumose seta. Accord- 146 Proceedings of the Biological Society of Washington Fic. 2. Male Bathyconchoecia hardingae, n. sp. a, Lateral view of right shell; b, Right clasping organ; c, Distal segment of endopodite of left second antenna, setae cut off; d, Endopodite of left second an- tenna, longer setae cut off; e, Endopodite of maxilla; f, Coxal and pre- coxal endites of maxilla; g, Sixth limb. Scale for a at bottom right, for g at bottom left, for d and e on upper left margin, for c and f at left center. Scales in mm. ing to Poulsen, the 5th segment bears the cluster of sensory filaments, which contains perhaps 15 rows of around 25 filaments per row, and there are therefore probably 300—400 filaments in the cluster. On the joint above the cluster there is a group of long hairs and a line of Two new ostracods from Caribbean 147 shorter ones. Each first antenna has a strong seta on the ventral surface of the segment proximal to the sensory cluster, with a group of long hairs distal to it. Angel (1973) described a similar ventral seta and cluster of hairs for his female B. sagittarius, but such a seta has not been noted in any other species. The first 2 segments have no hairs or setae, but the Ist segment has a disto-ventral bulge. No remnant of a frontal organ was observed. Second antenna (Fig. 2b-d): The length of the basal segment of the exopodite is around 63% of that of the shaft, the Ist segment of the endopodite around 43% of shaft length. The basal segment of the endopodite bears 2 plumose bristles distally, with long hairs proximally; the longer bristle was broken on both endopodites. The 2nd segment has distally 2 long setae with spinules at regular intervals, the longest broken at 3 mm length, the shorter 2 mm long, with a single spine proximally to their base. The right clasping organ (Fig. 2b) is large and sharply curved with 3 non-plumose setae of varying length, the longest broken at 0.8 mm, the 2nd 0.45 mm long, and the shortest only 0.25 mm long. The left clasping organ is much smaller (Fig. 2c, d) and bent back over its base, also with 3 setae of similarly varying lengths. The 2nd segment also has 2 short fat spinules near the base of the claspers. Mandible (Fig. 3b, c; Fig. 4b): There is a strongly chitinized knob- like articular process on the coxa (Fig. 3b, c), as in B. sagittarius and B. latirostris, which has not been reported for other Bathyconchoecia but is common in Conchoecia and Euconchoecia species. The coxal endite has several tooth rows, with serrations of varying size; the masticatory pad consists of rows of denticles and bristling spines. The basale has 6 teeth and 2 spine-teeth. As in most other species there are 3 plumose setae on the basale near the articulation of the endopodite. The Ist segment of the endopodite has 5 setae, 1 plumose, and many fine long hairs; the 2nd segment 4 setae, and the last 6 setae and a group of long hairs near the articulation with the 2nd segment. Maxilla (Fig. 2e, £): The endopodite (Fig. 2e) is similar to that of other species, except B. sagittarius and B. latirostris; the 1st segment has 10 plumose setae and the distal segment 2 claws and 4 setae. The precoxal endite (Fig. 2f) has 4 stout rather long bristles and 6 tubelike bristles; the coxal endite has some 14-17 bristles, including several tubelike ones. Fifth limb (Fig. 4a): The protopodite has 3 plumose and 3 bare setae, the endopodite 3 plumose and 5 bare setae and 2 claws. The Ist segment of the exopodite has 5 plumose and 4 bare setae, as well as the long dorsal seta characteristic of this genus; it appears partially segmented at the point where the long dorsal seta arises. The 2nd segment has 4 bare setae, and the last segment 2 clawlike setae and a bare seta. Sixth limb (Fig. 2g): The protopodite has 6 plumose setae, the Ist segment of the exopodite 6 plumose setae, 1 small bare seta, and the 148 Proceedings of the Biological Society of Washington Fic. 3. Bathyconchoecia hardingae n. sp., male. a, First antenna; b, Incisor and molar surfaces of coxa of mandible; c, Incisor surface of basale and another view of incisor and molar surfaces of coxa of mandible. Scale on center right margin for a, at bottom right for b and c. Scales in mm. long dorsal seta, broken on this specimen. The 2nd segment has 4 bare setae, the 3rd segment 3 bare setae, and the 4th 2 long slim clawlike setae and a bare seta. Seventh limb (Fig. 4e): This bears distally 2 bristles, 1 at least twice as long as the other. At least near the proximal end of the larger bristle there is a very fine double comb of tiny spinules. Two new ostracods from Caribbean 149 Fic. 4. Bathyconchoecia hardingae, n. sp., male. a, Fifth limb; b, Endopodite of mandible; c, Furca; d, Penis; e, Seventh limb. Scale in mm. Penis (Fig. 4d): This is long, straight and slim, bluntly rounded at the tip. Furca (Fig. 4c): As in all other described species except B. deeveyae there are 8 pairs of claws on this mature male. The Ist is 5-jointed, and all are very long, slim and weak. There is also a long slim unpaired bristle. The furcal lamella is partially covered with fine hairs. 150 Proceedings of the Biological Society of Washington Remarks: B. hardingae n. sp. differs from all other species except B. sagittarius and B. latirostris in having a highly chitinized knoblike articulation on the coxa of the mandible, but it is easily differentiated from these species by the size and shape of the shell, and by having only 3 or 4 setae on the penultimate segment of the 5th and 6th limbs. It is also distinguished from all species except Angel’s (1973) female B. sagittarius by the long ventral seta on the segment proximal to the one bearing the large cluster of sensory filaments on the Ist antenna. Bathyconchoecia diacantha new species Figures 5-6 Holotype: juvenile specimen 1.25 mm long by 0.5 mm high. Two slides, deposited in the National Museum of Natural History, Smith- sonian Institution (USNM 152438). Type locality: Putspury Cruise 6811, Sta. 12: 14°50’N, 80°45’W, 1100 m depth; bottom depth 1189 m. Figure 1: x 3. Etymology: The specific name is derived from the Greek ”acantha,” meaning spine or thorn, and refers to the 2 small points at the postero- dorsal corner of both shells. Description: Shell (Fig. 5a-c): The height of the shell is about 40% of the length, the anteroventral and posteroventral corners well rounded, anterior and posterior margins rounded, dorsal margin fairly straight. The rostrum is slim and pointed in lateral view, and projects 20% of the shell length beyond the anterior margin. There are 2 small sharp points at the posterodorsal corner of each shell, larger on the right shell (Fig. 5c). The shell glands are located beneath the posterodorsal points, and beneath these there is a single row of serrations or dentate structures extending to just anterior to the posteroventral corner. The sculpturing on the shell consists of faint vertical lines, as indicated in Figure 5b. First antenna (Fig. 5d): In this juvenile specimen the penultimate segment bears the stout plumose seta and the last segment 4 setae, the principal seta the longest with a few spinules on both sides near the distal end, the total number of setae therefore being the same as in ma- ture individuals. The cluster of sensory filaments is not as large as in the described mature specimens, possibly no more than 100 filaments being present. Only 1 of the 2 Ist antennae bear ventrally on what may become the segment proximal to the sensory cluster a short seta, and distal to this a bunch of long hairs. No trace of this seta was found on the other Ist antenna. Presumably this seta is comparable to the ventral seta just described for B. hardingae n. sp. and Angel’s female B. sagittarius, and it is probable that the mature individual will have a long ventral seta on each Ist antenna. No remnant of a frontal organ was noted. Second antenna (Fig. 6e): The Ist segment of the exopodite is around 55% of shaft length, the basal segment of the endopodite about Two new ostracods from Caribbean 151 Fic. 5. Bathyconchoecia diacantha, n. sp., juvenile. a and b, Lateral and dorsal view of shell; c, Dorsal view of posterodorsal corner of both shells opened out; d, First antenna; e, Fifth limb. Scale below b for a and b, below c for c, at upper right for d and e. Scales in mm. 35% of shaft length. The Ist segment of the endopodite has only 1 long spinose bristle, and the 2nd segment appeared to bear only 3 setae, the longest broken, with some spinules distally. Mandible (Fig. 6a, b): The basale has 6 teeth and 2 spine-teeth, the surface is covered with long hairs, and there are 4 long and 1 short setae, aside from 2 setae near the base of the endopodite, 1 of which is 152 Proceedings of the Biological Society of Washington WY Ui Fic. 6. Bathyconchoecia diacantha, n. sp., juvenile. a, Basale with endopodite and molar and incisor surface of coxa of mandible; b, Another view of incisor surface of basale; c, Sixth limb; d, Coxal and precoxal endites of maxilla; e, Endopodite of second antenna; f, Seventh limb; g, Endopodite of maxilla; h, Furca. Scale in mm. plumose. The endopodite has 1 seta on the Ist segment, 2 on the 2nd, and 6 spinous setae on the last segment. There is a small, strongly chitinized knoblike process on the coxa, 2 strong teeth rows, and the masticatory pad consists of rows of denticles and spinules. Two new ostracods from Caribbean 153 Maxilla (Fig. 6d, g): The Ist segment of the endopodite has 7 plumose setae, the distal segment 2 claws and only 3 bare setae. The precoxal endite has 7 bristles, of which one has long spines and 3 are tubelike; the coxal endites have a total of 14 bristles. These are relatively long and slim in shape, as in the group of larger species. Fifth limb (Fig. 5e): The protopodite has 2 plumose and 1 short bare seta, and is not clearly separated from the endopodite, which has 2 claws, 5 plumose and 4 bare setae. The exopodite has 2 slim claw setae on the distal segment, 3 bare setae on the 2nd segment, and 2 plumose, 4 bare setae, and the very long dorsal seta on the Ist segment. There are tufts of hairs on the protopodite, endopodite and the proximal ventral surface of the exopodite. - Sixth limb (Fig. 6c): This was segmented into only 3 distinct joints, the most distal with 2 very long slim claw-setae, the 2nd exopodite segment with 2 bare setae; the basal segment of the exopodite, not clearly separated from the protopodite, has only the long dorsal seta, broken on this specimen, and 2 setae, at least 1 plumose. The pro- topodite bears only 2 plumose setae and a spine. The epipodial ap- pendage has 3 groups of 6, 5, 5 setae each. Seventh limb (Fig. 6f): This has 2 relatively short setae, 1 much larger than the other, both bearing a row of hairs. Furca (Fig. 6h): This has 5 claws on each lamella and a single long unpaired bristle. The Ist claw is at least 3-jointed, and there are long hairs on the lamellae. Remarks: With 5 pairs of claws on the furca this specimen is 3 moults from maturity. Angel (1970) has described the juvenile development of B. subrufa on the basis of 11 females, 3 males and 87 juveniles at 4 dif- ferent stages of development. The males were 3.48-3.515 mm and the females 3.16-3.35 mm long. This 1.25-mm juvenile is at the same stage of development as the 1.0-mm B. subrufa juveniles, and should therefore be around 4 mm long when mature. Angel found that the adult seta- tion of the Ist antenna was present at this stage, but did not note whether the number of sensory filaments in the cluster changed during development. He also noted that the Ist segment of the endopodite of the 2nd antenna had 1 distal bristle at this stage, and 2 bristles were present in the last 2 juvenile stages. In B. subrufa the tooth lists of the mandible were similar at all stages, but the number of setae on on the basale and Ist 2 endopodite segments increased with growth; his 1.0-mm juveniles had the adult setation on the 3rd segment, as does this 1.25-mm juvenile. The number of setae on the Ist endopodite segment of the maxilla increased during development, but the distal segment had the adult setation in Angel’s earliest stage; the present specimen either lacks or has lost 1 seta of the adult number on the distal segment, although it has more setae on the Ist segment than the B. subrufa juvenile at the same stage. The setation of the 5th and 6th limbs increases during development, but Angel found that the long 154 Proceedings of the Biological Society of Washington dorsal setae were present on the earliest stage examined. With the exception of B. deeveyae which is reported (George, 1971) to have 7, all mature Bathyconchoecia have 8 pairs of claws on the furca. The last 4 juvenile stages of B. subrufa gained 1 pair/moult. Since in B. subrufa the Ist claw is apparently not jointed, there are no data to in- dicate whether the mature B. diacantha n. sp. would have a Ist claw with more than 3 joints, but it is probable that it would. This species is distinguished from all others by the 2 sharp points at the posterodorsal corners of both shells, and belongs in the group of large species. B. diacantha n. sp. differs from all species except B. sagittarius, B. latirostris, and B. hardingae n. sp. in having a knoblike articular process on the coxa of the mandible. It is also distinguished from the other species, except B. hardingae n. sp. and Angel’s female B. sagittarius, by having a short (at least at this juvenile stage) seta on the ventral surface of the segment proximal to the sensory cluster on the Ist antenna. Bathyconchoecia foveolata Deevey Figures 7-8 Bathyconchoecia foveolata Deevey, 1968, p. 565, Figs. 12-13. Material: 1 immature male, 2.4 mm long. 3 slides. Locality: Pruspury Cruise 6911, Sta. 1: 18°00’N, 64°44’W, 1371 m depth; bottom depth 3008 m. Figure 1: x 2. Description of immature male: Shell (Fig. 7a, b): The height of the shell is around 45% of the length, anteroventral and posteroventral corners rounded, dorsal and ventral margins fairly straight, postero- dorsal corner bluntly rounded with shell glands opening in depression below posterodorsal corners. Rostrum rounded, projecting around 21% of the total length beyond the anterior margin. A faint pattern of reticulations was visible on the shell, and the posterior margin and at least half of the ventral margin are fringed with a single row of serra- tions or dentate structures, decreasing in size anteriorly (Fig. 7b). First antenna: This was missing. Second antenna (Fig. 7c, d): The Ist segment of the exopodite is 51% of shaft length, the Ist endopodite segment 30% of shaft length. On the distal endopodite segment the incipient clasping organs are represented by similar small knobs, each bearing 3 setae of varying lengths, all shorter than the 2 longer setae borne at the distal end, which were broken on both endopodites. At the distal end of the Ist endopodite segment there are 2 bristles, one more than twice as long as the other with long hairs proximally and spinules more distally. There are 2 small fat spinules on the distal endopodite segment beside the knoblike precursors of the claspers. Mandible (Fig. 7e, £): The basale has a tooth row of 6 teeth and 2 spine-teeth somewhat removed from the tooth row; it has a total of 9 setae, 4 plumose, including 3 near the base of the endopodite, which Two new ostracods from Caribbean 155 Fic. 7. Bathyconchoecia foveolata, immature male. a, Dorsal view of shell opened out; b, Posterior and ventral margins of shell, showing fringe of serrations; c, Right second antenna, expodite setae not shown; d, Endopodite of left second antenna, setae cut off; e, Basale and endopodite and part of coxa of mandible; f, Incisor surface of basale and incisor and molar surfaces of coxa of mandible. Scale for a at top center, for b and c at bottom right, at left center margin for e, at bottom center for d and f. Scales in mm. 156 Proceedings of the Biological Society of Washington Fic. 8. Bathyconchoecia foveolata, immature male. a. Endopodite of maxilla; b, Coxal and precoxal endites of maxilla; c, Furca and im- mature penis; d, Seventh limb; e, Sixth limb. Scale for a and b at center right margin, for c, d, e at lower center. Scales in mm. is similar to that of the B. foveolata male except that the Ist segment has 4 instead of 5 setae. The incisor surface of the coxa has 2 strong tooth rows with 8-9 teeth each of varying size; the molar surface is strongly rounded and covered with denticles and spines. Maxilla (Fig. 8a, b): The precoxal and coxal endites have some 6, 7, and 6 long pointed bristles of which at least 9 are tubelike, the others spiny. The endopodite has 10 plumose setae on the basal seg- ment; this specimen had 2 claws and only 3 bare setae on the distal segment, 1 seta longer than the other 2. Two new ostracods from Caribbean 157 Fifth limb: This was missing. Sixth limb (Fig. 8e): The protopodite has 5 setae; the lst segment of the exopodite has 5 plumose setae, the long dorsal seta, and 1 small spine, the 2nd segment has 4 setae with tiny spinules, the 3rd 3 setae, and the 4th the 2 long slim claw-setae and a shorter seta with tiny spinules. The epipodial appendage has groups of 7 (1 tiny), 5, 5 setae. Seventh limb (Fig. 8d): This has 2 setae, 1 longer with tiny spinules. Penis and furca (Fig. 8c): The shape of the immature penis is shown in Figure 8e. The furca has 7 pairs of claws, the Ist 4-jointed, and a long unpaired bristle. The lamella is covered with fine hairs. The Ist claw of the male B. foveolata furca is 5-jointed. Remarks: This specimen is identified as B. foveolata because the shape of the shell, its sculpturing and the fringe of serrations on the posterior and ventral margin agree with that of this species. Judging from the growth data obtained by Angel (1970) for B. swbrufa a 2.4-mm immature male should be 3.5-4 mm long when mature. The described B. foveolata male was 5.2 mm long. The structure of the mandible appears similar to that of B. foveolata; the coxa also lacks the strongly chitinized process. The setation of the maxilla could be that of the immature male. The setation of the 6th limb is similar, as are the the relative lengths of the claws on the furca. This specimen has the characteristics of the group of large species: long hairs proximally on the larger bristle of the Ist segment of the endopodite of the 2nd antenna, a shell height of around 45% of its length, and coxal and pre- coxal endites with long slim bristles. LITERATURE CITED ANGEL, M. V. 1970. Bathyconchoecia subrufa n. sp. and B. septem- spinosa n. sp., two new Halocyprids (Ostracods, Myodocopida ) from the tropical North Atlantic and the description of the larval development of B. swbrufa. Crustaceana 19:181—199. ———. 1973. The description of the female of the ostracod Bathy- conchoecia sagittarius Deevey, 1968 (Myodocopida, Halo- cyprididae). Crustaceana 25:211-219. Dervey, G. B. 1968. Bathyconchoecia, a new genus of pelagic ostracods (Myodocopa Halocyprididae) with six new species from the deeper waters of the Gulf of Mexico. Proc. Biol. Soc. Wash. 81:539-570. GrorcE, J. 1971. On the occurrence of Bathyconchoecia deeveyae Kornicker (Ostracoda, Halocyprididae) in the Indian Ocean. Crustaceana 21:141-144. KornicxEr, L. S. 1969. Bathyconchoecia deeveyae, a highly orna- mented new species of Ostracoda (Halocyprididae) from the Peru-Chile Trench System. Proc. Biol. Soc. Wash. 82:403-408. 158 Proceedings of the Biological Society of Washington PoutsEN, E. M. 1969a. Ostracoda-Myodocopa, 1. Halocypriformes Thaumatocypridae. Dana-Report No. 75, 99 p. 40 figs. — —. 1969b. Ostracoda-Myodocopa from the eastern tropical At- lantic. Vidensk. Medd. Dansk Naturhist. Foren. Kjobenhavn 132:129-197. ——. 1972. On the Bathyconchoecia (Ostracoda Myodocopa) from the Azores collected by the Bathyscaph “Archimede” in 1969. Tethys 4:445—455. Scott, T. 1909. On some new and rare Entomostraca from the Scottish Seas. Ann. Mag. Nat. Hist. Ser. 8, 3:122—130, pls. 2-4. av (ee 7 Ch if Ht wf ? a iy d a ¥ r at a | ond Neri i 4 ‘i q ei, , th ) Gd, ii i i" yt F f i ‘ sey | ot 4 rehvas ey, a 1 , § uy Dek ay hay iy ' ‘ a] ’ : | 1%) y 4 f tw q : ft ‘ AY : rl J , ty a i ” , t u f Ms i ; id i ij 1 ) sy mg ay \ ‘ 4 14 eye ME, 4 4 C > hi A = ra 7) ‘ q ; fi i : T i, = A ia va a " ‘ 1 L . ki AT ie he No. 16, pp. 159-166 15 August 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON NOTES ON MEMBERS OF THE FAMILY PORCELLANIDAE (CRUSTACEA: ANOMURA) COLLECTED ON THE EAST COAST OF MEXICO By Jack A. RICKNER Texas A&I University, Kingsville, Texas 78363 In her review of the decapod crustaceans of the Gulf of Mexico, Behre (1954) stated “. . . examination of our records of the decapod Crustacea of the Gulf of Mexico indicates very clearly large gaps in our knowledge. These gaps are less evident along the northern shore and the Florida Keys than along the eastern part of Mexico and Yucatan.” In the subse- quent twenty years, little has been accomplished on the family Porcellanidae along the eastern coast of Mexico. Ives (1891) reported 21 decapods from Yucatan and 7 from Veracruz, but his list did not include porcellanids. Hildebrand (1954) listed only 1 porcellanid, Porcellana sayana, from catches by shrimp trawls on the brown shrimp grounds of the western Gulf of Mexico. Chavez, Hidalgo and Sevilla (1970) in their study of the lagoonal area of Lobos Reef listed no porcellanids among 22 species of crabs collected. My report will serve only as a preliminary survey to fill in a part of the gap in our knowledge. DESCRIPTION OF STUDY AREA Two study areas on the East Coast of Mexico were exten- sively sampled during late May and June, 1973: Lobos and Enmedio Reefs; a third area, El Moro, was only briefly sam- pled. Lobos and Enmedio Reefs are emergent, nearshore coral reefs with small associated islands. Lobos, located 120 km southeast of Tampico (21°27’N, 97°13’W), is one of a cluster of 3 reefs which comprises the northernmost emer- 16—Proc. Brow. Soc. WAsH., Vou. 88, 1975 (159) j { | a 1\ \\ fe \\ 160 Proceedings of the Biological Society of Washington gent coral reefs in the western Gulf. It is 9.6 km offshore Cabo Rojo in 28 m of water, is approximately 2930 m long and oriented in a north-south direction (Rigby and McIntire, 1966). It is situated near the boundary between 2 faunal provinces, tropical to the south and warm-temperate to the north (Briggs, 1974:218-221). Enmedio Reef is located 24 km southeast of Veracruz, (19°06’N, 95°56’W ), 6 km off Anton Lizardo in approximately 25 m of water. It is associated with the Veracruz-Anton Lizardo Reef complex. Enmedio is approximately 1500 m long and 850 m wide at its surface, oriented in a northwest- southeast direction (Tunnell, 1974). El Moro is an area of lava cliffs which project into the otherwise smooth, continuous sand beaches 80 km north of Veracruz (19°51’N, 96°24’W). These cliffs offer a unique habitat for many organisms which might otherwise be absent from this area. MATERIAL AND METHODS All collections of porcellanids on Lobos (30 May-10 June) and Enmedio (14-22 June) were made during extensive daylight searches using SCUBA, a snorkel and mask, or while wading. On Lobos, the coral boulder ridge and coral spoil islands along a small boat channel which provided suitable habitats for some porcellanids were examined while wading. The lagoon, back reefs, fore reefs, and reef crests were ex- amined using a mask and snorkel. Deeper areas of the reef were superficially sampled using SCUBA. By wading on an exposed reef crest and the coral beach during low tides, sam- ples were taken on Enmedio. The other areas were sampled in the same manner as the corresponding areas on Lobos. Because of poor diving conditions, El] Moro was sampled only by wading in the tide pools and climbing over rock cliffs in the area of the splash zone. In this splash zone sabellarid worms, chiefly Gunnarea sp., had built thick mats of tubes composed of sand grains. These tubes were constructed in such a way as to leave a chambered system in which the crabs were found. Five gallons of these tubes were collected and later examined for porcellanids. Porcellanid crabs—Mexican east coast 161 The crabs were placed in 10% formalin, labeled as to date and location, and placed in buckets for transport to the labora- tory. There they were identified with the aid of papers by Haig (1956 and 1960) and several were verified by Miss Haig. At least one specimen of each species was placed in the Texas A&I Crustacean Collection (TAIC) and several species unreported from the southwestern Gulf of Mexico were placed in the Allan Hancock Foundation Collection. ACKNOWLEDGMENTS I wish to express my thanks and sincere appreciation to Dr. J. W. Tunnell who made this project possible by furnishing transportation and most of the equipment used during this trip as well as valuable assistance. To Miss Janet Haig, Cu- rator of Crustacea, Allan Hancock Foundation, my apprecia- tion is given for help with the verification and identification of several specimens and review of the manuscript. For the loan of equipment during the field work I am indebted to Dr. R. B. Davis and the Caesar Kleberg Wildlife Foundation, Department of Biology, Texas A&I University. I am grateful to Lie. Rafael Vega Rivas, Director General de Regiones Pesqueras, Subsecretaria de Pesca, Secretaria de Industria y Comercio for issuing a collection permit to work in Mexico and to Ing. Graciano Bello, Pemex superintendent, Cerro Azul, for extending permission to be transported via Pemex crew boat. I also wish to express my thanks and appreciation to Dr. A. H. Chaney for his review of the manuscript and the encourage- ment he offered in the completion of this report. ACCOUNT OF SPECIES Haig (1956) listed 9 Western Atlantic genera of porcel- lanids. A subsequent revision (Haig, 1960) enlarged the num- ber to 10 through the creation of Neopisosoma and Clasto- toechus and the synonymizing of Pisosoma. Five of these genera containing 10 species were collected during this in- vestigation. Seven of these species have not previously been reported from the southwestern Gulf of Mexico and another has been reported only from Veracruz. 162 Proceedings of the Biological Society of Washington Pachycheles Stimpson, 1858 Two species of this genus were collected, both from El Moro. Pachycheles ackleianus A. Milne Edwards, 1880 Material examined: TAIC-380. El Moro. 1¢. Measurements: Male: length 3.1 mm, width 3.8 mm. Remarks: This species is known to occur in coral heads and sponges (Haig, 1956) and to a depth of 68 m (Schmitt, 1935). This individual was collected above the water level in the lower splash zone. In addi- tion to previous records from the West Indies, Florida, and Jolbos Islands (Haig, 1956), Gore (1974) reported the species from the southwestern Caribbean and Pernambuco, Brazil. Pachycheles monilifer (Dana, 1852) Material examined: TAIC-379. El Moro. 1¢. Measurements: Male: length 6.45 mm, width 6.55 mm. Remarks: To my knowledge this is the first published record of the species in the southwestern Gulf of Mexico. Two individuals have been collected in Florida; there is a single record from Isla Contoy, Mexico, several from the Virgin Islands, Venezuela and Brazil (Haig, 1962), and one from Trinidad (Gore, 1974). This species has been taken in the littoral zone and to depths of 33 m (Haig, 1956). Porcellana Lamarck, 1801 One species of this genus was collected. Porcellana sayana (Leach, 1820) Material examined: TAIC-429 (2). Enmedio Reef. 16, 1 (juv). TAIC-434,. Enmedio Reef. 19 (ov). Measurements: Male: length 7.7 mm, width 7.3 mm; ovigerous fe- male: length 8.9 mm, width 8.2 mm; juvenile: length 4.3 mm, width 3.95 mm. Remarks: One individual was taken from under coral rubble of the partly exposed lee reef crest during low tide. A second individual was taken from under a coral head in the lagoon in approximately 1.5 m of water. All were free living; however, Hildebrand (1954) found speci- mens attached to the decorator crab Stenocionops furcata, and P. sayana is known to be commensal with hermit crabs (Gore, 1974). This species occurs from Cape Hatteras, North Carolina to Florida, around the Gulf of Mexico and the Caribbean, south to Estado Rio Grande do Sul, Brazil (Gore, 1974). Porcellana sayana is known from shallow water to depths of 92 m and perhaps 713 m [?] (Gore, 1974). Clastotoechus Haig, 1960 This genus contains three species; one of the two that occur in the western Atlantic is represented in this collection. Porcellanid crabs—Mexican east coast 163 Clastotoechus nodosus (Streets, 1872) Material examined: TAIC-381 (14). El Moro. 56 6, 992 (8 ov). Measurements: Males: length 3.45 to 6.55 mm, width 3.0 to 6.15 mm; non-ovigerous female: length 4.8 mm, width 4.65 mm; ovigerous females: length 3.4 to 6.65 mm, width 3.1 to 6.3 mm. Remarks: Previously known from St. Martin and Curacao, Lesser Antilles, and Venezuela (Haig, 1956), its occurrence at El Moro, Mexico, extends its known range northward. Haig (1956) reported it from the littoral, among stones, and noted that it appeared to be an exceptionally small species, 2.6 mm?, 2.5 mmé. The individuals col- lected from El Moro are somewhat larger than those in previous reports. Neopisosoma Haig, 1960 There are five species in this genus, with two, N. angustifrons (Benedict, 1901) and N. curacaoense (Schmitt, 1924), occurring in the western Atlantic. Neopisosoma angustifrons (Benedict, 1901) Material examined: TAIC-433 (18). El Moro. 24 6, 169 2 (10 ov). Measurements: Males: length 2.25 and 2.45 mm, width 2.7 and 2.7 mm; non-ovigerous females: length 4.85 to 6.0 mm, width 5.0 to 6.7 mm; ovigerous females: length 2.25 to 6.15 mm, width 2.6 to 7.4 mm. Remarks: This was the most abundant species found at El Moro. It has been previously recorded from islands off the coast of Venezuela and the Lesser Antilles (Haig, 1956). Its range is now extended to El Moro, Mexico. The color of the El Moro specimens ranged from light cream to a grayish blue. Both chelipeds vary in the number, 4 to 5, and shape of the carpal teeth. The shape may be bifurcate or simple. Usually the first 2 teeth are bifurcate and the last 2 teeth spinous. Petrolisthes Stimpson, 1858 Five species of the genus were collected. Petrolisthes cessacit (A. Milne Edwards, 1878) Material examined: TAIC-373 (2). Lobos Reef. 16, 192 (ov). TAIC-374. Enmedio Reef. 19 (ov). TAIC-375. El Moro. 19 (ov). Measurements: Male: length 13.7 mm, width 12.0 mm; ovigerous females: length 11.8 to 16.45 mm, width 11.95 to 16.55 mm. Remarks: These are new records from the western Gulf of Mexico extending the known range to the north by 570 km. These individuals were taken from the splash zone at El Moro, from under coral rubble on the back reef crest on Lobos Reef, and from under coral rubble on Enmedio Reef in water depth of approximately 3 m. Gore (1974) 164 Proceedings of the Biological Society of Washington gave an account of some of the confusion in the literature concerning P. cessacii and P. marginatus. He pointed out that although P. cessacii from the western Atlantic was not actually reported in publication until 1970 it has probably been recorded in the literature since 1890 but incorrectly identified. Its distribution is now given as West Africa, Ascension Island, Brazil, Tobago Island, Trinidad, and the vicinity of Old Providence Island (Gore, 1974). Petrolisthes politus (Gray, 1831) Material examined: TAIC-376. Lobos Reef. 1¢. TAIC-377 (2). Lobos Reef. 16, 12 (ov). TAIC-378 (11). Enmedio Reef. 4¢6 6, 722 (ov). Measurements: Males: length 8.3 to 13.1 mm, width 8.4 to 14.1 mm; ovigerous females; length 7.0 to 9.95 mm, width 7.3 to 14.1 mm. Remarks: All individuals were collected on the reef crest where coral rubble was abundant. Schmitt (1935) reported this species to a depth of 201 m. It has been reported previously from the Florida Keys, throughout the West Indies, to Panama and Veracruz, Mexico (Haig, 1956). Petrolisthes jugosus (Streets, 1872) Material examined: TAIC-430. Enmedio Reef. 36 6. Measurements: Males: length 2.35 to 3.1 mm, width 2.45 to 3.2 mm. Remarks: All specimens were collected from water less than 3 m deep under coral rubble. Petrolisthes jugosus has been reported previously from Florida, West Indies, Panama, and Venezuela (Haig, 1956) in the littoral zone, under stones and associated with Zoanthus sociatus and corals including Maeandra. Petrolisthes quadratus Benedict, 1901 Material examined: TAIC-372 (17). Lobos Reef. 86 6,92 2 (ov). Measurements: Males: length 4.8 to 6.7 mm, width 5.0 to 6.55 mm; females: length 4.3 to 6.15 mm, width 4.5 to 6.6 mm. Remarks: The individuals were taken from tide pools on a coral boulder ridge and tide pools on a coral spoil island at the edge of a dredged channel. P. quadratus was previously known from only a few Caribbean localities (Haig, 1962). It has been reported from the littoral zone (Haig, 1956) and from algae along the shore and rock pools (Haig, 1962; Chace and Hobbs, 1969). Petrolisthes galathinus (Bosc, 1801 or 1802) Material examined: TAIC-370 (15). Enmedio Reef. 8646, 722 (3 ov). TAIC-371 (4). Lobos Reef. 24 6, 2992 (ov). Measurements: Males: length 6.05 to 11.4 mm, width 5.8 to 11.7 Porcellanid crabs—Mexican east coast 165 mm; non-ovigerous females: lengths 3.85 to 12.9 mm, width 3.4 to 13.0 mm; ovigerous females: length 6.65 to 8.65 mm, width 5.3 to 8.3 mm. Remarks: Considerable color variation exists between the individuals collected during this study, other specimens examined from the Texas A&I Collection and the color description given by Williams (1965). In the preserved specimens from both Lobos and Enmedio Reefs, the carapace and lower surface are maroon with intervening areas of light pale blue to very light yellow. Williams (1965) reported the colora- tion in life as grayish brown without markings, which is in agreement with specimens collected in South Texas. The extensive range of P. galathinus includes the western Atlantic from Cape Hatteras, North Carolina, to Brazil, intertidally to depths of 49 m (Haig, 1962). LITERATURE CITED Beure, E. H. 1954. Decapoda of the Gulf of Mexico. In Gulf of Mexico, its origin, waters, and marine life. U.S. Fish Wildl. Serv. Fish. Bull. 89:451—455. Briccs, J. C. 1974. Marine zoogeography. McGraw-Hill Book Co., New York, 475 p. Cuace, F. A., anp H. H. Hosss. 1969. The freshwater and _ ter- restrial decapod crustaceans of the West Indies with special reference to Dominica. U.S. Natl. Mus. Bull. 292. 258 p., figs. 1-76. CuAveEz, E. A., E. Hipatco, anp M. L. SEvitta. 1970. Datos acerca de las comunidades bentonicas del Arricife de Lobos, Veracruz. Rev. Soc. Mexicana Hist. Nat. 31:211—280, pls. 15, figs. 1-96. Gore, R. H. 1974. On a small collection of Porcellanid crabs from the Caribbean Sea (Crustacea, Decapoda, Anomura). Bull. Mar. Sci. 24(3):700-721, figs. 1-5. Haic, J. 1956. The Galatheidea (Crustacea Anomura) of the Allan Hancock Atlantic Expedition with a review of the Porcellanidae of the western North Atlantic. Rep. Allan Hancock Atlant. Exped. No. 8:1-45, pl. 1. —. 1960. The Porcellanidae (Crustacea Anomura) of the eastern Pacific. Allan Hancock Pacif. Exped. 24:1—440, pls. 1-41. ——. 1962. Papers from Dr. Th. Mortensen’s Pacific Expedition 1914-1916. 79. Porcellanid crabs from eastern and western America. Vidensk. Medd. Dansk Naturhist. Foren. 124:171- 192, figs. 1-5. Hi~peBRAND, H. H. 1954. A study of the brown shrimp (Penaeus aztecus Ives) grounds in the western Gulf of Mexico. Publ. Inst. Mar. Sci. Univ. Texas 3(2):233-366, figs. 1-5. Ives, J. E. 1891. Crustacea from the northern coast of Yucatan, the harbor of VeraCruz, the west coast of Florida and the Bermuda Islands. Proc. Acad. Nat. Sci. Philadelphia 43:176- 207, pls. 1-7. 166 Proceedings of the Biological Society of Washington Ricpy, J. K., anp W. G. MclIntime. 1966. The Isla de Lobos and associated reefs, Veracruz, Mexico. Brigham Young Univ., Geol. Stud. 13:1-46, pls. 1-8, text-fig. 1-9. ScumitT, W. L. 1935. Crustacea Macrura and Anomura of Porto Rico and the Virgin Islands. Scientific Survey of Porto Rico and the Virgin Islands. New York Acad. Sci. 15(2):125-227, 80 figs. TUNNELL, J. W. 1974. Ecological and geographical distribution of Mollusca of Lobos and Enmedio coral reefs, southwestern Gulf of Mexico. Doct. Diss., Texas A&M Univ. 158 pp. WiiuraMs, A. B. 1965. Marine decapod crustaceans of the Carolinas. Fish. Bull. U.S. Fish Wildl. Serv. 65(1):1-298, 252 figs. Ve , No. 17, pp. 167-178 15 August 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A REEVALUATION OF PRISTINA LONGISETA |\\= (OLIGOCHAETA: NAIDIDAE) IN NORTH AMERICA \% By WALTER J. HARMAN AND MicHAEL L. MCMAHAN Department of Zoology and Physiology Louisiana State University Baton Rouge, Louisiana 70803 Pristina longiseta Ehrenberg, 1828, long has been recognized as one of the most common aquatic oligochaetes in the world. Through the years, it has been determined to be cosmopolitan and has had several subspecies erected, usually coinciding with a continental distribution. The longest stable alliance of subspecies is: P. longiseta longiseta in Europe, P. I. sinensis in China and Africa, P. I. leidyi in North America, and P. I. bidentata in South America. Michaelsen (1905a:309) initiated the trinominal concept by recognizing as conspecific P. longiseta forma typica from Eu- rope and Zanzibar and P. longiseta var. Leidyi from North America and Paraguay. It was Sperber (1948:236), how- ever, who adapted the many names into the subspecies con- cept as required by the International Code of Zoological Nomenclature. In working with the Naididae, we have been impressed with the number of reports of the species in all its forms, but especially with the scarcity of morphometric data maintain- ing the undoubted validity of the subspecies concept. Several recent authors have raised questions regarding the validity of the subspecies, but no overt action was taken. Sperber (1948:236) said “. . . the distribution of the different forms of the longiseta complex indicates that they are geographical races, or perhaps even species, forming together a “‘Rassen- kreis, or an ‘Artenkreis. ” Brinkhurst and Jamieson (1971: 17—Proc. Brot. Soc. WasH., Vou. 88, 1975 (167) Alias. 168 Proceedings of the Biological Society of Washington 404) said “. . . it is almost impossible to separate leidyi from the combination of .. .” P. l. longiseta and P. I. sinensis, two subspecies which they synonymized. Working under the subspecies concept of Western Hemi- sphere forms, Harman (1966:241, 1973:160) and Harman and Platt (1961:93) reported P. 1. leidyi from several localities in North America. This was done with the idea in mind that any P. longiseta form in North America was automatically leidyi, and critical examination was foregone in the interest of expediency. Because of modest numbers of Central American specimens at hand and because the exact point of demarcation between the ranges of bidentata and leidyi was not known, every specimen from Central America received critical examination. They all were P. I. bidentata. When bidentata also was de- termined to be the subspecies present in the Rio Grande re- gion of Mexico, further interest in Texas specimens was stimulated. These also were determined to be bidentata. Be- cause both bidentata and leidyi then were known to be sym- patric, the prevailing concept of subspecies in the Western Hemisphere was under greater doubt. We set out to re- examine all specimens in our collection to determine where the form leidyi, characterized by simple-pointed needles, was to be found and how far northward the bidentata form, char- acterized by bifid needles, extended. To our great surprise and chagrin, no specimen taken in North America could be identified as leidyi. Critical examination of the needle setae revealed that every specimen had, in fact, a bifid needle. Therefore, the simple-pointed needle of leidyi, now regarded as the cardinal separatory character between bidentata of South America and leidyi of North America, did not exist. In defense of nearly a century of error, it can be stated that the bifid condition of the needle is an extremely difficult char- acteristic to see and that exact profile under the highest mag- nification using phase contrast microscopy usually is required. Even then, one is not able to see this characteristic in every setal bundle because of debris, broken setae, or the position of the mount. We now have critically re-examined every speci- men available to us (nearly 600) representing more than Oligochaete reevaluated 169 150 collections from 15 states in the United States and from Mexico. There is not a single specimen which does not have at least one setal bundle showing the bifid condition of the needle. Pristina longiseta was described by Ehrenberg (1828:112), who characterized it as having an anterior proboscis, 7-8 uncinate setae ventrally, and three setae per bundle dorsally, longest in III. Undoubtedly, this was an adequate descrip- tion at the time to separate this species from all others. Leidy (1850:44) reported a North American form which he placed in Ehrenberg’s species because it corresponded to it “_.. as far as the description goes.” To Ehrenberg’s descrip- tion, Leidy added that each worm possessed 16 segments, the ventral setae were retractile, and the dorsal setae of III were twice the length of the others. Vejdovsky (1884:31), providing the most complete descrip- tion of Pristina longiseta among the early authors, stated that the dorsal setae all were capilliform, not noting the existence of needles. Smith (1896:396) described Pristina leidyi from Illinois, stating that it probably was identical to Leidy’s form of P. longiseta. He expressed appreciation to Vejdovsky, whose tfa- miliarity with P. longiseta caused them to concur that the North American form was distinct from, but closely allied to, the European species. Smith, too, noted only hair setae in the dorsal bundles, but he did state that one hair in each bundle often was quite short, undoubtedly referring to the needle. Also, for the first time, serrations were seen on the hair setae. Smith saw no serrations on the proximal half of a fully-formed seta. In the middle of the hair, serrations were 6 pm apart, and the serrated condition was most evident distally. Michaelsen (1905b:357) noted serrations on hairs of speci- mens from Germany and Paraguay, which he therefore identi- fied as P. leidyi. Moore (1906:166) noted the absence of serrations on the hair setae of both II and III in P. leidyi although he did not detect the presence of needle setae. Piguet (1906:292) was the first author to observe that hair setae, with the exception of those in III, were serrated in P. 170 Proceedings of the Biological Society of Washington longiseta, and he was the first to note the presence of needles in the dorsal bundles of this species. The needles, 2-5 per bundle, were straight, tapering to a thin single point, and without a nodulus. Hayden (1914:137), describing a species which later was synonymized with P. leidyi, also observed the presence of serrated hairs, except in III, and needles in the dorsal bundles. Michaelsen (1905a:309) placed the North American and South American forms as a variety of P. longiseta, i.e. P. l. var. Leidyi, and other forms were designated as P. longiseta f. typica. Chen (1940:47) compared the two forms and restored them to species rank on the basis of differences in hair serrations and the ventrals of II and III. Cernosvitoy (1942:198), apparently unaware of Chen’s work, described a variety of P. longiseta from Argentina which he named P. longiseta, var. bidentata in reference to the bifid condition of the needles. In addition to the needles, this form differed from f. typica in having closer serrations on the hairs and a greater difference between the teeth of ventrals of II. Marcus (1943:107), probably not cognizant of the simul- taneous work of Cernosvitov, described P. longiseta f. typica from Brazil as possessing needles which were finely bifid. Sperber (1948:236) stated that P. longiseta in its various forms might consist of a number of geographical races or even species, recommending that additional work must be done to clarify the status of the group. She recognized four races of the Rassenkreis Pristina longiseta although giving them subspecific status: P. |. longiseta, P. |. sinensis, P. 1. leidyi, and P. |. bidentata. Brinkhurst and Jamieson (1971:403) reduced the list to three subspecies by the synonymy of P. I. sinensis with P. l. longiseta. They observed that there was difficulty in separat- ing leidyi from the new combination, indicating that further synonymy might be required. We now are abolishing the subspecies concept and report- ing Pristina leidyi Smith, 1896, to be the species name for all the forms reported from North America, South America, and Hawaii. The name is chosen because of priority held by Smith’s name over that established by Cernosvitoy (1942: Oligochaete reevaluated al 198). This is regrettable due to the fact that the description must now include bifid needles for the species, the charac- teristic upon which Cernosvitov’s name bidentata was based. It here is deemed prudent to return to specific status (with- out subspecies) Ehrenberg’s P. longiseta as the name for all those forms reported from Europe, Asia, and Africa. A form reported by Jackson (1931:74) from Australia cannot be identified with certainty, but it may fit better with P. longiseta. It is apparent that descriptions currently found in the literature for subspecies of Pristina longiseta are composites of data published over a span of nearly 150 years. It is evident that, far too frequently, published data were assumed to repre- sent fact without further investigation. The example at hand is the prevalent assumption that P. l. leidyi is distinguished from P. |. bidentata primarily by the existence of simple- pointed needles in leidyi, in contrast to the bifid needles of bidentata. Therefore, the following description of Pristina leidyi is based solely on our collection of this species to pre- vent the preservation of any erroneous information which might have been published in the past. Pristina leidyi Smith, 1896, new combination Prostomium forming a proboscis. Hair setae, 1-4 per bundle, begin- ning in IT, serrated, except in II and III, the latter extremely elongated. Serrations up to 10 wm apart on mid-shaft, becoming closer distally. Needles, 1-4 per bundle, beginning in II, fine and straight, finely bifid, without nodulus. Ventral setae, 2-10 per bundle anteriorly, 2-12 pos- teriorly; in II much longer and thicker than the rest, with distal tooth twice as long as the proximal; in III, slightly longer and thicker than the rest, with distal tooth 114 to 2 times as long as proximal; in the rest, teeth approximately equal; nodulus in II slightly proximal to median, median in III, becoming slightly distal posteriorly. Clitellum in %VII-'%4X. Genital setae in VI, 1-3 per bundle, bifid, with long converging teeth, enclosed within glands. Stomach in VIII. n = 12-25. s= 13-41. Distribution: North America, South America, Hawaii. Discussion: From our material and from synonymy, Pristina leidyi is seen to be a species with considerable variability across its distribution: however, such variability is within reason for the species. The ranges of setal characteristics for all specimens from nine geographical re- gions within the distribution of P. leidyi are found in Table 1. Dif- ferences are seen in setal numbers and lengths between different 172 Proceedings of the Biological Society of Washington 09-87 GI-P GO-GS My PS-OL Si TeMey 99-8P II-7 O8-LS L-¥ 96-89 NY puvyAreyy—eruesAsuuog Air S-P 89-6& 9-7 VL-89 L-G vIOSOUUTI—URS IYI 89-€¢ 6-F GL-€9 9-G €8-GL 8-g AyonyUsy—sessouus J, 9L-8P GI-F €8-G¢ 6-7 86-L9 OI-€ BIS1O95)—ePHOLy 89-GP OI-€ G8-SG NS 86-€L 8-G Tddisstsstjj—eueisino' Ty 08-0S 6-F 8L-9S 9-F 18-89 9-¥ BULOYL]YO-SEXO |, OL-€¢ 8-G 11-99 a 98-81 i OdIxa|\| IL-@S 8-€ 06-€9 GE PII-SL c-€ (PLET “ueuepy) weuring III % III 7 III u III uf II ut I] ut AW[LOOT IOLI9}SOg IOTIO}SOg (ui) yysue'T a[pungq zed ‘oN (wa) yysueT a[pung sed ‘oN (ui ) aTpunq 10d ‘on yjsue'T aves [vIJUOA €8-6Z (G8) 1 o-€ EPE-9El LI8-919 (ee) Gr eae G8-€e (€-2) #1 1-8 OSF-8P1 189-€0P (Ge) Sl purlAre—eraealAsuueg P9-8G (G6) FAT €-9 COP-9ET 118-709 (ee) Si B] OSOUUTIN—URSTYOII] IL-Se (G6) TT €-01 O€S-E9T 6Z8-GF9 (So) ill Ayonyuey—sessouus J, €8-0€ (G6) FI C6 PLS-IST Z66-80E (ee) Si vIS.109)—epHOo 6-08 (G=@) FT I-¢ 167-LGL OFL-8ZS (c=@) iI Iddississtj—eurismoT 88-GZ (G6) S11 Gao OGP-OET 89L-GOS (G6) 1 vULOYe[YO-SEX I, PL-GE (ee) I CG SGE-CET OSt-FIF (el) ei OO1XOJ] G1-06 eal I-S§ F6C-GFI OSL-SLE (GL) iA (FLEL “UeUUeEZ) WeuLINg (ua ) ( WOUTUIOD 3SsOUT ) (ua) III JO 9Arsnpoxe II ut ( WOUTUIOD 4sOULt ) AVRoO'T qysue', a[pung sted ‘on, SUOTZIIAS (utr) y{sue'T (uu) y{sue'T a[pung tod ‘on 8849S 2[PI9N ovqog Ivy UOLNGLYSIp S}I ULYA svoie Teorydeisoes ouIU WOT, Whpia? DUI4SI4g JO SOTSTIoJOvIeYO [eJOS JO sosueY “T ATAV], Oligochaete reevaluated 173 geographical areas, but such differences are slight. For instance, speci- mens from Mexico and Surinam characteristically usually have 1-2 hairs per bundle dorsally (in parentheses in Table 1), whereas worms from all other areas most often possess 2-3 hairs per bundle. This is a relatively minor difference, but it is one which occurs with regularity. Characteristic of P. longiseta and P. leidyi is the presence of elongate, thickened hairs in III. In material examined in this study, the hairs of III were elongate, measuring 375-992 um, and thickened in most cases. Occasionally, these hairs were shorter than those of other segments, especially those of mid-body, which characteristically are longest ex- cluding those of III. Such comparatively short hairs in segment III were determined to be ones which had not developed completely and therefore had not reached their full length. The longest hair of segment III of each worm which was examined is included in the range given in Table 1; therefore, this range overlaps that of hair lengths from all other segments (127-574 um). Previous reports which made careful study of the hairs noted serra- tions (Fig. 2) on the hairs of all segments except III. This condition was first noted in P. longiseta (by synonymy) by Michaelsen (1905b: 357) and in P. leidyi by Galloway (1911:302). We must concur with Moore (1906:166), who noted the absence of serrations on the hair setae of segments II and III. After examining hundreds of specimens from throughout the Western Hemisphere, we have yet to see a ser- rated hair on II or III. The hairs of II often are short (125-200 um) and covered in varying degrees with debris which might be construed to be serrations; however, critical examination at 1000 x (phase con- trast) reveals that these hairs indeed are smooth, i.e. non-serrated. Scanning electron micrographs up to a magnification of 1000 x also fail to show serrations on hairs of II or III. The teeth of the serrations in P. leidyi were stated by Smith (1896: 397) to be 6 um apart in mid-seta, more evident distally, and absent proximally. Later authors extended the inter-dental distance in mid- seta to 8-16 um in P. leidyi (Chen, 1940:47). Cernosvitov (1942: 199) found this distance to be 3.75 um in P. I. bidentata. With our synonymy of these two forms, the literature records a mid-setal inter- dental range of 3.75-16 wm, which compares favorably with our range of 3-10 um (Table 1). An interesting and quite evident feature is that, of our material, the inter-dental distance in mid-seta is shortest, 3 wm, in specimens from Texas—Oklahoma and Hawaii (Table 1), whereas the longest such distances occur in material from areas most distant from the Texas—Oklahoma area (Tennessee—Kentucky, Florida—Georgia, and Pennsylvania—Maryland). In all cases, the serrations become more closely spaced towards the distal tip of the hair, approaching 1 um, this being indicated by the latter figure given in the hair serrations column of the table. It is unknown by what method earlier authors measured the dis- tances between serrations, i.e. between what two points they measured. 174 Proceedings of the Biological Society of Washington Oligochaete reevaluated 175 Fic. 4. Ventral setae of Pristina leidyi. A, ventral seta of II; B, ventral seta of III; C, posterior ventral seta; D, genital seta. In determining these distances in our material, we determined the dis- tance between similar portions of two adjacent serrations (tips of teeth, center of portions of serration that contact hair shaft, etc.). It is possible that some earlier authors used similar techniques, whereas others < Fics. 1-3. Scanning electron micrographs of Pristina leidyi. 1, dorsal setal bundle with two needles and two hairs, x 5000; 2, hair seta near tip, < 5000; 3, hair seta of II, showing minute serrations, < 20,000. 176 Proceedings of the Biological Society of Washington may have measured the actual distance between the serrations, there- fore obtaining smaller measurements. Scanning electron microscopy at high magnifications reveals that, indeed, all hairs bear serrations. At a magnification of 20,000 diam- eters, serrations which are a mere 0.08 um apart are very evident (Fig. 3). It is believed that such structures, previously undetected, represent a spiral surface modification. Because of this discovery, the statement that “hairs of II and III are unserrated” must be qualified by adding “as observed at 1000 x with phase contrast microscopy.” Cernosvitov (1942:199), in his description of P. l., var bidentata, stated that the 1-3 needles per bundle were fine, without a nodulus, and gradually tapered to the tip, which was bifid. This characteristic in itself was sufficient to distinguish his variety from any other form or variety of P. longiseta. The teeth were stated to be less than 0.75 um in length, approximately equal, and spread at an obtuse angle. Marcus (1943:108) recorded a length of up to 100 um for needles. In our material, each dorsal bundle contains 1-4 (usually 2-3) needles which measure 20-93 yum, the shorter lengths coming from segments near the ends of the worms. The teeth are approximately equal and less than 0.75 um in our material (Fig. 1) although they may not always be apparent. More often than not the bifurcation is obscured by debris (Fig. 1) or by the position/orientation of the needle being such that the tip of the needle is seen on edge rather than in profile. In the latter case, the needle may appear to be single-pointed or to have an expanded tip. Although the teeth often are less than 0.50 wm in length, the bifurcation is evident when it is seen in good profile at magnifications as low as 400 x (phase contrast). The geometric knowledge of Cernosvitov is to be suspected concerning his descrip- tion of the needle bifurcation as being obtuse. In actuality, the teeth of the needles form an acute angle (Fig. 1). Ventral setae are fewer, thicker, longer, and have longer teeth in segments II and III than in following segments. In our material, there are 2-10, usually 3-8, ventrals in the first two setigerous segments. In II, the ventrals measure 67-114 um, and, in III, they range 52-90 wm. Posteriorly, the range is 42-80 um, there being up to 12 setae per ventral bundle, usually 4-10. Ventrals of III (Fig. 4B) are thinner than those of II (Fig. 4A) although thicker than those of more pos- terior segments (Fig. 4C); however, all ventrals are of the same general shape. In II, the distal tooth is approximately two to three times the length of the proximal, lengths being 5-7 um and 2.5-3 um respectively. In III, the distal is about 1144 to two times the length of the proximal, and, posteriorly, the teeth become subequal, approximately 3 wm in length. Beginning in VIII, the distal tooth is distinctly thinner than the proximal (Fig. 4C). The nodulus, often weak, is slightly proximal to median in IJ, approximately median in III, becoming slightly distal posteriorly. All characteristics of the ventral setae, and especially those Oligochaete reevaluated 177 of number and length (Table 1), are remarkably consistent throughout the distribution of P. leidyi. Of nearly 600 specimens examined in the course of this study, only two individuals are clitellate, having a clitellum in %VII-'2X. Two worms, only one of which is clitellate, bear genital setae in VI, 1-3 per bundle. These setae (Fig. 4D) are greatly thickened, and the range of lengths is 80-86 um. They are bifid, and the teeth are long, nearly equal in length, and converging at the tip. The length of the teeth is one-third to half the length of the whole seta. The proximal tooth, most dense at its edges, appears to be webbed medially due to the comparative thinness of this area. A _ genital gland encloses each bundle of genital setae. Twenty-six worms, representing 4.3% of those examined, were under- going asexual reproduction at the time of collection. The number of setigers anterior to the zone of fission varied from 12 to 25, usually 12-18. No worm had more than one budding zone. One worm which possessed a budding zone also bore genital setae in VI although it was aclitellate. Containing forty setigers, its budding zone followed XX, and it had three genital setae per bundle. The stomachal dilatation occurs in VIII although it appears to begin in VII because it pushes septum 7/8 forward to mid-VII. It is approxi- mately three times the diameter of the esophagus and of the anterior intestine and is nearly spherical with greatly thickened walls. Intestinal dilatation begins in IX, the diameter of this organ being greater than that of any other part of the digestive tract. ACKNOWLEDGMENTS We express appreciation to the Department of Geology, Louisiana State University, for permission to use their scanning electron mi- croscopy facility. Mr. Mike Turner and Mr. Allen Bourland assisted with photography. LITERATURE CITED Brinkuurst, R. O., AND B. G. M. JAMtEson. 1971. Aquatic Oligo- chaeta of the World. University of Toronto Press, Toronto and Buffalo, 860 p. Cernosvitov, L. 1942. Oligochaeta from various parts of the world. Proc. Zool. Soc. London (B) 111:197—236. CuHEeN, Y. 1940. Taxonomy and faunal relations of the limnitic Oli- gochaeta of China. Cont. Biol. Lab. Soc. China 14:1-132. EHRENBERG, C. G. 1828. Symbolae physicae —. Animalia evertebrata. Phytozoa. Berlin. GatLtoway, T. W. 1911. The common fresh-water Oligochaeta of the United States. Trans. Amer. Micros. Soc. 30:285-317. HarMan, W. J. 1966. Some aquatic oligochaetes from Mississippi. Amer. Midl. Nat. 76:239-242. 178 Proceedings of the Biological Society of Washington —. 1973. New species of Oligochaeta (Naididae) with ad- ditional distributional records from Oklahoma and Texas. Southwest. Nat. 18:151—-164. —. 1974. The Naididae (Oligochaeta) of Surinam. Zool. Bijdr. Leiden 133:1-36. ———, AND J. H. Puatr. 1961. Notes on some aquatic oligochaetes from Louisiana. Proc. Louisiana Acad. Sci. 24:90-95. Haypven, H. E. 1914. Further notes on Pristina and Naidium, with description of three new species. Trans. Amer. Micros. Soc. 33:135-138. Jackson, A. 1931. The Oligochaeta of south-western Australia. Jour. Proc. R. Soc. West. Australia 17:71-136. Ley, J. 1850. Descriptions of some American Annelida abranchia. Jour. Acad. Nat. Sci. Philadelphia (2), 2:43-50. Marcus, E. 1943. Sobre Naididae do Brasil. Bol. Fac. Fil. Ciéne. Letr. Univ. Sao Paulo, 32, Zool. 7:3—247. MICcHAELSEN, W. 1905a. Die Oligochaten Deutsch-Ostafrikas. Z. Wiss. Zool. 82:288—367. ——. 1905b. Zur Kenntnis der Naididen. In E. V. Daday, Unter- suchungen iiber die Siisswasser-Mikrofauna Paraguays. Zool- ogica 18, 44:350-361. Moore, J. P. 1906. WHirudinea and Oligochaeta collected in the Great Lakes region. Bull. U.S. Bur. Fish. Washington 25: 153-171. PrcueT, E. 1906. Observations sur les Naididées et revision sys- tématique de quelques espéces de cette famille. Rev. Suisse Zool. 14:185-315. SmitH, F. 1896. Notes on species of North American Oligochaeta. II. Bull. Illinois St. Lab. Nat. Hist., Urbana 4:396—413. SPERBER, C. 1948. A taxonomical study of the Naididae. Zool. Bidr. Uppsala 28:1-296. Veypovsky, F. 1884. System und Morphologie der Oligochaeten. Prag, 172 p. 74,0673 Vol. 88, No. 18, pp. 179-188 15 August 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON THE IDENTITY OF NANNARIA CONSERVATA CHAMBERLIN, WITH NOTES ON AN ABNORMAL MALE AND DESCRIPTIONS OF TWO NEW SPECIES OF NANNARIA FROM NORTH CAROLINA (DIPLOPODA: POLYDESMIDA: XYSTODESMIDAE) By RowLanp M. SHELLEY North Carolina State Museum, Raleigh, North Carolina 27611 The recent discovery of several different species of Nannaria in the mountains and western piedmont of North Carolina has compelled attention to the long-standing problem of N. conservata, whose identity has been uncertain since its original description by Chamberlin (1940). No illustrations were pro- vided, and repeated attempts by myself to secure topotypical material from Duke Forest, Durham County, North Carolina, have been unsuccessful. This species is significant and im- portant in the North Carolina milliped fauna because it is the easternmost Nannaria in the state and the only one au- thentically reported from a non-montane locality (Duke Forest is in the fall zone region of the eastern piedmont and over 100 miles from the Blue Ridge Front). In October 1973, I collected a male Nannaria from Rutherfordton, Rutherford County, in the western piedmont of the state, and could not be certain that this individual was different from Chamber- lin’s eastern piedmont species. Through the courtesy of Richard L. Hoffman, I have recently been able to examine the holo- type of conservata and to determine that the Rutherford specimen is quite different indeed. This new species and one from Blowing Rock, Watauga County, North Carolina, are described herein along with comments and drawings of con- servata, including an aberrant male with three pairs of 18—Proc. Brow. Soc. WaAsH., Vou. 88, 1975 (179) 180 Proceedings of the Biological Society of Washington gonopods. This male and the holotypes of the new species are deposited in the invertebrate collection of the North Caro- lina State Museum (NCSM). I am grateful to Richard L. Hoffman for the opportunity to examine the type specimen of N. conservata and for addi- tional locality records of the species in North Carolina. Recent collections from Raleigh, including the abnormal, trigono- podal male, were contributed by Martha and John Cooper, N. C. State Museum. Figure 6 was prepared by Renaldo G. Kuhler, Museum staff artist; the other illustrations are the work of the author. Nannaria conservata Chamberlin Figures 1-6 Nannaria conservata Chamberlin, 1940:56. Type-specimen: Male holotype (R. V. Chamberlin collection, now being accessioned by U. S. National Museum) collected 12 November 1939 from Duke Forest, Durham County, North Carolina, N. B. Causey, leg. The brief original description of this species lacks illustrations and did not mention the paramedian sternal knobs between the 4th pair of legs or the pointed spur at the base of the prefemoral process, al- though both are noted in the description of N. minor, the type species (Chamberlin, 1918). The sternal knobs are subconical and subequal in length to the width of the adjacent coxae (Fig. 1). Chamberlin (1940) did state in his description that the gonopods of conservata “conform closely in general form to those of N. minor,’ but as can be seen from the present illustrations (Figs. 2, 3) and that by Hoffman (1964) of minor (which was not originally illustrated either), the gonopods of the two species are quite different. Thus, the original description of conservata (Chamberlin, 1940) is confusing and does not adequately depict the characters of the species. The extremely short prefemoral process was noticed by Chamberlin, however, and it is clearly the diagnostic feature of conservata. I have recently received fresh material of conservata from Raleigh, Wake Co., (946, 29) and submit the following observations to supple- ment Chamberlin’s description. The species appears to be a small Nannaria and is considerably shorter in both length and width than the new species described later in this paper. Measurements in milli- meters are given below; they were made with vernier-scale calipers with the specimens lying flat, straight, and compressed. Averages were not computed for females, since only two were available. The ab- normal male, designated by an asterisk, is not included in the averages (NCSM 2829). Diplopods from North Carolina 181 Width Length W/L ratio Males 3.1 16.9 18.3% 3.0 15.8 19.0% 3.1 16.1 19.3% 2.8 14.4 19.4% *2.9 14.6 19.9% 3.3 16.5 20.0% 3.4 17.0 20.0% 3.4 16.6 20.5% 3.3 15.8 20.9% Averages 3.2 16.1 19.7% Females ail 18.2 20.3% 3.3 15.5 21.3% In color, conservata is olive-brown with a pink epiproct and paranota; some individuals have a darker base color, probably due to an earlier molt. A wide pink stripe extends across the anterior edge of the collum, and, on the lighter individuals, a faint, narrow pink stripe is visible across the caudal half of each metatergum. The latter stripe is noticeably removed from the caudal edge, as opposed to most striped xystodesmids I have seen. Ventrally, they are uniformly white in color. All the specimens are fragile and easily broken or crushed. The abnormal male possesses a condition which, to my knowledge, has not been reported before in any order of diplopods. It has a triple complement of gonopods, all located in fully formed apertures at the anterior position of segments 5, 7 (the normal location), and 9 (Fig. 6). No other external character is replicated, and the individual is normal in color pattern, although somewhat below the male average in physical dimensions (see table above). Due to the first “accessory” gonopods on segment 5, the paramedian sternal knobs of that segment are missing. The sternal spines begin as usual on the first post- gonopodal segment, but in this case it is segment 10 instead of 8. The gonopods on segments 5 and 7 are normal, but the pair on seg- ment 9 is severely reduced in overall size, and both the telopodite and prefemoral process are incompletely developed (Figs. 4, 5). The regu- larity of the pattern is impressive, that segments 5, 7, and 9 are involved rather than an irregular pattern such as 5, 7, and 10 or 6, 7, and 9, etc. Unfortunately, this male was preserved before its condition was de- tected, and mating experiments with normal females could not be at- tempted, although it might be expected to be sterile. A thorough search of this locality revealed no further abnormal millipeds. In the absence of a definitive explanation for this phenomenon, it seems best to label this male a “developmental accident,” caused by 182 Proceedings of the Biological Society of Washington Fics. 1-5. Nannaria conservata Chamberlin, 1940. 1-3, holotype: 1, profile of paramedian sternal knobs between 4th pair of legs, caudal view; 2, left gonopod, medial view; 3, left gonopod, lateral view. 4-5, left gonopod of 9th segment of abnormal male (NCSM 2829): 4, medial view; 5, lateral view. unknown genetic and/or environmental factors, and to monitor the site in future years to see if another abnormal individual appears. Distribution: Nannaria conservata is known only from the “triangle” (Raleigh-Durham—Chapel Hill) region and nearby localities in eastern piedmont North Carolina. In addition to the type-locality, known dis- tribution records include the following: Franklin Co., 1.5 miles SW Diplopods from North Carolina 183 Fic. 6. Ventral view of segments 5-9 of abnormal, trigonopodal male from Raleigh (NCSM 2829). The right gonopod of seg. 5 and the left of segs. 7 and 9 were removed prior to illustration, and the remain- ing gonopods have been slightly displaced medially. Pilot, 14 in collection of R. L. Hoffman, 2 May 1959, L. Hubricht, leg. Wake Co., Raleigh, several males and females in the Chamberlin collec- tion, collected on two unspecified dates by unspecified collectors; and 94 (one abnormal with three pairs of gonopods), 29, NCSM 2825, 2828 and 2829, 29-31 January 1975, M. R. and J. E. Cooper, leg. I have also collected several juveniles of Nannaria, both sexes, from Johnston Co., 184 Proceedings of the Biological Society of Washington 11.1 miles SW Smithfield. These are probably conservata, but no adults have been encountered at this locality. Ecology: No habitat descriptions have been published for con- servata, and the type locality, Duke Forest, includes a variety of bio- topes, some within the city limits of Durham. The 1975 material from Raleigh was taken from the yard of a residence in the northern part of the city. The species was concentrated primarily next to the house under a hedge and leaf clippings and was also found at the base of several trees under thick layers of pine needles. These were relatively dry locations; the milliped was not encountered in the more moist spots in the yard. The juveniles from Johnston Co. were found on sandy soil under a thin layer of leaves, in a small tract of hardwoods. This is a rather unusual habitat for xystodesmids, but Apheloria and Sigmoria were also collected there as was Ptyoiulus, a parajulid which I fre- quently find in sandy habitats. Remarks: Wray (1967) included the type locality for conservata in his list of millipeds of North Carolina, and Brimley (1938) and Wray listed N. minor as having been collected in Raleigh in December and February (no year given). Since minor is known only from Carter Co., Tennessee, and Madison Co., North Carolina (Hoffman, 1964), these Raleigh records are undoubtedly incorrect. Additionally, there are no specimens of minor from Raleigh in the Chamberlin collection, but it does contain the previously mentioned specimens of conservata, col- lected on two different dates. The latter are probably what Brimley and Wray had in mind, and it seems likely that Brimley collected the millipeds and sent them to Chamberlin, who misidentified them as minor. The Raleigh collections of conservata in 1975 indicate that the species is present in the city during the winter months and further suggest that the published listings for Raleigh refer to this species and not to minor. Thus, Nannaria conservata is the only Nannaria so far known in the eastern piedmont region of North Carolina. Nannaria rutherfordensis, new species Figures 7-9 Type-specimen: Male holotype (NCSM 2053) collected by the author on 15 October 1973, in Rutherfordton, Rutherford County, North Caro- > Fics. 7-9. Nannaria rutherfordensis, n. sp. (holotype, NCSM 2053). 7, profile of paramedian sternal knobs between 4th pair of legs, caudal view; 8, left gonopod, medial view; 9, left gonopod, lateral view. Fics. 10-12. Nannaria domestica, n. sp. (holotype, NCSM 1007). 10, profile of paramedian sternal knobs between 4th pair of legs, caudal view; 11, left gonopod, medial view; 12, left gonopod, lateral view. Diplopods from North Carolina 185 186 Proceedings of the Biological Society of Washington lina, along U. S. hwy. (Business) 74, 0.2 miles W U. S. hwy. (Bypass) 74, Female unknown. Diagnosis: Characterized by the following combination of char- acters which does not occur in any other known species of the genus: pointed spur at base of prefemoral process; prefemoral process sigmoidally curved and uniformly tapering to pointed tip, slightly shorter than acropodite; acropodite apically truncate and undivided, sharply bent mediad distally. Holotype: Length 22.9 mm, greatest width 4.7 mm, W/L ratio 20.5%. Color in life unknown; coloration of preserved specimen: metaterga brown, segments 1-6 lighter than others; all terga with nar- row dark stripe along posterior edge; paranota white; collum with wide white stripe along anterior edge, connecting with paranota; epicranium continuous color with anterior (lighter) segments; genal apices white; antennae light brown, distal two segments darker; venter and legs white; sterna slightly darker. Paramedian sternal knobs between 4th pair of legs subconical, sub- equal in length to width of adjacent coxae (Fig. 7); sternal spines beginning as faint blunt lobes on segment 7, larger from 8-17 and be- coming sharply pointed on 18 and 19; prefemoral spines short through segment 9, longer at 10 and curving caudad, longest on 17; hypoproct broadly rounded; paraprocts with margins strongly thickened. Gonopodal aperture broadly oval, edges slightly raised and thickened; gonopods (Figs. 8, 9) with pointed spur at base of prefemoral process; prefemoral process uniformly tapering and curving sigmoidally to pointed tip, slightly shorter than acropodite; acropodite bladelike, slightly wider at midlength, bending sharply mediad distally and overlapping near tip with acropodite of opposite gonopod, apex of acropodite blunt, truncate, not apically divided. Ecology: The type specimen was collected from under leaves against a log in an urban patch of hardwoods, a few feet from the pavement (elevation approximately 1000 ft. ). Distribution: Known only from type-locality. Remarks: The prefemoral spur distinguishes this species from N. simplex Hoffman, 1949, and the apically blunt, distally bent telopodite distinguishes it from N. minor Chamberlin, 1918, and N. fowleri Cham- berlin, 1947. The long, sigmoidally curved prefemoral process dis- tinguishes it from conservata, the other known piedmont species. Nannaria domestica, new species Figures 10-12 Type-specimen: Male holotype (NCSM 1007) collected by the author on 16 October 1971, in Blowing Rock, Watauga County, North Caro- lina, from yard of residence on Goforth Road, 0.5 miles N. U. S. hwy. 321. Female unknown. Diagnosis: Characterized by the strongly twisted prefemoral process with blunt spur at base; acropodite without apical division. Diplopods from North Carolina 187 Holotype: Length 24.5 mm, greatest width 4.6 mm, W/L ratio 18.6%. Color in life unknown; coloration of preserved specimen: metaterga dark brown, with narrow dark stripe along posterior edge of segments 5-18; paranota white; collum with wide white stripe along anterior edge, connecting with paranota; epicranium brown, frons and clypeus lighter; antennae brown; venter white. Paramedian sternal knobs between 4th pair of legs small, inconspicous, much shorter in length than width of adjacent coxae (Fig. 10); sternal spines beginning at posterior legs of segment 7, becoming spinelike on 8 and extending through 18; hypoproct with three weak lobes, one medial and one on either side; paraprocts strongly margined. Gonopodal aperture broadly oval, edges slightly raised and thickened; gonopods (Figs. 11, 12) with blunt spur at base of prefemoral process; prefemoral process strongly twisted at midlength and curving sigmoidally posteromediad to pointed tip, shorter than acropodite; acropodite blade- like, curving strongly anteromediad distally and becoming progressively narrower to pointed tip, not apically divided. Ecology: The type specimen was collected from under leaves along the edge of a paved driveway (elevation approximately 3600 ft.). Distribution: Known only from type-locality. Remarks: This species is very similar to N. laminata Hoffman, 1949. They both have twisted prefemoral processes, but laminata lacks the basal spur which is distinct on domestica. LITERATURE CITED Brimuey, C. S. 1938. The Insects of North Carolina. North Caro- lina Department of Agriculture, Division of Entomology, Raleigh. 560 pp. (millipeds pp. 498-499). CHAMBERLIN, RALPH V. 1918. New polydesmoid diplopods from Tennessee and Mississippi. Psyche 25:122—127. 1940. On some chilopods and diplopods from North Caro- lina. Can. Ent. 72:56—-59. HoFFMAN, RicHarRpD L. 1964. The status of Fontaria pulchella Boll- man, with the proposal of a new genus and tribe in the diplopod family Xystodesmidae. Proc. Biol. Soc. Wash. 77: 25-34. Wray, Davi L. 1967. Insects of North Carolina, Third Supplement. North Carolina Department of Agriculture, Division of En- tomology, Raleigh. 181 pp. (millipeds pp. 150-154). 188 Proceedings of the Biological Society of Washington i Pei : cies 8, No. 19, pp. 189-210 15 August 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON THE PYCNOGONIDA TYPES OF WILLIAM A. HILTON. — I. PHOXICHILIDIIDAE. = By C. ALLAN CHILD Department of Invertebrate Zoology (Crustacea) Smithsonian Institution, Washington, D.C. 20560 William Atwood Hilton (1878-1971) proposed 66 new species of Pycnogonida in 19 papers on the group, giving mostly preliminary descriptions based primarily on eastern and northern Pacific material. Since the few published figures are poorly drawn and all of the descriptions are inadequate and in part erroneous by today’s standards, it becomes in- creasingly necessary to redescribe and adequately figure most of his work before further Pacific taxonomic studies can pro- ceed. The 66 species, including two pairs named twice under different genera, are deposited in three institutions: 56 types in the U.S. National Museum of Natural History (USNM); nine in the B. P. Bishop Museum, Honolulu, Hawaii (BPBM); and one in the Pomona College Museum, Claremont, Cali- fornia (PCM). The state of preservation for these speci- mens ranges from good to disintegrated with one type known to be lost. Unfortunately, many of the USNM types were at some time squashed flat under cover glass so that, although permanent slides were never made, the specimens often ap- pear very different from fresh material of the same species. A note on the status of the type-specimen(s) will be found in the remarks under each species. This paper is the first of three on Hilton’s material. It treats 16 proposed species. including the two twice-named pairs, in the family Phoxichilidiidae. The other two papers will describe Hilton’s species of the family Ammotheidae and 19—Proc. Biot. Soc. WasH., Vou. 88, 1975 (189) 190 Proceedings of the Biological Society of Washington finally the remaining families except the Nymphonidae. The Nymphonidae have been described adequately by Hedgpeth (1949). Descriptions of the valid species will be given as if they were being described for the first time, unless subsequent authors have described the species in works cited here, and no reliance will be given upon Hilton’s original descriptions except for any applicable comments in the remarks. Half of the species treated in this section are synonyms of previously described species while the others appear to be valid or, in one case, an indeterminable juvenile. Hilton’s species Present designation Anoplodactylus intermedius Anoplodactylus batangensis ( Helfer ) A. nodosus valid A. oculospinus valid A. pacificus valid A. projectus A. portus Calman A. robustus A. portus Calman A. unospinus Phoxichilidium sp. juvenile Phoxichilus compactus (Endeis compacta) Phoxichilidium compactum (Halosoma compactum) P. micropalpidum P. parvum P. quadradentatum A. viridintestinalis (Cole) A. compactus (Hilton) valid valid valid P. truncatum Pigrogromitus robustus A. robustus (Dohrn) Pycnosoma strongylocentroti Losina-Losinsky A complete bibliography of Hilton’s pycnogonid literature is given at the end of this paper, along with other literature cited. Hilton’s literature will be omitted from the second and third papers as unnecessary duplication, and they will carry only other literature cited. I wish to thank Dr. D. M. Devaney of the Bernice P. Bishop Museum, Honolulu, Hawaii, for arranging the loan of several of Hilton’s Hawaiian types. Hilton's sea spider types 191 Family Phoxichilidiidae Sars, 1891 Genus Anoplodactylus Wilson, 1878 Anoplodactylus batangensis (Helfer) Anoplodactylus intermedius Hilton, 1942d: 44-45, fig. 2—1942f: 73. Anoplodactylus batangensis—Stock, 1968: 54 [literature]. Material examined: Female, holotype: Hawaiian Islands, coll. C. H. Edmondson. BPBM 4684. Hilton includes 2 other females and 2 males in his list of specimens. These were not seen. Remarks: Hilton probably did not have access to Helfer’s paper when he described this species although he seems to have ignored, at times, much of the non-American literature on the group. The species is now known to have an almost pantropical distribution. The type is in poor condition. Anoplodactylus compactus (Hilton) Figure 1 Phoxichilidium compactum Hilton, 1939a: 27-28.—1939b: 72.— Hedgpeth, 1941: 257 [key]. Halosoma compactum.—Marcus, 1940; 45-46 [list]—Hilton, 1942£: 74.—Hedgpeth, 1964: 205, fig. 93b. Material examined: Juvenile, holotype: Laguna Beach, southern California, intertidal, coll. Hilton, USNM 79424. Male, paratype: Pacific Grove, Monterey Bay, California, col. Hilton. USNM 80521. Description (paratype): Trunk only slightly longer than wide, un- segmented. Lateral processes contiguous for over half their length. Neck short, flanked by low tubercles above insertion of ovigers. Ocular tubercle large truncated cone, without spine or papillae. Eyes large, unpigmented in alcohol. Lateral processes armed with large blunt dorsodistal tubercle, about as wide as tall, each flanked by strong an- terior and posterior spine. Abdomen robust, blunt, carried almost vertically, shorter than ocular segment, not extending to tips of 4th lateral processes. Chelifores typical of genus with few scattered setae. Chela rather large, armed with few setae. Fingers as long as palm. Oviger 6-segmented with trace of articulation proximally on 3rd seg- ment. Second and 3rd segments subequal. Segments 4, 5, and 6 in- creasingly shorter, armed with setae equal to or longer than diameter of segments, on outer margins of segments. Legs short, robust, no segment longer than 3 times its diameter. Coxae armed with few lateral and ventrodistal setae. Coxa 2 with prominent ventrodistal tubercle only on 4th leg, bearing genital pore, armed with tuft of setae. Femur with few scattered short setae and single long seta on short dorsodistal tubercle. Femur % longer than tibiae. Femoral cement gland a very small low cup, halfway along 192 Proceedings of the Biological Society of Washington Fic. 1. Anoplodactylus compactus, paratype: a, Dorsal view of trunk; b, Lateral view of trunk; c, 3rd leg, femoral cement gland enlarged; d, 2nd coxa of 4th leg; e, Terminal segments of 3rd leg; f, Oviger. Hilton's sea spider types 193 segment dorsally. Tibia 1 slightly longer than tibia 2, both armed with few short dorsal and ventral setae and single long dorsal seta. Propodus with marked heel armed with 2 wide basal spines and field of 6 thin spines or setae. Sole with short spines increasing in number distally, flanking propodal lamina which extends entire length of sole. Small auxiliary claws present. Measurements of paratype, in mm: Trunk length (front of ocular segment to tip 4th lateral processes) _ 0.98 Trunk width (across 2nd lateral processes) ___.......----------------- 88 Eroboseismencthy (ventral) ee ee et $530 Third leg: ETT Seen) LS cea ee mia IV 09 (Cora ie eS RE 0.30 Dibiay 2) eee Bum eiaes auora 08 (Goxas oes ee 45 STRATESUIS be ee os eet UR Ny 13 CO se Lin oes at .30 Propod icy eee aaa wale a. .00 1 EXGSED DE eye a en SP Glew Peco is Bape a eee 31 Distribution: Mid- and southern California coast. Depth range: Intertidal, littoral. Remarks: Hilton lists several specimens in his three references to this species; one from Santa Cruz Island, one from Pacific Grove, and several from Laguna Beach, one of which he named as type. His description, which is not too incorrect for this species, is based on an adult male, but for some reason he named a juvenile as the type. The holotype specimen illustrated by Hedgpeth (1964: 205, fig. 93b) shows this juvenile and does not reveal the lateral process tubercles or the propodal lamina. Unfortunately, the type has been badly damaged under a cover glass. The above redescription is based on the well pre- served male paratype. This species is superfically very much like the sympatric species, A. viridintestinalis. It also resembles A. robustus (Dohrn) (non Hilton, 1939), A. arescus Marcus, and other species previously synonymized under Halosoma, It is set apart from each of these by its prominent lateral process tubercles which make it an easily recognized species. Like others of the old Halosoma group, it has extremely long propodal _ lamina extending the length of the sole. Anoplodactylus nodosus Hilton Figure 2 Anoplodactylus nodosus Hilton, 1939a: 29.—Marcus, 1940: 41 [list].— Hedgpeth, 1941: 257 [key].—Hilton, 1942f: 72. Material examined: Female, holotype, USNM 14260, and Male, paratype, USNM 124014: Catalina Harbor, Catalina Island, southern California, coll. William H. Dall, January, 1874. Description: Female: trunk distinctly segmented except for 3rd and 4th segments. Lateral processes separated by slightly less than their 194 Proceedings of the Biological Society of Washington Fic. 2. Anoplodactylus nodosus, paratype: a, Dorsal view of trunk; b, Lateral view of trunk; c, Oblique view of trunk, anterior; d, Ventral first lateral process with oviger implantation; e, 3rd leg; f, Terminal segments of 3rd leg; g, Chela. Hilton's sea spider types 195 diameters, with thin dorsodistal tubercles flanked by single anterior and posterior setae. Ocular tubercle moderately tall, pointed, carried obliquely forward; eyes large, unpigmented, with 2 tiny (sensory?) tubercles flanking tip and 2 slightly larger tubercles flanking ocular tubercle at base. Abdomen cylindrical, glabrous, not longer than ocular tubercle, carried obliquely toward posterior, at same angle as ocular tubercle. Proboscis cylindrical, flaring at midlength. Chelifore with scape overreaching proboscis, with 2 low tubercles along median dorsal surface and 2 dorsodistal low tubercles, glabrous. Fingers carried at slight angle to palm, overlap when closed. Paim slightly longer than fingers, armed with 3-4 short setae. Legs of moderate length. Second coxa with raised ventrodistal tubercle bearing genital aperture, on all legs. Femur and tibiae with many small dorsal and ventral tubercles, blunt or pointed, only a few armed with seta at tip. Second tibia longest, subequal to femur. Femur and first tibia armed with several short setae laterally and ventrally. Second tibia with row of ventral short setae. Tarsus very short, with 3-4 ventral setae. Propodus long, thin, with distinct heel bearing 2 basal spines and 2 pairs of setae. Sole armed with row of curved setae flanked by several small setae. Distal propodal lamina equal to less than half sole length. Claw long, equal to length of sole. Auxiliaries present. Male: slightly smaller than female, with female characters. Ovigers greatly reduced to single segment, as if atrophied (?). Femoral cement glands prominent on mid-dorsal surface, less than half segment diam- eter. Genital orifices on 2nd coxae of all available legs in same position as those of female. Propodal lamina slightly longer. Measurements of male, in mm: Trunk length (anterior of ocular segment to tip of fourth lateral j DEROGATION eR sb nei ee nes NN 2.36 Trunk width (across second lateral processes) _.....------------------ 1.44 Prolsosais Jemma «(aiterrallihy)) gba kees ee i ee 1.19 Third leg: Tilbiapil) yee emer Oa sae: 1.6 Poxawiliee ei kh as a 0.37 TU yy OAM RY eas GMaBs stems a 2.02 Coane eae es .89 Taguig ataats. | bain Baw elation Th 16 GOAN peers hw yao 8 en 69 Propodus) 245 rks orale .89 FF TU Sas aoe el 1.97 nape ho ie, alent Re .64 Distribution: Known only from the type-locality, Catalina Island, southern California. Depth range: Unknown, but probably littoral or sublittoral. Remarks: Hilton (1939a) listed the central California coast and islands as the distribution for this species, but never elaborated further on coastal records. It seems rather strange that this large species has never been collected since Dall’s collection, but the pycnogonids of the Cali- fornia coast are less than perfectly known and even less perfectly recorded. 196 Proceedings of the Biological Society of Washington The male specimen may be aberrant, but if not, it contributes to the growing suspicion that some pycnogonids may be hermaphrodites or perhaps gynandromorphs. The ovigers do not appear to have been broken off, but instead assume a withered aspect or one of atrophy. The leg integument is thick, but the cement gland is complete and probably functional. No eggs can be seen in the male femur, nor is the female ovigerous. The male has definite genital pores on the four legs attached to the trunk, and they appear on all second coxae as with those of the female. Hilton described the chelifores and legs as having more setae than they do. Very few of the leg tubercles have setae, but they may have broken off after many years in alcohol. These specimens are quite like A. erectus, particularly in the similar lateral process tubercles and propodus, but differ in the tuberculate legs. The type-specimens are in good condition except for the loss of some legs from each. Anoplodactylus oculospinus Hilton Figure 3 Anoplodactylus oculospinus Hilton, 1942f: 72-73. Material examined: Female, holotype: Off Moss Landing, Monterey Bay, California, from rocks in 91 m, coll. E. Ricketts, no. 78.1, USNM 81494. Description: Trunk fully segmented, covered with scattered tiny papillae. Lateral processes separated by slightly less than their diam- eter, armed with a pair of short thin tubercles distally, each bearing a seta, and a 3rd smaller tubercle, glabrous, between the 2 tubercles on the posterior 4 lateral processes. Neck widened laterally, almost to outer tips of Ist lateral processes. Ocular tubercle a large truncated cone capped with thin tubercle shorter than cone. Eyes large, unpigmented. Ocular tubercle flanked by pair of small lateral tubercles, each bearing single seta. Proboscis cylindrical, slightly bulbous proximally and flar- ing distally. Ventral proximal surface rough, but without tubercles. Abdomen short, curved posteriorly, reaching tips of 4th lateral pro- cesses, armed with distal seta. Chelifores with low tubercles on dorsal and lateral surfaces of scape, each bearing short seta. Chela palm longer than fingers, armed with several endal setae. Fingers curved, crossing at tips, armed with field of setae proximally on immovable finger and 2-3 setae on movable finger. Legs moderately long. First coxa with 2 dorsodistal thin tubercles matching those on lateral processes. Femur inflated, armed with few dorsal and lateral setae. First tibia slightly longer than second, armed with row of lateral setae and single long seta dorsodistally. Second tibia armed with rows of lateral and ventral setae and single long dorso- distal seta. Tarsus very short, armed with 2 ectal and 2 endal setae and Hilton's sea spider types 197 Fic. 3. Anoplodactylus oculospinus, holotype: a, Dorsal view of trunk; b, Lateral view of trunk; c, 3rd leg; d, Terminal segments of 3rd leg; e, Chela. single endal spine. Propodus moderately long, with distinct heel armed with 2 curved basal spines and 2 setae. Sole with 10-11 short curved spines, flanked by 7-8 thin setae, and short propodal lamina distally. Claw robust, equal to sole length. Auxiliaries short, robust. 198 Proceedings of the Biological Society of Washington Measurements, in mm: Trunk length (anterior of ocular segment to tip fourth lateral pro- CESSES!) thapebcartian Deis Gee ee a no 1.75 Trunk width (across second lateral processes) ____.--------------------- 1.0 Proboscis wlength) (laterally); 2s i Se ee ee 2, Abdomens length (laterally) 228 So se 3 ee eee .28 Third leg: Tibia J... 72 2 eee 1.51 (Cfo CF ROR eas oe 0.41 Tibia 2 2S 1.43 (Conca oe eat ee 6 Tarsus) 4... see 14 ( Clo cre 6 eS Se a ne he 44 Propodus. .......-3 Seen sa Henan escent awe 1.62 Claw 2.0 1 Sek 44 Distribution: Known only from its type-locality, Moss Landing, California. Depth range: 91 m. Remarks: Although described from a single female specimen that remains unique in spite of an active marine laboratory situated at its collecting site, this species is valid. The specimen is similar to A. nodosus, except for the double tubercles on the lateral processes and first coxae. The legs and scape are equally nodose, but the proboscis, chela and propodus are quite different in shape. The specimen is also about % smaller than A. nodosus. It is well preserved, although miss- ing three legs. Anoplodactylus pacificus Hilton Figure 4 Anoplodactylus pacificus Hilton, 1942f: 73.—Stock, 1955: 242-243. Material examined: Male, holotype: Dakins Cove off Avalon, Santa Catalina Island, southern California, Blake trawl in 86 m, coll. ALBA- Tross, 8 April 1896, sta. 3662. USNM 77078. Description: Trunk slender, segmented except for last 2 segments. Lateral processes long, smooth, separated by more than their diameter, armed distally with pair of dorsal setae. Neck long, posterior slightly raised above Ist trunk segment. Ocular tubercle large, conical, pointing toward anterior (tip broken off). Eyes large, slightly pigmented. First lateral processes with small tubercle above oviger implantation. Abdomen thin, vertical, armed with distal seta. Proboscis cylindrical, curved dorsally, inflated at midlength. Mouth truncated, lips distinct. Chelifores slender. Scape with row of dorsal setae and small tubercle at dorsodistal margin. Chela palm rectangular, with 2-3 dorsal setae. Fingers thin, as long as palm, curved, crossing at tips, armed with 8-9 dorsal and lateral setae on movable finger and single seta on immovable finger. Without teeth or spines. Oviger thin, with 6 segments. Second segment 9 times its diameter, half as long as 3rd segment. Fourth and 5th segments subequal, half Hilton's sea spider types 199 Fic. 4. Anoplodactylus pacificus, holotype: a, Dorsal view of trunk; b, Ventral oblique view of trunk anterior; c, 3rd leg, with enlargement of femoral cement gland; d, Terminal segments of 3rd leg, with en- largement of sole spine; e, Oviger; f, Oviger terminal segments; g, Chela. 200 Proceedings of the Biological Society of Washington as long as 2nd segment. Sixth segment ovoid, slightly longer than wide. All segments with few endal and ectal setae; 5th segment with field of endal setae, 6th with 2 endal setae. Legs slender, lateral moult sutures distinct proximally on longer segments. Coxa 2 with low mid-dorsal elevation and short ventro- distal tubercle. Femur longest segment, without distal tubercle or long seta. Femoral cement gland on mid-dorsal surface, the raised duct about half segment diameter, broad at base, flaring at tip. Tibia 1 and 2 slightly shorter than femur, without distal tubercles or long setae. Coxa 2, femur, and tibiae armed with short dorsal and ventral setae. Tarsus extremely short, merely a semicircle, armed with 3-4 short ventral setae. Propodus long, thin, without marked heel. Two curved spines and pair of setae on heel, 14—15 sole spines, bent anteriorly and slightly concave on outer surface, flanked by several lateral setae. Propodal lamina less than 4% sole length. Claw thin, 24 propodal length. Auxiliary claws minute. Measurements, in mm: Trunk length (chelifore insertion to tip fourth lateral processes) ___ 2.68 Trunk width (across first lateral processes) —...-------------2------- 1,92 Proboscis; lengthy (ventrally) 22) 4 ee ee 1.95 Abdomen length (laterally) ) EEE 48 Third leg: Tibia] 2 EEE 2.77 (Xo es Laie ee ey ew INN 48 Tibia+2: 2 a 2.68 Coxar 2 Ue aus ee eee 1.86 Tarsus ) 28.20.05... 27 Ca a ea Ue NN 87 Propodis, «=... aa 1.12 |S) 01 aa a Ree 2.82 Claw)... 32. oa .76 Distribution: Known only from Santa Catalina Island, southern California. Depth range: 86 m. Remarks: Hilton incorrectly published the type-locality as ALBA- TROsS station 4424, which is nearby, but labels with the type specimen give the locality and station number as shown above. Stock (1955: 242-243) compared this species with a very closely related species, A. erectus. The comparisons were correct except for the following two. The propodal sole spines are bent and concave on their ectal surface, but are not denticulate. The oviger segments are thin, but the 2nd segment is only 9 times its diameter, instead of 12 times as long. This species probably exists in several California collections as A. erectus, and has remained unrecognized because of Hilton’s sketchy preliminary description. Hilton (1942f: 73) listed “lateral spines and terminal spines on the ocular tubercle.” The lateral spines are missing and I can find no evidence of their ever having been present. The ocular tubercle spines (setae?) are missing because the tubercle tip Hilton’s sea spider types 201 has been broken off at some time. Verification must await recognition of this species in other collections. Anoplodactylus pacificus bears a superficial resemblance to A. gestiens in its long and widely spaced lateral processes, long neck and thin chelifores, and two heel spines on the thin propodus. It differs im- mediately from A. gestiens, in not having chela finger teeth, a straight cylindrical proboscis, and a low ocular tubercle. The type-specimen is well preserved, but missing some legs. Anoplodactylus portus Calman Anoplodactylus portus Calman, 1927: 405-408, fig. 103.—Stock, 1955: 238-239 [literature]. Anoplodactylus robustus Hilton, 1939a: 28-29 [non A. robustus (Dohrn, 1881) ].—Hilton, 1942b: 288-291, pl. 39.—Hilton, 1942f: 72.— Hedgpeth, 1941: 257 [key]. Anoplodactylus projectus Hilton, 1942d: 45-47, fig. 3—Hilton, 19428: oe Material examined: Anoplodactylus robustus, male and female, syn- types: Laguna Beach, southern California, coll. W. A. Hilton. USNM 79416. Remarks: Hilton’s well preserved specimens of A. robustus agree in all particulars with Calman’s figure 103 except that the female dorsodistal tubercle on the femur is slightly smaller, and are clearly Calman’s species as Stock (1955) points out. The female Hawaiian specimen of A. projectus was not seen, but was examined and synon- ymized by Stock for a work on Hawaiian pycnogonids soon to be pub- lished (Stock, in press). Anoplodactylus robustus (Dohrn) Phoxichilidium truncatum Hilton, 1942d: 48-49, fig. 5—Hilton, 1942f: 71.—Stock, [in press]. Anoplodactylus robustus.—Stock, 1955: [literature]. Material examined: Male, holotype: Oahu, Hawaiian Islands, coll. C. H. Edmondson?. BPBM 4691. Remarks: Close examination of Hilton’s broken up specimen reveals it to be A. robustus. Hilton listed a 6-segmented oviger, but may have confused the 3rd segment constriction with a segmentation line. Both ovigers are broken off at either the base or second segment and are missing. Hilton also described the propodus as having three heel spines, but all legs available have two propodal heel spines with a single seta distal to the spines. The specimen has been squashed under cover glass at some time. Hilton listed five other specimens in the type lot and another from Kawailoa, a beach on NW Oahu Island. None of these specimens were examined. 202 Proceedings of the Biological Society of Washington Anoplodactylus viridintestinalis (Cole) Phoxichilus compactus Hilton, 1939a: 35.—Hedgpeth, 1941: 254 [key]. Endeis compacta Hilton, 1943b: 19.—Hedgpeth, 1952: 430. Anoplodactylus viridintestinalis —Stock, 1955: 239 [literature]. Material examined: Female, holotype: Dillon Beach, Marin County, California, on bryozoans, coll. O. Hartman, 20 December 1934. USNM 81529. Remarks: Hilton was confused about this specimen. In 1939 he de- scribed it as Phoxichilus compactus, new species, and again in 1943 described it as another new species, Endeis compacta. The specimen was to be deposited in the USNM collections in 1939, but was withheld until after he had redescribed it in 1943. The single specimen is badly damaged, with the ocular tubercle and chelifores lacking. The remainder of the specimen conforms exactly with A. viridintestinalis. Hedgpeth (1952) was the first to recognize the specimen for what it is. It comes from Cole’s type-locality. Phoxichilidium micropalpidum Hilton Figure 5 Phoxichilidium micropalpidum Hilton, 1942f: 72.—Hedgpeth, 1949: 283 [text]. Material examined: Male, holotype: Off Cape Monati, Bering Island, Commander Islands, Russian Arctic, 54°36’15”N., 166°57’15”E., 132 m, coll. ALBaTRoss, sta. 4792, 14 June 1906. USNM 81522. Description: Trunk completely segmented, lateral processes separated by slightly less than their diameter, without tubercles, armed distally with 1-2 setae on anterior and posterior surfaces. Ocular segment with 2-3 setae laterally just posterior to chelifore insertion. Eye tubercle cylindrical with conical apex. Eyes slightly pigmented. Proboscis cylindrical, half trunk length, slightly constricted behind mouth, oral surface convex. Abdomen small, glabrous, blunt at tip. Scape downcurved, finely pilose distally, armed with few short setae dorsodistally. Chela of moderate size, fingers overlap when closed, palm with many fine setae distally, fringe of fine setae on ectal surface of movable finger. Fingers without teeth. Palp a small bud anterior to and above oviger insertion. Oviger 5-segmented, 3rd segment longest, constricted near proximal end, with scattered ectal and endal setae. Second and 5th segments equal, setose distally. Fifth segment with linear ectal, lateral, and endal setae, mostly recurved, terminal segment slightly spatulate. Third leg long, surfaces pilose with tiny setae. First coxa slightly over half length of 3rd. Second coxa longer than Ist and 8rd together, with low dorsal swelling toward proximal end and larger ventrodistal swelling with genital pore. Femur with 4—5 low tiny ventral tubercles, Hilton's sea spider types 203 Fic. 5. Phoxichilidium micropalpidum, holotype: a, Dorsal view of trunk; b, Lateral view of trunk; c, 3rd leg, with enlargement of one femoral cement gland; d, Terminal segments of 3rd leg; e, Oviger; f, Chela. 204 Proceedings of the Biological Society of Washington each bearing a short seta. Femoral cement glands a single row of 18 cribriform discs on almost entire dorsal length of femur. Tibiae with 7-8 low tiny dorsal tubercles with single short setae. Tibia 2 one- fourth longer than tibia 1. Terminal segments small; tarsus short, with single ventral spine. Propodus slightly curved, armed with 3 heel spines and 9-10 shorter sole spines, entire surface finely pilose. Claw moderately long, auxiliaries slightly less than half main claw length. Measurements, in mm: Trunk length (chelifore insertion to tip fourth lateral processes) _ 4.08 Trunk width (across first lateral processes) -....------------- Qt Proboscis Jength(laterally) ..2 0 eee 2.04 Abdomen: Jength, (aterally)) 22) 2 eee m5) Third leg: Tibiat] 23 5.06 Coxe gly at ekeet rina ane 0.78 Tibia, 2) ue)" 6.57 Coxaven) | a saeimnab es. 2.47 Tarsus eee 32 WOoxa 3. oP ee Wes hy/ Propodus. 2... 1.53 Pemur™ 2 ee Pe 6.14 Claw: . 0 73 Distributions Known only from its type-locality, Bering Island, Bering Sea. Depth range: 132 m. Remarks: Subsequent Russian records do not, to my knowledge, list this species or one like it as having been taken since the time of the Steamer ALBATRoss. Hilton listed the type-locality as the California coast and the type as a female measuring 1.9 mm long. His description is relatively accurate, but his locality and measurement beg the ques- tion of just what specimen he was looking at when he listed these data. This species bears a close resemblance to P. femoratum, and several others to a greater or lesser extent; P. capense Flynn, P. quadradentatum Hilton, P. tubulariae Lebour, and P. ungellatum Hedgpeth. Phoxichili- dium micropalpidum differs from P. femoratum in the following re- spects: a longer and less curved propodus with three instead of five (sometimes four) heel spines and more sole spines; a much more setose propodus; and auxiliary claws fully half the main claw length, whereas with P. femoratum, they are usually % or Y% the main claw length. Claw length varies, and Lebour (1945: 148, fig. 2a) figures auxiliary claws of 1% the main claw length on her British specimens. Most of these com- parisons only touch lightly on diagnostic characters and would possibly fall within the normal variability of P. femoratum, if it were completely known. Hilton’s type is much more pilose over all than are specimens of P. femoratum examined for comparison, and its size is larger. The movable finger of P. femoratum is consistently more swollen than the fingers of Hilton’s type, and the palp tubercles are smaller and of a different shape. The conical capped ocular tubercle is taller on P. micropalpidum, and the lateral processes are without tubercles. Several specimens examined of P. femoratum have small distal lateral process Hilton’s sea spider types 205 tubercles whereas others have none. This illustrates the disadvantage of having only one type-specimen rather than a type-series for comparison. The remaining characters of the two species are similar and it may be that when more of Hilton’s species are taken from the type-locality, it can be shown that the above variation will fall within that of P. femoratum. Phoxichilidium parvum Hilton Figure 6a Phoxichilidium parvum Hilton, 1939a; 28.—Hedgpeth, 1941: 257 [key]. —Hilton, 1942f: 71-72. Phoxichilidium hokkaidoense Utinomi, 1954: 4-7, figs. 2-3, Pl. I, fig. 1. Material examined: Male, holotype, USNM 81521, 18 male, female, and juvenile paratypes, USNM 124023: Three miles N of Santa Cruz, Santa Cruz County, California, coll. M. W. Williams, December, 1938. Remarks: Utinomi has adequately figured this species as P. hok- kaidoense, except for possibly the dorsal aspect. A figure of this is in- cluded here. Utinomi found as much variation in his specimens as is shown in the above type-lot. The holotype propodus more closely resembles his figure 3B (p. 6), while the oviger lacks the small terminal segment and is the same as figure 2H (p. 5). The leg has three dorsal femoral cement glands as in Utinomi’s figure 2F, and the leg is only slightly smaller than his measurements (p. 7). The second tibia of the type is subequal to tibia 1 instead of being 0.9 times its length. The type-lot males have either five or six oviger segments, a factor that may be related to age. Hilton erroneously published the type-locality as Vera Cruz, Cali- fornia. Williams (in USNM correspondence) corrected this to Santa Cruz, California. The type is partly squashed. Phoxichilidium quadradentatum Hilton Phoxichilidium quadradentatum Hilton, 1942f: 71.—Hedgpeth, 1963: 1336-1337, fig. 9. Phoxichilidium quadridentatum.—Hedgpeth, 1954: 204-205, fig. 93c. —MacGinitie, 1955: 171. Material examined: Female, holotype: Stewart Island, Alaska, 50°23.5’N., coll. W. Williams, 4 August 1937, Lewis Expedition. USNM 81520. Remarks; Latitude 50° N is not in Alaska, but corresponds with the latitude of Stuart Island, Strait of Georgia, British Columbia, Canada. There is also a Stuart Island, Bering Sea, Alaska, but Stewart Island eludes my geographical searches. Either location falls well within the species known distribution from San Francisco, California, to Point Barrow, Alaska. This species has been figured by Hedgpeth (1954, 1963, 1964), 206 Proceedings of the Biological Society of Washington Fic. 6. a, Phoxichilidium parvum, holotype: Dorsal view of trunk; b, Phoxichilidium sp.: terminal segments of 3rd leg; c, d, Pycnosoma strongylocentroti: c, Dorsal view of trunk; D, terminal segments of 3rd leg. Hilton’s sea spider types 207 but his propodus figure does not show the field of many median and lateral sole spines. The spines are mostly short and thick and extend proximally to the tarsus. Other specimens identified by Hilton and Hedgpeth were examined for comparison. The four heel spines appear to be constant, but the auxiliary claws vary from minute to just less than %4 the main claw length. The femur has from 17 to 18 tiny dorsal cement glands in a single row. Phoxichilidium sp. Figure 6b Anoplodactylus unospinus Hilton, 1942f: 73.—Stock, 1955: 243 [text]. Material examined: Juvenile male, holotype: Laguna Beach, southern California, coll. W. A. Hilton, low tide, August, 1917. USNM 81530. Includes another juvenile, paratype. Remarks: It is possible this is a new species. The holotype is prob- ably one moult away from an adult, and has most of the characters of the genus Phoxichilidium. The ovigers are about % formed, The specimen could be a juvenile P. quadradentatum, except for the single heel spine. The propodus should be fully formed by this late juvenile stage. There is no propodal lamina. Hilton lists this species as if he described it in 1939, but no literature has been found from that year in which A. unospinus is listed. This may be a lapsus due to haste in publishing new species. Pycnosoma strongylocentroti Losina-Losinsky Figure 6c, d Pycnosoma strongylocentroti Losina-Losinsky, 1933: 43-47, fig. 1— Hedgpeth, 1947: 7 [footnote 2]—Hedgpeth, 1949: 239 [text].— Losina-Losinsky, 1961: 86. Pigrogromitus robustus Hilton, 1942c: 40. Material examined: Female, holotype: Captains Harbor, Unalaska Island, Aleutian Islands, Alaska, 16 m, coll. William H. Dall, 1874, no. 850( 1189). USNM 13358. Remarks: Losina-Losinsky has adequately illustrated (p. 45, Fig. 1) this species, but her figures do not show trunk segmentation nor do they emphasize the peculiar propodal heel spines. I have included figures of the trunk dorsally, and propodus for this reason, and because her paper is not always available to western specialists. Hilton’s speci- men is very robust, as his proposed name suggests, and is entirely glabrous except for the heel spines. The species superficially resembles Pigrogromitus in some particulars. Both genera lack palps, are robust with closely spaced lateral processes, have short and thick leg segments, and lack auxiliary claws. Hilton’s specimen, as Hedgpeth (1947) pointed out, is not in the same genus because it has ovigers of six segments rather than 10, it has a one-segmented scape instead of two- 208 Proceedings of the Biological Society of Washington segmented, and a curved propodus with a very broad heel bearing six extremely broad blunt spines shaped like shovels. This last character alone is sufficient to distinguish P. strongylocentroti (and Losina- Losinsky’s other species, P. tuberculata) from any other pycnogonid known to me. Losina-Losinsky’ss types from Tatary Strait, Russian Siberia, and Hilton’s specimen from Unalaska, appear to be the only records of P. strongylocentroti. The two recorded depths are 16 and 75 m. LITERATURE CITED Catan, W. T. 1927. Zoological Results of the Cambridge Expedi- tion to the Suez Canal, 1924. XXVIII. Report on the Pycno- gonida. Trans. Zool. Soc. London . 22(3):403-410, figs. 102— 104. HepcretH, J. W. 1941. A key to the Pycnogonida of the Pacific Coast of North America. Trans. San Diego Soc. Nat. Hist. 9(26) :253-264, pls. 9-11. ——. 1947. On the evolutionary significance of the Pycnogonida Smithson. Mise. Coll. (no. 3866) 106(18):1-53, 16 figs., 1 pl. ———. 1949. Report on the Pycnogonida collected by the Albatross in Japanese waters in 1900 and 1906. Proc. U.S. Nat. Mus. 98 :243-321, figs. 18-51. — —. 1952. Class (or Subphylum) Pycnogonida, sea spiders. In: Ricketts and Calvin, Between Pacific Tides, 3rd ed., 430 pp. —. 1954. Class Pycnogonida. In: S. F. Light’s Manual of inter- tidal invertebrates of the Central California coast, R. I. Smith, ed. Univ. Calif. Press: i-xiv, 1-466, 138 figs. Second ed. ———. 1963. Pycnogonida of the North American Arctic. Jour. Fish. Res. Bd. Canada 20(5):1315-1348, figs. 1-12. Hitton, W. A. 1914. The central nervous system of the Pycnogonid Lecythorhynchus. Pomona Jour. Ent. Zool. 6(3):134—136, 1 fig. — —. 1915a. Pycnogonids collected during the summer of 1914, at Laguna Beach. Ibid 7(1):67—70. March. —. 1915b. Pycnogonids collected during the summer of 1915, at Laguna Beach. Ibid 7(3):201-206. September. —. 19l6a. A remarkable pycnogonid. Ibid 8(1):19—24, figs. 1-6. March. ——. 1916b. The life history of Anoplodactylus Erectus Cole. Ibid 8(1): 25-34, figs. 1-6. March. — —. 1916c. The nervous system of pycnogonids. Jour. Compar. Neurol. 26(5) :463—473, figs. 1-21. October. — —. 1920. Notes on Pacific Coast Pycnogonids. Pomona Jour. Ent. and Zool. 12(4):93. — —. 1934. Notes on parasitic pycnognids [sic]. Ibid, 26(4):57. —. 1939a. A preliminary list of pycnognids [sic] from the shores of California. Ibid 31(2):27-35. June. Hilton’s sea spider types 209 1939b. A collection of pycnognids [sic] from Santa Cruz Island. Ibid 31(4):72-74, 2 pl. December. —.. 1942a. Pantopoda chiefly from the Pacific. 1—Nymphonidae. Ibid 34(1):3-7. March. — —. 1942b. Pycnogonids from Allan Hancock Expeditions. Allan Hancock Pacific Expeds. 5(9):277-339, 14 pl. March. ——. 1942c. Pantopoda (Continued). II—Family Callipallenidae. Pomona Jour. Ent. and Zool. 34(2):38-41. June. ——. 1942d. Pycnogonids from Hawaii. Occ. Pap. Bernice P. Bishop Mus. 17(3):43-55, figs. 1-10. July. ——. 1942e. Pycnogonids from the Pacific. Family Tanystylidae. Pomona Jour. Ent. and Zool. 34(3):69-70. September. ——. 1942f. Pycnogonids from the Pacific. Family Phoxichilididae [sic] Sars 1891. Ibid 34(3):71-74. September. ——. 1942g. Pycnogonids from the Pacific. Family Ammotheidae. Ibid 34(4):93-99. December. —. 1943a. Pycnogonids from the Pacific. Family Colossendeidae. Ibid 35(1):2—4. March. ——. 1943b. Pycnogonids of the Pacific. Family Pycnogonidae. Family Endeidae. Ibid 35(2):19. June. Lesour, M. V. 1945. Notes on the Pycnogonida of Plymouth. Jour. Mar. Biol. Assoc. U.K. 26:139-165, 7 figs. Lostna-Losinsky, L. K. 1933. Pantopoda vostochnykh morei SSSR. [Die Pantopoden der éstlichen Meere der USSR.] Issledovanie Morei SSSR 17:43-80. [With German summary: 74—79]. ——. 1961. Mhnogokolenchatye (Pantopoda) dalnevostochnykh morei SSSR. [Pantopoda of the far eastern seas of USSR.] Issledovanie Dalnevostochnykh Morei SSSR 7:47-117. MacGrnitig, G. E. 1955. Distribution and ecology of the marine in- vertebrates of Point Barrow, Alaska. Smithson. Misc. Coll. (no. 4221) 128(9):1-201. Marcus, E. 1940. Os Pantopodia brasileiros e os demais sulameri- canos. Bol. Fac. Fil., Ciénce. Letr. Univ. Sao Paulo (19) (Zool. 4):3-179, pls. 1-17. Marcus, E. pu Bors-ReEyMonp. 1952. A hermaphrodite pantopod. Anais Acad. Bras. Ciénc. 24(1):23-30, figs. 1-9. Stock, J. H. 1955. Pycnogonida from the West Indies, Central America, and the Pacific Coast of North America. Vidensk. Medd. Dansk Naturh. Foren. 117:209-266, figs. 1-26. 1968. Pycnogonida collected by the GALATHEA and ANTON Bruun in the Indian and Pacific Oceans. Ibid 131:7-65, 22 figs. —. (Impress). Pycnogonida. In: Shore and reef fauna of Hawaii, 2nd ed. Utinomi, H. 1954. The fauna of Akkeshi Bay. XIX. Littoral Pycno- gonida. Publ. Akkeshi Mar. Biol. Sta. 3:1-28, 11 figs., 1 pl. 210 Proceedings of the Biological Society of Washington Lae} Yo. 20, pp. 211-216 15 August 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW GENUS AND SPECIES OF EURYURID MILLIPEDS FROM CHIAPAS (POLYDESMIDA: PLATYRHACIDAE) By RicHarp L. HorFMAN Radford College, Radford, Virginia 24142 _— ee In recently working through the euryurid material in my collection during revisionary studies, I came across two speci- mens which had been sorted out and tentatively identified as referable to the genus Polylepiscus (on the basis of external appearance) nearly ten years ago. When the male was un- rolled and the genitalia examined it was immediately obvious that a new generic type was represented, a conclusion rein- forced by the remarkable modifications of the anterior sterna. Since this species is excluded from the genera Amplinus and Pycnotropis, both currently being monographed, it is here described separately. As I have indicated elsewhere (in press), the euryurid milli- peds might best be considered a subfamily of the Platyrhacidae instead of a separate family. The known genera of this sub- family fall into three relatively discrete groups which were treated as subfamilies in my 1954 synopsis. Of these taxa, now regarded as tribes, one (Aphelidesmini) is monobasic, one (Euryurini) contains two genera, while the third is much larger and comprehends no fewer than eight genera in the Neotropical Region. It is this third tribe, Amplinini, that the new genus finds its place, in the general vicinity of Polylepiscus and Pycnotropis. The specimens described here are part of a large and ex- tremely important collection of Mexican diplopods assembled in 1965 and 1966 by Drs. George E. Ball and Donald R. White- head during their own studies on Mexican carabids. I am very 20—Proc. Biou. Soc. WAsH., Vou. 88, 1975 (SAUL) 212 Proceedings of the Biological Society of Washington much indebted to these assiduous collectors for their gen- erosity in placing this valuable material at my disposal. Genus Exallostethus, new genus Type species: E. thrinax, sp. nov., from Mexico (Chiapas). Diagnosis: Distinguished from the superficially similar genera Amplinus, Polylepiscus, and Pycnotropis (and all other known euryurid genera) by the singular form of the gonopods and by the presence of enormous median processes on the sterna of segments 4 and 5 in males, and 4 in females. Collum without anterior marginal rim, likewise a unique character in this group. Stigmata dissimilar in size, posterior stigmata smaller and set obliquely between dorsal coxal condyles. Gonopods without sclerotized median sternal remnant, but with a prominent thickened mass of membrane between the coxae; prefemur relatively short, its median face with greatly thickened setae, almost spurs, in addition to the usual slender hairs; acropodite elements elongated, much longer than prefemur, solenomerite very long, slender, and nearly straight, apically flattened and twisted; distal end of tibio- tarsus enlarged and divided into two subsimilar apically acute lobes; postfemur with a long, slender, distally falcate process, another unique feature within the subfamily. Distribution: Known so far only from the mountains of central Chiapas. Etymology: The name is derived from the Greek exallos, utterly dif- ferent + stethos, chest or sternum; bestowed in token of the remarkable modification of the anterior sterna in both sexes, not approximated in any other known euryurids. Gender masculine. Exallostethus thrinax, new species Figures 1-4 Diagnosis: With the characters of the genus. Females similar to those of Polylepiscus trimaculatus (Hoffman, 1954) in general ap- pearance but differ in that the posterior corners of the paranota of segments 3 and 4 are rounded instead of acute, the sides of the metaterga are coarsely granulate only along anterior and posterior margins, and in the presence of a prominent erect median process on the sternum of the 4th segment. Type-material: Male holotype and female paratype from Highway 195, 11.6 miles north of Pueblo Nuevo, 5200 ft., Chiapas, Mexico; April 27, 1966; G. E. Ball and D. R. Whitehead, legg. (Hoffman Collection ). Holotype: Adult male, ca. 55 mm in length, 9.3 mm in maximum width, W/L ratio, 17.0%. Body increasing gradually in width from collum back to posterior 2% of body, thence narrowing abruptly over the last few segments as shown by the following dimensions: New milliped from Chiapas 213 Segment 1-6.8 mm Segment 10-9.2 mm 2-7.6 12-9.3 4-8.3 14-9.3 6-9.1 16-8.8 8-9.1 18-7.1 Color of preserved specimen: dorsum of metaterga, most of head, and 6th antennomere rich maroon to chestnut brown; entire upper and lower surfaces of paranota and large median metatergal spot (median band on collum) nearly white; labrum and first four antennomeres yellowish; sterna light yellowish brown; legs uniformly brown; prozona and lower sides grayish. Head moderately convex, smooth and polished, epicranial groove deep and prominent, bifurcate between antennae, its branches con- tinuing ventrolaterad to ventral edge of antennal sockets. Latter set off below and laterally by a deep depression, subtended on each side by a large and prominent genal convexity, the median end of which is formed by a distinct vertical groove from lower edge of antennal socket. Lower half of genae planoconvex, without evident lateral mar- gin, edge shallowly emarginate. Side of head behind antennae con- spicuously ridged and grooved vertically. Frons elevated above level of genae and labroclypeus, in anterior aspect head appearing to be provided with two epicranial, a large triangular frontal, and two oval convex genal prominences. 0-0 epicranial setae; 1-1 interantennal; 1-1 widely spaced frontal; 4—4 clypeal, the outermost on each side remote from the median series of six; 10-10 labral; all setae multiple. Antennae of moderate length (6.6 mm) and proportions, extending back to caudal edge of paranota of 3d segment. Articles 2-5 sub- clavate, enlarged apically, 6th cylindrical, 7th small, truncate-conic in shape. Length relationships: 3>4=5=6>2>1>7. Collum evenly arched, laterally depressed, entire surface smooth and polished except for very indistinct sublateral corrugation; lateral ends symmetrically rounded; anterior margin without trace of groove, surface continuous with edge. Collum appreciably narrower than width of second segment. Body segments generally similar in structure; prozona and metazona separated by an evident ‘waist’ containing well-defined, faintly costulate stricture. Metazona dorsally slightly elevated above level of prozona. Surface of both subsegments smooth and polished middorsally, upper surface of paranota with several low and distinctly defined convex areas. Scapulorae marginal, those of segments 8 to 18 with an indistinct single row of minute, widely-spaced spicules forming a primitive strigil; peritreme strongly thickened near midlength, ozopore opening laterally in a large concavity (seen in dorsal aspect peritreme appears notched). Posterior corner of paranota right or obtusely angled back to segment 5, thence becoming increasingly acute and prolonged into a spiniform 214 Proceedings of the Biological Society of Washington process which on segments 12-15 is incurved. Posterior edge of paranota concave, with a single row of widely-spaced spicules extending in some cases to apex of peritrematic process. Epiproct spatulate in outline, sides very slightly arcuate but con- verging distally; caudal edge nearly transverse, with six small lobes, of which the median two carry two pairs of long setae, one pair placed apically and the other on the ventral side somewhat removed from the rear edge. Paraprocts of the usual form, distinctly convex, surface wrinkled-striate, with polished median discal knob. Hypoproct large, tumid, subhemispheric in outline; apically truncate; setiferous tubercules absent, paramedian setae originate from sockets on edge of sclerite. Podosterna strongly elevated, anterior and posterior faces nearly vertical; each divided by a deep transverse groove into two halves each with a shallower, broader longitudinal median depression, the quadrants thus formed not notably produced as subcoxal lobes or pro- jections; sternal surface smooth and glabrous. Stricture broad and poorly-defined down sides of segments, anterior edge fairly distinct across midventral region. Sides of metazona with an area of enlarged acute tubercules just behind stricture and another near caudal margin, beginning on underside of paranota and extending to supracoxal region where broadened and with tubercules larger and denser. Central area of metazonal sides with only a few small and scattered tubercules. Stigmata prominent, elongate and very slender, anterior stigmata placed entirely in stricture, in front of dorsal coxal condyle but well removed from it, rim only very slightly raised; posterior stigmata about half as long as anterior, placed between dorsal coxal condyles and slanted obliquely forward, edges slightly elevated with dorsal end somewhat expanded and subauriculate. Dorsal condyles relatively large and prominent, projecting laterally. Legs relatively large and robust, but short, only the three distalmost podomeres extend beyond edges of paranota, length at midbody about 7.8 mm, tarsal claw large, curved, about 0.8 mm in length. Podomeres nearly glabrous except for sparsely scattered microsetae, and with macrosetae dispersed as follows: one on ventral side of coxa, pre- femur, femur, and postfemur, ventral side of tibia with two, and of tarsus with three; tibia with one lateral macroseta on each side, and apical half of tarsus with about 10-12 dorsolaterally placed setae. Rela- tive lengths of podomeres: 3 >6>2>1=4>5. Anterior sterna as shown in Fig. 1: a large median process between > Fics. 1-4. Evxallostethus thrinax, sp. n. 1, sternal region and bases of legs of segments 4, 5, and 6, showing sternal modifications and reduc- tion of ventral coxal setae (small numbers at right refer to the leg pair). 2, gonopods, ventral (aboral) aspect, as would be seen in situ. 3, left gonopod, mesal aspect. 4, distal half of telopodite of right New milliped from Chiapas 215 My ‘ ; i ‘ fy Ey aw gonopod, dorsal (oral) aspect, to show relationships of terminal branches, enlarged. Figures 1-3 drawn x 45, Figure 4, x 90. Abbreviations: PFP, postfemoral process; SLM, solenomerite, TT: and TT2, apical lobes of the tibiotarsus. 216 Proceedings of the Biological Society of Washington legs of the 3rd pair, its surface closely and coarsely striate and with deep vertical groove on posterior side; segment 5 with massive median projection between 4th legpair and a low median ridge between 5th legs; sternum of segment 6 broad, medially planoconcave with prom- inently developed subcoxal lobes. Legs: Ist and 2nd pairs relatively setose, 3rd pair with only a field of ventral setae on coxae and several dispersed hairs elsewhere; coxal setae decrease as shown, to only one large seta on 6th pair. Anterior legs unmodified. Gonopod aperture small, oval, anterior edge reaching to stricture but slightly elevated above it; posterior edge somewhat elevated and thick- ened. Gonopod large, elongate, projecting forward to sternal process of segment 5. No sclerotized sternal remnant present, but coxae sepa- rated by a large median lobe of whitish heavy membrane; sternal apodemes slender, slightly longer than coxae. Gonopods as shown in Figures 2—4; prefemoral region short and set medially with unusually stout setae, almost spurs, of a form not found elsewhere in the Euryurinae; acropodite much longer than prefemur, composed of a long, attenuated, distally ribbon-like solenomerite (SLM), a large, apically broadened and falcate postfemoral process (PFP) like- wise unique in this group; and a medially curved tibiotarsus, apically divided into two similar uncate laminate terminal lobes. Paratype: Adult female, length about 60 mm (specimen broken), body outline similar to that of male, i.e., very narrow anteriorly, becom- ing broader back to about 14th segment, thence tapering abruptly posteriad. Width of collum, 7.2 mm; segment 2, 8.0 mm; segment 4, 9.0 mm; segment 6, 10.0 mm; segment 10, 10.2 mm; segment 14, 10.5 mm. W/L ratio, 17.5%. Color pattern similar to that of male, but median spots appreciably larger, occupying up to half of metatergal middorsal region. Structure generally as in male except: anterior edge of collum with marginal ridge indicated for a short distance just behind mandibular condyle; tergal sculpture more pronounced, upper side of paranota distinctly areate; costulation of stricture more distinct; posteriormost segments with low but evident transverse series of small tubercules; paranota appreciably smaller than in male but otherwise similar; legs more slender; sternum of 4th segment with an erect median process, somewhat smaller than that of male; sternum of 5th segment with two approximate para- median knobs between 4th pair of legs, and two lower, more widely spaced knobs between 5th pair. Etymology: The specific name is the Greek word for a three-parted implement, and relates to the three distal elements of the gonopod telopodite. LITERATURE CITED HoFFMAN, RicHAarD L. 1954. Further studies on American millipeds of the family Euryuridae (Polydesmida). Jour. Washington Acad. Sci. 44:49-58, figs. 1-4. mt i Tee MEL WPocacedtiae glia ag ie EUR aaa AT Poa ems a Troe wart ewey 7 i Vi f tet vie J ai ae 5 F 0 Wl one w Gk CTA f : ' i ov 7 ; ~ ve ! % ' Pes a) ® ae | PA] he Thun 1 eee a Scale ye, Aaa ta: Se - tek hr, Tee m me y Pra ! y v ; yes, Ae deeh Hi { ! v : ) : T i i | i i ‘ ( 0 Agcy ; { : LGieisiidy en ut Ss ts j \ i ; | \ } i € % bE “fi t " e i i i F 3 7 Hie b bi vi . he nan AP taps RS ha al abi ge i Cygne: 9 hea ae Nhs : Lei he, RY Py ®) i Wel i wee i fi ! 1 a 1 0. 21, pp. 217-232 15 August 1975 BAX NH PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON TAXONOMIC NOTES OF THE GENUS ONCAEA (COPEPODA: CYCLOPOIDA) FROM THE GULF OF MEXICO AND NORTHERN CARIBBEAN SEA By Frank D. FERRARI Smithsonian Oceanographic Sorting Center Washington, D.C. 20560 \\ —7 = 2h t+ 3 i 7 Copepods of the genus Oncaea are small cyclopoids between ; 0.2 mm and 1.4 mm long. They are abundant in the epipelagic zone of the tropical oceans and although they are often re- garded as free-swimming members of the zooplankton com- munity, the minute structure of the mouthparts (Sars, 1918), the reports of food procurement for some members of the genus (Wickstead, 1962; Heinrich (=Geinrich), 1957; Alldredge, 1972), and for the ecologically related genus Sapphirina (Heron, 1973), indicate that they probably main- tain semi-parasitic relationships with other members of the zooplankton community. Because of their small size, the ab- sence of obvious morphological differences, and the intra- specific size variability, the species are difficult to identify. In this paper the common larger species of the Gulf of Mexico and Northern Caribbean Sea are described and some of the variability encountered in these species is discussed. Forty-one zooplankton samples were collected in July, 1969, September, 1971, and May-June, 1972, from the upper 500 m in the area beyond the continental shelf in the Gulf of Mexico and Northern Caribbean Sea (hereafter the Gulf area). The mesh width of the finest nets used was 216 microns. Problems with retention limited the study to the larger members of the genus Oncaea but a single Bongo-net tow (mesh width 183 microns) taken in March, 1973, from the northeastern Gulf 21—Proc. Biot. Soc. WAsH., Vou. 88, 1975 (217) 218 Proceedings of the Biological Society of Washington of Mexico, provided females of three small species. Much of the information given here, including comprehensive station details, were initially reported in a Ph. D. dissertation (Ferrari, 1973). Specimens were measured and dissected in latic acid and transferred to glycerol to be drawn with the aid of a camera lucida. Fixing with formaldehyde causes various degrees of flexation of the prosome and urosome and telescoping of the urosome segments in Oncaea. By positioning the animal for measurement, errors caused by flexation can be avoided. Adjustment for telescoping of segments is more tedious. The total length of a specimen was calculated simply as sum of the measurements of the prosome and urosome and was not adjusted for telescoping of segments. However measure- ments made for comparing the relative lengths of different abdominal segments were adjusted for the telescoping effect. Oncaea conifera Giesbrecht 1891 Figures 1A—2G Females of Oncaea conifera (length 1.01-1.26 mm) represent one of the four large species (total length greater than 0.70 mm) in the Gulf area. These females and those of O. similis can be distinguished from all other females by the distal conical projection of third segment of the endopod of the fourth swimming leg. Greater total length and, generally, some development of a dorsal projection on the second pedigerous seg- ment separates O. conifera from O. similis. Males of O. conifera were the only males found in this study with a conical projection on the third segment of the endopod of the fourth swimming leg. Males of O. similis share this trait but were not collected. For purposes of identi- fication, Olson (1949), who first described the males of O. similis, states that the terminal seta of the third endopodal segment of the second swimming leg is more prominent than the two external setae. In the males of O. conifera this terminal seta is less prominent. O. conifera poses the most complex problem of variation of any species in the genus. Farran (1936) described size variation from the more limited area of the Great Barrier Reef. He divided 40 females into three groups one of which he called a variety. These three groups were often found in the same sample. Moulton (1973) described morphological variation of this species in the Indian Ocean. His study included a com- prehensive numerical analysis of 23 females from two stations. From this analysis he recognized four groups, three of which were referable to those of Farran. Specimens separated and counted from the Gulf area exhibited a wide Taxonomic notes on Oncaea 219 oe t GA eg SN ane mt Sy Frimaues oP Fic. 1. Oncaea conifera: A, Small male-female pair, lateral; B, Large male-female pair, lateral; C, Female Ur, dorsal; D, Male Ur, dorsal; E, Female Enp3 P4; F, Male Enp3 P4; G, Male A2; H, Female, Mxpd. degree of variation. Some individuals could be easily placed in either of Farran’s groups, others could not; no systematic analysis by depth, water mass, or season was attempted. However, during the study, at seven different stations, seven male-female pairs of O. conifera were found in what is considered a mating posture (Fig. 1A, 1B). In these instances the male had grasped the female in the area of the articula- tion of the genital segment and the preceding urosomal segment. Grasp- ing was done only with the large distal claws of the maxillipeds. In ad- 220 Proceedings of the Biological Society of Washington dition some accessory structures were found in one pair (Fig. 1B). The genital segment and the tip of the endopod of the fourth swimming leg of the female was encircled by a line composed of twisted strands. Both ends of this line passed outside and over the maxilliped of the the male and into the area of the mouthparts. From this area both lines passed tautly back to an irregular mass of tubules or cells ventral to the male urosome. One small line, originating from this mass, seemed to wrap loosely around the caudal ramus and caudal setae of the male and terminate in the mass. Another thicker line dangled from the mass. As these two animals were collected with a net, held in a catch bucket for a period of time until the tow was completed and then exposed to the shock of fixation with formaldehyde, it is difficult to assess the significance of the position of these structures. Since they have not been recorded previously they have been included here. Below is a list of the lengths (in mm) of the seven male-female pairs of O. conifera. females males 1.06* 0.60* 1.09 0.63 1.09 0.65 1.11 0.60 1.11 0.82 1.12 0.64 1.25+ 0.79+ *Fig. 1A; -Fig. 1B Had these 14 individuals occurred separated in a sample they could easily have been grouped into two size classes for both sexes (females: 1.06—1.12 and 1.25 mm; males: 0.60-0.65 and 0.79-0.82 mm). In six of the seven cases the data indicate that the females and males paired according to these size classes; the fifth pair is the exception. Incom- patability of size could be a premating isolation mechanism but cau- tion should be exercised in extrapolating from this mating posture to actual reproduction. The females of these pairs all have a well developed dorsal projec- tion of the second pedigerous segment and the body is well sclerotized. The posterio-lateral extensions of the last prosomal segment are parallel. The anal segment is not quite as long as the preceding two abdominal segments together and the caudal rami are set wide apart. Morphologi- cally these females seem to belong to the “a” group of Farran (1936) although their total lengths lie outside the range reported by him. They also fit the “stocky” group of Moulton (1973) which, Moulton 6c 9? notes, is very similar to the “a” group of Farran. Taxonomic notes on Oncaea 221 Fic. 2. Oncaea conifera: A, Female Ur, dorsal; B, Male Ur, dorsal; C, Female A2; D, Male A2; E, Female Mxpd; F, Female Enp P4; G, Male Enp P4; Oncaea media: H, Female, lateral; I, Male, lateral; J, Female Ur, dorsal, flexed dorsally. Two male-female pairs were dissected and certain structures drawn. Fig. 1C-H is the small female-small male pair illustrated in Fig. 1A; Fig. 2A—G is the large male—large female of Fig. 1B. The only notable difference between these two pairs is the slight serration on the second segment of A2 in both male and female of the large female—large male pair. 222 Proceedings of the Biological Society of Washington Oncaea media Giesbrecht 1891 Figures 2H-3D In the Gulf area the females of this species are easily separated from all others encountered by their deep blue coloration in reflected light. Farran (1929) remarks that females found around New Zealand were crimson while Sewell (1947) notes an orange-red color for some of his specimens from the northern Arabian Sea. The males are characterized, as are those of O. venusta, by the short curved spine in the proximal group of spines and setae of the third segment of second antenna (Fig. 3B). The males are separated from those of O. venusta by their smaller size and smaller length of the caudal ramus relative to the anal segment (compare Fig. 3A with Fig. 6B). Females of O. media were collected extensively in the Gulf area in July, 1969. The individuals were referable to the major form of Sewell (1947) on the basis of length range, 0.72-0.81 mm. The species was not found in September, 1971, or May—June, 1972, but was found again in the single sample collected in March, 1973. The length range of these specimens, 0.73-0.93 mm, exceeded that of the specimens col- lected in July, 1969. Below is a list of lengths for a sample of 30 fe- males from each year. Length (mm) 72) 73) 74715 16 77.7.8) 99) SO Glee 1969 2 2) Bee, 6S oo) Ti eko pa 1973 Se ak Segoe Acre eee 1 gut Ze Length (mm) .83 84 85 .86 .87 .88 .89 .90 .91 .92 .93 1969 —- = FRR RSS 1973 2 - 3 1 1 i 1 1 1 82 Small individuals which could be ascribed to Sewell’s minor form, length range 0.58-0.65 mm, were not collected from the Gulf area at any time. The nets of the 1969 survey, mesh width 333 microns, may not have re- tained specimens of this size but if present, they should have been found in the sample of 1973, mesh width 183 microns, because smaller specimens of the genus were collected. A single male 0.63 mm long was found in the 1969 samples; the length range of 10 males from the sample of 1973 was 0.56—0.62 mm. Oncaea mediterranea Claus (1863) Figures 3E-5D Females of Oncaea mediterranea can be separated with O. venusta from other oncaeids by the length of the caudal ramus, more than 2.5 times longer than wide (Fig. 3G). Females can be separated from those of O. venusta by their longer fifth leg (Fig. 3G) which is re- duced in O. venusta (Fig. 5J), and by the structure of the maxilliped, especially the distal spine on the second basipod segment, which is Taxonomic notes on Oncaea 293 Fic. 3. Oncaea media: A, Male Ur, dorsal; B, Male A2; C, Female Mxpd; D, Female P4; Oncaea mediterranea typical group: E, Female, lateral; F, Male, lateral; G, Female Ur, dorsal; G, Male Ur, dorsal; I, Male A2; J, Female Mxpd; K, Female, Enp3 P4. toothed on O. mediterranea (Fig. 3J) but setose on O. venusta (Fig. 6C). Males of O. mediterranea and O. venusta have distinctive caudal rami that are more than 3 times longer than wide and longer than the anal segment. The two species can be separated by the presence of a seta instead of a curved spine on the third segment of the second antenna in O. mediterranea (compare Fig. 3I with 6D). 224 Proceedings of the Biological Society of Washington There are three distinct size groups of both males and females. of O. mediterranea from the Gulf area. The length range of the typical group, females 1.07-1.22 mm and males 0.76—0.89 mm, falls within the range for females reported by other authors (Giesbrecht, 1892; Rose, 1933; Farran, 1936). In samples from the Gulf area this group always makes up a relatively greater number of individuals than the large group. The length range of the larger group, females 1.29-1.37 mm and males 1.06—-1.12 mm, exceeds the upper limit usually reported for the females, although Olson (1949) reported the length of O. mediterranea from the east coast of the Pacific Ocean as 1.38 mm. There are no morphological differences in the appendages of these two groups but the larger group is easily distinguished by its greater total length, more heavily sclerotized exoskeleton, and greater amount of orange pigment on the extremities of the body and appendages. Farran (1929) reported “a small colorless form” of O. mediterranea in a sample taken south of New Zealand but gave no further description of the specimen. A similar group of O. mediterranea was collected abundantly throughout the Gulf area during the four years studied. In- dividuals of this third group are smaller, females 0.85-0.91 mm and males 0.69-0.75 mm, than previously reported for O. mediterranea. This small group differs from the preceding two in its lack of pigment and its more weakly sclerotized exoskeleton. Its caudal rami are parallel to one another, not divergent as in the other two groups (compare Fig. 3G with Fig. 5C). The caudal ramus length to width ratio is 3.0 to 3.5 while for the other two groups it is greater than 4.0; the ratio of caudal ramus length to anal length is 1.0 in the small group and 1.6 in the other two groups. In Fig. 4 measurements of 30 females, randomly selected from a general collection of O. mediterranea, are used to indicate these three separate groups. Males of the small group can be distinguished most easily by the shorter total length, lack of pigment and minimal sclerotization of the exoskeleton. The small group was tentatively treated as a new species by Ferrari (1973) because it differed in relative lengths of the body segments as well as in total body length, whereas the large and typical groups dif- fered only in total body length. However, other species of Oncaea, specifically O. conifera, also exhibit a great degree of relative size variation. Since this variation in O. mediterranea is not accompanied by structural differences in the appendages, it is not considered great enough to warrant recognition as a species. Oncaea similis Sars 1918 Figures 5E-H Due to its very small size Oncaea similis was collected only in the March, 1973, sample (mesh width 183 microns) and only females of this species were found. Separation of these females has been discussed under O. conifera. Taxonomic notes on Oncaea 995 ONCAEA MEDITERRANEA 99 (2) 2= = a ° = n 5) ° =F ‘ 5 i ol ° ® ce) c ar ° Q 2 © i 2 8 az 4 a fe} 5 = gs) @ a ° oS © oe Oo 3 = ° = < fe 2 n rn rs n bn ; 42 44 46 48 50 52 54 56 58 LENGTH OF ABDOMEN (MM X 10?) aaa ———F ee SMALL TYPICAL LARGE Fic. 4. Length of the abdomen plotted against length-width ratio from 30 females randomly selected from a general collection of Oncaea mediterranea from the Gulf area. Specimens from the Gulf area differ slightly from those described by Sars (1918) but agree well with the description of Olson (1949) for O. similis from the coastal waters of the eastern Pacific Ocean. Sars lists the total length of the adult females as 0.78 mm while those from the Gulf area are 0.62-0.65 mm. The genital segment does not gradually taper posteriorly from its widest to narrowest width; rather the anterior area is more rounded and the posterior portion distinctly cylindrical (Fig. 5F) as in Olson (1949). The second abdominal segment is longer than the third; the medial spine of P5 is much thicker than the lateral. The spines on the second basipodal segment of the maxilliped are dis- similar, the distal one being longer, stouter, and toothed. Oncaea venusta Philippi 1843 Figures 5I-6D The separation of Oncaea venusta has been discussed with O. media and O. mediterranea. Males of the former species and females of the latter are the only ones which might cause confusion. Two size groups were found in the Gulf area, females 0.92-0.99 mm, males 0.57-0.63 mm and females 1.10-1.20 mm, males 0.76-0.86 mm, throughout the period sampled. These correspond respectively to forma venella and forma typica described by Farran (1929) and noted by Sewell (1947). Although there are slight differences in the length ranges reported in all three cases, the presence of distinct size groups is consistant. On page 226 Proceedings of the Biological Society of Washington Fic. 5. Oncaea mediterranea new group: A, Female, lateral; B, Male, lateral;! C, female Ur, dorsal; D, Male Ur, dorsal; Oncaea similis: E, female, lateral; F, Female Ur, dorsal; F, Female Mxpd; H. Female, P4; Oncaea venusta typical group: I, Female, lateral; J, Female Ur, dorsal; K, Female A2. 228 is a list of the lengths of 30 females and 30 males randomly selected from a general collection of the species. Both groups were present in the Gulf area during all four years studied. In all stations the smaller group was by far the more abundant, which contradicts the findings of Farran (1929) from New Zealand and Sewell (1943) from the Arabian Sea. Both groups from the Gulf area, Taxonomic notes on Oncaea DAT F Fic. 6. Oncaea venusta typical group: A, Male, lateral; B, Male Ur, dorsal; C, Female Mxpd; D, Male, A2; Oncaea sp. 1: E, Female, lateral; F, Female Pr, dorsal; Oncaea sp. 2: G, Female, lateral; H, Female Pr, dorsal. when viewed under reflected light, are tinged with red on the ex- tremities of the body and appendages. This coloration agrees with speci- mens described by Farran (1929) from the Great Barrier Reef but Giesbrecht (1891), Rose (1933), and Owre and Foyo (1967) describe this species as purple. The following two females were recovered from the sample collected in the net with mesh width of 183 microns. In the keys of Rose (1933) and Shmeleva (1969), Oncaea sp. 1 and Onceae sp. 2 fall be- tween O. media and O. curta. 228 Proceedings of the Biological Society of Washington Length (mm) 93 94 95) 96 87 98 99. == IO mem females 3 3 5 6 3 2, 1 - 2: Length (mm) 12S V4 LIS 6) aie as eet Seog females 1 _ 1 1 1 - 1 - - Length (mm) 57 58 59 .60 61 62 63 — .76 77 males =- = 1 ee, 8 - = - Length (mm) 8 <9 .80" 481-82) (83) (S4eeoaeros males 1 ee 1 = ee Oncaea sp. 1 Figures 6E, F and 7A—D Length 0.68 to 0.71 mm. Prosome 68% of total length. Third segment of second antenna shorter than second. Spines of second basipodal seg- ment equal; both spines with a row of short, fine hairs. Fine hairs on medial margin of terminal hook of maxilliped. First three swimming legs as figured for O. curta in Sars (1916); fourth without terminal cone on endopod; third segment of its endopod with two lanceolate setae; the proximal external seta naked. Genital segment longer than rest of urosome; caudal ramus longer than anal segment. Urosome formula 8,53,8,6,8,17 = 100. This female resembles O. curta Sars (1916). Third segment of second antenna is proportionally shorter than second, as compared to Sars’ drawing. Other differences: the setae on second basipod of maxilliped are unequal, genital segment proportionally wider, genital segment longer than rest of urosome.. In the description of O. curta Sars states “. . . tail does not exceed half the length of the anterior division and having a shorter genital segment than O. media being in length equal to the rest of the tail.” However, Olson (1949) figures O. curta with the genital segment longer than the rest of the urosome. Until the males can be found and described no decision can be made as to whether this is a new species or another size group of O. curta. Oncaea sp. 2 Figures 6G, H and 7E-H Total length range of these females, 0.54-0.58 mm. Prosome twice urosome. Caudal ramus slightly longer than anal segment which is longer than third abdominal segment. Genital segment longer than rest of urosome. Urosome formula 6,53,10,7,10,14 = 100. Appendages simi- lar to those of Oncaea sp. 1 but smaller and narrower. These females are easily separated from those of Oncaea sp. 1 by the shorter total length, narrower prosome and urosome, and the less arched aspect of the prosome. DISCUSSION Due to their small size, variation in the species of the genus Oncaea is often overlooked. Several types of variation have been observed or Taxonomic notes on Oncaea 929 Fic. 7. Oncaea sp. 1: A, Female Ur, dorsal; B, Female, A2; C, Fe- male Mxpd; D, Female P4; Oncaea sp. 2: E, Female Ur, dorsal; F, Fe- male A2; G, Female Mxpd; H, Female P4. cited here. Color variation within the same species from different geographic regions is noted in O. media and O. venusta. Simple size variation, in which only the total length of the specimens varies, is found in O. media, O. mediterranea, and O. venusta from the Gulf area and reported for O. venusta by Farran (1929) and Sewell (1947) and for O. media by Sewell (1947). Relative size variation, in which the size of certain body segments varies in relation to one another, is demon- strated for O. mediterranea from the Gulf area and reported for O. conifera by Farran (1936) and Moulton (1973). Both kinds of size variation can produce definite size groups of specimens; these size 230 Proceedings of the Biological Society of Washington groups can be consistant over space and time, e.g., O. mediterranea and O. venusta, or change temporally, e.g. O. media. If these groups are different species, distinctions in appendage morphology should be noticable, especially in those appendages of the male involved in re- productive behavior. However, distinctions in the male mazxillipeds have not been noted so far. It is possible that the behavior of the male in securing the female with accessory reproductive structures (noted in O. conifera) may act as a premating isolation mechanism. Finally it should be noted that variation in seven male-female pairs of O. conifera, captured in what is believed to be a mating posture, is not broad as the variation in the rest of the population in the Gulf area. These seven females fit into only one of the several previously described groups of O. conifera. It is possible in this species that only certain females are actually capable of attracting males for the purpose of reproduction. KEY TO THE SPECIES OF ONCAEA Ur 5 segmented _____.. females Ur 6 segmented _______ males Females I Eps oft sP4 with) ay distal conical sprojection\ee ae 2 1Banyorsy (ye TEL Ceyitdevaymle yornoyereulovY ae 3 2. Th2 usually with pronounced hump at the dorsal midline; total lengthy greater than) 0:90)immy ee ee O. conifera Th2 without this hump; total length less than 0.80 mm __. O. similis 3. Caudal ramus less than 3 times as long as wide _____---- O. media Caudaleramus)s—5. timesyasslonpyasi wide se 4 4. Body heavily sclerotized; Pr strongly arched; P5 short _ O. venusta Body not as sclerotized; Pr not arched; P5 elongated ____.__________- Ey AIS bt oh a ET SANA gs ORE Lt On 8 he ee O. mediterranea Males 1. Enp3 of P4 with a terminal rounded projection —_-._....- O. conifera EMpoOhP4 without sthismsprojecti onan ane 2 2. Seg3 of A2, proximal set of spines and setae with 2 thick spines and, QithinsGetae (a ves ee ah 3 Seg3 of A2, proximal set of spines and setae with 1 thin seta, ithickispinesanda2athinusetac years atu ae sree O. mediterranea 3. Ratio, caudal ramus length to Ansg length, less than 2 _.. O. media Mihis eration ore ate rect bierryy acess eames es en ea O. venusta ACKNOWLEDGMENTS I would like to thank Drs. John Wormuth and Leo Berner of Texas A&M University and Dr. Thomas Bowman of the National Museum of Natural History. Taxonomic notes on Oncaea 931 LITERATURE CITED ALLDREDGE, A. 1972. Abandoned larvacean houses: a unique food source in the pelagic environment. Science 177:885-887. Farran, G. 1929. British Antarctic (“TERRA NOVA”) Expedition, 1910. Natural History Report. Zoology VIII (3) :203-306. 1936. Great Barrier Reef Expedition, 1928-1929. Scientific Reports V(3):126-128. Ferrari, F,. 1973. Some Corycaeidae and Oncaeidae (Copepoda: Cyclopoida) from the epipelagic waters of the Gulf of Mexico. Ph. D. Dissertation. Texas A&M University. 214 pp. GIESBRECHT, W. 1892. Systematic und Faunistic der pelagischen Copepoden des Golfes von Neapel. Fauna u Flora des Golfes von Neapel. XIX:590-604. Heinrich, A. 1957. On the nutrition of marine copepods in the tropical region. Dokl. Akad. Nauk. S.S.S.R. 119:1029 (in Russian ). Heron, A. 1973. A specialized predator-prey relationship between the copepod Sapphirina angusta and the pelagic tunicate Thalia democratica. Jour. Mar. Biol. Ass. U.K. 53:429-435. Moutton, T. 1973. Principle component analysis of variation in form within Oncaea conifera Giesbrecht, 1891, a species of copepod (Crustacea). Syst. Zool. 22(2):141-156. Orson, F. 1949. The pelagic cyclopoid copepods of the coastal waters of Oregon, California and lower California. Ph. D. Thesis, University of California, Los Angeles. 208 pp. Owre, H. and M. Foyo. 1967. Copepods of the Florida Current. Fauna Caribbaea, No. 1, Crustacea, Part 1: Copepoda p. 110- 112. Rose, M. 1933. Copepodes pelagiques. Faune de France 26:295- 303. Sars, G. 1916. Liste systematique des Cyclopoides, Harpacticoides et Monstrilloides recueillis pendant les campagnes de S.A.S. le Prince Albert de Monaco, avec descriptions et figures des especes nouvelles. Bull. Inst. Oceanogr. Monaco 323:1-15. 1918. An Account of the Crustacea of Norway. VI Copepoda, Cyclopoida 193 pp. SEWELL, R. 1947. The John Murray Expedition 1933-34. The Free- Swimming Planktonic Copepoda. Systematic Account VIII(1): 258-264. SHMELEVA, A. 1969. Espéces nouvelles du genre Oncaea (Copepoda, Cyclopoida) de la mer Adriatique. Bull. Inst. Oceanogr. Monaco 68( 1393) :1—28. WicksTEAD, J. 1962. Food and feeding in pelagic copepods. Proc. Zool. Soc. London 139:545-555. 932 Proceedings of the Biological Society of Washington No. 22, pp. 233-248 15 August 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON REVIEW OF THE GENUS HERMENIA, WITH A DESCRIPTION OF A NEW SPECIES (POLYCHAETA: POLYNOIDAE: LEPIDONOTINAE) as By Marian H. PETTIBONE Smithsonian Institution, Washington, D.C. 20560 As a part of my revisionary work on the aphroditoid poly- \ chaetes, a study was made of the members of the genus Hermenia Grube (Polynoidae: Lepidonotinae). Three species have been described previously, one of them both a homonym and a synonym. A new species is added herein, based on material previously confused with one of the currently recog- nized species. The species, treated in this report are: H. verruculosa Grube, 1856. Virgin Islands [Includes Polynoe nodosa Treadwell, 1901 (= homonym). Puerto Rico] H. acantholepis (Grube, 1876), as Polynoe. Philippine Is- lands H. neoverruculosa new species, as H. verruculosa by Monro, 1939. Seychelles In addition to the specimens in the Smithsonian Institution, Washington (USNM), types and additional material were bor- rowed from the following Museums: Bernice P. Bishop Mu- seum, Honolulu (BPBM), through D. M. Devaney; British Museum (Natural History), London (BMNH), through J. D. George; Museum National d’Histoire Naturelle, Paris (MNHNP), through J. Renaud-Mornant; Rijksmuseum van Natuurlijke Historie, Leiden (RMNHL), through J. van der Land; University Zoological Museum, Copenhagen (UZMC), through J. B. Kirkegaard; Zodlogisch Museum, Amsterdam (ZMA), through S. van der Spoel, and Zoologisches Institut 22—-Proc. BroL. Soc. WaAsuH., Vou. 88, 1975 (233 ) awe 234 Proceedings of the Biological Society of Washington and Museum, Hamburg (ZMH), through G. Hartmann- Schréder. The manuscript benefited from the suggestions of H. H. Hobbs, Jr. and M. L. Jones, both of the Smithsonian Institution. Genus Hermenia Grube, 1856 Type-species: H. verruculosa Grube, 1856, by monotypy. Gender: feminine. Diagnosis: Body short, subrectangular, segments 26. Elytra 12 pairs, on segments 2, 4, 5, 7, and on alternate segments to 23, with dorsal cirri on posterior 3 segments; elytra firmly attached to indistinct elytrophores; first 1-3 pairs moderately large, remaining ones very small (easily overlooked) to moderately so; elytra with fringes of short papillae and spiny macrotubercles. Prostomium bilobed, with 2 pairs of eyes, paired palps, and 3 antennae; lateral antennae inserted terminally on anterior continuations of prostomium, i.e., lepidonotoid. Tentacular segment (1) lateral to prostomium, with few notosetae, 2 pairs of tentacular cirri, and bulbous facial tubercle. Buccal segment (II) with ventral buccal cirri longer than those following and without distinct nuchal fold. Parapodia subbiramous; notopodia forming small lobes on anterodorsal faces of stout neuropodia; neuropodia truncate and papillate distally. Notosetae few to absent, slender, finely spinous. Neurosetae stout, with falcate tips and 1 or 2 stout teeth. Dorsal cirri with cylindrical cirrophores and rather short styles. Ventral cirri short, subulate. Pygidium with pair of anal cirri. Pharynx with 11 pairs of papillae (somewhat variable?) and 2 pairs of jaws. Nephridial papillae short, cylindrical, beginning on segments 6, 7, or 8. Integument tuberculate and/or papillate. KEY TO THE SPECIES OF HERMENIA 1. First 2 or 3 pairs of elytra moderately large; remaining ones smaller but none so small as to be easily overlooked (Fig. 4a-c). Both dorsal and ventral surfaces papillate, not tuber- culate (Fig. 3a,c,d). Neurosetae with 2 stout accessory teeth (HST PURSE) (7, aE WOE v's Hate ered 2 Ge H. acantholepis (Grube) 1’. First pair of elytra large; remaining ones very small, deeply em- bedded, and easily overlooked (Figs. la—c, 5a,c,d). Dorsal Surhacetuberculatem(Higs- silage) es 2 2. Neurosetae with single stout accessory tooth (Fig. 2h). Ventral surface densely papillate (Fig. 2f,¢) —__... H. verruculosa Grube 2’. Neurosetae with 2 accessory teeth—one stout and one more slender (slender one sometimes missing; Fig. 6f). Ventral surface wrinkled, not thickly papillate (Fig. 6c,d) — 4B. SS Te EE oc EEN os RE H. neoverruculosa new species Review of polychaete Hermenia 235 Hermenia verruculosa Grube Figures 1, 2 Hermenia verruculosa Grube, 1856: 44.—Treadwell, 1911: 9, figs. 23- 26; 1924: 5; 1928: 452; 1939: 187, fig. 19.—Seidler, 1923: 261, fig. 6; 1924: 95.— Augener, 1925: 4; 1927: 43.—Hartman, 1939: 4.— Fauvel, 1953a: 4.—Bellan, 1964; 307—Ebbs, 1966: 500, fig. 4, a-g.— Devaney, 1974: 122, 140.—Rullier, 1974: 19. Polynoe nodosa Treadwell, 1901: 187, figs. 8,9. [Not Sars, 1861.] Lepidonotus verruculosus.—Horst, 1922: 198. Material examined: West INDIES AND CARIBBEAN SEA: Virgin Is- lands: St. John, 9 June 1846, Krgyer, collector—holotype of Hermenia verruculosa (UZMC). Off St. Thomas 36.5-42 m, Fish Hawk sta. 6079, 6 February 1899—holotype of Polynoe nodosa (USNM 16013). St. James Island near St. Thomas, 4.5 m, 10 July 1915, C. R. Shoe- maker, collector—1 specimen (USNM 46912). Virgin Islands, 91 m, S.S. ADVANCE m1, sta. 1, September 1970, J. Clark, collector—1 specimen (USNM 46922). Barbuda: near Spanish Point, shore, on beach wrack and coral, 28 April 1958, W. L. Schmitt, collector—1 specimen (USNM 46915). Barbados: 3 miles west of Needham Point, University of Iowa Barbados-Antigua Expedition—1l specimen (USNM 20272; reported by Treadwell, 1924). Puerto Rico: Fajardo, Fish Hawk 1898-99— paratype of Polynoe nodosa (USNM 16014). Majimo Reef, rather com- pact calcareous sand of Mangrove Island, 1-2 feet, 9 March 1967, M. E. Rice, collector—1l specimen (USNM 46920). Caribbean: 09°32’N, 79° 54’W, 62 m, Avpatross sta. 2146, 2 April 1884—1 specimen (USNM 951). 18°30’N, 66°25’W, 73 m, Johnson-Smithsonian Deep Sea Expedition, 7 February 1933—1 specimen (USNM 20058). Ba- hamas: Golding Key, Andros Island, 13 May 1912, P. Bartsch, collector— 1 specimen (USNM 16492). Off North Point, Gibson Cay, Andros Island, 24°19.8’N, 77°40.8’W, 6 March 1966, M. L. Jones, collector— 1 specimen (USNM 46918). Outer reef, north of Clarencetown, Long Island, 15 May 1970, C. Riser, collector—1 specimen (USNM 46921). CENTRAL AND SourH AMERICA: British Honduras: on brittle star, Ophiocoma pumila Liitken, 7 March 1969, D. M. Devaney, collector— 1 specimen (BPBM). Carrie-Bow Cay, 27 m, 11 June 1972, M. E. Rice, collector—2 specimens (USNM 50112). Colombia: Old Providence Island, shore, Presidential (Roosevelt) Cruise sta. 30, 6 August 1938, W. L. Schmitt, collector—1l specimen (USNM 20489; reported by Hartman, 1939) Fiorina AND GuLF oF Mexico: Florida: Dry Tortugas, June—July 1914, A. L. Treadwell, collector—3 specimens (USNM 17722, 17733). Loggerhead Key, Dry Tortugas, A. L. Treadwell, collector, Fauvel col- lection—1 specimen (BMNH 1928: 4: 236). Dry Tortugas, 1930, W. L. Schmitt, collector—2 specimens (USNM 46913; BMNH 1972: 106). Loggerhead Key, from matrix of large brain coral, 3 m, 19 August 1966, 236 Proceedings of the Biological Society of Washington Fic. 1. Hermenia verruculosa (Paratype of Polynoe nodosa Tread- well, USNM 16014): a, Prostomium and anterior four segments, dorsal view; first right elytron removed; b, First elytron (segment II), with detail of tubercles and sensory papillae; c, Elytron from middle of body, with detail of tubercles and sensory papillae. R. F. Cressey, collector—2 specimens (USNM 46919). Off Louisiana: 28°06’N, 91°02’W, 53 m, OreEGonN sta. 1416, 21 September 1955—1 specimen (USNM 46914). S.E. Mexico: Yucatan peninsula, Bahia de la Ascensién, Quintana Roo, Nicchehabin Reef, cracked from coral pieces and sea fans, 1-2 m, Smithsonian-Bredin Expedition, stations 67 and 72, 13/14 April 1960, W. L. Schmitt and E. L. Bousfield, col- lectors—2 specimens (USNM 46916-7). NortH ATLANTIC: Off New Jersey, 39°56’N, 70°20’W, 223 meters, ALBATROSS station 2246, 26 September 1884—1 specimen (USNM 3296). Description: Length of holotype 20 mm, width including setae 8 mm, segments 26. Length of largest specimen from Tortugas (USNM 17733) 32 mm, width including setae 9 mm, segments 26. Body short, subrectangular, flattened ventrally, arched dorsally, slightly tapered an- teriorly and posteriorly. First pair of elytra moderately large, nearly covering prostomium; following elytra very small, partially embedded in integument of body (easily overlooked; Figs. la—c, 2f). Elytra oval; margins with short, cylindrical, sensory papillae; surface with small, globular, sensory papillae and spiny, globular macrotubercles of various sizes. Review of polychaete Hermenia 237 Prostomium (Fig. la) bilobed; ceratophore of median antenna cylindrical, papillate, inserted in anterior notch; style long, smooth, with subterminal enlargement and filiform tip; lateral antennae similar to but shorter than median antenna; palps long, stout, tapered, smooth; anterior pair of eyes situated anterolaterally, larger than posterodorsal pair. Tentacular segment (1) distinct dorsally but sometimes retracted within segment II; parapodia lateral to prostomium, bearing 2 pairs of tentacular cirri, resembling median antenna, ventral pair slightly shorter than dorsal pair; few (1 or 2) notosetae on mesial side; bulbous facial tubercle projecting anteriorly on ventral upper lip. Buccal segment (II) bearing first pair of elytra, subbiramous para- podia, and ventral buccal cirri similar to tentacular cirri (Figs. la, 2a). Third segment with first pair of dorsal cirri (Figs. la, 2d). Neurosetae of these two segments differing from those of following in being more slender, straight, and usually provided with more numerous accessory teeth (Fig. 2c,e). Parapodia subbiramous (Fig. 2f,g). Notopodia represented by small bulbous acicular lobes on anterodorsal sides of neuropodia. Notosetae greatly reduced (0-1), slender, curved, tapering to acute tips, with numerous spinous rows (Fig. 2b). Neuropodia stout, thick, truncate and distally papillate, with upper distal part distinctly notched; its dorsal surface covered with globular tubercles and its ventral surface papillate. Neurosetae (about 10), stout, amber-colored, falcate, with single lateral stout tooth (Fig. 2h). Cirrophores of dorsal cirri stout, cylindrical, papillate; styles moderately short, smooth, with subterminal enlarge- ments and filiform tips. Dorsal tubercles on cirrigerous segments in- conspicuous. Ventral cirri short, subulate, smooth. Pygidium with anus situated medial to parapodia of posterior seg- ment and bearing pair of long anal cirri on distinct papillate cirrophores. Pharynx usually with 11 pairs of papillae (10/11 or 11/12) and 2 pairs of amber-colored jaws. Nephridial papillae short, cylindrical, beginning on segment 7 or 8 (Fig. 2g). Dorsal surface of body and parapodia covered with globular tubercles and covered ventrally with fine papillae (Figs. la, 2f,g). Biology: In the Tortugas, H. verruculosa was observed by Treadwell (1911: 11) living in the interstices of coral rocks, and, when disturbed, moving into deeper crevices with considerable rapidity, clinging with great tenacity to the rock, and thus making it difficult to remove them without injury. In British Honduras, Devaney (1974: 122, 140) col- lected a specimen of H. verruculosa that was attached to the ophiuroid, Ophiocoma pumila Liitken. The two animals were both transversely banded, brown and white, and the dorsal papillae of the polynoid were similar in size and color to the granules of the disc of the ophiuroid, giv- ing the worm a camouflaged appearance against the ophiuroid and sug- gesting a possible commensal relationship. 238 Proceedings of the Biological Society of Washington O.Jmm Fic. 2. Hermenia verruculosa (Paratype of Polynoe nodosa Tread- well, USNM 16014): a, Left parapodium from segment II, anterior view; internal acicula dotted; b, Notoseta from same; c, Two neuro- setae from same; d, Left parapodium from segment III, anterior view; internal acicula dotted; e, Three neurosetae from same; f, Left ely- tragerous parapodium, anterior view; internal acicula dotted; g, Left cirrigerous parapodium, posterior view; h, Neuroseta. Distribution: West Indies and Caribbean Sea, Bahamas, South and Central America (Colombia, British Honduras), Gulf of Mexico (S.W. Florida, off Louisiana, Yucatan Peninsula, Mexico), North Atlantic (off New Jersey). In low water to 223 m. Review of polychaete Hermenia 239 Hermenia acantholepis (Grube) Figures 3, 4 Polynoe acantholepis Grube, 1876: 61. Polynoe (Lepidonotus) acantholepis—Grube, 1878: 24, pl. 2: Fig. 1. Lepidonotus acantholepis——Michaelsen, 1892: 95.—Hornell, 1903: 25.— Horst, 1917: 67, pl. 15: Figs. 3, 4—Fauvel, 1922: 490, Fig. 1, a—d. Hermenia acantholepis—Seidler, 1923: 262; 1924: 94.—Monro, 1924: 39, Figs. 2, 3; 1939: 169.—Pruvot, 1930: 11, pl. 1. Figs. 27-32.— Fauvel, 1933: 41; 1935: 296; 1953b: 38, Fig. 14,ab. Material examined: Japan, Goto Island—2 specimens (BMNH 1925: 1: 28: 4-5; reported by Monro, 1924). Philippine Islands, Upolu—holotype of Polynoe acantholepis (ZMH 504). Indonesia (reported by Horst, 1917): Bay of Badjo, west coast of Flores, shore, Srsoca sta. 50—1 specimen (ZMA 547a). Anchorage off Pasir Pandjang, west coast of Binongka, reef, sta. 220—2 specimens (ZMA 547b; RMNHL 1124). Nalahia Bay, Nusa Laut Island, reef, sta. 2341 specimen (RMNHL 1125). South coast of Timor, 08°39.1’S, 127°04.4’E, 34 m, Lithothamnion, sta. 285—1 specimen (ZMA 547c). Annam, Honlohé, Tle Mamelles, Fauvel collection—1 specimen (MNHNP; reported by Fauvel, 1935). Ceylon, Driesch, collector—2 specimens (ZMH 505; reported by Michaelsen, 1892). Ceylon, Talili Bay, 73 m, Willey collection—1 specimen (BMNH 1973: 12: reported by Hornell, 1903). Maldives, Felidu, J. S. Gardiner, collector—1 specimen (BMNH 1941: 4: 4: 195; reported by Monro, 1939). Description: Length of holotype 25 mm, width including setae 8 mm, segments 26. Length of two specimens from S1Boca stations 220 and 234 (RMNHL 1124/5) 13-25 mm, width including setae 6-9 mm, seg- ments 26. Length of specimen from New Caledonia, according to Pruvot (1930: 11), 33 mm, width including setae 10 mm, segments 26. Body short, subrectangular, flattened ventrally, arched dorsally, tapered slightly anteriorly and posteriorly. First 2-3 pairs of elytra moderately large, touching medially; follow- ing elytra smaller (not so small as in H. verruculosa; Figs. 3d, 4a—c). Elytra rounded to oval; margins and surfaces with short cylindrical sensory papillae and globular to oval spinuous macrotubercles of various sizes. Prostomium (Fig. 3a) bilobed; ceratophore of median antenna cylin- drical, distally papillate, inserted in anterior notch; style long, smooth, with subterminal enlargement and filiform tip; lateral antennae similar to and slightly shorter than median antenna; palps long, stout, tapered, smooth; anterior pair of eyes situated anterolaterally, slightly larger than posterodorsal pair; prostomium often partly withdrawn, resulting in posterior pair of eyes being hidden from view by fold of segment II. 240 Proceedings of the Biological Society of Washington Fic. 3. Hermenia acantholepis (ZMA 547b): a, Prostomium and an- terior three segments, dorsal view; elytra removed; prostomium partly withdrawn, resulting in posterior pair of eyes being hidden from view by fold of segment II; b, Two neurosetae from segment II; c, Left cirrigerous parapodium, posterior view; d, Left elytragerous parapodium, anterior view, with detail of elytral tubercles and papillae and neuro- podial papillae; internal acicula dotted; e, Notoseta; f, Three neurosetae. Review of polychaete Hermenia 241 Fic. 4. Hermenia acantholepis (from Annam, MNHNP): a, First right elytron (segment II), with detail of tubercles and papillae; b, Second right elytron (segment III); c, Middle right elytron. Parapodia of tentacular segment (I) lateral to prostomium and bearing 2 pairs of tentacular cirri resembling median antenna; ventral pair slightly shorter than dorsal pair; few (1-2) notosetae on mesial side; bulbous facial tubercle projecting anteriorly on ventral upper lip. Buccal segment (II) bearing first pair of elytra, subbiramous parapo- dia, and ventral buccal cirri similar to tentacular cirri. Neurosetae of segments II and III differing from those of following in being more slender, straight, and provided with more numerous accessory teeth (Fig. 3b). Parapodia subbiramous (Fig. 3c,d). Notopodia forming small bulbous, papillate acicular lobes on anterodorsal sides of neuropodia. Notosetae few in number, slender, curved, tapering to blunt tips, with numerous spinous rows (Fig. 3e). Neuropodia stout, thick, papillate, with diag- onally truncate presetal acicular lobes and shorter, rounded postsetal ones. Neurosetae (about 18), stout, amber-colored, slightly falcate, with pair of large conical teeth (Fig. 3£). Cirrophores of dorsal cirri stout, cylindrical, papillate; styles relatively short, smooth, with subterminal enlargements and filiform tips. Dorsal tubercles on cirrigerous segments inconspicuous. Ventral cirri short, subulate, smooth. Pygidium with anus situated medial to parapodia of posterior segment and bearing pair of long anal cirri on distinct papillated cirrophores. Pharynx not extended. Nephridial papillae short, cylindrical, beginning on segment 6 (Fig. 3c). Body thickly covered with short cylindrical papillae (Fig. 3a,c,d). Distribution: Widely distributed in Indo-Pacific Region: Japan (Goto Island, Seto), Indochina, Philippines, Indonesia, Australia (Pel- sart group), New Caledonia, Ceylon, Maldives, Madagascar, Gulf of Suez. Intertidal to 73 m. © 242 Proceedings of the Biological Society of Washington Hermenia neoverruculosa new species Figures 5, 6 Hermenia verruculosa.—Monro, 1924: 40; 1939: 169, fig. 2. [Not Grube, 1856]. Material examined: INDIAN OcEAN: Cargados, Carajos, 55 m, SEA- LARK, 28 August 1905, Gardiner, collector—holotype (BMNH 1941: 4: 4: 197). Seychelles, 68 m, SEALARK, 30 October 1905, Gardiner, col- lector—paratype (BMNH 1941: 4: 4: 196). Darros Island, the Amirantes, H.M.S. “Alert”—1 specimen (BMNH 1925: 1: 28: 6). Description: Length of holotype 29 mm, width including setae 8 mm, segments 26. Length of paratype 22 mm, width including setae 7 mm, segments 26. Body short, subrectangular, flattened ventrally, arched dorsally, tapered slightly anteriorly and posteriorly. First pair of elytra large, covering prostomium; following elytra very small (Figs. 5a,c,d; 6d). Elytra oval; margins with short cylindrical papillae; surface with short, globular, sensory papillae and globular to oval, spiny macrotubercles of various sizes. Prostomium (Fig. 5a, b) bilobed; ceratophore of median antenna cylindrical, somewhat wrinkled or annulate, inserted in anterior notch; style long, with subterminal enlargement and filiform tip, smooth; lateral antennae similar to but shorter than median antenna; palps missing; anterior pair of eyes situated anterolaterally, subequal in size to postero- lateral pair. Tentacular segment (1) dorsally distinct, forming inflated area medial to elytrophores of segment II; parapodia lateral to prosto- mium, bearing 2 pairs of tentacular cirri resembling median antenna; single notoseta on mesial side; bulbous facial tubercle projecting anteriorly on ventral upper lip (Fig. 5a,b). Buccal segment (II) bearing first pair of elytra, subbiramous parapodia, and ventral buccal cirri similar to tentacular cirri (Figs, 5a,b; 6a). Neurosetae few in number (some lost?), differing from those of follow- ing in being more slender, curved, and provided with few spinous rows (Fig. 6b). Parapodia subbiramous (Fig. 6c,d). Notopodia forming small sub- conical acicular lobes on anterdorsal sides of neuropodia. Notosetae greatly reduced (0-2), slender, curved, tapering to blunt tips, with nu- merous spinous rows (Fig. 6e). Neuropodia stout, thick, truncate and dis- tally papillate, upper distal part distinctly notched. Neurosetae (about 12), stout, amber-colored, slightly faleate, with 2 teeth—one stouter and one more slender, latter one sometimes missing (Fig. 6f). Cirrophores of dorsal cirri stout, cylindrical; styles relatively short, smooth, with sub- terminal enlargements and filiform tips. Dorsal tubercles on cirrigerous segments inconspicuous. Ventral cirri short, subulate, smooth. Pygidium with anus situated medial to parapodia of posterior seg- ment and bearing pair of long anal cirri on distinct cirrophores. Pharynx not extended. Nephridial papillae short, cylindrical, beginning on seg- ment 7. Dorsal surface of body covered with globular tubercles; ventral Review of polychaete Hermenia 243 Fic. 5. Hermenia neoverruculosa (Holotype, BMNH 1941: 4: 4: 197): a, Dorsal view of prostomium and anterior two segments, palps missing; right elytron removed; left buccal cirrus smaller, regenerating; b, Same, ventral view; position of missing palps cross-hatched; c, First right elytron, with detail of tubercles and papillae; d, Elytron from middle of body, with detail of tubercles and papillae. 244 Proceedings of the Biological Society of Washington Fic. 6. Hermenia neoverruculosa (Holotype, BMNH 1941: 4: 4: 197): a, Right parapodium from segment II, anterior view; internal acicula dotted; b, Two neurosetae from same (tips broken); c, Right cirrigerous parapodium, posterior view; d, Right elytragerous parapodium, anterior view, with detail of tubercles from dorsal surface; internal acicula dotted; e, Notoseta; f, Three neurosetae. Review of polychaete Hermenia 245 surface and parapodia wrinkled, with only few scattered small papillae ( Figs. 5a,b; 6c,d). Remarks: In many respects, H. neoverruculosa is intermediate be- tween H. verruculosa and H. acantholepis, as indicated in the key to the species above. It lies within the geographic range of the widely dis- tributed Indo-Pacific H. acantholepis but appears to resemble more closely the Caribbean H. verruculosa, as indicated by Monro (1924, 1939), when he referred the specimens from the Indian Ocean, with some hesitancy, to H. verruculosa. Distribution: Indian Ocean. Intertidal to 68 m. LITERATURE CITED AucENER, H. 1925. Uber Westindische und einige andere Polychaeten- typen von Grube (Oersted), Krgyer, Mérch und Schmarda. Publ. Univ. Zool. Mus. Kgbenhavn No. 39:1-47, 3 figs. 1927. Polychaeten von Curacao. Bijdragen tot de Dierkunde, Amsterdam 25:39-82, 9 figs. BELLAN, G. 1964. Campagne de la Calypso dans YAtlantique. An- nélides polychétes. Ann. Inst. Oceanogr. N.S. 41:301-314, 1 fig. Devaney, D. M. 1974. Shallow-water echinoderms from British Hon- duras, with a description of a new species of Ophiocoma (Ophiuroidea). Bull. Mar. Sci. 24(1):122-164, 16 figs. Esss, N. K., Jk. 1966. The coral-inhabiting polychaetes of the north- ern Florida reef tract. Part 1. Aphroditidae, Polynoidae, Amphinomidae, Eunicidae, and Lysaretidae. Bull. Mar. Sci. 16:485-555, 12 figs. FauveEL, P. 1922. Annélides polychétes de l’Archipel Houtman Abrolhol (Australie occidentale) recueillies par M. le Professeur W. G. Dakin, F.L.S. J. Linn. Soc. London 34:487-500, 2 figs. —. 1933. Mission Robert Ph. Dollfus en Egypte. Annélides polychétes. Mémoires présentés a l’institut d’Egypte et publiés sous les auspices sa Majesté Fouad I°’, Roi d’Egypte 21:31-83, 6 figs. ——. 1935. Annélides polychétes de !Annam. Mem. Pont. Accad. Sci. Nuovi Lincei ser. 3, 3:279-354, 18 figs. ———. 1953a. Annélides polychétes de la Croisiere du “President Theodore Tissier” aux Antilles (1951). Bull. Inst. Océanog. Monaco No. 1033:1—23. ——. 1953b. Annelida Polychaeta. The Fauna of India including Pakistan, Ceylon, Burma and Malaya. Allahabad, The Indian Press. 1-507, 250 figs. GrusBE, E. 1856. Annulata Orstediana. Enumeratio Annulatorum, quae in itinere per Indiam occidentalem et Americam centralem annis 1845-1848 suspepto legit cl. A. S. Orsted, adjectis speciebus nonnullis a cl. H. Kréyero in itinere ad Americam meridionalem 246 Proceedings of the Biological Society of Washington collectis. Vidensk. Meddl. naturh. Foren. Kjobenhavn 1856: 44-62. ——. 1876. Bemerkungen iber die Famillie der Aphroditeen (Gruppe Polyonina, Acoétea, Polylepidea). Jahresber. Schles. Gesells. Vaterl. Kultur, Breslau, vol. for 1875: 46-72. — —. 1878. Annulata Semperiana. Mém. Acad. Imp. Sci. St. Péters- bourg ser. 7, 25(8):1-300, 15 pls. HartTMan, O. 1939. The polychaetous annelids collected by the Presi- dential Cruise of 1938. Smithsonian Misc. Coll. 98:1-22, 3 figs. Hornet, J. 1903. In Herdman, W. A. Report to the Government of Ceylon on the Pearl Fisheries of the Gulf of Manaar, with Supplementary Reports upon the Marine Biology of Ceylon by other Naturalists, Pt. 1:1—97. Horst, R. 1917. Polychaeta errantia of the Siboga-Expedition. Pt. 2. Aphroditidae and Chrysopetalidae. Siboga-Exped. Leyden 24b: 1-140, 5 figs., pls. 11-29. . 1922. On some polychaetous annelids from Curacao. Bijdragen tot de Dierkunde Amsterdam. Feest-nummer: 193-201, 2 figs. MICHAELSEN, W. 1892. Polychaeten von Ceylon. Jahrb. Hamburg. wiss. Anstalten 9(2):93-113, 18 figs. Monro, C. C. A. 1924. On the Polychaeta collected by H.M.S. ‘Alert, 1881-1882. Families Polynoidae, Sigalionidae, and Eunicidae. Jour. Linn. Soc. Zool. 36:37-64, 24 figs. 1939. On some tropical polychaetes in the British Museum mostly collected by Dr. C. Crossland at Zanzibar, Tahiti, and the Marquesas. I. Families Amphinomidae to Phyllodocidae. Ann. Mag. Nat. Hist. ser. 11, 4:161—-184, 7 figs. Pruvot, G. 1930. Annélides polychétes de Nouvelle-Calédonie re- cueillies par M. Francois. Arch. Zool. Exp. Gen. Paris 70:1—94, 8 figs., 3 pls. Ruiter, F. 1974. Quelques annélides polychétes de Cuba recueillies dans les éponges. Trav. Mus. Hist. Nat. “Gr. Antipa” 14:9-77, 9 figs. Sewer, H. F. 1923. Uber neue und wenig bekannt Polychiten. Zool. Anz. Leipzig 56:254—-264, 8 figs. 1924. Beitrige zur Kenntnis der Polynoiden, I. Arch. Naturg. 89A (11):1-217, 22 figs., 2 pls. TREADWELL, A. L. 1901. The polychaetous annelids of Porto Rico. Bull. U.S. Fish Comm. Washington, Bull. for 1900 2:181- 201, 81 figs. —. 1911. Polychaetous annelids from the Dry Tortugas, Florida. Bull. Amer. Mus., Nat. Hist., N.Y. 30:1-12, 29 figs. —. 1924. Polychaetous annelids, collected by the Barbados- Antigua Expedition from the University of Iowa in 1918. Univ. Iowa Stud. 10(4):1-23, 2 pls. Review of polychaete Hermenia 247 1928. Polychaetous annelids from the Arcturus Oceano- graphic Expedition. Zoologica, N.Y. 8:449-485, figs. 177-179. 1939. Polychaetous annelids of Porto Rico and _ vicinity. Scientific Survey of Porto Rico and the Virgin Islands. N. Y. Acad. Sci. 16(2):151-319, 118 figs. 948 Proceedings of the Biological Society of Washington No. 23, pp. 249-252 15 August 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON EURYSQUILLA PACIFICA, A NEW STOMATOPOD CRUSTACEAN FROM NEW BRITAIN By RaymMonp B. MANNING Smithsonian Institution, Washington, D.C. 20560 Among the unidentified stomatopods examined in the collec- tion of the Zoological Museum, Berlin, in 1971, was a small gonodactylid which proved to be the first species of Eury- squilla to be identified in the central Pacific and only the second species to be found in the Indo-West Pacific region. This species is described below. Studies of Indo-West Pacific stomatopods have been sup- ported by the Smithsonian Institution through its Research Awards Program. I thank Dr. G. Hartwich of the Zoological Museum, Berlin, for providing working space there in 1971 and for the loan of this specimen. The illustrations were prepared by my wife Lilly. Eurysquilla pacifiea, new species Figure 1 Holotype: 192, total length 17 mm; Ralum, New Britain; F. Dahl, col., 1896; ZMB 18097. Description: Comea broadened, faintly bilobed, set obliquely on stalk. Eyes extending beyond end of second segment of antennular peduncle. Ocular scales low, appearing fused in midline, produced laterally. An- terior margin of ophthalmic somite unarmed. Antennular peduncle shorter than carapace. Antennular processes produced into slender spines directed anterolaterally. Antennal scale short, less than half as long as carapace. Antennal protopod with 1 ventral papilla. Rostral plate cordiform, broader than long, rounded anterolateral mar- gins converging on obtusely pointed apex. Carapace smooth, strongly narrowed anteriorly, anterior parts of lateral plates projecting slightly beyond base of rostral plate. 93—Proc. Biot. Soc. WaAsH., Vou. 88, 1975 (249) 250 Proceedings of the Biological Society of Washington Fic. 1. Eurysquilla pacifica, new species, female holotype, TL 17 mm: @, Anterior part of body; b, Eye; c, Sixth abdominal somite, telson, and uropod, dorsal view; d, Telson, lateral view; e, Telson, ventral view; f, Uropod, ventral view. Setae omitted in all figures. Raptorial claws missing. Mandibular palp and 5 epipods present, third to fifth very small. Exposed thoracic somites lacking carinae or spines. Fifth thoracic somite lacking distinct lateral process or ventrolateral spine. Lateral processes of sixth and seventh thoracic somites rounded anteriorly and posteriorly, that of sixth somite smaller, more truncate laterally. Eighth thoracic somite lacking distinct median ventral tubercle. Endopods of walking legs slender, two-segmented. Basal segments of walking legs unarmed. Anterior 4 abdominal somites smooth, lacking carinae or spines. Fifth somite with low, unarmed longitudinal carina above lateral mar- gin. Sixth somite with 4 pairs of spines: submedians on margin, lacking New stomatopod from New Britain 251 sharp dorsal carina; intermediates submarginal, with short dorsal carina; accessory spine on longer carina present between intermediates and laterals, latter on normal sharp lateral carina. Sixth somite with obtuse lobe ventrally in front of articulation of each uropod. Telson broader than long, with entire median and dorsal submedian carinae, each terminating in long spine. Accessory median carinae each a row of 4—5 tubercles converging under apical spine of median carina. Tubercles present on each side of submedian carina, in a line mesially, in a patch laterally. Three pairs of marginal spines present: submedians, with bases appressed and long, movable apices; intermediates, each with large, rounded, dorsally-projecting dorsal lobe and low lobe on outer margin; and marginals, with smaller, rounded lobe on inner margin. Sub- median denticles absent; 2 intermediate denticles present, inner with apical spinule, outer triangular, with submarginal spinule; lateral den- ticle with submarginal spinule. Uropod with 7 movable spines on outer margin of proximal segment of exopod, distalmost extending beyond midlength of distal segment; inner distal angle of proximal segment with 3 stiff setae. Basal pro- longation of uropod broad, terminating in 2 spines, inner much the longer, with 20-22 sharp spinules on inner margin; outer spine of basal prolongation with slender spinule on inner margin. Color pattern completely faded. Measurements: Female only known, total length 17 mm. Other measurements, in mm: carapace length 2.8; cornea width 1.1; rostral plate length, width 0.7, 1.1; antennal scale length 1.2; antennular peduncle length 1.7; fifth abdominal somite width 3.4; telson length, width 1.6, 2.8. Discussion: Eurysquilla pacifica, only the second species of the genus to be found in the Indo-West Pacific region, differs from E. sewelli (Chopra, 1939) from the Gulf of Aden as well as from 3 American species, E. plumata (Bigelow, 1901), E. veleronis (Schmitt, 1940), and E. delsolari Manning, 1970, in having a broad rather than a very long slender basal prolongation of the uropod; in those 4 species the basal prolongation is Y-shaped and has at most 1 additional spine or lobe on its inner margin. E. pacifica resembles 3 western Atlantic species, E. maiaguesensis (Bigelow, 1901), E. chacei Manning, 1969, and E. holthuisi Manning, 1969, in having a broad basal prolongation with 5 or more spinules on the inner margin. The new species differs from E. maiaguesensis in having a mandibular palp, 8 rather than 6 posterior spines on the sixth abdominal somite, more dorsal tubercles on the telson, and a spinule rather than an obtuse lobe between the spines of the basal prolongation of the uropod. E. pacifica differs from E. chacei in having armed carinae on the fifth abdominal somite; only 1 pair of marginal spines on the sixth abdominal somite; fewer tubercles in the accessory median row on the telson, and, on the basal prolongation of the uropod, 20-22 rather than 10 slender spinules on the inner margin 252 Proceedings of the Biological Society of Washington and a spinule rather than a rounded lobe between the apical spines. E. pacifica and E. chacei are the only species in this section of the genus in which the sixth abdominal somite is armed posteriorly with 8 spines; in E. chacei the accessory spine is between the submedian and the intermediate on each side, whereas in E. pacifica it is between the intermediate and lateral spines. The new species differs from E. holthuisi in having rounded posterolateral angles on the lateral processes of the sixth and seventh thoracic somites, 8 rather than 6 posterior spines on the sixth abdominal somite, more dorsal tubercles on the telson (E. holthuisi lacks the accessory median row), and, on the basal pro- longation of the uropod 20-22 rather than 10-12 inner spines, and a spinule rather than a rounded lobe between the apical spines. All 3 of the similar western Atlantic species have a sharp apical spine on the rostral plate. The unique specimen of E. pacifica is young, and older specimens may well have an apical spine on the rostral plate. The specific name alludes to the occurrence of the species in the central Pacific ocean. LITERATURE CITED BicELow, R. P. 1901. The Stomatopoda of Puerto Rico. Bull. U-S. Fish Comm. 20(2):149-160, figs. 1-13. Cuopra, B. 1939. Stomatopoda. The John Murray Expedition, Sci. Reps. 6(3):137—181, figs. 1-13. MANNING, RAyMonp B. 1969. Stomatopod Crustacea of the western Atlantic. Stud. trop. Oceanogr. Miami 8: viii + 380, figs. 1-91. 1970. Nine new American stomatopod crustaceans. Proc. Biol. Soc. Washington 83(8):99-144, figs. 1-9. Scumitt, Waxtpo L. 1940. The stomatopods of the west coast of America based on collections made by the Allan Hancock Expeditions, 1933-1938. Allan Hancock Found. Pacif. Exped. 5(4):129-295, figs, 1-33. No. 24, pp. 253-262 15 August 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON TWO NEW SPECIES OF THE INDO-WEST PACIFIC. GENUS CHORISQUILLA (CRUSTACEA, ) STOMATOPODA), WITH NOTES ON C. EXCAVATA (MIERS) By Raymonp B. MANNING Smithsonian Institution, Washington, D.C. 20560 In 1972 I visited the Zoological Survey of India (Indian Museum ), Calcutta, to study their extensive collection of types and other specimens of Indo-West Pacific stomatopods. Among the materials examined were 3 lots from the Andaman Islands that had been identified by Kemp (1913) with Gonodactylus excavatus Miers, 1880, the type-species of Chorisquilla. Earlier I had examined material from several Indo-Malayan localities in the Zoological Museum, Amsterdam that had been identified with the same species by Hansen (1926) and Holthuis (1941), and I had noted discrepancies in the published accounts of the species. In April 1973 I studied the holotype of C. excavatus, from an unknown locality, and additional specimens from the South China Sea that had been identified with that species by Pocock (1893) in the collection of the British Museum (Natural History). The specimens reported by Kemp proved to be a new species and those reported by Pocock, Hansen, and Holthuis represent a second new species. These two new species are described below, and a brief description of the unique type of C. excavata is included. In the descriptions given below, particular attention is given to characters afforded by the sixth abdominal somite and telson. Except for those parts of the body, all of the species of Chorisquilla are very similar. Characteristics of the genus were summarized in an earlier paper (Manning, 1969). 24——Proc. Biot. Soc. WAsH., Vou. 88, 1975 (253) Mf =] 254 Proceedings of the Biological Society of Washington Studies on Indo-West Pacific stomatopods have been sup- ported by the Smithsonian Institution through its Research Award and Foreign Currency Programs. I thank K. K. Tiwari and H. C. Ghosh, Zoological Survey of India (IM), for provid- ing working space and assistance in 1972 as well as for a subsequent exchange of material; and R. W. Ingle, British Museum (Natural History) (BMNH), and J. Stock, Zoological Museum, Amsterdam (ZMA), for providing working space and for the loan of specimens. Figures 1 and 2 were prepared by my wife Lilly. Chorisquilla excavata (Miers, 1880) Figure la—b Gonodactylus excavatus Miers, 1880: 123, pl. 3, figs. 11-12. Protosquilla excavata.—Brooks, 1886: 78 (listed; diagnosis from Miers, 1880). not Gondactylus excavatus——Gravier, 1937: 209, fig. 23—Seréne 1939: 349 (discussion) —Dawydoff, 1952: 145, 146 (discussion ).—Seréne, 1953: 507; 1954: 53 (discussion) (= Haptosquilla glyptocercus ( Wood- Mason) ). Material: Locality unknown; holotype, 14, total length (TL) 19 mm; BMNH 75.14. Diagnosis: Sixth abdominal somite (Fig. la) with 6 longitudinal bosses, submedians and intermediates unarmed, intermediates irregular in shape. Posterolateral angles of sixth abdominal somite forming sharp spines. Telson (Fig. la) slightly broader than long, with 5 smooth longitudinal bosses dorsally. Median boss (Fig. la) tapering pos- teriorly, extending almost to median cleft in posterior margin. Sub- median bosses smooth, terminating at base of submedian marginal tooth. Marginal bosses smooth, extending to apex of intermediate teeth. Lateral margins of dorsal bosses smooth, lacking any trace of vertical furrows characteristic of C. pococki, new species and C. andamanica, new species. Marginal teeth of telson well separated, with broad but sharp apices; submedians with movable apices arising submarginally. 7 pairs of sub- median (Fig. la) and 2 pairs of intermediate denticles present. Remarks: Chorisquilla excavata differs from both C. pococki, new species and C. andamanica, new species in several characters: there are smoother bosses on the sixth abdominal somite and telson; the median boss of the telson extends almost to the anterior edge of the median incision in the telson margin; the apices of the marginal teeth of the telson are sharp; and there are only 7 pairs of submedian denticles. It further differs from C. andamanica in having sharp posterolateral spines on the sixth abdominal somite. New species and notes on Chorisquilla 255 ) Fic. 1. Chorisquilla excavata (Miers), ¢ holotype: a, Outline of sixth abdominal somite and telson; c, Outline of rostral plate and ocular scales. Chorisquilla pococki, new species, 6 paratype, Siboga sta. 133: b, Outline of sixth abdominal somite and telson. (Setae omitted). C Miers (1880, p. 123) clearly described the smooth longitudinal bosses on the telson which are shown in his illustration of the telson (pl. 3 fig. 12); however, his illustration was inaccurate in showing the _ inter- mediate marginal teeth of the telson extending well beyond the sub- medians. A camera lucida sketch of the telson of the holotype is given here (Fig. la); the marginal teeth, separated by a deep incision, extend to the same level. The male holotype is now preserved dry; it could not be measured in detail. The type-locality is unknown. The specimen was among ma- terials collected during the voyage of H.M.S. “Herald.” Other material collected by the “Herald” and reported by Miers was from Bau, British Solomon Islands; Matuka and Conway Reef, Fiji Islands; and Shark’s Bay, Western Australia. Presumably the species lives in the southwestern Pacific Ocean, but its range is unknown. The specimens from Viet Nam reported and figured by Gravier (1937) are identifiable with Haptosquilla glyptocercus (Wood-Mason), as indicated by Seréne (1954, p. 53). Inasmuch as Seréne (1954, p. 6) did not include this species in his list of stomatopods from Viet Nam, it seems likely other earlier records of C. excavata from Viet Nam (Seréne, 1939, 1953; Dawydoff, 1952) are based on Gravier’s er- roneous identification. 256 Proceedings of the Biological Society of Washington Chorisquilla pococki, new species Figures 1b, 2 Gonodactylus excavatus.—Pocock, 1893: 476.—Odhner, 1923: 15.— Hansen, 1926, 36, pl. 2, fig. 4a.—Holthuis, 1941, 290. Material: Macclesfield Bank, South China Sea; ¢ holotype, total length (TL) 17 mm; ¢ paratype, TL 12 mm; BM(NH) 92.8—28.9-10.— Same; 2 paratype, TL 18 mm; BM(NH) 93.11.3.153.—Off Ampenan, west coast of Lombok, Indonesia; 13 fms (24 m); Van de Sande, col.; May 1909; ¢ paratype, TL 34 mm; ZMA.—Anchorage off Lirung, Salibabu Island, Talaud Islands, Indonesia; to 36 m; mud and hard sand; “Siboga” sta. 133; 4g paratype, TL 14 mm; ZMA.—Anchorage between Nusa Besi (Ilhea de Jaco), and the northeast point of Timor, Portuguese Timor; 08°25.2’S, 127°18.4’E; 27-54 m; sand, coral and lithothamnion; “Siboga” sta. 282; ¢ paratype, TL 15 mm; ZMA. Diagnosis: Fifth abdominal somite (Fig. 2b) smooth dorsally, with acute or sharply-spined posterolateral angles and 3 longitudinal grooves above margin. Sixth abdominal somite (Figs. 1b, 2b) with 3 pairs of dorsal bosses, submedians and intermediates unarmed, smooth dorsally, with irregular, eroded margins; lateral bosses smoother, each terminating in sharp posterolateral spine. Telson (Figs. 1b, 2b) length and width subequal (length greater in some specimens, width in others), appearing elongate, with 5 prominent longitudinal bosses on dorsal surface, smooth dorsally, margins of median and submedians very irregular, eroded. Median boss of telson not extending to anterior margin of median in- cision in telson. Submedian bosses (Figs. 1b, 2b) extending almost to bases of submedian teeth, posteriorly with transverse ridge extending to midline behind apex of median boss. Marginal bosses smooth, rounded dorsally, not furrowed laterally or mesially. Telson with 2 pairs of mar- ginal teeth, apices well-separated by intermediate incision. Submedian teeth with distinct, sharp dorsal carina in small specimens, carina ob- scure in larger examples, apices of submedian teeth sharper than those of intermediates; submedian teeth each with movable submarginal spine. Broad intermediate tooth fusing with marginal carina in adults. 7—12 slender submedian denticles (Figs. 1b, 2b), fewer in larger specimens, and 2 intermediate denticles present (1 on largest male). Uropodal exopod with 9-10 blunt, movable spines on outer margin of proximal segment of exopod. Measurements: Males (5), TL 12-34 mm; only female examined, TL 18 mm. Odhner (1923) reported a female 26.5 mm long. Other measurements, of male, TL 34 mm, in mm: carapace length 7.4; cornea width 1.7; rostral plate length 2.3, width 2.7; fifth abdominal somite width 6.2; telson length 5.7, width 5.9. Remarks: Chorisquilla pococki resembles C. excavata in basic facies and it is not too surprising that earlier authors did not recognize it as New species and notes on Chorisquilla 257 Fic. 2. Chorisquilla pococki, new species, @ paratype, Macclesfield Bank, South China Sea: a, Anterior portion of body; b, Posterior 2 abdominal somites, telson, and uropod; c, Outline of submedian denticles of telson, ventral view; d, Uropod, ventral view. (Setae omitted). 258 Proceedings of the Biological Society of Washington a distinct species. The characters that can be used to separate the two species are listed under the account of C. excavata. Smaller specimens differ from the large male, TL 34 mm, in having more distinct carinae on the submedian and intermediate teeth of the telson (Fig. 1b) and in having more (9-10 rather than 7) submedian and intermediate (2 rather than 1) denticles on the telson. The sharp carina of the intermediate teeth merge with the marginal carinae in large specimens, but even in the largest male there is a faint sulcus where the carinae meet. I have not seen the material reported from the Bonin Islands by Odhner (1923), but inasmuch as he mentions the posterolateral spine on the sixth abdominal somite and refers to Kemp’s illustration (in which furrowed dorsal carinae are shown) I believe that his material should be identified with this species rather than with C. excavata. Pocock (1893) recorded one specimen of this species; there are three specimens from Macclesfield Bank in his material which was deposited in the British Museum (Natural History). Hansen (1926) recorded three specimens, from “Siboga” stations 133, 282, and 299. The vial from station 299 was found to include a thalassinid, not a stomatopod. The stomatopod from station 299 may now be lost. The species is named for R. I. Pocock. Distribution: Hatsune-ure, Bonin Islands (Odhner, 1923); Maccles- field Bank, South China Sea, 26 fms (48 m) (Pocock, 1893); off Ampenan, Lombok, Indonesia, 13 fms (24 m) (Holthuis, 1941); off Lirung, Salibabu Island, Talaud Islands, Indonesia, 36 m (Hansen, 1926); off northeastern Timor, Portuguese Timor, 27-54 m (Hansen, 1926); and off Roti Island, Lesser Sunda Islands, Indonesia, 34 m (Hansen, 1926). Chorisquilla andamanica, new species Figure 3 Gonodactylus excavatus—Kemp, 1913: 187, pl. 10 figs. 122, 123.— Manning, 1969: 158 (listed). Chorisquilla lenzi—Manning, 1969: fig. 6 (error in figure legend). Chorisquilla excavata.—Manning, 1969: 159 (key). Material: Off Andaman Islands: 20 fms (37 m); “Investigator”; 2 6, 49 (6, TL 18 mm, is holotype, others are paratypes); IM 9177/6 (1g, TL 13 mm, paratype, USNM 143562)—Off Little Andaman Island; 10.5 fms (19 m); “Investigator”; paratypes, 1¢, 39, TL 16— 26 mm; IM 9850-3/6.—Off Table Island, Coco group, Andaman Islands; 9.5 fms (17 m); “Investigator”; paratypes 16, 39, TL 13-16.5 mm; IM 9846-9/6. Diagnosis: Fifth abdominal somite (Fig. 3b) smooth dorsally, with sharp posterolateral angles and 2 longitudinal grooves above margin. Sixth abdominal somite (Fig. 3b) with 3 pairs of dorsal bosses, smooth dorsally, none armed, margins of intermediates irregular in outline. New species and notes on Chorisquilla 259 Fic. 3. Chorisquilla andamanica, new species (from Kemp, 1913, pl. 10 figs. 122, 123): a, Anterior part of body; b, Posterior 2 abdominal somites, telson, and uropod. Telson (Fig. 3b) length and width subequal, appearing elongate, with 5 prominent longitudinal bosses, smooth dorsally, on dorsal surface, margins of median and submedians very irregular, eroded; submedians with smaller, separate lobe posteromesially. Median boss (Fig. 3b) of telson not extending to anterior margin of median posterior incision, separated from incision by transverse ridge extending from each dorsal submedian carina. Submedian bosses extending almost to bases of sub- median teeth. Marginal bosses smooth, rounded dorsally, not furrowed laterally or mesially. Telson with 2 pairs of marginal teeth, separated by shallow incision, submedians extending slightly posterior to inter- mediates. Submedian tooth smooth dorsally, lacking distinct dorsal carina, apices acute but blunt, with movable, submarginal apical spine. Intermediate tooth, broader than submedian, apices blunt, obtuse. 11-12 slender submedian (Fig. 3b) and 2 slender intermediate denticles present. Uropodal expod with 9-11 blunt, movable spines on outer margin of proximal segment. 260 Proceedings of the Biological Society of Washington Measurements: Kemp (1913) reported 14 specimens with total lengths ranging from 11.5 to 26 mm. Other measurements of a female, TL 18 mm, in mm: carapace length 2.7; cornea width 0.8; rostral plate length 1.2, width 1.3; antennal scale length 1.1; fifth abdominal somite width 2.1; telson length 2.3, width 2.1. Remarks: Chorisquilla andamanica is very similar to both C. excavata and C. pococki, agreeing with the latter and differing from the former in having laterally eroded and irregular median and dorsal submedian bosses on the telson. It further agrees with C. pococki in having a relatively short median boss on the telson which does not extend to the anterior margin of the median incision. It differs from both species in lacking posterolateral spines on the sixth abdominal somite and in having much smaller marginal incisions between the submedian and _inter- mediate marginal teeth. Kemp's figure of this species is so good that I have decided to re- produce it here (Fig. 3) rather than have a new illustration prepared. It was also reproduced in Manning (1969, Fig. 6) where it was inad- vertently labelled Chorisquilla lenzi. The name is derived from the type-locality, the Andaman Islands. Distribution: Known only from the Andaman Islands, in depths between 9.5 and 20 fms (17 and 37 m). LITERATURE CITED Brooxs, W. K. 1886. Report on the Stomatopoda collected by H.M.S. “Challenger” during the years 1873-1876. Rep. Voy. Challenger, Zool. 16:1—116, pls. 1-16. DawyporF, C. 1952. Contribution a ]’étude des invertébrés de la faune marine benthique de |l’Indochine. Bull. Biol. France Belgique, supp. 37:1—-158, maps 1-2. Gravier, C. 1937. Stomatopodes des cétes dIndo-Chine. Ann. Inst. Océanogr. Monaco 17(3):175—-211, figs. 1-23. Hansen, H. J. 1926. The Stomatopoda of the Siboga Expedition. Siboga Exped., Monogr. 35:1-48, pls. 1-2. Hottuuts, L. B. 1941. The Stomatopoda of the Snellius Expedition. Biological results of the “Snellius” Expedition, no. 12. Tem- minckia 6:241—-294, figs. 1-9. Kemp, S. 1913. An account of the Crustacea Stomatopoda of the Indo-Pacific region, based on the collection in the Indian Mu- seum. Mem. Indian Mus. 4:1—217, 10 text-figs., pls. 1-10. MANNING, RAYMOND B. 1969. Notes on the Gonodactylus section of the Gonodactylidae (Crustacea, Stomatopoda), with descrip- tions of four new genera and a new species. Proc. Biol. Soc. Washington 82:143-166, figs. 1-8. Miers, E. J. 1880. On the Squillidae. Ann. Mag. Nat. Hist. (5) 5:1-30, 108-127, pls. 1-3. New species and notes on Chorisquilla 261 OpuNER, T. 1923. Indopazifische Stomatopoden. Géteborg Vetensk. Samh. Handl. (4) 27(4):1-16, figs. 1-10. Pocock, R.I. 1893. Report upon the stomatopod crustaceans obtained by P. W. Basset-Smith, Esq., Surgeon R.N., during the cruise, in the Australian and China Seas, of H.M.S. “Penguin,” Com- mander W. U. Moore. Ann. Mag. Nat. Hist. (6):11:473-479, pl. 20B. SERENE, R. 1939. Note sur les stomatopodes des eaux indochinoises. Bull. Soc. Zool. France, Paris 64:343-349, figs. 1-4. 1953. Sur la collection des stomatopodes de _ I'Institut Océanographique de l’Indochine. Proc. 7th Pacific Sci. Congr., 1949, 4, Zool. p. 506-508. 1954. Observations biologiques sur les stomatopodes. Mém. Inst. Océanogr. Nhatrang 8:1-91, text-figs. 1-15, pls. 1-10. 262 Proceedings of the Biological Society of Washington No. 25, pp. 263-280 15 August 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON NOTES ON INDO-PACIFIC PIPEFISHES | (PISCES: SYNGNATHIDAE) WITH DESCRIPTION || OF TWO NEW SPECIES By C. E. Dawson aAnp J. E. RANDALL Gulf Coast Research Laboratory Museum, Ocean Springs, | Mississippi 39564, and Bernice P. Bishop Museum, P. O. Box 6037, Honolulu, Hawaii 96818 Among Indo-Pacific pipefishes recently received for study are two undescribed species and several specimens that provide important additions to our knowledge of five poorly known species. The new species are described here and descriptive notes and comment are included for other material. Measure- ments are in millimeters (mm); body proportions are referred to standard length (SL) or head length (HL); unless other- wise stated, coloration is described from specimens preserved in alcohol; depths are reported in meters (m), distances in kilometers (km). Study materials are deposited in collections of the Academy of Natural Sciences of Philadelphia (ANSP), Bernice P. Bishop Museum (BPBM), California Academy of Sciences (CAS), Gulf Coast Research Laboratory Museum (GCRL), Museo Storia Natural, Genoa (MSNG), Rijksmu- seum van Natuurlijke Historie (RMNH), University of Guam (UG) and National Museum of Natural History, Smith- sonian Institution (USNM). Appreciation is expressed to S. S. Amesbury (UG), M. Boeseman (RMNH), J. E. Bohlke and W. F. Smith-Vaniz (ANSP), B. A. Carlson (Univ. S. Pacific, Fiji), W E. Esch- meyer and Pearl Sonoda (CAS), L. W. Knapp (Smithsonian Oceanographic Sorting Center), M. McCoy, E. Tortonese (MSNG) and V. G. Springer (USNM) for making materials available for study. The holotype of Micrognathus boothae 25—Proc. Bron. Soc. WAsH., Vou. 88, 1975 (263 ) 264 Proceedings of the Biological Society of Washington Fic. 1. Delineation of head of Micrognathus boothae USNM 213481. was reexamined by J. R. Paxton, Australian Museum. Data on the holotype of Halicampus macrorhynchus were provided by A. C. Wheeler (BMNH), British Museum (Natural His- tory). Dr. Springer’s Indonesian material was collected dur- ing Rumphius Expedition I of the Indonesian Institute of Oceanology. Seychelles and Amirantes collections were by J. E. Bohlke and associates during The International Indian Ocean Expedition—-Seychelles Island Program 1964 (IIOE). Specimens of Corythoichthys amplexus from the Solomon Is- lands collected by the junior author and G. R. Allen were ob- tained during an expedition supported by the National Geo- graphic Society. Drawings are by Anne Langenfeld (GCRL). This study was in part supported by National Science Founda- tion Grant GB-31053X to the senior author. Micrognathus boothae Whitley Figure 1 Material examined: USNM 213481 (1, 130 mm SL); Fiji Is., Wailangilala Is., 16°45’S, 179°02’E, coral reef, 0.3-6.4 m, TE VEGA Cr. 7, Sta. 295, 26 May 1965. Counts and measurements: Dorsal fin rays 20; caudal rays 10; pectoral rays 10-11; anal rays 4; rings 14 + 41; subdorsal rings 0.5 + 4 = 4.5; brood pouch occupies anterior 13 tail rings, contains eggs over 11 rings. Head length 7.8 mm (16.1 in SL); snout length 2.3 (3.4 in HL); eye diameter 1.7 (4.6 in HL); dorsal fin base 10.0 (0.8 in HL); caudal fin length 2.9. Descriptive notes: Median frontal and nuchal ridges slightly de- veloped, prenuchal ridge obsolete. A rudimentary opercular ridge crosses anterior 14 of opercle; distinct ridges absent from pectoral cover plate but both opercle and cover plate crossed by striae so as to impart a pocked or waffle-like surface configuration. Lateral tail ridge terminates below dorsal fin origin; scutellar width about equals 1% distance between Notes on Indo-Pacific pipefishes 265 proximal margins of adjacent scutella. Brood pouch contains single layer of eggs arranged in four transverse rows, with outermost row containing 29-30 eggs. Pouch folds voluminous, will meet on ventral midline but margins not thickened and there is nothing to indicate that folds overlap in life. Ground color dark brown, venter and distal 24 of brood pouch folds lighter. Snout tip tan; with indications of widely spaced narrow pale bars crossing dorsum: three on trunk, five behind dorsal fin. Pectoral and anal fins pale; dorsal fin with narrow shading of melanophores proximad, elsewhere pale. Discussion: Whitley (1964) described M. boothae from a single specimen (presumably female) from Lord Howe Island. Meristic data were given as: D 21, A 2, P 11, C 7, rings 14+ 42 and subdorsal 1+ 5. Dr. J. R. Paxton has reexamined this fish at our request and finds 22 dorsal, 10 caudal and 4 anal rays. Whitley (p. 163) described the dorsal fin base as “extending over almost six tail-rings” and it ap- pears that relative dorsal fin lengths and positions are very similar in these specimens. The Fiji specimen represents the second known specimen, and only known male, of M. boothae, and extends the species range more than 2200 km to the northeast of the type-locality. A month of intensive col- lecting of fishes at Lord Howe Island in 1973 by Randall and nine other ichthyologists resulted in 278 species of fishes (189 were new records), but failed to produce additional specimens of M. boothae. Micrognathus brocki Herald Material examined: Indonesia: USNM 209722 (1, 64 mm SL), Tandjung Suli, 11 Jan. 1973. Mariana Is.: UG 5776 (+ 58, tail damaged), Guam, 22.9 m, 27 Dec. 1971. Counts and measurements: (USNM 209722): Dorsal fin rays 21; pectoral rays 12-11; caudal rays 10; anal rays 4; rings 14 + 35; sub- dorsal rings 0.5 + 4.5 = 5.0. Head length 6.1 mm (10.5 in SL); snout length 2.2 (2.8 in HL); eye diameter 1.3 (4.7 in HL); dorsal fin base 5.6 (1.1 in HL); caudal fin length 2.1. Descriptive notes: Median snout ridge tripartite (see Herald 1953: 259, fig. 39g); 5 spinules on anterior, 4 on middle and 2 on posterior segments (8, 3 and 2 in Guam specimen). Supraorbital cirri branched, right cirrus about 2.4 mm in length, left somewhat shorter; suborbital cirri shorter than eye diameter, distally branched; eye with about 12, equally spaced, fleshy flaps or cirri; cirrus on frontal crest about 3.5 mm long, profusely branched proximad but distal half simple; prenuchal crest with short, broad, leaflike cirrus; single branched cirrus near middle of each opercle. Most trunk rings bear minute cirri near dorsal and ventral margins; 4th, 8th, 12th and 16th rings with expanded, leaflike cirri on each lateral and dorsal ridge (4 per ring). Ground color white; lower half of opercle brownish; two brown bands pass ventrad between lateral midlines of snout; dorsum and sides of body crossed by about 9 266 Proceedings of the Biological Society of Washington indistinct brownish bands (3-4 rings wide) each alternating with 1-2 pale rings; venter with brownish shading on pectoral ring followed by narrow brown bars between each of next 12-13 rings. Discussion: | Micrognathus brocki was described from a 110 mm SL female (?) taken at Bikini Atoll, Marshall Islands. A second specimen (82 mm SL) was subsequently reported from the Ryukyu Islands (Herald and Randall, 1972), but egg bearing males are yet to be recognized. The present Indonesian juvenile extends the known range more than 5500 km west southwest of the type-locality and suggests widespread distribution in central Pacific waters. Micrognathus edmonsoni (Pietschmann ) Material examined: ANSP 128417 (1, 62 mm SL); Fiji Is., Viti Levu, E side Mbengga Is., ca. 37 km SW of Suva harbor, 24.4 m, 14 Apr. 1974. Counts and Measurements: Dorsal fin rays 19; caudal rays 10; pectoral rays 12; anal rays 4; rings 14 + 38; subdorsal rings 0.25 + 3.75 = 4.00 Head length 5.4 mm (11.5 in SL); snout length 2.3 (2.4 in HL); eye diameter 1.1 (4.9 in HL); dorsal fin base 4.4 (1.2 in HL); caudal fin length 2.4. Descriptive notes: Median snout ridge low, entire, more or less bi- partite, the shorter anterior portion slightly notched behind; opercular ridge indistinct, extends across 7% of opercle; remaining head ridges low, with entire margins; one ridge on pectoral cover plate. Superior trunk and tail ridges prominent, margins minutely serrate over all but last 10-12 tail rings; other trunk and tail ridges entire. Ground color pale; sides with traces of about 13 widely spaced brownish blotches; ventral part of head lightly shaded with brown; venter elsewhere with brown shading on Ist and 2nd trunk rings and indications of narrow brown bars between remaining trunk rings. Discussion: Compared with a 65 mm SL fish from the Marquesas Is. (CAS 13977), the present specimen has more prominent superior trunk and tail ridges with somewhat more pronounced serrations; pectoral fin rays are 12 in both. Herald (1953) reported M. edmonsoni to have rounded body ridges without serrations and 10 pectoral rays. Our ob- servations show pectoral rays to range from 10 to 12 and that superior trunk and tail ridges are serrate in juveniles. Originally described from Hawaii, Micrognathus edmonsoni was recently reported from the Mar- quesas Is. (Herald and Randall, 1972). The present record extends the known range some 5144 km southwest of the type-locality. Syngnathus banneri Herald and Randall Figures 2 and 3 Material examined: Amirante Isles: ANSP 128817 (1, 36 mm SL), Remire Reef, NE of Eagle Is., ca. 9.1 m, 4 Mar. 1964, IIOE Sta. F-82; ANSP 128818 (1, 43), off E side D’Arros Is., 15.2-27.4 m, 5 Mar. Notes on Indo-Pacific pipefishes 267 Fic. 2. Head and anterior body of Syngnathus banneri GCRL 13350; 52 mm SL male. 1964, ILOE Sta. F-87. Seychelles: ANSP 128819 (1, 58), GCRL 13350 (1, 52), off S shore of Faon Is., 10.7-13.7 m, 22 Jan. 1964, IIOE Sta. F-11; ANSP 128816 (1, 52), vicinity of Praslin Is., S of Round Is., 12.2-15.2 m, 22 Feb. 1964, ITOK Sta. F-61. Mauritius: BPBM 16343 (2, 41-46), W coast, cave in 30 m, 20 Nov. 1972. Indonesia: USNM 213482 (1, 35), Saparua, S of Kampungmahu, 9.1 m, 18 Jan. 1973. Taiwan: USNM 213810 (1, 44), SW _ shore off Ch’uan-fan-shih, 21°55’48”"N, 120°48’48”E, 7.6-8.5 m, 2 May 1968. Counts and measurements: Dorsal fin rays 17-20; pectoral rays 12-14; anal rays 3-4 (usually 4); caudal rays 10; trunk rings 15; tail rings 27-29; subdorsal rings 0.5—1 + 3.25 = 44.75; head in SL 7.1-8.2; snout in HL 2.2—2.5; dorsal fin base in HL 1.4—1.7. Male brood pouch on anterior 11-14 tail rings. Descriptive notes: Median snout ridge (Fig. 3) well developed, be- gins at rear of upper jaw, deeply incised to form 3 or 4 distinct, semi- isolated, subtriangular sections; broadly truncate lateral projection located below level of nares and near middle of snout length; most speci- mens with small lateral spine just behind angle of gape. Orbital ridge expanded dorsad to form a broadly rounded or truncate dorsolateral pro- jection; postorbital ridge with a rather broad lateral projection just be- hind eye; median frontal, prenuchal and nuchal ridges elevated; supra- orbital ridge represented by a short knob-like projection; opercular ridge crosses about %-% of opercle; remaining opercular surface pocked between radiating striae; upper and lower pectoral cover plate ridges distinct in most specimens. Lateral trunk and tail ridges overlap on anal ring in one specimen, fail to overlap in other material; superior and inferior trunk and tail ridges somewhat elevated, well indented between rings of trunk and anterior 1% of tail, distinctly notched between remaining tail rings; body surfaces concave between ridges; head and body ridges rough to minutely serrate in all specimens. Scutella rather broad, greatest width about % distance between proximal margins of 268 Proceedings of the Biological Society of Washington mm Fic. 3. Syngnathus banneri. Top: lateral aspect of head and an- terior trunk rings. Bottom: Section of body illustrating ridges, dorsal and anal fins. Drawn from 43 mm SL female, ANSP 128818. adjacent scutella; neuromasts indistinct but apparently present on tail rings. Eye with ring of 10-12, equally spaced, minute flaps; cirri other- wise absent. Brood pouch folds of egg bearing males do not meet on ventral mid- line, and margins appear to be neither thickened nor otherwise modi- fied. A 46 mm SL fish (BPBM 16343) has the pouch extending over 12 rings and about 29 eggs in two longitudinal rows. A 52 mm speci- men (GCRL 13350) has two (occasionally three) transverse rows of eggs over 11 rings, about 17 in outer row and 41 in pouch. Most specimens are mainly pale; well marked individuals have faint indications of a narrow bar from posteroventral margin of eye across anterior portion of opercle; dorsum crossed by faint dark blotches (bars) on about every 4th ring behind pectoral ring; ventral portion of pectoral ring crossed by dusky bar; sides below lateral trunk ridges and venter crossed by indications of narrow bars on trunk rings 2 through 10; body and head without other markings; fins pale. In some specimens, bars on lower sides and venter interrupted, forming a series of vertically- Notes on Indo-Pacific pipefishes 269 oriented, oblong spots on each lateral trunk ridge, each inferior trunk ridge and on ventral ridge. One 41 mm SL fish (BPBM 16343) is mainly brownish, the dorsum crossed by 9-10 broad bars (bars to about 5 rings in length) separated by mottled pale areas (0.5-1 ring in length) and lower sides and venter ringed by narrow brown bars on each trunk ring. Discussion: Herald and Randall (1972) described S. banneri from two specimens (26.5-39 mm SL) from the Ryukyu and Marshall Islands. Reexamination shows 4 anal rays in each (rather than 2 as described) and 12 rather than 11 rays in the left pectoral fin of the paratype. AlI- though omitted from the original description and figure, the 39 mm holotype bears a minute lateral spine behind angle of gape and a larger lateral projection near middle of snout length; both processes are reduced but definitely present in the paratype. The condition of the holotype precludes observation of eye flaps, but some are present on the paratype. Pectoral rays are modally 13 in our material but meristic and morpho- metric characters fall within expected variation of banneri. Among Indian Ocean specimens, larger fish have shorter opercular ridges and greatest development of other head and body ridges. Herald and Randall speculated that banneri would have an overlapping brood pouch closure and thereby fall within the subgenus Microsyngnathus Herald. Out ma- terial shows that Syngnathus banneri has the “semi-brood-pouch closure” of Herald (1959) wherein eggs are arranged in one layer and margins of pouch folds do not meet on ventral midline. Pouch closure and egg arrangement suggest that banneri is intermediate between two of the four nominal subgenera of Syngnathus: Microsyngnathus (overlapping pouch folds, anal fin present) and Bryx Herald (non-overlapping folds, anal fin absent). Our specimens extend the known distribution to Indonesia and the western Indian Ocean. All specimens, including type material, have been taken in coral habitats within the 6.1-30 m depth range. Syngnathus darrosanus, new species Figures 4 and 5 Holotype: ANSP 128815 (47.4 mm SL); Amirante Isles, off E side of D’Arros Island (Admiralty Chart No. 724), 0-2.1 m, 6 Mar. 1964, IIOE Sta. F-91. Paratype: GCRL 13302 (43.3 mm SL), collected with holotype. Diagnosis: Dorsal fin rays 23; pectoral rays 12; anal rays 3; caudal rays 10 (in holotype); trunk rings 16; tail rings 29-30; subdorsal rings 6.25 (2 trunk + 4.25 tail rings); head in SL 7.9; snout in HL 2.8-2.9; dorsal fin base in HL 1.1. Body ridges indented between rings; all ridges smooth, neither serrate nor spinulose. Ridge pattern typical of Syng- nathus, i.e., lateral trunk ridge discontinuous with lateral tail ridge near anal ring, inferior trunk and tail ridges continuous and superior 270 Proceedings of the Biological Society of Washington Fic. 4. Syngnathus darrosanus ANSP 128815; holotype; 47 mm SL. trunk and tail ridges discontinuous near rear of dorsal fin. Trunk V- shaped ventrad; brood pouch unknown. Description: Measurements (mm) are as follows: head length 6.0 (5.5 in paratype); snout length 2.1 (2.0); eye diameter 1.1 (1.0); length of dorsal fin base 5.5 (5.2); pectoral fin length 1.2 (1.1); length of pectoral fin base 0.9 (0.8); caudal fin length 1.1 (damaged in para- type). Median snout ridge (Fig. 5) smooth, begins near middle of snout length with two well separated, subtriangular, dorsal expansions, not deeply incised; snout without lateral spines; orbital ridge low, not ex- panded dorsolaterad; median frontal, prenuchal and nuchal ridges low, indistinct. Opercular ridge extends over % to % of opercle, opercle elsewhere crossed by radiating striae imparting a waffle-like surface con- figuration; upper and lower pectoral cover plate ridges present. Trunk and anterior 14 of tail with slightly elevated, rounded ridges, only slightly indented between rings. Ridges on remainder of tail deeply incised between rings, margins elevated, tail surface concave between; head and body ridges entire. Lateral trunk and tail ridges overlap on 2nd tail ring in holotype; lateral trunk ridge ends on Ist tail ring in paratype and falls short of lateral tail ridge, which ends near middle of 2nd tail ring. Eye bears circle of 17-18 minute, equally-spaced, pale flaps or cirri; head and body elsewhere devoid of cirri; scutella oval, rather broad, width about 1% distance between proximal margins of adjacent scutella; neuromasts in short series on posterior half of each tail ring. Predominantly brown with faint mottling of tan; upper half of snout and tip of lower jaw pale; dorsum of trunk and tail with indications of several pale bars of irregular width and spacing; dorsal and pectoral rays streaked with brown, membranes mainly pale; caudal fin dark brown with narrow pale margin. Etymology: Named darrosanus, in reference to the type-locality, D’Arros Island. Discussion: Both specimens are assumed to be adult or subadult fe- males and future collections of egg bearing males will probably show this species to be a member of the poorly defined Bryx-Microsyngnathus group of subgenera. Syngnathus darrosanus seems most closely related Notes on Indo-Pacific pipefishes Dall mm Fic. 5. Syngnathus darrosanus. Top: Lateral aspect of head and anterior trunk rings. Bottom: Section of body illustrating ridges, dorsal and anal fins. Drawn from holotype. to such small forms as S. balli (Fowler) and S. banneri Herald and Randall. The combination of 16 trunk rings, 29-30 tail rings, 23 dorsal fin rays, smooth ridges, and short snout clearly separates S. darrosanus from congeners. Corythoichthys amplexus, new species Figures 6 and 7 Holotype: USNM 213479 (66 mm SL, male), Fiji Islands, Beqa Lagoon, patch reef between Stuart and Yanuca Islands, 6.1 m, rotenone, 16 Jan. 1974, B. Carlson and B. Goldman. Paratypes: Fiji Is.: BPBM 17949 (2, 44-54), GCRL 13337 (5, 44— 62), USNM 213480 (2, 45-54), taken with holotype. Solomon Is.: BPBM 15614 (2, 59-71), GCRL 13336 (1, 68), Alite Reef off Malaita, la- goon coral head, 3 m, 24 July 1974, J. E. Randall and G. R. Allen. Celebes Is.: USNM 213510 (8, 49-69), GCRL 13358 (2, 54-64), Kabena Is., 272 Proceedings of the Biological Society of Washington Fic. 6. Corythoichthys amplexus GCRL 13336; paratype; 68 mm SL. Tallabassi Bay, off NE tip of Big Damalawa Is., ca. 05°17’20"S, 122°04’00"E, 2-15 m, 25 Feb. 1974, V. G. Springer. New Guinea: USNM 213511 (1, 66), Madang harbor, inside S tip of Paeowai Is., 9.1-10.7 m, 26 May 1970, B. B. Collette. Seychelles: ANSP 110058 (8, 55-70), Mahé Is., NW end Beau Vallon Bay, isolated coral outcrop in sand, 6.1-7.5 m, 15 Mar. 1964, IIOE Sta. F-114; ANSP 110017 (1, 59), Beau Vallon Bay, NNW of Hotel des Seychelles, 12.2-15.2 m, 19 Mar. 1964, IIOE Sta. F-119; ANSP 108993 (1, 68), Mahé Is., W side of N islet, 7.6-18.3 m, 16 Mar. 1964, ITOK Sta. F-115; ANSP 110053 (5, 68-88), vicinity Praslin Is., S of Round Is., to 15.2 m, 22 Feb. 1964, IIOE Sta. F-61; ANSP 108972 (63, 50-89), BPBM 18033 (4, 65-70), GCRL 13463 (6, 70-88), Curieuse Is., S end Laraie Bay, to 9.1 m, 23 Feb. 1964, IIOE Sta. F-64. Amirante Isles: ANSP 110050 (5, 49-64), off E end D’Arros Is., 21.3-30.5 m, 6 Mar. 1964, IIOE Sta. F-89; ANSP 110036 (4, 55-76), African Is., SW of North Is., to 3.7 m, 3 Mar. 1964, IIOE Sta. F-80. Other material: GCRL 13338 (2, with regenerated tails), taken with holotype. ANSP 110052 (1, anomalous), Seychelles, off N tip of Mahé Is., ca. 18.3 m, 14 Feb. 1964, IIOK Sta. F-50. Notes on Indo-Pacific pipefishes 213 Fic. 7. Corythoichthys amplexus. Top: Lateral aspect of head and anterior trunk rings. Middle: Section of body illustrating ridges, dorsal and anal fins and portion of brood pouch. Bottom: Same section showing color pattern. Drawn from holotype. Diagnosis: Dorsal fin rays 24-29 (modally 26); pectoral fin rays 12-15 (14); caudal rays 10; anal rays 4; trunk rings 15-16 (15); tail rings 35-39 (37); subdorsal rings 4.5-6.25 (average 5.2); dorsal fin origin on Ist or 2nd tail ring; snout short, averages 2.5 in head length. Ridge pattern typical of Corythoichthys, i.e., lateral trunk ridge discon- 274 Proceedings of the Biological Society of Washington tinuous with lateral tail ridge near anal ring, inferior trunk and tail ridges continuous, superior trunk and tail ridges discontinuous near rear of dorsal fin. Trunk slightly V-shaped ventrad; brood pouch of males without protective plates, located on anterior 11-13.5 tail rings. Body ringed by 11-13 (usually 12) broad brown bands behind head, 4 bands from pectoral ring to 2nd—4th tail ring, 7-9 on remainder of tail; sides of head without wavy lines or stripes; no dark blotch near anus. Description: Counts and measurements (mm) of holotype: Dorsal fin rays 27; pectoral rays 14(2); caudal rays 10; anal rays 4; rings 15 + 38; subdorsal rings 5.5 from anterior edge of 2nd tail ring; brood pouch on 13 rings. Head length 6.9; snout length 2.8; eye diameter 1.7; length of pectoral fin base 0.9; caudal fin length 2.5. Range and means (in parentheses) of principal proportions for holotype and paratypes: head length in SL 7.9-12.0 (9.5); snout length in HL 2.2-2.8 (2.5); dorsal fin base in HL 1.0-1.4 (1.1). Trunk rings 16 in one fish, 15 in remainder of 124 examined. Median snout ridge, frontal, prenuchal and nuchal ridges smooth and moderately elevated in holotype (Fig. 7); opercular ridge almost reaches posterior margin of opercle; pectoral cover plate and opercle pocked between radiating striae; trunk ridges distinct, superior ridges mainly entire, somewhat elevated with dorsum concave between, dis- tinct notch between rings; superior tail ridges lower, indented between rings; brood pouch folds broad, their depth at 8th tail ring about % greater than remaining depth of tail; pouch with single layer of eggs in 4 transverse rows over 12 rings; egg diameter about 0.6 mm, 65 eggs in pouch. Ground color light tan, markings brown. Side of snout with dark blotch behind angle of gape, another close to eye; top and sides of head from interorbital to rear of frontal crest and between upper, an- terior part of opercles capped with brown; two narrow, dark blotches near ventral margin of opercle, another on gill membranes below; head elsewhere immaculate or peppered with dark chromatophores. Body circled by 4 brown bands from pectoral ring to middle or end of 3rd tail ring, pale interspaces about one ring wide dorsad, somewhat wider below; 7 bands on remainder of tail, last 3 with pigmentation faint and largely restricted to localized concentrations of dark chromato- phores in irregular circlets; color bands continue to margins of brood pouch folds but supramarginal pale areas included within 2nd and 3rd color bands crossing brood pouch (Fig. 7). Pectoral rays with scatter- ing of basal chromatophores, the fin otherwise pale; dorsal rays above color bands lined proximad with chromatophores, remainder of fin pale; caudal fin pale, with a single narrow dark bar near base. Dorsal fin origin between anterior margins of Ist and 2nd tail rings, most frequently near middle of Ist ring. Scutella rather broad, maximum width more than half distance between proximal margins of adjacent scutella. Head crests and body ridges most prominent in Solomons, Notes on Indo-Pacific pipefishes 275 Celebes and western Pacific material; superior trunk ridges minutely denticulate in most specimens examined; median snout ridge usually with slight dorsal emargination (Fig. 7). Details of coloration vary but basic broadly banded pattern is con- stant; snout lightly shaded with chromatophores in some fish, 1-2 lateral rows of up to 6 small brown spots in others; body bands usually less distinct across venter. Males usually have 2-4 moderately large, dark brown blotches on venter of Ist—3rd trunk rings, whereas females most frequently have several small, tan spots arranged in 2, more or less parallel, longitudinal rows. A color transparency of a fresh Solomon Islands specimen shows pale areas as bluish-white, the markings reddish- brown; dark bands are sprinkled with white and all bands are distinct rather than faded caudad as in most preserved material. A small species, which probably seldom exceeds 100 mm SL; among our material, the smallest male with brood pouch eggs is 55 mm SL. Etymology: amplexus, encircling, in reference to the broad encircling color bands; here used as a noun in apposition. Discussion: Herald (1953) recognized five species and two sub- species within Indo-Pacific populations of Corythoichthys. Subsequently, Smith (1963) showed C. intestinalis Ramsay to be a junior synonym of C. haematopterus (Bleeker), and both Smith (1963) and Klausewitz (1972) questioned the validity of Herald’s subspecies: C. intestinalis intestinalis (E. Africa to New Hebrides) and C. intestinalis waitei (Jordan and Seale) from New Hebrides, Fiji and other central Pacific localities. With the exception of C. ocellatus Herald, a long-snouted form with 30-31 tail rings described from the Solomon Islands, all nominal species exhibit considerable overlap in meristics and identifi- cation is largely based on differences in proportional characters and coloration. Corythoichthys amplexus is readily distinguished from all congeners by the combination of short snout, modally 15 trunk rings and broadly banded color pattern. Trunk rings are modally 16-18 in other Pacific species and counts of 15 have occurred in only 20 of 570 specimens examined (Herald, 1953 and our unpublished data). Frequency of trunk rings is one of the most stable meristic features of pipefishes and the modal count of 15 in amplexus is distinctive. The number of pectoral fin rays (13 or 14 in 92% of 235 counts) also differs from that of Pacific congeners wherein modal values are 15 or 16 (detailed meristic comparisons will be provided in a review of the genus now in preparation by Dawson). Other short-snouted species, C. flavofasciatus Riippell and C. haematopterus, also have banded coloration but this is rather poorly defined; the bands and pale interspaces are often about equal in length and there are either distinctive stripes or wavy lines on the head (absent in amplexus). In addition, males of flavofasciatus have a dark, perianal blotch and males (and some females) of haematopterus usually have prominent, dark bars crossing venter of the anterior 3-5 trunk rings. Available 276 Proceedings of the Biological Society of Washington Fic. 8. Head and anterior body of Halicampus macrorhynchus BPBM 17354; 149 mm SL. data suggest that C. amplexus prefers protected coral habitats within the 2-31 m depth range. Halicampus macrorhynchus Bamber Figures 8 and 9 Material examined: BPBM 17354 (1, 149 mm SL), Solomon Is., Guadalcanal, Point Cruz, Honiara, among “algae” covered rocks, 25 m, Jan. 1974. Counts and measurements: Dorsal fin rays 18; pectoral fin rays 19-18; caudal rays 10; anal rays 4; rings 14 + 25; subdorsal rings 2.75 (1.5 trunk + 1.25 tail rings). Head length 30.0 mm (5.0 in SL); snout length 18.5 (1.6 in HL); eye diameter 3.0; length of dorsal fin base 7.6 (4.0 in HL); pectoral fin length 2.5; length of pectoral base 3.6; caudal fin length 6.1. Description: Median snout ridge (Fig. 9) low, represented mainly by a series of distinct spines and short elevated ridges; additional spines laterad on snout; orbital ridge spiny anteriad, expanded above eye to form a large subtruncate, dorsolateral projection; suborbital ridge produced into a large, somewhat spiny, knoblike, lateral pro- trusion; opercular ridge anteriorly expanded to form a flat, medially- notched, lateral projection, ridge low behind, directed dorsad, crossing entire opercle; supraopercular ridge low in front, expanded into a broad, flat projection behind; pectoral cover plate with broad, distally curved, shelflike projection near lower pectoral angle and a small, flat, spinelike projection just before upper angle of fin; a strong ridge runs ventrad from anterior edge of cover plate, terminating in a substantial, recurved, hooklike knob at anterolateral margin of pectoral ring. Median frontal ridge rather low and denticulate in front, higher behind; pre- nuchal and nuchal ridges lower, more or less subconical; interorbital deep between supraorbital ridges. All body ridges distinct, their margins granular, clearly notched or indented between rings; dorsum and sides concave between ridges; trunk V-shaped ventrad. Inferior trunk and tail ridges discontinuous; lateral trunk ridge continuous with inferior tail ridge; superior trunk and tail ridges discontinuous below dorsal fin. Lateral trunk ridge produced Notes on Indo-Pacific pipefishes 207 Kt Fic. 9. Halicampus macrorhynchus BPBM 17354. Top and middle: Lateral and dorsal aspect of head and anterior trunk rings. Bottom: Section of body illustrating ridges, cirri, dorsal and anal fins. on each ring to form strong, recurved, hooklike projection. Superior trunk and tail ridges produced laterad, margins of rings 3-4, 7-8, 11, 13, 18 and 23 produced as retrorose hooks, remaining rings rounded or bluntly pointed; inferior trunk and tail ridges produced laterad, mar- 278 Proceedings of the Biological Society of Washington gins straight or slightly rounded. Dorsal fin base somewhat elevated; scutella suboval, width on trunk about % distance between proximal margins of adjacent scutella, about %4 this distance on tail; neuromasts present but indistinct. Two pairs of branched cirri on snout; branched cirri also present on: supra and suborbital projections, anterior part of opercular ridge, median frontal ridge; large feather-like cirrus im- mediately behind each hook like projection on superior trunk and tail ridges; dorsum with rounded, fleshy pads on scutella of 7th and 11th trunk rings, which may represent persistent bases of additional cirri; eye bears circle of 10-12 small, white tabs, with single large, brown flap above and below. Ground color light tan. Snout, dorsum and upper sides of head lightly and irregularly shaded with brown; pectoral ring mainly brown across upper sides and dorsum; about 4 indistinct bands on trunk, 6 on tail, bands most prominent across dorsum between lateral trunk ridges, less distinct or obsolete on lower sides; interspaces variously flecked with brown; pectoral and dorsal fin bases brownish, fins otherwise pale; caudal fin brown at base, traces of two narrow bars distad; venter immaculate. Discussion: Duncker (1940) described Phanerotokeus (type-species P. gohari Duncker) from Ghardaqa in the Gulf of Suez without men- tion of Bamber’s (1915) description of Halicampus macrorhynchus. Subsequently, Fowler (1956) erected Halicampoides on the basis of Bamber’s description and figure, but omitted reference to Duncker (1940). Smith (1963) treated Halicampoides as a junior synonym of Phanerotokeus and placed P. gohari in synonymy with macrorhynchus. Other references to this species include the description of a dried speci- men (as Phanerotokeus macrorhynchus) from Eilat by Tortonese (1968), and Botros’ (1971) listing as a possible Red Sea endemic (as both P. gohari and Halicampus macrorhynchus). Duncker (1940) failed to compare his material with Halicampus Kaup (1856), and the only char- acters differentiating Phanerotokeus (and Halicampoides) appear to be relative lengths of snout and the presence or absence of dermal cirri. Length of snout alone is not a meaningful basis for generic separation among pipefishes and the presence, state of development, or absence of cirri is variable within a species. Herre and Herald (1951) remarked on the relatively long snouts of their specimens of Halicampus grayi Kaup [= H. koilomatodon (Bleeker)] compared with the speci- men figured by Weber and DeBeaufort (1922:103, fig. 43). Snout length is about 2.7 in head length in the figured specimen whereas Herre and Herald reported this ratio as 2.0 for their material. Our examination of H. koilomatodon (9, 66-94 mm SL; including two Bleeker specimens, RMNH 7226) shows a 2.1-2.5 range for this character. In addition, there are hooklike protrusions on lateral trunk ridges, ventrolateral projections on the pectoral ring and projections on pectoral cover plate and dorsum of head, which are essentially similar Notes on Indo-Pacific pipefishes 279 to those on the fish described here. Body ridge configurations agree in both nominal genera; dorsal fin bases are elevated and the male brood pouch is subcaudal in both, with eggs deposited in cutaneous cells protected by short, lateral folds and poorly developed, protective plates. In the absence of substantial differentiating characters we place Phanerotokeus in synonymy with Halicampus Kaup. The Guadalcanal specimen is more robust and deeper bodied than examined Red Sea material [USNM 166931 (8, 120-145 mm SL); MSNH 40551 (1, 155)]. Compared with a 145 mm SL male from Ghardaqa, minimum snout depth is 2.4 mm (against 1.6), minimum snout breadth 1.8 (1.3), maximum head breadth at opercular ridge 8.2 (5.9), maximum breadth at lower pectoral ridge 10.5 (8.6), maxi- mum trunk depth 9.2 (7.3) and maximum trunk breadth 9.0 (7.6). Head and body ridges are much larger and more prominent in the Pa- cific fish and there are no cirri on the snout of any Red Sea specimens although body cirri are variously present or absent. Mr. Wheeler advises that the holotype of Halicampus macrorhynchus (BMNH 1915.10.25.2 ) lacks cirri on the snout and has somewhat reduced hooklike lateral trunk ridges. Similar hooks occur on some of the examined Red Sea fish, but lateral trunk ridges are mostly straight with but a slight notch near the posterior margins. The holotype and other Red Sea specimens have 26-27 tail rings (25 in Pacific specimen) and subdorsal rings total 3-3.25 (against 2.75), but other meristics of the Pacific speci- men fall within the range of examined Red Sea material. We consider the Guadalcanal specimen conspecific with H. macrorhynchus; it con- stitutes the first Pacific record of a species previously thought to be a Red Sea endemic. Additional material may prove observed differences to be constant and subspecific treatment may be warranted for the Pacific population. LITERATURE CITED BAMBER, R. C. 1915. Reports on the marine biology of the Sudanese Red Sea, from collections made by Cyril Crossland, M. A., D. Sc., F. L. S.—XXII. The fishes. Jour. Linn. Soc., Zool. 31:38- 485, pl. 46. Botros, G. A. 1971. Fishes of the Red Sea. Oceanogr. Mar. Biol. Ann. Rev. 9:221-348. Duncker, G. 1940. Ueber einige Syngnathidae aus dem Roten Meer. Publ. Mar. Biol. Sta. Ghardaqa (Red Sea) 3:83-88. Fow.ier, H. W. 1956. Fishes of the Red Sea and southern Arabia. Vol. 1. Weitzmann Science Press, Jerusalem. 240 pp. Heratp, E. S. 1953. Family Syngnathidae: Pipefishes. In: L. P. Schultz et al. Fishes of the Marshall and Marianas Islands. Bull. U. S. Nat. Mus. 202(1):231-278, figs. 36—44. 1959. From pipefish to seahorse—a study of phylogenetic relationships. Proc. Calif. Acad. Sci., Ser. 4, 29(13):465—473. 280 Proceedings of the Biological Society of Washington AND J. E. RANDALL. 1972. Five new Indo-Pacific pipefishes. Proc. Calif. Acad. Sci. 39(11):121-140. Herre, A. W. ano E. S. HERALD. 1951. Noteworthy additions to the Philippine fish fauna with descriptions of a new genus and species. Philippine Jour. Sci. 79(3):309-339. KLaAusewitz, W. 1972. Litoralfische der Malediven. I. Einleitung und Fische der Ordnung Syngnathiformes (Pisces: Teleostei). Senckenbergiana Biol. 53( 3/4) :199—217. Kaup, J. J. 1856. Catalogue of lophobranchiate fish in the collection of the British Museum. Taylor and Francis, London. 76 pp. SmirH, J. L. B. 1963. Fishes of the family Syngnathidae from the Red Sea and the Western Indian Ocean. Ichthyol. Bull. Rhodes Univ. 27:515-543. TorTONESE, E. 1968. Fishes from Eilat (Red Sea). Bull. Sea Fish. Res. Sta., Haifa 51( 40) :6-30. Weber, M. anp L. F. DEBEAurorr. 1922. The fishes of the Indo- Australian Archipelago. E. J. Brill, Leiden 4:i-xiii + 1-410, 103 figs. WuitLey, G. P. 1964. Fishes from the Coral Sea and the Swain Reefs. Rec. Austr. Mus. 26(5):145-195. = i 0 7 : : ’ y ; ' f i _ a ae ee i ). 26, pp. 281-304 15 August 1975 B44 NH PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON THE RELATIONSHIPS OF HEMICAULODON EFFODIENS COPE 1869 (MAMMALIA: ODOBENIDAE) \\cv- By Crayton E. Ray Smithsonian Institution, Washington, D.C. 20560 INTRODUCTION In 1965, while reviewing the older literature in connection with studies of fossil pinnipeds of the eastern United States, I happened upon the type-description of Hemicaulodon ef- fodiens Cope 1869, described on the basis of an incomplete tusklike tooth thought to be an upper incisor, from a marl pit at Shark River, Monmouth County, New Jersey, as an Eocene sea cow, but obviously, from a glance at Cope’s excellent illustrations, the basal part of the tusk of an advanced, prob- ably Pleistocene, walrus. As the name seemed not to have en- tered into the literature of fossil walruses, a search was made to learn its disposition, revealing that for the most part it had been overlooked or ignored, for example in otherwise com- prehensive modern works on the Mammalia (Simpson, 1945), on the Sirenia (Simpson, 1932; Reinhart, 1959), and on fossil vertebrates of New Jersey (Miller, 1955). When noticed at all, it generally has been mentioned or listed as an Eocene sirenian: without question, by Gervais (1872a, p. 169; 1872b, p. 352), Cope (1883, p. 52; 1890, p. 699, as Hemicaulodon effodens), Heilprin (1887, p. 340), Roger (1887, p. 30, as from South Carolina, in error), Trouessart (1898, p. 1006; 1905, p. 751), Hay (1902, p. 583; 1930, p. 657), Rhoads (1903, p. 246), and Palmer (1904, p. 316); with question, by Lepsius (1882, p. 185), Zittel (1893, p. 201, listed incertae sedis, as an inade- quately founded genus), Roger (1896, p. 249, incertae sedis, as from South Carolina, in error), and Allen (1923, p. 231). Leidy (1869, p. 440), followed by Brandt (1873, p. 290), 26—Proc. Brow. Soc. WaAsH., Vou. 88, 1975 (281) 282 Proceedings of the Biological Society of Washington placed Hemicaulodon effodiens among the toothed whales, but with characteristic astuteness, implied that it might represent a walrus-like animal, just as might Ontocetus emmonsi, about which he was quite correct also (Ray, in press). Kellogg (1924, p. 757) regarded the allocation of H. effodiens as “too uncertain at present for any discussion of its possible relation- ships,’ but seemed to imply cetacean or sirenian affinities, whereas he later (1966, p. 78) regarded it definitely as a dugongid sirenian. Reinhart (in press) concurs that it is a walrus. To my knowledge the above include all published indications of relationship of Hemicaulodon effodiens. Several citations of the name (Roger, Trouessart, and Rhoads) give its year of publication erroneously as 1871, which is the year indicated on the title page of whole volume XI of the Proceed- ings of the American Philosophical Society, but not of part No. 82 therein, which was published in 1869. The present notes are published now in order to make them available to others engaged in the completion of relevant com- prehensive studies, and to call attention to the taxon in the hope that its type-specimen may be rediscovered. FATE OF THE HOLOTYPE Sporadic efforts from 1965 to the present to find the holo- type have thus far failed to locate it in the collections of the following institutions: Academy of Natural Sciences of Phila- delphia (ANSP); American Museum of Natural History (AMNH); Monmouth County Historical Association Museum; National Museum of Natural History (USNM); New Jersey State Museum; Peabody Museum of Natural History, Yale Uni- versity; Princeton University, Department of Geology; Rutgers, the State University, Department of Geology. The specimen was originally in the private collection of the Reverend Samuel Lockwood, an avid amateur naturalist, who was for 15 years pastor in Keyport, New Jersey, before re- signing in August, 1869, and moving to Freehold in 1870 (John Enright, 1885, p. 449). There is no certainty that Cope ever had the specimen in his possession, and, although explicit information is lacking, there is considerable circumstantial evi- dence that he did not, based upon the sequence of events re- Relationships of a Pleistocene walrus 283 lating to the description of Ornithotarsus immanis, a very large hadrosaur, that commanded greater attention in publication, correspondence, and records. Although quoted already by Lull and Wright (1942, p. 4), Lockwood’s colorful account of a visit from Cope, apparently only a few days prior, will stand repeating for its bearing on Hemicaulodon: Just as I was about packing the bone [of Ornithotarsus immanis|] to send to Mr. H. [Waterhouse Hawkins] (I actually had it in the box) a gentleman knocked at the door. He had a very large carpet bag and announced him- self as Prof. Cope. He had smelt the bone & come for it. But no sir that could not be. Well, he would see it & see it he did, & took drawings of it. Thus [?] I found this stranger a remarkable man. If the sibyl could de- clare in dubious language Alexander is invincible, I can say literally as regards Prof. Ed. C. that this humble individual could not cope with him. He is the man to seize your very bones. (Marsh correspondence, Peabody Mu- seum, letter from Lockwood to Marsh of May 20, 1869). Cope exhibited the drawings and introduced the scientific name at a meeting of the Academy of Natural Sciences of Philadelphia on June 1, 1869. He introduced Hemicaulodon in the Proceedings of the American Philosophical Society for July 16 of the same year, probably after having made draw- ings and notes during the same visit to Lockwood. At about the same time Lockwood did however lend Cope for descrip- tion a plesiosaur vertebra, which became the type of Plesio- saurus lockwoodii (see Welles, 1952, p. 113, for citations to the confusing literature). Thus Cope described at least three taxa from Lockwood's collections. Of these, at least one, Ornithotarsus immanis, be- came part of the Marsh collections at the Peabody Museum, having been sent by Lockwood on March 16, 1886, and re- ceived on March 18, 1886, according to the records under Accession No. 1830. The shipment consisted of only “one box (express $1.00) containing type specimen of Ornithotarsus immanis Cope. . . .,” with no reference to other material. The 284 Proceedings of the Biological Society of Washington only explicit reference to Hemicaulodon found in the Marsh correspondence is in a list of specimens enclosed with a letter from Lockwood to Marsh of March 21, 1886, in which Plesio- saurus lockwoodii is listed as well, indicating that both were still in Lockwood’s possession at that time. Whether these were purchased subsequently by Marsh seems to be un- known, though probable. Kellogg’s (1966, p. 78) footnote in- dicating that the type was purchased by Marsh in March 1886 was based upon my findings in the Marsh correspon- dence, which were not as conclusive as his statement suggests. In any case, none of the specimens has been found in the Pea- body Museum in repeated searches by several individuals since 1965. Rhoads (1903, p. 247) stated, “Type may be in Amer. Mus. Nat. Hist., N. York,” perhaps on the basis of the purchase of the Cope Collection of North American fossil mammals by the American Museum of Natural History in 1895 (Osborn, 1931, p. 446). There is no positive evidence that the specimen was included in this collection, and apparently it is not now in the American Museum. CHARACTERISTICS OF THE HOLOTYPE Although, without the holotype in hand (and possibly even with it), a firm decision as to its specific affinities cannot be made, its identity as the basal, intra-alveolar portion of the tusk of a modernized walrus is evident by inspection of Cope’s excellent illustrations, here reproduced, enlarged to natural size (Pl. 1 and Pl. 6, Fig. 3). Cope (1869, pp. 190-191) noted essentially all of the salient features of the holotype that make it distinctively odobenid, and distinctively not sirenian or cetacean, and even compared it to the odobenids Trich- ecodon Lankester and Ontocetus Leidy, the latter unfortu- nately regarded by him at the time as sirenian. His failure to reach the correct conclusions from his keen observations of pertinent characters may be traced primarily to the assump- tion of Eocene age, to the erroneous belief that the tooth was essentially complete with a part of the triturating surface pre- served, and to the paucity of comparative material. The 285 Relationships of a Pleistocene walrus () here enlarged ial s in med 1 icaulodon effodien The holotype of Hem After Cope to approximately x 1 PLATE 1. aspect. b) 2 riginal x O Fig. 6. 5) Vv Pl 1869 y) 286 Proceedings of the Biological Society of Washington characteristics of the tooth may be discussed conveniently under the general topics of structure, form, and _ surficial features. Structure: As Cope illustrated and aptly stated, “The sub- stance of the tooth is composed of a large axis of osteo-dentine surrounded by a broad cylinder of dentine, which is in turn surrounded by a thick stratum of cementum. . . . The cemen- tum is everywhere entire, and is thicker distally. . . . The dentinal layer is about one half the thickness of the osteodental axis, and three times that of the cementum.” The discrete, flattened-cylindrical core of globular osteodentine extending the length of the tusk is alone enough to exclude the specimen from all other Mammalia, and to identify it conclusively as odobenid (Pls. 3 and 6). Globular osteodentine probably oc- curs sporadically in many, if not most, groups of mammals, but is especially common in some forms with large tusklike teeth, including wart hogs (Penniman, 1952, Pl. XIII), dugongs (PI. 5, Fig. 2), elephant seals (Pl. 5, Fig. 1), and odontocetes, especially sperm whales (Boschma, 1938, PI. XIII; Pl. 4). Its occurrence and abundance in these groups seems to be variable, but, in elephant seals and sperm whales at least, it tends to be more prevalent in old individuals. In sperm whales the globules occur singly or in intergrown clumps in the soft tissue of the pulp cavity, where they may ultimately become imbedded here and there within successive layers of the dentinal laminae, possibly in part in annual pulses, but apparently never as a well-defined subcylindrical mass extending the length of the tooth. Occurrence in the modern dugong is negligible. The canines of old male elephant seals show the closest approach to those of walruses in de- velopment of globular osteodentine, but even there it occurs mostly as isolated lumps within the laminar dentine with at most a cone-shaped mass in part filling the pulp cavity in old > PLate 2. Basal part of right tusk of Pleistocene Odobenus rosmarus in medial aspect, USNM 24886, dredged off North Carolina. Cementum spalled away from most of surface, revealing transverse annular ridges. S< Il, 287 Relationships of a Pleistocene walrus Tiers oC 288 Proceedings of the Biological Society of Washington age (as in a sectioned right upper canine of USNM 265353, an old male of Mirounga angustirostris, Pl. 5, Fig. 1). In the course of his original description of Trichecodon huxleyi, Lankester (1865, pp. 228-230, Pl. 10, Figs. 4-6) clearly defined and illustrated the unique development of globular osteodentine in walrus tusks, noting that it was described as early as 1805 by Cuvier. Although Cope obviously had Lankester’s paper at hand while describing Hemicaulodon, neither he nor apparently most authors who have studied isolated fossil teeth of marine mammals to date recognized the utility of this feature in identifying walrus tusks. As Cope pointed out, the tusk of Hemicaulodon differs from those of dugong and Rytiodus in having a complete sheath of cementum and no enamel band. The tusks of modernized walruses have only a short apical cap of enamel that is worn entirely away within the first few years of life (Mansfield, 1958, pp. 25-28, Fig. 5). Form: Cope recognized the following salient features: “The form of the tooth is that of a compressed, slightly curved cylinder, with distal and proximal vertical diameters equal. . . . The transverse diameter posteriorly is less than that an- teriorly. .. . The widths of the tooth increase very little from the basis to the worn surface.” All of these characters and those revealed in the illustrations are appropriate to walrus tusks, and for the most part not to sirenian tusks. The pos- terior narrowing of the holotype in cross section is atypical of walrus tusks, but not unknown. The tusks of Hemicaulodon effodiens and Rytiodus capgrandi, with which Cope compared his specimen especially, are shown together in lateral and cross-sectional aspects by Gervais (1872b, Pl. XIX, Figs. 1, la, 2, 2a), revealing clearly their dissimilar shapes, the tusk of Rytiodus being irregular and noncylindrical in shape with varied transverse dimensions along its length, curved con- spicuously in two planes, and extremely compressed. Tusks > Piate 3. Longitudinally sectioned basal part of right tusk of Recent Odobenus rosmarus in internal aspect, USNM 21667, showing core of globular osteodentine. x 1. Relationships of a Pleistocene walrus 289 290 Proceedings of the Biological Society of Washington of modern dugongs are superficially similar to the incisors of rodents in general form, with more nearly equal transverse diameters, none of which approach those of H. effodiens or modern walrus tusks in size. The ratio of least to greatest diameters (.61) of the holo- type tusk is rather low for Odobenus rosmarus, though not outside its observed range, reflecting a degree of flattening more characteristic of tusks from late Tertiary beds (Ray, in press ). Cope contrasted the acuminate form of the tusks of Trichecodon and Ontocetus with the cylindrical form of that of H. effodiens. This unjustified emphasis of correct observa- tions was based on the fact that his specimen represented the proximal, and the others, the distal, parts of walrus tusks. This misinterpretation is directly related to another, i.e., the supposition that his specimen represented a nearly complete tooth in which “a portion of the triturating surface is pre- served; it truncates the tooth upwards and backwards in re- lation to its axis, as would be anticipated in a superior in- cisor [of a sirenian].” In fact the holotype represents only the proximal, intra-alveolar end of a walrus tusk. The “triturat- ing surface” could indeed have developed in life in an in- dividual walrus with a broken canine, a not uncommon ac- cident among living walruses (Cousteau and Diolé, 1974, p. 174). The polished surface could equally well have been a postmortem phenomenon, judging from experience with other fossil walrus tusks. The durable rostrum of the skull will often remain intact through extensive rolling and abrasion, after the cranium has been separated or destroyed and the tusks broken off near the alveolar borders, and their stubs polished, yielding a “triturating surface” in approximately the right position and orientation for H. effodiens. Further tumbling might then break up the rostrum or cause the basal tusk frag- > Puate 4. Longitudinally sectioned tooth (distal 6 cm not shown) of Recent Physeter catodon in internal aspect, USNM 25653, showing irregularly distributed globules of osteodentine. x 1. 291 2 Relationships of a Pleistocene walrus 292 Proceedings of the Biological Society of Washington ment to slip out of its alveolus. Whether broken and polished in life or death, the proximal end of the tusk represented is typically walrus-like and could not be expected to show the acuminate form of the extruded distal end. Surficial Features: These include “two open grooves on the inner and one similar on the outer side, the former enclosing a broad bead... . The dentine .. . marked at regular distances by annuliform ridges, which are more or less undulate. They become gradually more distant distally. These ridges can be traced through the cementum” (Cope, 1869, p. 190). The open grooves, two inner and one outer, are highly characteristic of walrus tusks, as are the transverse dentinal ridges. These ridges were described by Tomes (1904, pp. 195-196) who seems to have been the first to recognize that they reflect an- nual increments of growth, farther apart in youth (distal) and progressively closer in old age (proximal). They tend to be masked by cementum in unweathered tusks, both fossil and modern, except toward the proximal, intra-alveolar end—hence not apparent in the distal fragments of tusks of Trichecodon of Lankester (1865), but reflected in the cementum of Cope’s proximal fragment. This condition is shown exceptionally well in modern tusks illustrated by Scheffer (1950, Fig. 3) and by Mansfield (1958, Fig. 7), but the ridges show best in weathered tusks in which some of the cementum has spalled cleanly away from the underlying dentine (Pls. 1 and 2; Brooks, 1954, Fig. 7; Mansfield, 1958, Fig. 8a). Only Kellogg’s (1966, p. 78) comparisons and conclusions regarding Hemicaulodon now require comment. He recorded three isolated teeth as incisor tusks of dugongids: USNM > PuLaTE 5. Fig. 1, longitudinally sectioned right upper canine of Re- cent old male Mirounga angustirostris in internal aspect, USNM 265353, showing isolated globules of osteodentine, and basal cone of osteo- dentine filling pulp cavity, x 1. Fig. 2, longitudinally sectioned left tusk (upper incisor) of Recent adult male Dugong dugon in internal aspect, USNM 284441, showing small amount of osteodentine embedded in laminar dentine near apex of pulp cavity, x l. of a Pleistocene walrus Relationship 294 Proceedings of the Biological Society of Washington 9457, possibly from the Calvert Formation, Tar Bay, Virginia; AMNH 9852 (cast in USNM), dredged off Fort Myers, Florida; USNM 23110 (cast), probably from the Duplin For- mation, near Savannah, Georgia. He felt that the major fea- tures, except size, of Hemicaulodon effodiens could be matched by USNM 9457 and by AMNH 9852. In fact, in addition to disparate size, these teeth differ from the holotype of H. ef- fodiens also in lacking the osteodentinal core, the strong, widely spaced transverse dentinal ridges (fine, closely spaced trans- verse striae are present), and the compressed cylindrical shape with two inner and one outer groove (the transverse dimen- sions are subequal). All of these teeth can be matched within the variation of a large series of physeterid teeth from the Yorktown Formation of the Lee Creek Mine, North Carolina, and in my opinion represent sperm whales, not sirenians, and in any case bear little resemblance to H. effodiens. Dr. Kellogg (1966, p. 78) perpetuated Cope’s comparison of Hemicaulodon effodiens to Rytiodus capgrandi, but he in- formed me in subsequent discussion that he had been misled by the unfortunate juxtaposition in Cope’s Plate V, of Figure 6, the holotype of H. effodiens, and Figure 4a, a human imple- ment made from the shell of Strombus gigas from Anguilla, Lesser Antilles. This error caused him to visualize the former as much more compressed and therefore more like the tusk of R. capgrandi than is the case. When shown the illustrations of H. effodiens enlarged to natural size alongside known walrus tusks with cementum exfoliated, he readily agreed that H. effodiens is indeed a walrus (personal communication, 1966). > Pate 6. Fig. 3, the holotype of Hemicaulodon effodiens in distal aspect, after Cope, 1869, Pl. V, Fig. 6a, original x %4, here enlarged to approximately x1. Fig. 4, left tusk of Pleistocene Odobenus rosmarus transversely sectioned 11 cm from base, in distal aspect, ANSP uncatalogued specimen, dredged off New Jersey, X 1. Fig. 5, left tusk of Recent O. rosmarus transversely sectioned 15 cm from base, in distal aspect, USNM 219054, x 1. Fig. 6, right tusk of Pleistocene O. rosmarus transversely sectioned 16 cm from base, in distal aspect, USNM 24886, x 1. 295 Relationships of a Pleistocene walrus 296 Proceedings of the Biological Society of Washington Typr-LOcALITY With regard to the type-specimen Cope (1869, p. 191) stated that Lockwood “obtained it from the Eocene mar] pits at Shark River, Monmouth co.” Lockwood also indicated the locality as “Shark River” in his list sent to Marsh on March 21, 1886. Neither specified whether he referred to the River itself or to the settlement of the same name. However, in a letter to Marsh of November 17, 1881, Lockwood wrote, “I send you a small bone by mail. It was obtained in the Miocene marl, or green sand, at Shark River, Monmouth Co., N. Jersey (now known as Hamilton).” The village continues to be known as Hamilton, and is situated on the north side of the Shark River at approximately 40°12’30” N, 74°05’ W, in the Asbury Park, New Jersey, quadrangle, U.S. Geological Survey, 7.5 minute series. Hamilton, formerly Shark River, is not to be confused with Shark River Station, which appears on some contempo- rary maps, sometimes with the word Station omitted. Shark River Station was a stop on the Central of New Jersey Rail- road. It no longer appears as such on maps, but was located some 5.5 km WNW of Hamilton, where the still-existing Shark River road terminates at the railroad, in the northwest corner of the township of Wall, 1.1 km northward from Earle, in the Farmingdale quadrangle. Hamilton is somewhat less than 3.2 km upstream from the Shark River estuary, and the principal 19th century marl pits nearest to Hamilton were con- centrated near the River some 1.6-3.2 km still farther upstream, immediately SW and W of interchange no. 100 of the Garden State Parkway (Cook, 1868, pp. 275-276; Ralph Johnson and Robert W. Purdy, personal communication). The locality probably cannot be established more accurately than “marl pit, near Hamilton, Monmouth County, New Jersey. GroLocic AGE The general geology along the Shark River near Hamilton is probably permissive of a source for Hemicaulodon effodiens in Eocene, Miocene, or Pleistocene strata. Cook (1868, pp. 275-276, Fig. 71) gives a section of a typical marl pit of the time along the Shark River, the J.T.L. Tilton pit, apparently Relationships of a Pleistocene walrus Zo the pit closest to Hamilton (Robert W. Purdy, personal com- munication). The Eocene deposit here is the Shark River marl, an argillaceous glauconite sand, regarded by some authors in recent years as a member of the Manasquan Formation, but separated and redefined as the Squankum Member of the Shark River Formation by Richard Enright (1969, p. 18). The Eocene strata are overlain disconformably in the area by the Asbury Park and Grenloch Sand Members of the Miocene Kirkwood Formation, generally regarded on sparse paleonto- logical evidence as in part equivalent to the Calvert Forma- tion of Maryland (Isphording, 1970). The Grenloch Sand Member may not be readily separable from the overlying Cohansey Sand, which may or may not extend into Mon- mouth County (Markewicz, 1969). Pleistocene (?) gravels probably pertaining to the Pennsauken Formation are wide- spread in the area (Salisbury and Knapp, 1917, pp. 155- 156; Bowman and Lodding, 1969) and at least around the Shark River estuary, as at Neptune City, there are thick late Pleistocene marine deposits. The recognition, distribution, geologic age, correlation, and paleoecology (marine vs. non- marine) of all strata from the Kirkwood Formation upward remain uncertain, but all have been regarded at times as in part or in whole nonmarine, based in large measure upon the dearth of fossils. In the original publication on Hemicaulodon effodiens Cope (1869, p. 191) indicated only that it was from the “Eocene marl pits,” but in 1872 (p. 14) he noted that the Eocene “near to Shark River especially, [is] overlaid by a thin stratum of loamy sand, which contains fragmentary remains of terrestrial vertebrates of the Miocene period. ... Whether . . . Hemi- caulodon (Cope) and Anchippodus (Leidy) were derived from this or from the Eocene bed remains uncertain.” Kellogg (1966, p. 78), unduly influenced by supposed similarities of the type of H. effodiens to later Tertiary specimens from eastern United States and to the tusks of Rytiodus capgrandi, stated as follows: “It is now suggested without hesitation that the type tooth was not derived from the middle Eocene Shark River marl... . Inasmuch as the Miocene Kirkwood formation overlaps the Shark River marl (Manasquan Formation), the 298 Proceedings of the Biological Society of Washington association of any specimen picked up on the surface along- side or in a marl pit being worked could readily be misin- terpreted.” Lockwood (1885, p. 15), in his account of the paleontology of Monmouth County, stated that, “In the so- called Drift, one phase of the glacial period, we have collected in Monmouth County relics of the reindeer, walrus, and even a species of dugong.” Throughout the publication Lockwood referred to himself in the plural, including explicit reference to his collecting the types of Ornithotarsus immanis and Plesiosaurus lockwoodii. The “species of dugong” could hardly apply to anything but Hemicaulodon effodiens, and suggests that Lockwood had come to suspect a Pleistocene age. Remains of Pleistocene walruses were recognized early along the coast of New Jersey (Leidy, 1857), and they con- tinue to accumulate in substantial numbers (Kardas, 1965, and unpublished specimens). Donald Baird (personal com- munication) has pointed out that bones of Mammut ameri- canum were recovered from the same spoil heap as the Cretaceous reptiles of New Jersey reported by Baird and Case (1966). This situation is familiar to anyone accustomed per- force to working with vertebrate remains from ditches, road cuts, dredgings, and spoil in the outer coastal plain with its low relief, deep weathering, poor exposures, and thin, un- consolidated deposits reflecting complex geologic history. My interpretation of the affinities of H. effodiens would suggest a maximum age no earlier than the deposition of the Yorktown Formation, the source of the otherwise oldest known walruses in the Atlantic Coastal Plain. Thus the Shark River Formation, which antedates the origin of pinnipeds, is out of the question as a possible source, and even the Calvert- equivalent Kirkwood Formation is too old, barring the im- probable possibility that H. effodiens is the oldest North At- lantic walrus by a considerable margin. The morphology of the specimen, so far as revealed in the original illustrations, does not serve to differentiate among Yorktown and younger walruses. In view of this and of the inexact type-locality and of uncertain identification and correlation of the later Ceno- zoic beds in the area, it is futile to speculate further about the geologic age of H. effodiens, at least until the type is found. Relationships of a Pleistocene walrus 299 CONCLUSION Hemicaulodon effodiens is definitely an advanced walrus on the basis of all details of structure, shape, size, and an- nulations of the type tusk. Similar tusks are known from the Yorktown to the Recent in and around the North Atlantic. The most convenient and reasonable disposition of the taxon for the present is as a Pleistocene (?) representative and junior synonym of Odobenus rosmarus, with the proviso that its place- ment should be reappraised on the basis of morphology and possibly adherent matrix (noted by Cope, 1869, p. 190, in the pulp cavity) when the type is relocated. ACKNOWLEDGMENTS I wish particularly to acknowledge my debt to the late Remington Kellogg who listened patiently, responded flexibly, and encouraged publication of these findings. I wish also to thank the curators and staff of the various museums and in- stitutions in which the type-specimen of H. effodiens or in- formation about it has been sought. In addition certain in- dividuals must be thanked individually for their extraordinary assistance. Foremost among these is Robert W. Purdy, who has shared his special knowledge of the paleontology and his- tory of Monmouth County and has explored numerous leads in our search. Similarly, Donald Baird, James A. Hopson, Ralph Johnson, and John S. McIntosh, have provided useful information and suggestions. Louise Jost, Librarian of the Monmouth County Historical Association, Harmony Coppola, Secretary to the Archivist, Rutgers, the State University, and Melvin C. Krampf, historian, and columnist of the Freehold Transcript, have been extraordinarily helpful in seeking infor- mation about Samuel Lockwood. Charles O. Handley, Jr., and James G. Mead made the collections of Recent marine mam- mals of the National Museum of Natural History available, and permitted sectioning of selected specimens. The plates were prepared by Lawrence B. Isham from photographs taken by Victor E. Krantz. The manuscript was reviewed by Roy H. Reinhart and Frank C. Whitmore, Jr. The work on which this paper is based was supported by the Smithsonian Institu- 300 Proceedings of the Biological Society of Washington tion, in part through the Remington Kellogg Memorial Fund and in part through the Smithsonian Research Foundation. LITERATURE CITED ALLEN, G. M. 1923. Additional remains of the fossil dugong of Florida. Jour. Mammal. 4(4):231-239, pl. 26, 1 text-fig. Barrp, D. anp G. R. Case. 1966. Rare marine reptiles from the Cretaceous of New Jersey. Jour. Paleontol. 40(5):1211-1215, text-figs. 1-3. BoscuMa, H. 1938. On the teeth and some other particulars of the sperm whale (Physeter macrocephalus L.). Temminckia 3: 151-278, pls. 10-13, text-figs. 1-19. Bowman, J. F., II ann W. Loppinc. 1969. The Pennsauken For- mation—A Pleistocene fluvial deposit in New Jersey, pp. 3-6, text-fig. 1. In S. Subitzky (ed.). Geology of selected areas in New Jersey and eastern Pennsylvania and guidebook of ex- cursions. Rutgers Univ. Press, New Brunswick, New Jersey. Branpr, J. F. 1873. Untersuchungen iiber die fossilen und subfos- silen Cetaceen Europa’s. Mém. Acad. Imp. Sci. St.-Pétersbourg, Ser. 7, 20(1):I-VII, 1-372, pls. 1-34. Brooks, J. W. 1954. A contribution to the life history and ecology of the Pacific walrus. Alaska Coop. Wildl. Res. Unit, Spec. Rep. 1:viii + 103 pp., text-figs. 1-9. Cook, G. H. 1868. Geology of New Jersey. [New Jersey State] Board of Managers, Newark, xxiv + 899 pp., figs. Corre, E. D. 1869. Synopsis of the extinct Mammalia of the cave formations in the United States, with observations on some Myriapoda found in and near the same, and on some extinct mammals of the caves of Anguilla, W. I., and of other lo- calities. Proc. Amer. Philos. Soc. ~11(82):171-192, pls. 3-5. —. 1872. List of the Reptilia of the Eocene formation of New Jersey. Proc. Acad. Nat. Sci. Phila. 1872:14-18. 1883. On a new extinct genus of Sirenia, from South Caro- lina. Proc. Acad. Nat. Sci. Phila. 1883:52-54. ——. 1890. The extinct Sirenia. Amer. Natur. 24(284):697—702, pls. 25-26, text-figs. 1-3. Cousteau, J.-Y. AND P. DioLt. 1974. Sea lion, elephant seal, walrus. Cassell & Co., London, 304 pp. Enricut, J. 1885. Samuel Lockwood, Ph.D., pp. 444-450, 1 text- fig. In F. Ellis, History of Monmouth County, New Jersey. R. T. Peck & Co., Philadelphia. [Reprinted 1974. Polyanthos, Inc., Cottonport, Louisiana. ] EnricHt, R. 1969. The stratigraphy and clay mineralogy of the Eocene sediments of the northern New Jersey coastal plain, pp. 14-20, text-figs. 1-2. In S. Subitzky (ed.). Geology of selected areas in New Jersey and eastern Pennsylvania and Relationships of a Pleistocene walrus 301 guidebook of excursions. Rutgers Univ. Press, New Bruns- wick, New Jersey. Gervais, P. 1872a. Analyses d’ouvrages et mémoires. XVIII.—Cope (E.D.): diverses notes relatives aux mammiféres fossiles et aux myriapodes des Etats-Unis. Jour. Zoologie 1:168-170, 1 text- fig. . 1872b. Analyses d’ouvrages et mémoires. XXXVIII.—Travaux récents sur les sirénides vivants et fossiles. Jour. Zoologie 1:332-353, pls. 18-19, text-figs. 1-4. Hay, O. P. 1902. Bibliography and catalogue of the fossil Vertebrata of North America. U.S. Geol. Surv. Bull. 179:868 pp. 1930. Second bibliography and catalogue of the fossil Verte- brata of North America. Carnegie Inst. Washington Publ. 390, vol. 2:XIV + 1074 pp. Heitprin, A. 1887. The geographical and geological distribution of animals. Internat. Sci. Ser., vol. 57, D. Appleton and Co., New York, xii + 435 pp. IspHORDING, W. C. 1970. Petrology, stratigraphy, and re-definition of the Kirkwood formation (Miocene) of New Jersey. Jour. Sed. Petrol. 40(3):986—-997, text-figs. 1-8. Karpas, S. J., Jk. 1965. Notas sobre el género Odobenus (Mammalia, Pinnipedia). 1. Una nueva subespecie fésil del Pleistoceno superior-Holoceno. Bol. Real Soc. Espanola Hist. Nat. (Biol.), Madrid. 63(4) :363-380, text-figs. 1-7. Kewtoce, R. 1924. Tertiary pelagic mammals of eastern North Amer- ica. Bull. Geol. Soc. Amer. 35(4) :755-766. —. 1966. New species of extinct Miocene Sirenia. U.S. Nat. Mus. Bull. 247(3):65-98, pls. 33-43, text-figs. 32-38. LANKESTER, E. R. 1865. On the sources of the mammalian fossils of the red crag, and on the discovery of a new mammal in that deposit, allied to the walrus. II. Trichecodon Huxleyi, a new mammalian fossil from the red crag of Suffolk. Quart. Jour. Geol. Soc. London 21(1) :226—232, pls. 10-11, figs. 1-2. Lempy, J. 1857. Notice of remains of the walrus discovered on the coast of the United States. Trans. Amer. Philos. Soc., New Ser. 11(5):83-86, pls. 4-5, fig. 1. 1869. Synopsis of extinct Mammalia of North America. Jour. Acad. Nat. Sci. Phila. Ser. 2, 7:363-446, pls. 28-30. Lepsius, G. R. 1882. Halitherium Schinzi, die fossile Sirene des mainzer Beckens. Abhandl. Mittelrheinischen Geol. Vereins TeIy 2900 98 .Vil pp. pls, 1210) Locxwoop, S. 1885. Archaeology and Paleontology, pp. 7-16. In F. Ellis. History of Monmouth County, New Jersey. R. T. Peck & Co., Philadelphia. [Reprinted 1974. Polyanthos, Inc., Cotton- port, Louisiana. ] Luti, R. S. anp N. E. Waricur. 1942. MHadrosaurian dinosaurs of North America. Geol. Soc. Amer. Spec. Pap. 40:xi + 242 pp. 302 Proceedings of the Biological Society of Washington MANSFIELD, A. W. 1958. The biology of the Atlantic walrus Odobenus rosmarus rosmarus (Linnaeus) in the eastern Canadian Arctic. Fish. Res. Board Canada, Ms. Rept. Ser. ( Biol.) 653:xiii + 146 pp., text-figs. 1-35. MarKEwicz, F. J. 1969. Ilmenite deposits of the New Jersey coastal plain, pp. 363-382, text-figs. 1-13. In S. Subitzky (ed.). Geology of selected areas in New Jersey and eastern Penn- sylvania and guidebook of excursions. Rutgers Univ. Press, New Brunswick, New Jersey. Mituer, H. W., Jr. 1955. No. 28, pp. 315-318 15 August 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ADENOMERA LUTZI (AMPHIBIA: 4 LEPTODACTYLIDAE), A NEW SPECIES i OF FROG FROM GUYANA — By W. RonaLtp HEYER National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560 Three specimens of a distinctive frog were catalogued as. Leptodactylus hylaedactylus in the collections of the British Museum (Natural History) (BMNH). Through the courtesy of Alice G. C. Grandison I have recently been able to examine the specimens and partially dissect one of them. The three specimens represent a distinctive new species of the genus Adenomera. The three specimens were catalogued in 1872 and 1905. Interestingly, I was not able to locate further speci- mens in the extensive Guyana collections at the American Mu- seum of Natural History. The genus Adenomera has re- cently been reviewed (Heyer, 1973, 1974); the purpose of this paper is to call attention to the addition of this distinctive new species to the genus. Adenomera lutzi, new species Figure 1 Holotype: BMNH 1905.11.1.17, adult male from Guyana, Chinapoon R., upper Potaro (probably Chenapowu River). Dr. Bovallius, collector. Paratypes: BMNH 1872.10.16.43—44, females, from Guyana, Demerara Falls. Diagnosis: Adenomera lutzi differs from all other species in the genus by its large size (male 30 mm, males of other species range from 23.4—27.0 mm maximum SV; females to 34 mm, females of other species range from 24.5-31.0 mm maximum SV), presence of a dark tri- angular seat patch (absent in other species), and distinct spotting on the posterior face of the thigh (mottled or uniform in other species). 28—Proc. Brot. Soc. WAsH., Vou. 88, 1975 (315) 316 Proceedings of the Biological Society of Washington x ‘ Fic. 1. Dorsal and ventral views of the holotype of Adenomera lutzi. Description of holotype: Dorsal outline of snout subelliptical; snout profile rounded-acute; canthus rostralis rounded; loreal slightly concave in cross section; tympanum distinct, greatest diameter 14 eye diameter; vomerine teeth in two transverse series posterior to choanae; vocal slits elongate, paired; no external vocal sac; finger lengths in order of decreasing size III > 1 > II > IV; first finger just longer than second; large ovoid inner metacarpal tubercle smaller than ovoid outer meta- carpal tubercle; dorsal surfaces including eyelids profused with small warty tubercles; short glandular ridge behind each eye; faint indications of mid-dorsal and dorsolateral glandular folds; belly fold distinct; toe tips expanded into small disks, disks not flattened, no circumferential groove; toes lacking fringe; subarticular tubercles pronounced; distinct oval inner metatarsal tubercle larger than distinct rounded outer meta- tarsal tubercle; tarsal fold not continuous with inner metatarsal tubercle, extending % length of tarsus; sole of foot with large white tubercles, lower surface of tarsus profused with small white tubercles. Snout-vent 30.1 mm; head length 11.1 mm; head width 11.2 mm; interorbital distance 2.4 mm; greatest diameter of tympanum 1.7 mm; A new frog from Guyana 317 eye diameter 3.0 mm; eye-nostril distance 2.8 mm; femur 13.4 mm, tibia 14.6 mm; foot 14.9 mm. Dorsum tan in preservative with darker brown markings on ridges behind eyes, chevron in shoulder region and two round sacral spots above groin; lips and dorsal limb surfaces barred; circumferential parts of venter mottled, belly lacking melanophores; dark brown triangular seat patch; posterior surface of thigh dark brown with bold white round and elongate spots. Variation: The female paratypes measure 33.6 and 34.1 mm SV. The vomerine teeth are contiguous in one of the specimens. The color patterns are similar to the holotype with the following differences: dark interorbital bars are distinct in the paratypes, indistinct in the holotype; one paratype has dark brown stripes along %4 the body length in the dorsolateral fold region; the posterior face of the thighs are not as boldly marked in the paratypes; in the most differently marked paratype, the dark brown seat patch continues as a uniform brown on the thighs with small light spots above and below the uniform brown. Etymology: The species is named in honor of the late Adolfo Lutz, a Brazilian pioneer in herpetology. Dr. Lutz was the first worker with field experience to call attention to the distinctiveness of the frogs com- prising the genus Adenomera. ANATOMICAL SUMMARY One paratype, BMNH 1872.10.16.43, was partially dissected to de- termine certain muscle and skeletal character states. The other para- type was X-rayed to determine the shape of the terminal phalanges. Origin of depressor mandibulae muscle mostly from dorsal fascia, with some fibers originating from the squamosal bone and the annulus tympanicus. Adductor mandibulae posterior subexternus present, ad- ductor mandibulae externus superficialis absent. Geniohyoideus medialis muscle continuous medially (Heyer, 1974, Fig. 2, O). Geniohyoideus lateralis muscle lacking attachment or slip to hyale. Anterior petro- hyoideus muscle inserting on ventral surface of alary process of hyoid (Heyer, 1974, Fig. 1, G). Sternohyoideus muscle insertion on ventral body of hyoid (Heyer, 1974, Fig. 1, I). Omohyoideus muscle inser- tion on ventral surface of hyoid plate (Heyer, 1974, Fig. 4, 2). Iliacus externus muscle extending full length of iliac bone. Tensor fasciae latae muscle inserting posterior to iliacus externus muscle on iliac bone. Exterior head of semitendinosus muscle absent. Sartorius muscle broad, covering adductor longus muscle. Accessory head of adductor magnus muscle ending proximally in a tendon. Adductor longus muscle well developed, inserting on knee. Anterior process of hyale absent. Alary process of hyoid broad and winglike (Heyer, 1974, Fig. 1, G). Omo- sternum slightly expanded anteriorly. Mesosternum a bony style. Xiphi- sternum cartilaginous, entire, single. Terminal phalanges T-shaped. 318 Proceedings of the Biological Society of Washington All known myological and osteological character states of the new species agree with the previous definition of the genus (Heyer, 1974). The new species extends the known snout-vent length from 27.0 mm to 30.1 mm in males, and 31.0 to 34.1 mm in females. DISCUSSION The relationships among the species of the genus Adenomera remain obscure. There are few morphological specializations shared by two or more species on which to base an analysis of relationships. The most suggestive characters are in the hyoid complex. The new species and A. martinezi share the states of the anterior petrohyoideus inserting on the ventral surface of the hyoid apparatus and a broad winglike alary process of the hyoid. Previously, I considered these as derived states (Heyer, 1974). If these states are derived, then A. lutzi and martinezi are the most closely related species in the genus Adenomera. My subjective impression is the opposite, however. The four species A. andreae, bokermanni, hylaedactyla, and marmorata form a _ tight cluster and are distinguished from each other by subtle morphological differences. On the other hand, A. lutzi and martinezi are the most dis- tinctive species in the genus and very distinctive from each other. In the case of Adenomera, it appears that a broad winglike alary process and ventral insertion of the anterior petrohyoideus muscle are primitive states. Resolution of the intrageneric relationships may well require bio- chemical and/or more karyotypic information. ACKNOWLEDGMENTS Alice G. C. Grandison and A. F. Stimson, British Museum (Natural History) allowed me to examine the specimens in their care and pro- vided information on the geographic origin of the specimens. George R. Zug, Smithsonian Institution, read the manuscript. The Smithsonian Research Foundation supported the research for this paper. LITERATURE CITED Heyer, W. R. 1973. Systematics of the Marmoratus group of the frog genus Leptodactylus (Amphibia, Leptodactylidae). Con- trib. Sci., Los Angeles County Mus. Natur. Hist. 251:1-50. ——. 1974. Relationships of the Marmoratus species group (Amphibia, Leptodactylidae) within the Subfamily Lepto- dactylinae. Contrib. Sci., Los Angeles County Mus. Natur. Hist. 253:1—46. vi 4 , No. 29, pp. 319-330 15 August 1975 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON LABIDOCHROMIS TEXTILIS, A NEW CICHLID // FISH (TELEOSTEI: CICHLIDAE) FROM LAKE MALAWI By Micuaret K. OLIiver Department of Ichthyology, American Museum of Natural History, New York, N.Y. 10024 An aquarist (Johnson, 1974a) recently published brief de- scriptions of two new species of Labidochromis, a genus of cichlid fishes endemic to Lake Malawi, Africa. He based one of them, L. joanjohnsonae, on a holotype and paratype be- longing to two species that differ strikingly from each other in coloration and oral dentition. Although he gave conflicting statements both of standard length and total length of the paratype in different paragraphs, Johnson (1974a: 15) clearly designated the smaller of the two specimens as the holotype, and since they differ in standard length by nearly 10 mm there is no doubt as to which this is. The number of dorsal fin spines (sixteen in the holotype, seventeen in the para- type) corroborates his designation. It might be noted, how- ever, that Johnson’s counts of segmented dorsal rays are in- accurate for both specimens. I have examined the types of all species described by Johnson (1974a and 1974b) and found that the holotype of L. joanjohnsonae Johnson is a specimen of L. fryeri Oliver in Davies (1973); L. joanjohnsonae is therefore a junior synonym of that species. The manuscript name Labidochromis fryeri and a diagnosis of this taxon were unfortunately published in an aquarium magazine in a letter from T. E. (Peter) Davies, who specifically (and correctly) attributed to me the name, the diagnosis, and my intention eventually to publish them. Regrettably, by Articles 11 and 13 (a) (i) of the International 29—Proc. Brot. Soc. WAsH., Vou. 88, 1975 (319) 320 Proceedings of the Biological Society of Washington Code of Zoological Nomenclature the name Labidochromis fryeri is clearly available, and by Article 50 I am its author. I am preparing a full description of that species, with a desig- nation of type-specimens, for publication elsewhere. The paratype of L. joanjohnsonae, figured by Johnson (1974a: 16, fig. 2) but wrongly captioned the holotype, is recognizable in his color photograph because of its proportions and fin markings. It represents an undescribed species of Labidochromis discovered by Mr. Richard Furzer on the Mozambique shore of Lake Malawi and first exported by him to the American aquarium trade in January 1973. The same species has since been collected at Likoma Island by T. E. Davies and party. I was already preparing a description of this Labidochromis before Johnson became interested in it. Since Johnson (1974a: 14) expressed a wish to avoid “even greater duplication of Mr. Oliver’s forthcoming work on this genus,” I do not hesitate to describe it at this time. It is hoped that this comely species will not continue to circulate in the aquarium trade under the name L. joanjohnsonae, junior syn- onym of the quite distinct species L. fryeri. Counts and measurements herein employed follow Green- wood (1973), except that vertebral counts include the fused first preural and ural centra supporting the parhypural and hypurals; my counts of caudal and total vertebrae are there- fore one greater than his. Pectoral rays, scales, gill rakers, and teeth were counted on both sides of each fish. Standard deviation (SD) and standard error (SE) were calculated for the principal morphometric ratios. Labidochromis textilis, new species Brocade Cichlid Figures 1-5 Labidochromis joanjohnsonae Johnson, 1974a: 15, in part, paratype only, by original designation, p. 15; incorrectly designated as holo- type in fig. 2 (Likoma Island). Labidochromis sp. Axelrod, 1974: 224, in part, lower photograph only (Likoma Island). Fish in upper photograph possibly L. caeruleus Fryer, 1956. Labidochromis caeruleus “likomae” Scheuermann, 1974: 441, trade name only, in part, “weibchen” only, lower photograph, non caeruleus A new cichlid fish from Lake Malawi 321 Fryer, 1956 (Likoma Island). Fish in upper photograph possibly L. caeruleus. Labidochromis spec. nov. (marineatus) Neergaard, 1974: 74-75 and photograph p. 75, trade name only (Likoma Island). Notes on synonymy: The collectors who export aquarium fishes from Lake Malawi commonly coin pseudo-scientific trade names for species they cannot identify. This practice is strongly to be dis- couraged because it promotes nomenclatural instability and confusion when, as frequently happens, such names are published in the aquarium literature. An example is the name Labidochromis caeruleus “likomae.” Scheuermann (1974) did not intend to propose this as a formal scientific name; as the following evidence shows he: (1) attributed the name to an unspecified dealer of aquarium fishes; (2) always included it between quotation marks whereas he used no quotation marks for the established species Labidochromis caeruleus and its implied sub- species L. c. caeruleus; (3) did not write in the style of an original description; (4) provided no formal diagnosis; (5) designated no type- material; and (6) did not clearly assign a rank to the taxon, saying only that its treatment as a subspecies is “an assumption to which I also incline” (eine Annahme, zu der auch ich neige) but later in the same sentence that it is one of several “phenotypes” (Erscheinungsformen ) of L. caeruleus. These facts indicate that L. c. “likomae” is merely a trade name, without standing in formal zoological nomenclature. A second trade name mentioned by Scheuermann (1974), L. c. “nkatae,” is apparently not a synonym of L. textilis; rather, it is a nomen nudum since it was published without any diagnosis or indication and from the text it cannot be identified with any species. Neergaard (1974) also employed several names that are clearly not available. For the present species he used Labidochromis spec. nov. as the scientific name and the parenthetical marineatus as the trade name, not as a formal specific epithet (Neergaard, 1974: 26-27 and personal commun.); moreover the term marineatus is not associated, in the meaning of the Code, with a generic name because of the inter- vening words. Collectors of aquarium fishes would serve the best interest of aquarist and ichthyologist alike if they proposed a simple common name such as “Brocade cichlid” for each unidentified species or form, instead of a troublesome and deceptive pseudo-scientific term. Holotype: British Museum (Natural History) (BMNH) Reg. No. 1975.5.27: 9, a mature male 76.5 mm standard length (SL), an aquarium specimen exported from Lake Malawi and probably collected at Likoma Island by T. E. Davies and party; donated by Mrs. Virginia Egolf. Paratypes: Aquarium specimens exported from Lake Malawi by T. E. Davies and R. E. Furzer: BMNH Reg. Nos. 1975.5.27: 10-12 (8 speci- mens, 43.0, 45.5, 53.0 mm SL); 1975.5.27: 14-15 (2, 47.0-50.0 mm); 1975.5.27: 18 (1, 58.5 mm); 1975.5.27: 16-17 (2, 33.0-87.0 mm). 322 Proceedings of the Biological Society of Washington VSI OST SEL TST SSI TI SOL LET LST SSI VFI TST OLI 230 S60 T'St (+) wWsue, epunped jepnep SLG 66S OLE T9S O9G FSS HLS FOS BLS 69S 60S LES 129 LLO 99% SLG (+) Yue, oseq uy [esi0q eee BIE OSE TES SSE BE SBE SES FOE HEE GOE GOS O'SS SGO IBT SEE (+) Yasue, [estope1g LOE OPE OGE OGE BEE TIE LOS OFE TOE SEs SE OES O'FE LEO YET OEE (%) sue, Mel 1omoT €83 [LG TLS I'LZ 88S 66S 18S LS O'8S FSS LOS GSS FS IFO PHI VLZB (%) sjeorped Are[preurrg 883 G9 LZ VSS TLS 183 GLZ LHS BSS 79S LOS 9S F9S SEO ZIT L193 (%) Ysue, wef reddy Ltr Vrr Loy Sib Vor Ter Vor TIP GSh OG OOF SLY OGF S90 LIZ Ger (%) YWuE, peoy [eIqG10}s0g €93 906 006 SLI SIZ S6Il FBS ELI OZ GOS ELI SSI GAT FLO LYS 0'0Z (%) dep yeqD Occ LG LOS 663 TIE TIE SES BOE SBE SIE OVE GFE BIE 860 OE BOS (%) Yrsue, 3qI0 GLE OEE SHE 666 DIE ECVE FES TE SOS 16S O'ZE 16S LSS 780 063 SIE (%) Ysuez yous 636 OCS SIS SEI SEI OIG SHS SGI TES GSI OSI 60S FSI 990 SET 2:02 (%) pi [erq10107 0] S'S 10S FOG SEI S6I V0 YOGI T6l SSI 6ST LSI FOL MFI 290 STS Z6I (%) Ydep [e1qr1oe1g Lop LOb VhrR Thr Thr T9F OIG TSh O'6F BSF L@F 60S SOS 180 08S LOF (%) Ypreiq peoH VIE GI€E TPE Tes VES Tee QO LZE FOE GE BE LEZ TIE PHO FST BSE (+) Yasusy peop 9IE TES OVE OE TE GOE OVE BIE TIE SOE FOE HIE LE Eso 280 LIE (+) widep Apog GoL 089 099 SEG O'SS GOS 00S S6r OLY GSP O'EF OLE OEE (uiur) YASUE] prepurrs add} te sodAjeieg aS Gs uroyy ToqeieyD apuosuyoluvol suuo1ysopiqvT jo odAyered = , “ySueT piepuejs jo Juoo1ed = |} “YJsueT peoy Fo yUsoIed = % ‘syizxaz SMuOLyDOpIqvT JO sonel oIMeuI0Yydiour [edioullg ‘[ aATaV], A new cichlid fish from Lake Malawi 323 American Museum of Natural History (AMNH) 32413 (2, 49.5-66.0 mm); AMNH 32414 (1, 50.5 mm); AMNH 33465 (1, 68.0 mm, para- type of L. joanjohnsonae). Etymology: The trivial name, from the Latin for woven, refers to the life coloration which suggests brocade cloth. Diagnosis: Labidochromis textilis differs from all other species of Labidochromis in having about 12 alternating horizontal stripes of orange and pale blue or blue-green on side of body in life, and distinct, oval orange spots (brown in preserved fish) on dorsal and caudal fins of adults. Description: Based on holotype (Fig. 1) and twelve paratypes, 33.0-76.5 mm SL. Principal morphometric ratios given in Table 1. Dorsal profile of head usually straight from tip of snout to over rear of orbit in mature fish; snout slightly concave in large specimens. Dorsal margin of orbit lying below dorsal profile of head when fish viewed laterally; premaxillary pedicels not entering profile. Snout 0.9-1.25 times broader than long. Angle of premaxillary pedicels 30°—40° to horizontal; frontal angle 25°-40°. Cephalic lateral line pores and canals not hypertrophied. Posterior tip of maxilla usually reaching level of anterior orbital or ocular margin. Jaws pointed or narrowly rounded anteriorly when fish viewed from above or below. Lower jaw projecting slightly, 1.1-1.4 times longer than broad, its angle 30°-40°. Caudal peduncle 1.02-1.30 (mean, M=1.17) times longer than deep. Fins: Dorsal with 25 (in 8 specimens), 26 (4), or 27 (1) elements, comprising 16 (2), 17 (10), or 18 (1) spines and 8 (6) or 9 (7) seg- mented rays; last spine 13.6-16.2 (M = 15.0) % of SL. Anal with 10(9) or 11 (4) elements, comprising 3 spines and 7 (9) or 8 (4) segmented rays; last spine 14.2-16.7 (M=16.0) % of SL. Pectorals 20.2-25.3 (M = 23.4) % of SL; with 13-15 (mode 14) rays, comprising 2 upper unbranched, 9-11 (mode 10) branched, and 1-3 (mode 2) lower un- branched rays. Caudal slightly rounded to emarginate, lobes rounded and subequal; scaled on proximal 4-74 (medially) or 24—4% (dorsally and ventrally), extent of scale coverage positively correlated with standard length. Pelvics 23.5-29.7 (M= 26.0) % of SL, longest in mature males, first ray slightly produced in both sexes; spine 13.5- 16.8 (M = 15.0) % of SL. Scales: Ctenoid. Lateral line with 30-32 (mode 31) scales, upper part with 20-24 (mode 24), lower with 8-12 (mode 10). (50.0 mm paratype abnormal in lacking lower part of right lateral line; scales in that area appear normal and unregenerated, completely lacking canals and pores. Upper part of right lateral line comprising 20 scales; on left side upper part with 22, lower 10, total 32.) Cheek with 4-6 (mode 5) rows. About 5-8 scales between dorsal fin origin and upper part of lateral line. Seven to 9 scales between pectoral and pelvic fin 324 Proceedings of the Biological Society of Washington rR Weng yi NNN Gi | } AW Ha \ WE HHT y \ § i Hs SL Hey ne Labidochromis textilis. Holotype. Fic. 1. A new cichlid fish from Lake Malawi 325 a) 1 mm ee eee Fic. 2. Oral dentition of Labidochromis textilis, drawn from 53.0- mm paratype. (A) Right premaxilla, viewed from a lateral and slightly dorsal position. (B) Anteriormost tooth of right premaxilla. Lingual view at left, lateral view at right. (C) Fourth tooth from posterior tip of right premaxilla. Lingual view at left, lateral view at right. (D) Tooth of outermost inner row of right dentary, anterior in position. Lingual view at left, lateral view at right. bases; small thoracic scales grade gradually into larger scales of belly. Sixteen scales around caudal peduncle. Gill rakers: 2-4 (mode 3) on epibranchial + 1+ 8-10 (usually 9 or 10) on ceratobranchial of outer arch. Upper ceratobranchial rakers simple, rather short; lower 2 or 3 rudimentary. Teeth (Fig. 2): Outer row of upper jaw with a total of 27-33 (M = 29.4) teeth in all but two smallest specimens (33 and 37 mm SL), both with 19. About 4 anterior teeth on each side slightly pro- cumbent and markedly longer and coarser than lateral and posterior teeth, into which they grade in size. All teeth unicuspid (a single exceptional unequally bicuspid tooth present laterally in 50.0 mm para- type), crowns acute, isoscelene in outline, compressed in curved plane of premaxilla, and somewhat incurved. Teeth of outer row of lower jaw similar in form to those of upper jaw, but anterior ones strongly procumbent. Inner teeth of each jaw miniatures of outer teeth, gen- erally unicuspid but sometimes weakly tricuspid. Crowns compressed in curved plane of jaws, and incurved. Anterior inner teeth markedly larger than adjacent lateral teeth of inner rows and grading into them in size; 1 or 2 inner rows in each jaw, and sometimes a short third row anteriorly in large specimens. Lower pharyngeal bone (Fig. 3) triangular in outline, 1.19-1.46 (M = 1.28) times broader than long, its length 20.0-24.0 (M = 22.0) % of head length, its breadth 51.9-67.2 (M=60.4) % of head breadth. Joint uniting two halves straight, not sinuous; blade shallow, not angled 326 Proceedings of the Biological Society of Washington Sn S BLS: & lass7 Fic. 3. Labidochromis textilis. Lower pharyngeal bone of holotype, in occlusal view. relative to plane of dentigerous surface. Dentigerous area 1.4—1.6 (rarely almost 1.8) times broader than long. Teeth rather small but not densely crowded; crowns compressed, bicuspid. About 4 postero- median teeth slightly or markedly coarser than adjacent anterior and lateral teeth but crowns always compressed; 22-32 (M = 29.0) teeth in posterior row, 7-11 in an antero-posterior row, and 5-8 in an oblique row. Vertebral counts in 13 specimens radiographed: 29 (in 3), 30 (9), or 31 (1), comprising 13 (4) or 14 (9) abdominal and 15 (1), 16 (9), or 17 (3) caudal centra. Modal combination 14+ 16 (in 7). Caudal skeleton: Patterns of hypural fusion include no fusion (in 6), 1 and 2 fused (2), 3 and 4 fused (2), and 1 fused with 2 as well as 3 fused with 4 (3). No correlation between length of fish and degree of fusion. Coloration in life: Live fish of both sexes, adult but sexually quiescent, light blue to blue-green on the head and body, becoming white on ventral surfaces. Distinctive pattern of about 6 to 9 irregular, horizontal orange stripes on the flanks superimposed on this ground color (Fig. 4). No apparent correlation between number of stripes and length of fish. Head with irregularly shaped orange stripes and blotches, among which invariably occurs an oblique stripe that extends from upper end of operculum upwards and forwards to nape. Blackish markings on head, generally indistinct, include a lachrymal stripe, an opercular spot, and 1 horizontal stripe crossing interorbital region and another crossing snout. About 8 dusky vertical bars on flanks below dorsal fin base, rarely apparent except as a “fright pattern.” Dorsal fin is light iridescent blue, frequently with a narrow, broken orange stripe basally; lappets are orange distally, bluish white proximally. A broad submarginal band in spinous portion sometimes dark brown and composed of vertical marks between spines, forming an almost continuous horizontal stripe, A new cichlid fish from Lake Malawi 327 Fic. 4. Orange coloration (stippled) of live specimens of Labido- chromis textilis. The pattern shows much individual and ontogenetic variation, particularly on fins and head. Semidiagrammatic. or sometimes consisting of isolated oval orange spots between spines. (It is possible that there is an ontogenetic development from the former condition to the latter, or that the two conditions represent sexual di- morphism, but neither possibility can be verified with the limited material available.) Soft portion of dorsal fin with 2 to 4 rows of oval orange spots, relatively smaller and more numerous in larger fish; fin sometimes with a narrow orange to brick-red margin. Caudal iridescent pale blue with bluish white dorsal and ventral edges and sometimes an orange to brick-red posterior margin. Proximal portion of fin with several vertical rows of orange spots; distal half to 84 sometimes spotted, or rays orange posteriorly, producing horizontal orange stripes alternating with blue of membrane. Anal fin whitish to iridescent pale blue, some- times with a fairly distinct, dusky submarginal stripe on spinous por- tion. The anal ocelli numbering 1 to 5 and deep yellow. Pelvic fins dirty white to dusky, darkest anteriorly except for bluish white leading edge. Coloration of breeding individuals unknown. Preserved coloration: Both sexes with head and body brown, darker dorsally. Orange markings of body and head usually lost, but a few specimens retaining horizontal stripes of flanks in dark brown contrasted to paler ground color. Orange markings of dorsal and caudal fins always retained, becoming dark brown. About 8 dark vertical bars on flanks usually evident. Ecology: Nothing is known of the ecology of L. textilis except that it inhabits shallow inshore water. Breeding: Scheuermann (1974) reported that L. textilis is a female mouth brooder (see below under Relationships, however, uncertain as 328 Proceedings of the Biological Society of Washington 25 nN (fo) BODY DEPTH (MM) 30 40 50 60 70 80 STANDARD LENGTH (MM) Fic. 5. Relationship of body depth to standard length in Labido- chromis textilis (OQ) and L. caeruleus (x). Data for L. caeruleus re- plotted from Fryer (1956). to the identity of the male parent of the broods he observed). L. textilis matures at a small size. The smallest adult male is the 53.0 mm SL paratype, ripe or nearly so. The smallest mature female is the 43.0 mm SL paratype, a ripe individual in which the left ovary is undeveloped and the right ovary contains seven ripe ova, all of about the same size and bright orange, the largest measuring 4.0 x 2.8 x 1.8 mm. These ova so distend the ovary that it nearly fills the abdominal cavity. Relationships: An analysis of the phylogenetic relationships of L. textilis must await the description or redescription of all other known species of Labidochromis, a project which I have undertaken as part of a larger study of the systematics of Lake Malawi Haplochromis and related species. For the present it is necessary only to examine the statement of Scheuermann (1974) that L. textilis represents the female of a color form or subspecies of L. caeruleus Fryer of which the male is blue and lacks orange markings. Axelrod (1974: 224) also believed that a similar blue fish and the orange striped fish represent male and female of a single Labidochromis species, although he did not identify the species by name. This suggestion is refuted by the finding of both male and female fish among the type-series of L. textilis, all of which had orange stripes in life. Scheuermann claimed that one of his orange striped females spawned in his aquarium with an unstriped blue male; however, he stated (1974: 440) that he did not observe the spawning itself. It is therefore possible (a) that the male parent was also one of his orange striped fish (i.e. was also L. textilis), or (b) that only A new cichlid fish from Lake Malawi 329 females of L. textilis were present and that one of them spawned with a blue fish, which, to judge from Scheuermann’s photograph, may pos- sibly be the true L. caeruleus. Interspecific and even intergeneric hybrids of Lake Malawi cichlids have been produced in aquaria when a fish is not given a conspecific mate (D. H. Eccles, personal commun. ). It is not possible to decide between these alternatives since Scheuer- mann’s specimens are not available for examination. In any case, L. textilis does not appear to be a color form of L. caeruleus. That species resembles L. textilis meristically but differs from it in several morphometric ratios. L. caeruleus has a deeper body (Fig. 5), a longer head (34.25-36.0 % of SL, cf 29.7-34.9, M = 32.2 % in L. textilis), a longer snout (34.0-38.9 % of head length, cf 25.7- 32.3 % in L. textilis excluding the three largest types, all of which are of greater total length than any of Fryer’s L. caeruleus specimens), and a longer snout relative to postorbital head length (snout “a trifle shorter than post, orbital [sic] part of head” [Fryer, 1956: 88], cf snout = 0.56— 0.80, M = 0.70 times postorbital head in L. textilis). Such differences are conceivably ascribable to geographic variation, but I have not en- countered comparable geographic variation in morphometrics among other species of Malawi cichlids. ACKNOWLEDGMENTS Drs. C. Lavett Smith and Donn E. Rosen provided valuable com- mentary on the tangled nomenclatural problems encountered in this study. Mrs. Virginia Eglof, Mr. Robert C. Brooks, and Mr. Paul V. Loiselle supplied most of the type-specimens of L. textilis. Mr. Brooks also provided color transparencies of living specimens, as did Mr. Rick Johnson. Mr. Sgren Neergaard furnished a copy of his book and comments on the names therein employed. Finally, Dr. James W. Atz gave incisive criticism of the manuscript and much additional help. To all of these generous persons I am most grateful. LITERATURE CITED AxELroD, H. R. 1974. African Cichlids of Lakes Malawi and Tan- ganyika. Second ed. T.F.H. Publications, Inc. Ltd., Hong Kong. 256 pp. Davies, P. 1973. [Letter] p. 65. In H. R. Axelrod, African cichlids of Lakes Malawi and Tanganyika. Trop. Fish Hobbyist 22(3): 42-43, 45-48, 52-55, 57-59, 61-65. Fryer, G. 1956. New species of cichlid fishes from Lake Nyasa. Rev. Zool. Bot. Afr. 53(1-2):81-91. GrEeENWwoop, P. H. 1973. A revision of the Haplochromis and re- lated species (Pisces: Cichlidae) from Lake George, Uganda. Bull. Brit. Mus. nat. Hist. (Zool.) 25(5):139-242. Jounson, D. S. 1974a. New cichlids from Lake Malawi. Today’s Aquarist [Norwalk, Connecticut] 1(1):12, 14-17, 4 figs. 330 Proceedings of the Biological Society of Washington 1974b. Three new cichlids from Lake Malawi. Today’s Aquarist [Norwalk, Connecticut] 1(3):38—42. NEERGAARD, S. 1974. Nyasa-Cichlider. J. Fr. Clausens Forlag, Copen- hagen. 88 pp. SCHEUERMANN, H. 1974. Beobachtungen bei Labidochromis caeruleus “likomae.” Aquarium Aqua Terra [Wuppertal] 8(64):439- AAI. GH i) 4% MH No. 30, pp. 331-344 22 January 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW GENUS AND SPECIES OF CHARACID FISH FROM THE BAYANO RIVER BASIN, PANAMA (PISCES: CYPRINIFORMES) By WituiaM L. FINK Division of Fishes National Museum of Natural History Washington, D.C., 20560 and Department of Biology George Washington University Washington, D.C., 20006 As a part of their pioneering work on the fish fauna of Pan- ama, Meek and Hildebrand (1916) described the fauna of the lower Bayano river. At that time, travel to the upper trib- utaries of the river was very difficult and was not attempted by them; however, recent new roads have opened previously poorly-collected areas of the region to reveal a unique fauna in the upper Bayano tributaries. Dr. J. D. McPhail, University of British Columbia, is preparing a study on the ecology and distribution of the fishes of the area and has made specimens of an unusual characid available for the following description. I wish to thank Dr. Donald L. Kramer of the Smithsonian Tropical Research Institute (STRI) and Dr. McPhail for spec- imens of the new species, along with information regarding its distribution and ecology. Dr. Kramer also made valuable comments on the manuscript. Richard Goodyear of the Uni- versidad de Panama has aided in acquisition of specimens and supplied valuable information on the Bayano river. Dr. Stanley H. Weitzman kindly read the manuscript and has participated in numerous discussions on characids. Methods are those used in Fink and Weitzman (1974). 30—Proc. Biou. Soc. WaAsH., Vou. 88, 1976 (331) 332 Proceedings of the Biological Society of Washington TaBLE 1. Morphometrics of E. bayano in rfercent of standard length males, females, male, Characters range range holotype Standard length 43.2-52..1 40.4-55.9 Slo Greatest depth 2 IS Boll 31.3-34.7 Sloy Snout—dorsal fin origin 48.9-53.0 50.2—54.0 49.5 Snout—pectoral fin origin 24,7-27.1 25.0—27.8 24.7 Snout—pelvic fin origin 40.6—45.3 A42.0—46.4 42.5 Snout—anal fin origin 57.6—61.7 60.8—64.0 58.0 Caudal peduncle depth 12.3-15.8 11.6-13.8 14.7 Caudal peduncle length 11.9-14.7 11.6-13.4 14.7 Pectoral fin length 22.2-24.3 21.1-23.5 23.0 Pelvic fin length 16.2-18.2 14.8-16.6 18.0 Dorsal fin length 22,.9-25.2 23.1-26.7 24.0 Head length 94.2-97.6 25.9-27.8 iS Eye diameter 9.1-10.3 9.5-11.1 9.1 Snout length 6.1— 7.6 6.5— 8.1 6.9 Interorbital width 7.5— 8.2 7.7— 8.8 7.9 Upper jaw length 10.4—12.4 10.9-12.9 10.8 Eye—dorsal fin origin 35.0-38.9 36.4-38.3 35.4 Dorsal fin origin— caudal fin origin 51.4—55.0 50.6—54.1 52.8 Morphometric characters are given as percent of standard length (SL). In the description, range is given for morpho- metric characters; range for males and females separated by sex and the morphometric characters of the holotype are given in Table 1. Meristic characters are given in the text, holotype first, followed in parentheses by the range. Paratypes have been deposited in the National Museum of Natural History, Washington, D.C. (USNM), British Museum (Natural His- tory) (BMNH), Academy of Natural Sciences, Philadelphia (ANSP), California Academy of Sciences, San Francisco (CAS), and Zoological Museum of the University of Amster- dam (ZMA). Eretmobryeon, new genus Type-species: Eretmobrycon bayano, new species. Diagnosis: Pelvic fin with i,8 rays. Caudal fin of mature males asymmetrical, lower lobe much enlarged. Ventral rays of lower caudal fin lobe thickened and expanded dorsoventrally (Fig. 1). Caudal fin Characid fish from Panama 333 Fic. 1. Caudal fin of E. bayano, male, 52.3 mm, in breeding condition, from USNM 214006. of large adult females often asymmetrical but not as much as in males. Monotypic. Other characters of the genus include: premaxillary with an irregular row of 4-6, usually 5, tricuspid outer row teeth and four inner row teeth with 5-6 cusps. Lateral line complete. Third infraorbital well ossified, contacting horizontal limb of preopercle ventrally and with a naked area between itself and vertical limb of preopercle. Dorsal fin, pelvic fins, pectoral fins, and anal fin with spinules in breeding males. Caudal fin with irregular scales at base, some somewhat enlarged, es- pecially on the lower caudal fin lobe. Caudal peduncle of males much deeper than in females. Precaudal vertebrae 17-18; total vertebrae 37— 38. Etymology: From eretmon, meaning paddle, referring to the paddle- like lower caudal fin lobe, and Brycon, a genus of American characids. 334 Proceedings of the Biological Society of Washington Fic. 2. Holotype of E. bayano, male, 51.7 mm, USNM 213842. Eretmobrycon bayano, new species Figures 1-4 Holotype: USNM 213842, male 51.7 mm SL; Panama, Bayano river basin, pool in small stream about 19 km along road from E] Llano to Carti, about 0.5 km downstream from the road. Collected 10 March, 1973 by J. D. McPhail and R. Dressler. Paratypes: All with same data as holotype: USNM 213843, 22 speci- mens 32.1-55.9 mm; ANSP 130186, 2, 46.5-53.6 mm; CAS 31895, 2, 43.2-46.8 mm; BMNH 1975.2.3.1-2, 2, 43.6-46.9 mm; ZMA 113.490, 2, 49.0—52.1 mm. Other specimens: USNM 214006, Panama, Bayano river basin, first stream on road from El Llano to Carti, approximately 11 km from main road between Chepo and Bayano Dam sites, collected 24 Jan. 1974 by D. L. Kramer, 7 specimens 41.9-52.9 mm (1 specimen cleared and stained); USNM 214007, 40 specimens 30.4—-57.5 mm, same data as USNM 214006, but collected 15 Jan. 1975; USNM 214664, 36 speci- mens 16.8—53.9 mm, same data as USNM 214006, but collected 23 April 1975. Description: The following description is based on specimens selected over the size range from the types and USNM 214006. Standard length of examined specimens 40.4—55.9 mm. Body elongate, compressed lat- erally, greatest body depth 27.7—34.7 percent. Predorsal profile convex with a slight concavity at nape. Profile between posterior dorsal fin base and anterior adipose fin base slightly convex, with a slight dip at adipose fin base. Posterior to adipose fin, profile slightly concave to upper procurrent caudal fin rays. Distance from eye to dorsal fin 35.0— 38.9; distance from dorsal fin origin to end of caudal peduncle 50.6—55.0. Ventral profile gently rounded from jaws to anus; steepest inclination ventral to jaws. Ventral profile with greatest protrusion just anterior to pelvic fin. Profile along anal fin base straight in females, slightly concave anteriorly and slightly convex posteriorly in males; between anal fin and procurrent caudal rays, body profile concave in females, convex in males. In males, lower procurrent caudal fin rays protrude more than upper procurrent rays. Caudal peduncle depth 11.6—15.8 (Fig. 5); peduncle length 11.6-14.7. Characid fish from Panama 335 Fic. 3. Female paratype of E. bayano, 54.3 mm, USNM 213843. Head length 24.2-27.8. Eye diameter 9.1-11.1. Snout length 6.1- 8.1. Least bony interorbital width 7.5-8.8. Maxillary sloping ventrally and posteriorly, forming an angle of 50-60 degrees to longitudinal body axis; upper jaw length 10.4-12.9. All teeth with median cusp longest. Premaxillary with 4 inner row teeth each with 5-6 cusps, and 4-6, usually 5 tricuspid outer row teeth; first, third, and fifth outer row teeth set slightly anterior to other outer row teeth. Maxillary with 2-3, usually 2, teeth each with 5-7 cusps. Dentary with 4 large quinque- cuspid teeth anteriorly, 4-6 tricuspid to conical teeth posteriorly; sec- ond tooth from symphysis offset slightly anterior to tooth row; fourth tooth from symphysis curves slightly posteriorly. No teeth present on vomer, palatines or pterygoids. Fontanels moderately long, that part anterior to epiphyseal bar about half the length of that part posterior to bar. Gill rakers moderate, 15 counted in the cleared and stained specimen. Circumorbital bones well ossified; infraorbital 2 wide, no fleshy area ventral to it; infraorbital 3 wide, contacting preopercle ventrally; a fairly wide area of skin pos- terior to infraorbital 3. Scales moderately large, cycloid, with concentric circuli and 0-10 radii on exposed posterior field. Lateral line complete, with 37-39, usually 37-38 perforated scales. Lateral line often with a slightly ir- regular ventral curve on side of body from origin to below dorsal fin, then continuing with a gentle irregular dorsal curve to caudal fin base. Scales above lateral line 6-7 (rarely 7); scales below lateral line 5. Predorsal scales 12-13. Scale sheath at anal fin base of about 8-10 scales in a single row. Axillary scale present dorsal to pelvic fin in- sertion. Base of caudal fin with irregular scales, some somewhat en- larged, especially on lower caudal fin lobe (Fig. 1). Dorsal fin with 2 anterior unbranched rays and 9 (8[4], 9 [22]) branched rays. Dorsal fin origin anterior to anal fin origin, posterior to pelvic fin origin, nearer eye than caudal fin base. Distance from tip of snout to dorsal fin origin 48.9-54.0. Third or fourth ray of dorsal 336 Proceedings of the Biological Society of Washington Fic. 4. Live specimen of E. bayano, male, 52.9 mm, same locality data as USNM 214006, collected 23 April 1975 by D. Kramer. fin longest, posterior rays shorter, forming rounded posterior margin of fin; length of longest ray 22.9-26.7. In breeding males minute spi- nules present, sparingly, on first through fourth branched dorsal fin rays. The spinules occur singly, primarily on the posterior main ray branch segments, more numerous distally. Anal fin with v (vi[9], vii[16]) unbranched rays and 28 (26[1], 27[2], 28[13], 29[7], 30[2]) branched rays. First through third or fourth un- branched rays usually visible only in radiographs or cleared and stained specimens. Origin of anal fin posterior to midpoint of standard length 57.6-64.0. Posteriormost unbranched ray and first through seventh or eighth branched rays longer, with successive posterior rays shorter, forming an abruptly protruding fin margin anteriorly and a straight margin posteriorly. Dorsally recurved anal fin spinules present in adult breeding males; these occur on the posteriormost unbranched ray and on as many as all branched rays in males in full breeding condition. Spinules present on posterior edge of fin rays, usually in bilateral pairs, with one spinule pair per bony ray segment. Spinules less numerous on anterior branches of branched rays than on posterior branches. On ex- treme distalmost segments of anterior ray branches, spinules may be on anterior margin of ray segments instead of posterior margin. Non-breed- ing males may have small nubbins instead of spinules. Fin often fleshy around spinules. Anterior branched anal fin rays in all males somewhat thicker than corresponding rays in females. Pectoral fin with one unbranched ray and 10 (9[1], 10[15], 11[11]) branched rays. Pectoral fins reach slightly beyond pelvic fin origin. Distance from tip of snout to dorsal end of pectoral fin base 24.7—27.8 and length of pectoral fin from base to tip of longest ray 21.1-24.3. Small spinules present in fully developed breeding males on first un- branched ray and first through fifth to seventh branched rays; most spinules point ventral-ward and are on posterior edges of rays, one or Characid fish from Panama 337 two to a bony ray segment. A few dorsal-pointing spinules on longest rays, one to a ray segment (these may occur on ray segments along with two ventral-pointing spinules). Pelvic fin with i,8 rays in all specimens, distal fin tip reaching just before or just to anal fin origin. Distance from tip of snout to pelvic fin origin 40.6-46.4; pelvic fin length 14.8-18.2. Males with antero- ventral pointing spinules on all pelvic fin rays. Spinules occur primarily distal to branching point of branched rays on anteriormost four fin rays, but occur proximal to that point in more posterior rays, area cov- ered progressively increasing with each successive ray. Spinules more numerous on posterior main branch of ray; when main ray branches branched, spinules more numerous on posterior branch of secondary branch. Spinules occur singly or in pairs on each bony ray segment. When only one spinule present, it occurs on posterior edge of segment; when in pairs, spinules on both anterior and posterior edge of ray seg- ment (posterior spinule of pair is first to develop). Paired spinules oc- cur primarily on posterior main ray branch, but are commonly found on anterior main ray branch on more posterior fin rays. Caudal fin with 10/9 principal rays in all specimens; fin forked, not split to base. In adult males and large adult females, lower lobe much larger than dorsal lobe; posteriormost two ventral procurrent rays and rays 9-4 (numbering ventralmost principal caudal ray as 9) enlarged and dorsoventrally expanded; this condition more extreme in males than in females (Figs. 1-3). In small females and immature males caudal fin symmetrical. No spinules on caudal fin rays. Precaudal vertebrae 17 (17[12], 18[12]). Total vertebrae 38 (37([6], 38[19]). Color in alcohol: Ground color cream brown. Scales of back and sides to midline with large dark chromatophores on inner surface of free margin forming a reticulate pattern. In dermis adherent to inner scale surface, anterior to large chromatophores, occur numerous smaller black chromatophores. Numerous, less dark chromatophores approximately size of marginal scale chromatophores, in dermis medial to scales (and medial to layer adherent to scales). Anteriorly, ventral to lateral mid- line, small black chromatophores less numerous; more ventrally, large chromatophores less regular and less numerous with loss of reticulate pattern. No chromatophores on belly. Above anal fin, ventral to lateral midline, most chromatophores black, of dermal type, irregularly placed, partially following myomere junctions. Dark stripe posteriorly along lateral midline, becoming concentrated into a rounded caudal blotch. Dark chromatophores extend to distal ends of middle caudal fin rays. Numerous small dark chromatophores cover top of head and lower lip, extend along maxillary, around orbital rim, anteriorly along vertical pre- opercle limb, and on opercular skin flap. Larger, less dark chromato- phores loosely scattered under infraorbital bones and opercle. No well defined humeral spot. Dorsal fin with numerous dark chromatophores along fin rays and on 338 Proceedings of the Biological Society of Washington interradial membrane immediately adjacent to fin rays, often more con- centrated at fin base. Adipose fin with numerous small dark brown chromatophores. Caudal fin with dark chromatophores along dorsal and ventral ray surface and on interradial membrane along posterior fin margin; numerous small dark chromatophores along dorsal margin of dorsal lobe and ventral margin of ventral lobe. Anal fin with small brown chromatophores along rays and on interradial membrane. Pelvic fins and pectoral fins with dark chromatophores along fin rays; no chromatophores on interradial membrane. Color in life: Body light olive brown above and silvery white below a mid-lateral band of silver or greenish gold (color of band depends on angle of light reflection). Band along posterior two-thirds of body. Dark caudal spot and faint humeral spot present. Males with yellow at base of upper caudal fin lobe and in middle of lower caudal fin lobe; faint black streaks along fin rays on lower caudal fin lobe. Dorsal fin slightly dusky; other fins nearly hyaline. Iris of eye silvery except for yellow spot dorsally. Female coloration similar to males but yellow on caudal fin fainter. Color description based on information supplied by D. Kramer and from live specimens kept in aquaria; see Fig. 4 for color pattern in life. Ecology: According to Drs. McPhail and Kramer (in litt.), the water- shed of the habitat of E. bayano consists of heavily forested steep hills. The type-collection was made in a small stream, in a rocky-bottomed pool about 15 m long, 8 m wide, and less than 50 cm deep. Dr. Kramer collected in a small stream of moderate gradient with pools alternating with riffles and small falls; mean width of the stream is about 3 m, depth rarely over a meter. The stream bed is sandy with some stones and ledge (except where thick red clay eroded from the road has covered the bottom) and moderate cover in the form of forest debris; the stream has no aquatic macrophytic vegetation. Stream temperature in mid- January (early dry season) in mid-afternoon was 24.5° C.; stream flow was estimated by Kramer at about 0.05 m°/sec. Other genera of fishes collected with E. bayano included Brachyraphis, Brycon, Bryconamericus, Chaetostoma, Hypopomus, Poecilia, Pygidium, and Rivulus. Dr. McPhail examined stomach and gut contents of E. bayano collected at midday. Although most stomachs were empty, the intestines of many specimens contained seeds, terrestrial insects, and in two individuals a few fish scales (not of Eretmobrycon). In five specimens Dr. Kramer found aquatic insect larvae, terrestrial insects, and algae to be impor- tant components of the stomach contents. Sexual dimorphism and reproduction: Mature non-breeding males of E. bayano can be distinguished from females by the enlarged rays of the lower caudal fin lobe. In some large females these rays are some- what enlarged but not as much as in males. Sexually ripening or second and third year males which have bred, have poorly developed or re- duced spinules on the anal fin rays. Breeding males are easily identified Characid fish from Panama 339 USNM 214007 — usamales — ofemales 213842-3 score emales 6.0 f ------ o females 4.0 CAUDAL PEDUNCLE DEPTH 30 25.0 30.0 35.0 40.0 45.0 50.0 55.0 60.0 STANDARD LENGTH Fic. 5. Caudal peduncle depth related to SL in two populations of E. bayano, USNM 214007, breeding; USNM 213842-3, non-breeding. Linear regressions are plotted for males and females of each population. by their well-developed complement of spinules on all fins but the caudal fin, and large testes; breeding females have plump bellies and ripe ova. Caudal peduncle depth is greater in males than in females (Fig. 5), and the limited data suggest that peduncle depth in males increases dur- ing sexual ripening. The non-breeding males of USNM 213843 and USNM 214664 (the latter not shown in Fig. 5) have virtually identical caudal peduncle depths (these populations come from close, but not identical localities). The breeding males of USNM 214007 have sig- nificantely greater caudal peduncle depth than the non-breeding males just mentioned. Note that USNM 214664 and USNM 214007 are from the same locality and presumably represent the same population in breeding and non-breeding condition. Also note (Fig. 5) that caudal peduncle depth of females is very similar in breeding and non-breeding populations. Dr. Kramer (in litt.) indicates that the smallest maturing male he has examined was 43 mm SL and the smallest maturing female was 40 mm SL. His studies suggest that breeding is seasonal (January to Feb- ruary), and that the breeding population consists primarily of two-year fishes, with a few three-year fish breeding a second time. A large number of specimens in the population samples seem to be immature one-year fish. Relationship: A thorough phylogenetic analysis of characters that could be used to infer the relationships of Eretmobrycon cannot be undertaken at this time. As discussed by Fink and Weitzman (1974) 340 Proceedings of the Biological Society of Washington and Weitzman and Fink (in press), current knowledge of the distribu- tion and lability of many morphological structures traditionally used in characid systematics is woefully inadequate. Determination of the use- fulness in classification of such characters as number and shape of pre- maxillary and maxillary teeth, fin specializations, and size of the third infraorbital is a long-term project and involves examination of many characids from Central and South America and Africa. Such studies are underway. Given this disclaimer, it is possible to speculate on the affinities of E. bayano and the validity of the genus. The characters of E. bayano that have prompted erection of a new genus, i.e., the pelvic fin ray count and the specialized caudal fin struc- ture, are of unknown phylogenetic significance. The species cannot be placed in any existing genus without causing a major redefinition of that genus. It may be that future studies will indicate that E. bayano should simply be placed in Bryconamericus. Until those studies, including a thorough revision of Bryconamericus and related genera, are possible, it seems best to recognize the unique specializations of E. bayano and place it in a monotypic genus. Of the genera in Central America and northern South America, Eretmobrycon seems most probably related to Bryconamericus and Astyanax. Members of these genera are rather morphologically and ecologically generalized and widely distributed, especially in South America. The supposed differences between Astyanax and Bryconameri- cus may eventually be considered unimportant at the generic level. There are four characters which have been used to separate the two genera: (1). The number of inner row premaxillary teeth (5 in Astyanax, with a few species having 4; 4 in Bryconamericus). It becomes evident in reading Eigenmann (1921), that in a small number of species of Astyanax the tooth counts cannot be used in generic placement. This is a common problem in systematics of characids; many characters can be used to differentiate species or genera in limited geographical areas, but nearly every systematic character used is labile to such a degree that in some geographical areas generic or species definitions based on these characters break down. Many characid genera “grade” into one another and until a thorough analysis of systematic characters used is completed, this unaccept- able situation will remain. (2). Regular, in Astyanax, or irregular, in Bryconamericus, place- ment of the outer row premaxillary teeth was used by Meek and Hildebrand (1916) for separation of these two genera. The outer row teeth may be in a regular series or some may be slightly dis- placed relative to the other outer row teeth. Although this char- acter is useful in separating the genera in at least some parts of Panama (Kramer, pers. comm.), Eigenmann (1927) states (and I concur) that among species of Bryconamericus over a wide geo- Characid fish from Panama 341 graphical range, the outer tooth row may be either regular or ir- regular depending on the species. (3). Width of third infraorbitals (Astyanax has a narrow fleshy area between the third infraorbital and the horizontal limb of the preopercle while Bryconamericus has the third infraorbital touching the horizontal limb of the preopercle). This kind of difference has been used to differentiate several characid genera. Eigenmann (1921, 1927) numbered the infraorbitals differently than is done currently; his suborbital 2 is actually infraorbital 3 (see Weitzman, 1962: 28 for a discussion of the osteology of characid orbital bones). Eigenmann diagnosed Astyanax as having a naked area ventral to the “second suborbital” (infraorbital 3 of Weitzman) and Brycon- americus as having the “suborbital” touching the preopercle ven- trally. The character is useful, at least in Panama. He further de- fined Astyanax as having a naked (fleshy) triangle below the joint between the “first and second suborbitals” (= 2 and 3 infraorbitals ) while in Bryconamericus there is no naked triangle below the “sub- orbitals.” I have found in cleared and stained specimens that the amount of “naked” area is somewhat variable and simply a function of the size of the third infraorbital; when the third infraorbital is large enough to touch the preopercle ventrally, it is also large enough to touch the second infraorbital along the posteroventral edge of that bone. (4). Kind and distribution of adult male anal fin hooks and spi- nules are characters I have found to be of use in Central America. Mature Astyanax males have bony hooks on the segments of the fourth unbranched ray and the first through seventh to ninth branched rays, the hook tips pointing toward the fin base. In the sexually ripe Bryconamericus males examined, spinules, not hooks as found in most characids, are present on the last unbranched ray and all branched rays, from near the ray bases to their distal tips. These spinules are very thin and point toward the distal ends of the rays in some species and toward the fin base in other species. In fact, on some Panamanian Bryconamericus, these spinules cov- ered all of the fins, including pelvic and pectoral fins, dorsal fin, and dorsal and ventral caudal fin lobes. Previous investigations (Fink and Weitzman, 1974), and information supplied by D. Kramer (pers. comm.) indicate that presence of fin spinules or hooks is de- pendent on breeding season in at least some characids and that different species may be sexually active at different times of the year in the same geographical area. Therefore, care must be exer- cised in interpreting presence or absence of the structures in these fishes. Of the four characters mentioned above, Eretmobrycon shares (1), (2), (3), and (4) in part, with Bryconamericus and not with Astyanax. 342 Proceedings of the Biological Society of Washington In dentition and shape of the infraorbital bones, Eretmobrycon is vir- tually identical with Bryconamericus. In character (4), Eretmobrycon agrees with Bryconamericus in having spinules on the last unbranched and all branched anal fin rays. However, the spinules of Eretmobrycon are slightly “heavier” (thicker) than the spinules of Bryconamericus (although not nearly as massive as the hooks of Astyanax). The spinules of Eretmobrycon point toward the fin base as in Astyanax, not the distal fin edge as in some Bryconamericus. Again, we are dealing with a character of unknown phylogenetic significance. Hooks, as found in Astyanax, are more commonly found in characids than spinules. The distribution of spinules in characid species clearly not closely related (e.g., Bryconamericus, Carlana [Fink and Weitzman, 1974], and some “glandulocaudin” genera [Weitzman and Thomerson, 1970], indicates that their presence is probably a convergent specialization appearing throughout the Characidae. Therefore, their presence in Bryconamericus and Eretmobrycon, while suggestive, cannot be taken as evidence of close relationship. The presence of four premaxillary inner row teeth is another char- acter common to Eretmobrycon and Bryconamericus. Astyanax usually has five inner row teeth. I expect that four teeth is more specialized than five teeth, and is probably related to mouth size and feeding spe- cializations. As mentioned above, this character should be used with caution because of its lability. Tubercles (hypertrophied patches of epidermal cells) are present on breeding Bryconamericus males from Panama and are absent on Astyanax and Eretmobrycon. These structures may be found to be im- portant characters but their evaluation is beyond the scope of this paper. It is reasonable to question the establishment of a new genus for E. bayano in view of the overabundance of generic names of Neotropical Characidae. Eretmobrycon bayano has one specialization found in no other characids (the hypertrophied lower caudal fin lobe) and another (the pelvic fin ray count of i,8) found only in the genus Crenuchus, a highly specialized form belonging to a very different characoid lineage. Since these characters are autapomorphic at this level they are of no use in determining relationships. Traditionally important systematic characters suggest relationship between Eretmobrycon and Bryconamericus and the latter may even- tually be where E. bayano is placed. Nevertheless, my examination of many nominal species of Bryconamericus strongly suggests that that genus is polyphyletic and consists of a number of difficult to define lineages. For instance, at least some Central American species of Bry- conamericus may someday be placed together as a natural group, based in part on the presence of breeding tubercles. I do not believe that EF. bayano can be precisely related to any species now currently placed in Bryconamericus and hesitate to place it in an already polyphyletic genus. It is possible that E. bayano is not related to any existing species of Bryconamericus but is a specialized relict of a Bryconamericus-like form Characid fish from Panama 343 that entered Panama before the ancestor(s) of Bryconamericus now living there. Alternatively, E. bayano may be descended from the com- mon ancestor of one of the lineages of Bryconamericus. If Eretmobrycon bayano did arise from a Bryconamericus-like an- cestor, subsequent evolution has resulted in specializations which clearly set it apart from the known species of that genus, i.e. the increase of one pelvic fin ray and the hypertrophied lower caudal fin lobe. The caudal fin lobe hypertrophy in adults, especially males, is perhaps re- lated to breeding behavior. Determination of the function of this par- ticular morphological specialization must await aquarium or field ob- servations of spawning. Etymology: The specific epithet bayano refers to the river basin in which the species seems to be endemic. Resumen En el presente trabajo se describe Eretmobrycon bayano, neuvo género y nueva especie de la familia Characidae del rio Bayano, Panama. El género Eretmobrycon se caracteriza por tener i,8 radios pélvicos, radios caudales del lébulo inferior muy hipertrofiados (particularmente en los machos) y dos filas de dientes premaxilares. Los machos de E. bayano que no han alcanzado la época de la re- produccién, se distinguen de las hembras por poseer radios caudales hipertrofiados y un pediinculo caudal mas alto. En contraste con los machos antes descritos, los machos reproductores tienen los radios in- feriores de la aleta caudal mas alargados, el pedunculo caudal mas alto y los radios de todas las aletas, excepto la caudal, recubiertas de espinulas. E] desove tiene lugar en enero y febrero de cada afio y la mayoria de la poblacién reproductora esté formada por individuos de dos afios y algunos de tres anos que posiblemente desovan por segunda vez. El habitat propio de la especie son arroyos de montafias poco pro- fundos, en vertientes con una vegetaciOn boscosa. Su dieta basica con- siste en semillas, larvas de insectos acuaticos, insectos terrestres y algas. Con la finalidad de determinar las relaciones del género Eretmo- brycon, se presenta un breve analisis filogenético de varios caracteres sistematicos usados para definir a Astyanax y Bryconamericus. El nu- mero de dientes premaxilares en la fila interna (5 en Astyanax, aunque algunas especies tienen 4; 4 en Bryconamericus) es un caracter util para la separacién de estos géneros en una regidn geografica determinada; sin embargo, la similitud del numero de dientes entre algunas especies de los dos géneros impide el empleo de este caracter para una diferenciacién generica. La regularidid (en Astyanax) o irregularidad (en Brycon- americus ) de la fila externa de dientes premaxilares es un caracter bueno para distinguir los dos géneros en algunas localidades de Panama, pero en especies de Bryconamericus de un ambito geografico mas amplio, la fila externa de dientes premaxilares puede ser regular o irregular de- pendiendo de la especie. El ancho del tercer infraorbitario (= segundo suborbitario de Eigenmann) y el area “desnuda” asociada, son utiles 344 Proceedings of the Biological Society of Washington para la separacién de los dos géneros en muestras de Panama, pero son demasiado variables para el diagndstico, cuando se incluyen especies de la América del Sur. El] tipo y distribucién de ganchos y espinulas en las aletas, son caracteres Utiles, por lo menos en la América Central. Los machos adultos de Astyanax tienen ganchos éseos en los radios pélvicos y en los radios anteriores de la aleta anal, mientras que en Bryconameri- cus presentan espinulas mas delicadas, que a veces se encuentran en todas las aletas. Aunque los ganchos y espinulas de las aletas aparecen sdlo en la época de la reproduccién, son eventualmente utiles para de- finir, por lo menos, algunos de los géneros de la familia Characidae, in- cluyendo linajes actualmente ubicados en Astyanax y Bryconamericus. Eretmobrycon y Bryconamericus son muy similares en los caracteres discutidos anteriormente, difiriendo sdlo en detalles morfolégicos de las espinulas. E. bayano podria ser considerado como una especie de Bry- conamericus, pero si aceptamos que este ultimo género es polifilético y que las posibles relaciones de E. bayano con las otras especies del género son atin ignoradas, es mejor reconocer por el momento que el numero peculiar de radios pélvicos y las especializaciones de la aleta caudal de la nueva especie son al mismo tiempo caracteres de un nuevo género. LITERATURE CITED EIGENMANN, C. H. 1921. The American characidae. Mem. Mus. Comp. Zool. 43 (3):209-310. 1927. The American characidae. Mem. Mus. Comp. Zool. 43(4):311—-428. Fink, W. L., anp S. H. WeirzMan. 1974. The so-called cheirodontin characids of Central America, with descriptions of two new species (Pisces: Characidae) Smithsonian Contrib. Zool. No. 172:46 pp. Merk, S. E., anp S. F. Hixpespranp. 1916. The fishes of the fresh- waters of Panama. Field Mus. Nat. Hist., Zool. Ser. 10(15): 217-374. WerrzMan, S. H. 1962. The osteology of Brycon meeki, a generalized characid fish, with an osteological definition of the family. Stan- ford Ichthyol. Bull. 8(1):1-77. WEITZMAN, S. H., anp J. E. THomerson. 1970. A new species of glandulocaudine characid fish, Hysteronotus myersi, from Peru. Proc. Calif. Acad. Sci. 38(8):139-156. WEITZMAN, S. H., AND W. L. Finx. In press. The interrelationships of the neon tetras, a unique group of South American characids (Pisces ), with comments on the phylogeny of new world chara- coid fishes. Smithsonian Contrib. Zool. GH 1 Bax MH No. 31, pp. 345-350 22 January 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON GLYPHIDODONTOPS SPRINGERI, A NEW SPECIES, / OF DAMSELFISH (POMACENTRIDAE) FROM [i THE PHILIPPINE AND MOLUCCA ISLANDS By Geratp R. ALLEN AND RocEeR LUBBOCK | =. Department of Fishes, Western Australian Museum 4\ ~ Perth 6000, Australia, and \ Zoology Department, University College, London, Gower Street, London WC1 E6BT The damselfishes (Pomacentridae) are one of the dominant fish groups of tropical reefs in both species and individuals. In spite of their abundance the classification of these fishes remains poorly understood. The majority of species have been assigned by most previous authors to Chromis, Abudefduf or Pomacentrus. The taxonomy of the latter two genera is partic- ularly confusing. Allen (1975) provisionally recognizes 21 genera from the Indo-West Pacific. His classification rep- resents a modification of that proposed by Bleeker (1877). In his study, the genus Abudefduf is restricted to relatively large species characterized by 3 to 4 scale rows above the lateral-line and uniserial teeth. Many of the species have al- ternating light and dark bars on the sides. The genus is typi- fied by the well known circumtropical species, A. saxatilis. The majority of species previously included in Abudefduf are assigned to Glyphidodontops Bleeker. The salient features of this genus include a relatively small size (usually less than 80 mm standard length at maturity) and an elongate body (depth generally in excess of 2.0 in standard length). Many of the species are brightly colored and most exhibit biserial dentition, at least at the front of the jaws. This paper describes a member of this genus which was recently collected in the Philippine and Molucca Islands. 31—Proc. Bion. Soc. WAsH., Vou. 88, 1976 (345) 346 Proceedings of the Biological Society of Washington TasLeE 1. Counts from 30 type-specimens of Glyphidodontops springeri Dorsal rays Gill rakers Locality XI XII XIII 10 10% 11 11% 21 22 23 24 25 Molucca Islands Qe NS eel Tey ee Ay ede io PDs Onl ORS Philippine Islands 9 Oni bas 3) uo mee Tubed lateral-line scales Anal rays Pectoral rays LO NS 4 TS G6 LO LO Le al i Se Ae enc Molucca Islands TOM ON fo. I apo 5 Glib iee Philippine Islands 1 We chy ateh hares 3. 3 9 The methods of counting and measuring are the same as those described by Allen (1972), except the length of the dorsal and anal spines are measured proximally at the base of the spine rather than the point at which the spine emerges from the scaly sheath. Measurements were made with needle- point dial calipers to the nearest millimeter (mm). Standard length is abbreviated as SL. The fraction % appearing in the dorsal and anal fin ray formulae refers to a bifurcate condi- tion of the last ray. The counts and proportions which appear below in paren- theses apply to the paratypes when differing from the holotype. A summary of counts for the dorsal, anal, and pectoral fin rays, gill rakers on the first arch and tubed lateral-line scales is presented in Table 1. Proportional measurements of the holo- type and several paratypes are expressed as percentages of the standard length in Table 2. Type-material has been deposited in the following institu- tions: Bernice P. Bishop Museum, Honolulu (BPBM); British Museum (Natural History), London (BMNH); National Mu- seum of Natural History, Washington (USNM); Western Australian Museum, Perth (WAM). Damselfish from East Indies 347 TaBLE 2. Morphometric proportions (in % of SL) of Glyphidodontops springeri Paratypes Holotype ———-AA USNM BMNH USNM USNM USNM Character 213577 1974.8.22.1-9 209652 209838 209652 Standard length (SL) SBMS | Cyd) USS eh VERO) lbtsyho) ND w Body depth 494 483 466 494 518 £500 Head length 333 SILA nosou 14 2nN G42, Snout length 68 on 68 78 67 68 Eye diameter 110 115 119 121 141 137 Interorbital width 92 88 96 93 99 107 Least depth caudal peduncle 149 155 155 149 144 150 Length caudal peduncle 137 147 153 130° 9116) 124 Snout to origin dorsal fin 366 349 376 391 £380 385 Snout to origin anal fin 664 659 + 641 696 701 658 Snout to origin pelvic fin 381 432 390 404 430 397 Length dorsal fin base 586 = =587 613 590 574 564 Length anal fin base 250 232 24a) eo u ea nl 1256 Length pectoral fin 268 245 198* 280 296 282 Length pelvic fin DS BBB) BIS | BULL 246* 291 Length pelvic spine 152 179 184 186 180 205 Length Ist dorsal spine 65 64 76 78 TA U0 Length 5th dorsal spine TS I APE cs Pilsy PALE! Length 12th dorsal spine 131 149 147 140 123 128 Length longest soft dorsal ray 202 205 eins Se OTM Di Ni 2 Se Length Ist anal spine hid 96 102 96 85 81 Length 2nd anal spine 149 ISO MODE SS TO) M50 Length longest anal ray PAL 165* 124* 905 292 235 Length middle caudal rays S04) GES Tule BHI BIB BAP} * damaged Glyphidodontops springeri, new species Figure 1 Holotype: USNM 213577, 33.6 mm SL, collected with rotenone in 8 m off east shore of Piru Bay, Tandjung Liang, Ceram, Molucca Islands by V. Springer and M. Gomon, 10 January 1973. Paratypes: BMNH 1974.8.22.1-9 (9 specimens, 24.2-37.5 mm SL) collected with dipnets in 20-30 m off Cebu, Philippine Islands, 1974; BPBM 18618 (3, 18.9-33.6 mm) collected with holotype; USNM 209652 (45, 9.6-34.0 mm) collected with holotype; USNM 209838 (5, 17.0- 28.4 mm) collected with rotenone in 5-7 m at Piru Bay, Ceram, Molucca 348 Proceedings of the Biological Society of Washington hava } Fic. 1. Glyphidodontops springeri, holotype, USNM 213577, 33.6 mm SL, Ceram, Molucca Islands (drawing by Helen K. Larson). Islands by V. Springer and M. Gomon, 9 January 1973; WAM P25127-001 (3, 18.9-33.6 mm) collected with holotype. Diagnosis: A species of Glyphidodontops with the following combina- tion of characters: dorsal spines usually 12; pectoral rays usually 15; tubed lateral-line scales usually 12 to 14; color mostly blackish (brown in preservative) with bright blue streaks on most scales of lower half of sides. Description: Dorsal rays X1I,10% (XI to XIII,10 to 111%); anal rays IZ,11 (JL,11 or 11%); pectoral rays 15 (14 to 16); pelvic rays I,5; branched caudal rays 13; gill rakers on first arch 23 (21 to 24); tubed lateral-line scales 12 (11 to 15); vertical scale rows from upper edge of gill opening to base of caudal fin 27 (27 or 28); horizontal scale rows from base of dorsal fin to posteriormost scale of tubed lateral-line (exclusive of sheath scales at dorsal base) 1; from lateral-line to anal fin origin 8; predorsal scales about 17 (16 to 18), extending to front margin of orbits; teeth of jaws biserial (inner row of teeth poorly de- veloped and apparent only under high magnification; best observed at middle of lower jaw), more or less incisiform with truncate to slightly notched tips. Body ovate, laterally compressed, greatest depth 2.0 (1.9 to 2.1) in SL. Head profile conical; head length 3.0 (2.9 to 3.2) times in SL. The following characters are expressed as proportions of the head length: snout 4.9 (4.0 to 5.1), eye diameter 3.0 (2.4 to 2.7); interorbital width 3.6 (3.2 to 3.6), least depth caudal peduncle 2.2 (2.0 to 2.4), length caudal peduncle 2.4 (2.1 to 2.9), pectoral fin length 1.2 (1.2 to 1.3), pelvic fin length 1.2 (0.9 to 1.2), caudal fin length 1.1 (1.1 to 1.2). Damselfish from East Indies 349 Single nasal opening on each side of snout; mouth oblique, terminal; lateral-line gently arched beneath dorsal fin, terminating at scale row below base of ninth dorsal spine; preorbital, suborbital, tip of snout, lips, chin and isthmus naked; remainder of head and body scaled; scales finely ctenoid; preopercular scale rows 2, an additional row of smaller scales covering anterior part of inferior limb; small sheath scales cover- ing approximately one-third to one-half of membranous basal portions of dorsal, anal, and caudal fins; margin of preorbital, suborbital, and pre- opercle entire; margin of opercle with flattened spine at angle and another near upper corner of gill opening. Origin of dorsal fin above third tubed scale of lateral-line; anterior spines of dorsal fin gradually increasing in length to about fifth or sixth spine, remaining spines gradually decreasing in length. The following characters are expressed as proportions of the head length: length first dorsal spine 5.1 (4.1 to 4.9), fifth dorsal spine 1.8 (1.5 to 1.6), last dorsal spine 2.5 (2.1 to 2.8), longest soft dorsal ray 1.6 (1.5 to 1.6), first anal spine 4.3 (3.3 to 4.2), second anal spine 2.2 (1.6 to 2.3), longest soft anal ray 1.5 (1.5 to 1.6). Color in alcohol: Head and body of holotype mostly uniform brown- ish-blue, lighter on breast and caudal peduncle; spinous dorsal and an- terior portion of pelvic and anal fins brownish-blue or dusky; remainder of fins pale; pectorals with black spot covering base. Paratypes from the Molucca Islands are similar in color although many were damaged and exhibit missing scales and tan splotches on the head and body. The largest paratypes (33.5-37.5 mm SL) from Cebu, Phil- ippine Islands are more colorful with many of the body scales pale blue. There are also scattered blue spots on the head and three pale blue stripes; one extending from top of eye to snout tip; another, very short, from anterior edge of eye to upper lip; and one on suborbital. A small black spot present at upper comer of the gill opening and another on upper pectoral base. Smaller specimens from Cebu are similar to those from the Moluccas. Color in life: Based on a 35-mm transparency of a paratype from Cebu, photographed in an aquarium: Ground color of head and body blackish; head and lower two-thirds of body with numerous blue scales, those on body giving the appearance of narrow blue streaks; blue lines passing through upper and lower portions of eye; similar blue lines on snout; spinous dorsal fin blue; soft dorsal, caudal, and posterior portion of anal fin pale or slightly dusky; anterior part of anal fin bluish-black with bright blue anterior margin; pelvic fins blackish with bright blue anterior margin. Remarks: The count of 12 dorsal spines separates G. springeri from other known species of Glyphidodontops. Most of the members of this genus have 13 spines and at least one other, G. caeruleolineatus (Allen), has 14. Specimens of G. springeri from Cebu, Philippine Islands tend to have slightly higher gill raker and lateral-line scale counts in com- parison with those from the Moluccas (Table 1). G. springeri belongs 350 Proceedings of the Biological Society of Washington to a complex which includes G. azurepunctatus (Fowler and Bean), G. hemicyaneus (Weber), G. rollandi (Whitley), G. talboti Allen, and G. traceyi (Woods and Schultz). These species share a similar dentition and have relatively low soft dorsal, anal, and pectoral counts. They are relatively deep bodied or ovate shaped Glyphidodontops and the dorsal profile is distinctive with the membranes between the spines (at least anterior ones) deeply incised. The closest relative of G. springeri is G. hemicyaneus, differing only in color pattern and dorsal spine count. The latter species generally possesses a variable amount of bright yellow color on the lower half of the body or frequently on the caudal fin and peduncle. It has the normal Glyphidodontops compliment of 13 dorsal spines (62 specimens counted at the Australian Museum, Sydney). G. hemicyaneus is known from the Ryukyu Islands, Philippine Islands, East Indies, New Guinea, and the Solomon Islands. Its distribution overlaps that of G. springeri in the Molucca Islands (based on specimens at the National Museum of Natural History, collected by V. Springer). Named springeri in honor of Dr. Victor G. Springer, Curator of Fishes, Smithsonian Institution. Dr. Springer and M. Gomon collected the holo- type and most of the paratypes. These were kindly lent to the senior author along with many other pomacentrids from the Molucca Islands. LITERATURE CITED ALLEN, G. R. 1972. The anemonefishes, their classification and biol- ogy. T.F.H. Publications, Inc., Neptune, New Jersey. 288 pp., 140 figs. 1975. Damselfishes of the South Seas. T.F.H. Publica- tions, Inc., Neptune, New Jersey. 240 pp. BLEEKER, P. 1877. Mémoire sur les Chromides marins ou Pomacen- troides de lInde archipélagique. Nat. Verh. Holl. Maatsch., Haarlen 3, Ser. 2(6):1-166. i eer irate: ‘t aint many if Vie i he Diy 7 ei Rec he ty vi 4 Reenne ene Sah hi oh ‘aa a gay Dacian: aioe MANDAN ane Cer Ml Petre ONS fe LL oe ny a bi | 5) he sks avdorth,. "PRM ee 0 te iva “e ee: (iver ay Ae eo Pe ral “ ‘ Tr iyo, wel Pa TONY Aee corn ‘ A 7 Key et wit A ‘ j . na p i * Dib all hay 7 a FUG DR Ae 4 id Tied Vy i 2 ¢ 4 i " ope a ad i , No. 32, pp. 351-354 22 January 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON can A NEW HIGH ANDEAN SLOPE SPECIES // OF ELEUTHERODACTYLUS = (AMPHIBIA: LEPTODACTYLIDAE) it Se FROM COLOMBIA AND ECUADOR sj, = Wee By Joun D. LyncH ade School of Life Sciences, The University of Nebraska, Lincoln Since redescribing Eleutherodactylus glandulosus (Boulen- ger) from the Papallacta valley in Napo Prov., Ecuador (Lynch, 1970), specimens of a closely related new species have come to my attention. Through the courtesies of Dr. William E. Duellman, Museum of Natural History, The University of Kansas (KU), and Drs. W. Ronald Heyer and George R. Zug, National Museum of Natural History, (USNM) I have exam- ined 28 specimens of the new species from the high Amazonian slopes of the Andes along the border of Colombia and Ecuador. Eleutherodactylus leucopus, new species Holotype: USNM 197927, an adult male from a series collected at Santa Barbara, Napo Prov., Ecuador, 2590 m, by R. Mullen, M. Olalla, and J. A. Peters, 22 June 1962. Paratypes: USNM 197928-36, topotypes; 21-22 June 1962. Diagnosis: (1) skin of dorsum feebly granulate, that of venter coarsely areolate; no dorsolateral folds or enlarged tubercles; (2) tympanum round, prominent, 44-14 eye diameter; (3) snout subacuminate in dorsal view, tip pointed, protruding in lateral profile; (4) upper eyelid slightly nar- rower than interorbital distance; cranial crests present, not prominent; (5) prevomerine odontophores well elevated, oval to triangular in out- line; (6) males with vocal slits and external subgular vocal sac; (7) first finger shorter than second; all digits bearing discs on broad pads; (8) fingers bearing lateral fringes; (9) small ulnar tubercles, ante- brachial large, non-conical; (10) heel bearing small, non-conical tu- bercles; inner tarsal fold ridge-like, extending %#%—-% length of tarsus; outer tarsal tubercles small, non-conical; (11) two metatarsal tubercles, inner oval, 3—4 times size of non-conical outer; plantar surface areolate; 32—Proc. Biou. Soc. WAsH., VoL. 88, 1976 (351) 352 Proceedings of the Biological Society of Washington (12) toes bearing prominent lateral fringes, no webs; toes bearing large pads and discs; (13) dark gray above, venter cream flecked with gray; fingers and toes pale cream; posterior thighs, groin dark gray; (14) adults moderate size, 6 ¢ 30.0-37.8 mm, 2 9 42.3-44.0 mm SVL. E. leucopus is most similar to and presumably closely related to E. glandulosus. E. glandulosus is smaller (6 6 21.5-26.2, 299 29.4-38.3 mm SVL); has a rounded snout in dorsal view, truncate in profile; has less prominent lateral fringing; and lacks tubercles on the heel and outer edge of tarsus. E. leucopus is markedly bicolored with a dark gray to black dorsum instead of brown to yellow-tan of E. glandulosus; E. glandulosus does not have markedly pale digits. E. glandulosus has dark brown areas in the groin and concealed portion of limbs enclosing large yellow spots—neither feature is seen in E. leucopus. I know of no other Eleutherodactylus species lacking a dorsal pattern and having the mark- edly bicolor appearance. Few species have protruding snouts (Fig. 1); among Group II Eleutherodactylus the feature also occurs in the small Amazonian species E. acuminatus Shreve and E. paululus Lynch. Description: Statements listed in the diagnosis are not repeated here unless there is some qualifying condition or variation. Head slightly narrower than body, wider than long, notably flat; head width 35.9-42.0 (x = 39.2, N= 28) per cent SVL; tip of snout extending well beyond lower jaw; snout short, E-N 72.9-102.4 (x= 90.6, N=18) per cent eye length in males, 91.7-109.5 (x= 100.5, N= 10) in females; nos- trils not or only weakly protuberant, directed laterally; canthus rostralis relatively sharp, straight; loreal region weakly concave, sloping abruptly to lips; lips not flared; interorbital space gently furrowed, furrow ex- tending anteriorly onto snout; edges of frontoparietals upturned but cranial crests not prominent; upper eyelid width 77.1-104.3 (x = 90.7, N = 28) per cent IOD; supratympanic fold prominent, obscuring upper edge of tympanum, tympanum separated from eye by distance equal one and one-half tympanic diameters; tympanum size not sexually di- morphic, its length 33.8-50.0 (x = 41.8, N = 28) per cent eye length; choanae round, relatively small, not concealed by palatal shelf of max- illary arch; prevomerine teeth and odontophores present, latter median and posterior to choanae, separated by distance equal to choanal breadth, each bearing a transverse row of 4—8 blunt teeth along posterior edge; tongue slightly longer than wide, posterior one-third not adherent to floor of mouth, posterior edge bearing shallow notch; males with long vocal slits lateral to tongue. Skin of dorsal surfaces feebly granulate (least so on head, most no- ticeable on posterior portion of body and limbs); no distinct tubercles on head except for one postrictal tubercle; body lacking ridges or folds other than supratympanic fold; throat, venter, and underside of thighs coarsely areolate; discoidal folds prominent; no enlarged subanal warts; forearms robust; two palmar tubercles, outer one smaller, neither as large as oval thenar tubercle; numerous flat supernumerary palmar tubercles; subarticular tubercles round, non-conical, simple; fingers High Andean slope frog 303 Fic. 1. Head and hand (palmar view) of Eleutherodactylus leucopus new species (USNM 197927, holotype). Line equals 10 mm for head, 20 mm for hand. bearing prominent fringes (Fig. 1), fringes present along outer edge of fingers I and IV as well as outer edge of palm; all fingers bearing discs (broader than long) on large pads; pads round, 2—214 times digit width below pad except that on thumb which is smaller than pads on fingers II-IV. Heel bearing one or two small, non-conical tubercles; two metatarsal tubercles, outer longer than wide; inner metatarsal tubercle non-com- pressed, its length twice its width; many supernumerary plantar tu- bercles (sole is areolate), all small, flat; subarticular tubercles round, non-conical, simple; toes bearing prominent lateral fringes but lacking distinct webs other than that produced by coalescing lateral fringes; hindlimbs short, heels of legs flexed at right angle to sagittal line just overlap; heel of adpressed hindlimb reaches to between posterior edge of eye and a point between eye and nostril; shank 47.5-56.7 (x = 52.5, N = 28) per cent SVL. Ground color dark gray; some creamy suffusion on flanks; venter cream with gray flecking; undersides of limbs gray with creamy blotches; posterior surfaces of thighs and groin dark gray; distal por- 304 Proceedings of the Biological Society of Washington tions of fingers and toes pale cream; no bars, bands, or dark spots on dorsal surfaces. The spots in the groin and on the posterior surfaces of the thigh are seen in USNM 197939 (JAP 4589) and 197940 (GOV 8534). The latter has a few spots on the dorsum. Neither is otherwise different from the other specimens examined. In life, E. leucopus was “jet black dorsally, with yellowish white pads and toes.” Two individuals were noted as having orange or yellow spots on the thighs. Measurements of holotype in mm: SVL 34.4; shank 17.8; head width 13.6; head length 12.6; upper eyelid width 4.0; IOD 4.4; tympanum length 1.6; eye length 4.1; E—N 3.4. Natural history: No calling males were so noted by the collectors nor were amplectant pairs found. Adults were found at night sitting on leaves and branches in heavy undergrowth in the montane forests. By day, specimens were found beneath logs and other debris in a clear- ing of good forest. Three adult females were collected. All have exten- sive convolution of the oviducts and large, yellow ovarian eggs. One adult female was taken in January and the other two in June. The largest immature female (USNM 197937) is 38.5 mm SVL. A slightly smaller female (KU 140306, 37.2 mm SVL) has moderate size eggs and weakly convoluted oviducts. All males examined appear mature (large testes, vocal slits and sacs present). E. leucopus is known from intermediate elevations (2440-2700 m) in the valley of the Rio Chingual at localities in both Colombia and Ecua- dor. In addition to the holotype and paratypes, the following were examined: COLOMBIA, Depto. Narifio: La Victoria, 2700 m, KU 140303-07, 140312. ECUADOR, Prov. Napo: Santa Barbara, USNM 197940(7); 1 km NW Santa Barbara, 2590 m, USNM 197938(2); 1 km SW Santa Barbara, 2590 m, USNM 197939(2); 3 km SW Santa Barbara, 2440 m, USNM 197937. Etymology: Greek, meaning white-footed, in reference to the col- oration. Remarks: E. glandulosus and E. leucopus seem to be geminate spe- cies. Neither has apparent close relatives on the Andean slopes or paramo and I am aware of no lowland species sufficiently similar to be construed a close relative. The two species are geographically separated by some 120 km of unexplored Andean slopes in Ecuador. LITERATURE CITED Lyncu, J. D. 1970. Identity of two Andean Eleutherodactylus with the description of a new species (Amphibia: Leptodactylidae). Jour. Herpetol. 3:135—43. 38, No. 33, pp. 355-366 22 January 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ie A NEW SPECIES OF MILVAGO FROM HISPANIOLA/ WITH NOTES ON OTHER FOSSIL CARACARAS FROM. THE WEST INDIES (AVES: FALCONIDAE) 9 & r | By Storrs L. OLson le Smithsonian Institution, Washington, D.C. 20560 | Extensive collections of vertebrate fossils from caves in\, Haiti have thus far yielded remains of two extinct birds—a ~* giant barn owl Tyto ostologa Wetmore (1922b), and a flight- less rail Nesotrochis steganinos Olson (1974). In this same ma- terial I have recently discovered a single well-preserved and diagnostic bone representing a new species of the falconid genus Milvago—a genus hitherto unknown outside of South America and Panama. Study of this new form has prompted me to investigate the two other fossil caracaras known from the West Indies, Polyborus latebrosus Wetmore (1920) of Puerto Rico and P. creightoni (Brodkorb, 1959) of the Bahamas. Within the Falconidae the caracaras form a distinctive group of vulture-like New World falcons, often accorded the status of a subfamily (Polyborinae or Daptriinae) or a tribe (Polybori). Of the other members of the Falconidae, the monotypic Argentine genus Spiziapteryx appears to be the closest relative of the caracaras. Its intermediate nature may make it difficult in the future to recognize the caracaras as more than an informal group of the Falconidae (Olson, in press ). Nearly a hundred years ago, Ridgway (1876) reviewed the systematics of the “Polybori” in an excellent and meticulous monograph. Using a wide variety of osteological and external characters, he recognized four genera of caracaras—Polyborus (Caracara auct.), Phalcobaenus (including Senex), Daptrius (including Ibycter), and Milvago. These four genera have 33—Proc. Brow. Soc. WaAsH., Vou. 88, 1976 (355) € g a | PAs)! 356 Proceedings of the Biological Society of Washington been generally recognized, but Vuilleumier (1970) presented a revised classification of the group in which only two genera, Daptrius and Polyborus, were admitted, the latter encompass- ing Phalcobaenus and Milvago. Vuilleumier appears to have overlooked Ridgway’s study. He does not consider many of the characters used by Ridgway and as a result his conclusions as to generic limits seem to have a less solid basis than do Ridgway’s. For the present I prefer to follow Ridgway in con- tinuing to recognize four genera of caracaras. Because two of these genera, Daptrius of neotropical forests and Phalco- baenus of the Andes and Patagonia, are not known in the West Indies, they will not be considered here. MitvaAco Spix The genus Milvago is composed of two very distinct species. The southern one, M. chimango, ranges from southern Brazil and Chile south to Tierra del Fuego. In M. chimango the juvenal plumage is similar to that of the adult, the tarsus is long and slender, and the skin about the orbit is feathered. The other species, M. chimachima, is found from Panama and northern and eastern South America east of the Andes, south to northern Argentina, sympatric in the southern parts of its range with M. chimango. In M. chimachima the adult and juvenal plumages are very different, the tarsus is short, and there is a considerable area of bare skin around the orbit. I agree with Vuilleumier (1970:17) that it is inappropriate to consider these two quite different and partially sympatric species as members of a superspecies, as was suggested by Brown and Amadon (1968). To the two living species we may now add a third, equally distinctive, extinct Antillean form. Milvago alexandri, new species Figure 1 Holotype: Nearly complete right tarsometatarsus, lacking part of the hypotarsal crest and part of the posterior process of the external cotyla, > Fic. 1. Milvago alexandri, new species, holotype tarsometatarsus, USNM 214573: a, anterior view; b, lateral view; c, posterior view; d, medial view; e, proximal view; f, distal view; a—d stereo pairs, natural size; e and f, twice natural size. < S S SS S ~~ Ss S as <= — > wm NN) = — == a) is) Ry 358 Proceedings of the Biological Society of Washington TaBLE 1. Length of tarsometatarsus (mm; from skeletons) in three species of Milvago. n range mean sD M. chimachima 8 49.7-54.5 52.3 1.49 M. alexandri il —— 56.2 — M. chimango 9 57.8-63.9 60.1 1.88 USNM 214573 (Fig. 1). Collected in Pleistocene cave deposits (Cave Tin Miller’s notes) near St. Michel de L’Atalaye, Département de L’Arti- bonite, Haiti, Hispaniola, in 1925 by Gerrit S. Miller. The specimen is orange buff in color and was formerly encrusted with reddish limey matrix and old shellac, thus being identical in preservation and prepara- tion to specimens from the older layers of the St. Michel caves. Measurements of holotype: Overall length 56.2 mm, proximal width through anterior portions of cotylae 8.2, depth of internal cotyla 4.3, least width of shaft 3.4, least depth of shaft 2.6, distal width 9.1, depth through arc of trochleae 6.1, width of middle trochlea 3.4. Etymology: I take great pleasure in dedicating this interesting new species to my close friend and associate Alexander Wetmore, in recog- nition of his many contributions to avian paleontology and to West Indian ornithology, particularly that of Hispaniola. Diagnosis and description: The holotype of Milvago alexandri is at once separable from Polyborus and Phalcobaenus by its much smaller size. From Daptrius it may be distinguished by its slender proportions, particularly the narrower proximal end, and by the narrower trochleae, less excavated posterior surface of the proximal end of the shaft and the lower and more elongate tubercle for the tibialis anticus. In size and proportions the tarsometatarsus of M. alexandri is dis- tinctive in being intermediate between the two living species, M. chi- mango and M. chimachima. In the last two there is no overlap in the length of the tarsometatarsus of the available skeletal specimens (Table 1). M. alexandri differs further from either of the living species as follows: the proximal foramina are larger and more widely spaced; in anterior view the external edge of the shaft bears a more distinct ridge, about as well developed as that of the internal edge, with the result that the anterior face of the shaft appears more deeply excavated; in external view the proximal portion of the external ridge of the shaft is thicker and the papilla for the tendinal attachment on it is better de- veloped; the external trochlea is smaller, projects straight posteriorly rather than curving inward, breaks away more abruptly from the shaft, and has a noticeably deeper pit on the external face; the middle trochlea is proportionately smaller, with a deeper posterior groove. In the following respects, M. alexandri differs from M. chimango and more closely resembles M. chimachima: shaft more robust; ridge of Fossil West Indian falcon 359 hypotarsus extending farther down the shaft; internal cotyla in proximal view not as deep and with posterior border less rounded; wing of inner trochlea much less pronounced and not as distinctly set off from the body of the trochlea. Remarks: Morphologically and geographically, Milvago alexandri is closest to M. chimachima. There are two possible explanations for its having a longer tarsus than that species. Hither M. alexandri was de- rived from M. chimachima stock before the two continental species had diverged greatly in tarsal length, or the longer tarsus of M. alexandri evolved from the shorter one of M. chimachima after the colonization of Hispaniola. Vuilleumier (1970) suggested that a shorter tarsus in caracaras may be correlated with more arboreal habits and noted that the short-legged M. chimachima seems to prefer more wooded habitats than the long-legged M. chimango. Both species prefer open rather than than densely wooded terrain, and this is about all that can be inferred of the habitat of M. alexandri. The discovery of M. alexandri adds a hitherto unknown element to the avifauna of the West Indies. Unless some form of Milvago once extended farther north in Central America or through the Lesser Antilles, it must be assumed that M. alexandri was derived directly from northern South America. Although it is probable that the flightless rail Nesotrochis steganinos was exterminated by introduced predators, and that the giant barn owl Tyto ostologa vanished with the extinction of the large rodents that formed its principal prey, it is not so easy to account for the disappear- ance of Milvago alexandri. Bond (1974) has indicated that Hispaniola is the least known, ornithologically, of the Greater Antilles, thus it is remotely possible that this bird might yet survive there. POLYBORUS VIEILLOT All the extant forms of the genus Polyborus are now con- sidered to belong to a single species, P. plancus, which ranges from the southern United States to Tierra del Fuego. The pop- ulations of southern South America, P. p. plancus, differ from the northern forms in their barred rather than uniformly black dorsum and in their finer ventral barring. The northern forms range from northern South America through Central America to northern Baja California, southern Arizona, and southern Texas. Disjunct populations, probably relicts, are found in south Florida, Cuba and the Isle of Pines, but are not distin- guishable from those to the west. The northern forms have at times been divided into three subspecies(cheriway, audu- boni, and ammophilus) and were long regarded as a species (cheriway) distinct from plancus. Hellmayr and Conover 360 Proceedings of the Biological Society of Washington (1949) demonstrated that the two “species” intergrade in Brazil and regarded them as races of a single species. For the purposes of the following discussion, the three northern forms will be referred to collectively as Polyborus plancus cheriway. The population of the Tres Marias Islands, Mexico, is distinctly smaller and somewhat paler than the mainland forms and is recognized as a separate subspecies, P. ». pallidus (Grant, 1965). Of greater interest is the extinct form P. lutosus from Guadelupe Island off Baja California, which in its plum- age, and to a lesser extent the shape of the rostrum, is quite distinct. Vuilleumier (1970) regarded P. lutosus as a sub- species of P. plancus, but most authors have accorded it full specific rank. What is most interesting about P. lutosus is that its plumage pattern is more similar to the South American race P. p. plancus than to the race of the adjacent mainland. Wetmore (1920, 1922a) described a new species of caracara, Polyborus latebrosus, from Pleistocene cave deposits in Puerto Rico, based on the proximal end of a right carpometacarpus (holotype) and the proximal end of a right ulna. It was said to be intermediate between P. cheriway and P. plancus (then considered separate species) but distinct from either. Howard (1938) analyzed the abundant caracara remains from the Pleistocene tar pits at Rancho la Brea in southern California and concluded that they represented an undescribed species which she named Polyborus prelutosus. In her studies, P. plancus and P. cheriway were regarded as distinct species and P. prelutosus, like P. latebrosus, was said to exhibit char- acters intermediate between the two forms and others in which it was more similar to P. plancus. Howard concluded that P. plancus, P. lutosus, P. prelutosus, and P. latebrosus, were more primitive and more closely related to each other than any were to P. cheriway, which was presumed to have replaced the more primitive forms in the areas where it presently oc- curs. I am basically in agreement with this conclusion, my main reservation being the level at which the taxa are to be recognized. Howard stated (1938:239) that P. prelutosus un- doubtedly “represents a form not far removed from, if not actually the ancestral type of, the Recent caracaras,’ but she rejected the idea of its being a temporal subspecies on the Fossil West Indian falcon 361 grounds that temporal variation “cannot be evaluated in the same manner as intergradation of contemporaneous geographic races. Pleistocene remains of caracaras from New Mexico and Florida, previously referred to P. cheriway, were reiden- tified as P. prelutosus by Howard (1938). Records of P. pre- lutosus from other localities, including Texas, are listed by Brodkorb (1964). A series of specimens of P. prelutosus from San Josecito Cave, Nuevo Leon, Mexico, was found to aver- age smaller than the series from Rancho La Brea, some speci- mens being smaller than the minimum for that locality. Al- though there was broad overlap in measurements, Howard (1940) designated the Mexican bones as a new subspecies, Polyborus prelutosus grinnelli. From a Pleistocene deposit on New Providence Island, Ba- hamas, Brodkorb (1959) described yet another species, Cara- cara (= Polyborus) creightoni. This was based entirely on a fragmentary distal portion of a left carpometacarpus consist- ing mostly of the second metacarpal. I have studied the type-material of Polyborus latebrosus and P. creightoni. Comparisons were made with 7 skeletons of P. plancus plancus, 11 of P. p. cheriway, 1 of P. lutosus, and 10 carpometacarpi of P. prelutosus. It must be emphasized that within the modern and fossil series there is a very great deal of individual variation in size, in proportions, and in the conformation of minor features of the bones. The holotype carpometacarpus of P. latebrosus was said to be slightly larger than in P. p. cheriway (Wetmore, 1922a). As shown by Howard (1938), who also studied the type, the bone is near the maximum or slightly larger than in P. p. cheriway but falls well within the range of P. p. plancus and P. prelutosus. The referred ulna of P. latebrosus was described by Wetmore (1922a) as being larger than in P. cheriway but smaller than in P. plancus. In reality, however, it is matched only by smaller specimens of P. cheriway. In view of the seeming discrepancy between the size of the ulna and that of the carpometacarpus of P. latebrosus, it may be worth not- ing that Howard (1938) found the carpometacarpus of P. prelutosus to be proportionately larger than would be expected from the size of the other elements. The supposed distinctions 362 Proceedings of the Biological Society of Washington in the type of P. latebrosus are very minor features, most of which are highly variable in modern forms. The variation in the shape of the first metacarpal, for instance, is so great in the comparative series that given the same level of criteria as used to differentiate P. latebrosus it would be possible to name nearly each specimen as a different species. In her compari- sons, Howard (1938:225) commented that “latebrosus reveals marked overlapping similarities” with P. prelutosus. The only distinguishing feature of P. latebrosus was said to be in the “unusual position of the notch between the process of meta- carpal 1 and the trochlea,” which was said to be at the junc- tion of those two features rather than anterior to that junc- tion. This was deemed sufficient to separate “latebrosus from the Rancho La Brea bird as well as from all the other species” (Howard 1938:225). Wetmore (1922a:305) also mentions a “notch at the outer margin of the proximal condyle [being] lower so that it is found at the line of ankylosis of the first and second metacarpals instead of above that point.” Even after prolonged scrutiny of the type, the feature was not apparent to me and the distinction would seem to be a very minor one. It is difficult not to doubt the validity of Polyborus creigh- toni solely on the basis of its being based on such a fragmen- tary specimen almost wholly lacking in diagnostic features. Nevertheless, it cannot, in fact, be summarily dismissed. Be- cause the type is so imperfect, it occurred to me that it might be from some bird quite unrelated to Polyborus, but on com- parison no more suitable placement for it could be found. In contrast to the type of P. latebrosus, the type of P. creightoni is small. Although it falls within the size range of P. p. cheri- way, only one of the specimens examined was smaller, this being number 28302 in Dr. Brodkorb’s collection (distance from distal end of metacarpal II to metacarpal I = 42.9 mm as opposed to 43.8 mm in P. creightoni). In two of the char- acters given by Brodkorb (1959), P. creightoni does appear to be distinct; the other differences cited do not stand up on comparison with the series of Polyborus at hand. These two characters are: “metacarpal III with its base nearly straight, without medial angulation proximal to intermetacarpal tu- berosity; tuberosity of metacarpal II, in medial view, with its Fossil West Indian falcon 363 outline more angular and less rounded, and its base more deeply excavated dorsally and laterally” (Brodkorb, 1959: 303). It is difficult to determine just how much consideration should be given to such minor differences. The only fossil form of Polyborus for which there is ade- quate material to make any determination of its status is P. prelutosus. Howard (1938) has demonstrated that this form is not identical with its geographically nearest modern con- geners. Yet when P. cheriway and P. plancus are combined, virtually all of the variation seen in P. prelutosus falls within the range of the expanded species P. plancus. It appears that only in its smaller and proportionately wider bill is P. pre- lutosus recognizably different from P. plancus and in this respect it resembles P. lutosus of Guadelupe. I do not consider that an adequate basis exists for regard- ing any of the named fossil forms of Polyborus as being spe- cifically distinct from P. plancus, particularly given the great variability of that species. P. prelutosus is almost surely the temporal equivalent of, and on a direct genetic line with, P. plancus. The present status of the West Indian forms P. late- brosus and P. creightoni seems unsatisfactory when one rec- ognizes that both geographic as well as temporal variation may be involved. I believe it is best to regard all the known fossils of Polyborus as belonging to the species plancus. Within that species, prelutosus is a valid temporal form. The names creightoni and latebrosus are available if further material should indicate that either of the populations they represent are subspecifically distinct, although the possibility exists that such material might show latebrosus to be inseparable from prelutosus. ACKNOWLEDGMENTS For the loan of types and comparative material I am greatly indebted to Malcolm C. McKenna, American Museum of Nat- ural History; S. David Webb and Pierce Brodkorb, University of Florida; Robert McKenzie and David P. Whistler, Natural History Museum of Los Angeles County. I am likewise grate- ful to G. B. Sullivan for his preparation of the type of M. alexandri and to Victor E. Krantz for photographing it. For their useful comments on the manuscript I thank Isabel Canet, 364 Proceedings of the Biological Society of Washington Robert J. Emry, John Farrand, Jr., Larry D. Martin, Clayton E. Ray, and Alexander Wetmore. RESUMEN Se describe una nueva especie de caraira (Falconidae), Milvago alexandri, encontrada en depositos pleistocenicos de la Espanola. El género Milvago hasta ahora se conocia solo de América del Sur y Panama, por tanto la especie M. alex- andri representa un elemento completamente nuevo para la fauna antillana. Otras dos especies fdésiles de carairas antil- lanas, Polyborus latebrosus Wetmore de Puerto Rico y P. creightoni (Brodkorb) de las Bahamas, no fueron descritas en base de restos suficientes de modo que puedan ser dis- tinguidas de la especie viviente P. plancus. Aqui consideramos que pueden ser solo subespecies de P. plancus. LITERATURE CITED Bonp, J. 1974. Nineteenth supplement to the Check-list of Birds of the West Indies (1956). Acad. Nat. Sci. Philadelphia. Bropxors, P. 1959. Pleistocene birds from New Providence Island, Bahamas. Bull. Fla. State Mus. Biol. Sci. 4:349-371. ——. 1964. Catalogue of fossil birds: Part 2 (Anseriformes through Galliformes). Bull. Fla. State Mus. Biol. Sci. 8:195-335. Brown, L. anp D. AMapon. 1968. Eagles, hawks and falcons of the world. New York, McGraw-Hill. 2 vols. Grant, P. R. 1965. A systematic study of the terrestrial birds of Tres Marias Islands, Mexico. Postilla 90:1—106. HeiiMayr, C. E. anp B. Conover. 1949. Catalogue of birds of the Americas. Zool. Ser. Field Mus. Nat. Hist. 13 (Pt. 1, No. 4): 1-358. Howarp, H. 1938. The Rancho La Brea caracara: a new species. Carnegie Inst. Wash. Publ. 487:217—240. ——. 1940. A new race of caracara from the Pleistocene of Mexico. Condor 42:41—44. Otson, S. L. 1974. A new species of Nesotrochis from Hispaniola, with notes on other fossil rails from the West Indies (Aves: Rallidae). Proc. Biol. Soc. Wash. 87:439—450. —. in press. The affinities of the falconid genus Spiziapteryx. Auk. Ripcway, R. 1876. Studies of the American Falconidae. Monograph of the Polybori. Bull. Geol. Geogr. Surv. Terr. Ser. 2, 6:451- 473. VurLLEumMier, F. 1970. Generic relations and speciation patterns in the caracaras (Aves: Falconidae). Breviora 355:1-29. Fossil West Indian falcon 365 1920. Five new species of birds from cave deposits WETMoreE, A. in Porto Rico. Proc. Biol. Soc. Wash. 33:77—-82. Bird remains from the caves of Porto Rico. Bull. ——. 1922a. Amer. Mus. Nat. Hist. 46:297-333. ———. 1922b. Remains of birds from caves in the Republic of Haiti Smithson. Misc. Collect. 74(4):1—4. 366 Proceedings of the Biological Society of Washington , No. 34, pp. 367-372 22 January 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW SPECIES OF SPHAERODACTYLUS (SAURIA: GEKKONIDAE) FROM HISPANIOLA By ALBERT SCHWARTZ AND RICHARD FRANZ Miami-Dade Community College North, Miami, Florida 33167, and Florida State Museum, University of Florida, Gainesville, Florida 32611 Among the Hispaniolan species of Sphaerodactylus, Sphaero- dactylus copei Steindachner is one of the most distinctive. It reaches a snout-vent length of 42 mm; has very much en- larged and almost boss-like dorsal scales in contrast to much smaller, smooth ventral scales; a middorsal zone of tiny gran- ules; and four supralabial scales to the center of the eye. The females are brightly patterned in contrast to generally more drab and unpatterned males (Schwartz, 1975). Nine sub- species of S. copei are now recognized. The species is re- stricted to the Haitian Tiburon Peninsula and to Ile de le Gonave, Ile Grande Caymite, and Ile-a-Vache. S. copei has never been recorded from the Republica Dominicana (where it is expected in the Valle de Neiba, the eastern extension of the intermontane arid Cul de Sac Plain) despite having been taken within a very few kilometers of the Dominico-Haitian border (near Fond Parisien in the Cul de Sac Plain, Haiti). Although widely distributed on the Tiburon Peninsula, large areas still remain whence S. copei is unknown. The Peninsula de Barahona and the associated southern foot- hills of the Sierra de Baoruco and the Massif de la Selle which border the xeric lowlands of the peninsula in the Republica Dominicana on the north have been revealing an interesting array of herpetological species which are limited to, or have their centers of distribution in, this arid region. This is surely 34—Proc. Biou. Soc. Wasu., Vou. 88, 1976 (367 ) 368 Proceedings of the Biological Society of Washington in part due to the fact that these lowlands are completely cut off in the north from other arid areas by the very mesic Sierra de Baoruco which reaches the shore along the east coast of the peninsula. Thus, animals on the Peninsula de Barahona are “trapped” in and severed from their more northern rela- tives by the intervening massif of the Sierra de Baoruco; this “Barahona Entrapment” has played a significant role in the evolution of the amphibians and reptiles of the Peninsula de Barahona, where such species as Eleutherodactylus alcoae Schwartz, Leptotyphlops pyrites Thomas, Typhlops syntherus Thomas, Anolis longitibialis Noble and Ameiva leberi Schwartz and Klinikowski are localized. A host of other reptiles, includ- ing amphisbaenids, lizards, and snakes, have subspecies which are endemic to this region. The Sphaerodactylus of the Peninsula de Barahona are poorly known. S. difficilis randi Shreve has been collected at Oviedo and its vicinity near the eastern coast of the penin- sula, and at and near Pedernales on the Dominico-Haitian border on the west. A second unnamed species, with primarily Haitian distribution, crosses the border and occurs at Peder- nales and in the Sierra de Baoruco at low to intermediate elevations. The high uplands of the Sierra de Baoruco are occupied by S. armstrongi Noble and Hassler. No Sphaero- dactylus are known from the lowlands of the Peninsula de Barahona between Oviedo in the east and Pedernales in the west, although the 65 kilometer distance between these two extremes seems eminently suitable for members of this genus. In the east, the peninsula was originally covered with fairly high canopied xeric forest on a rocky substrate; this forest was virtually destroyed by Hurricane Inez in 1966. Farther west, at about mid-peninsula, the road gradually ascends a forested limestone ridge; its western escarpment is steep and borders the very hot and dry plain on which Pedernales is situated in the westernmost corner. Much of the latter plain, as well as the limestone ridge, is rocky or rock-strewn and provides a suitable habitat for Sphaerodactylus. These same conditions persist onto the southern slopes of the Massif de la Selle above Pedernales and the Sierra de Baoruco above Cabo Rojo, but in both cases, despite often intensive search Gecko from Hispaniola 369 and the taking of other cryptic or burrowing reptiles, no sphaerodactyls have been secured. Between 20 and 24 March 1974, the junior author, with Sylvia Scudder and Fred G. Thompson of the Florida state Museum, collected in the region between Cabo Rojo and Pedernales. Their collections include two specimens of a new species of Sphaerodactylus, one of which is from the xeric lowlands of the Peninsula de Barahona and the other from low elevations on the southern Sierra de Baoruco slopes. This new species is allied to the Haitian S. copei; the latter species (subspecies enochrus Schwartz and Thomas) occurs in south- eastern Haiti from southwest of Jacmel and the Vallée de Trouin, east to the vicinity of Marigot, about 60 kilometers west of the Dominico-Haitian border. Thus there is a hiatus of about 75 kilometers between the known distributions of S. copei and the new species; since the extreme southeastern portion of Haiti remains unknown as far as sphaerodactyls are concerned, it is likely that one or the other species will be collected at Saltrou, Grand Gosier, or Anse-a-Pitre, all along the southeastern Haitian littoral. Although the new species is allied to S. copei, it differs in several trenchant characters and we have no doubt that it represents still another Peninsula de Barahona endemic, derived from more western S. copei. For this new species, in honor of Dr. Thompson whose efforts were responsible for securing the two known specimens, we propose the name Sphaerodactylus thompsoni, new species Holotype: UF/FSM (Florida State Museum, University of Florida) 21555, adult female, from 6 km SW Las Mercedes, 60 m, Pedernales Province, Republica Dominicana, taken 21 March 1974 by Fred G. Thompson. Original number FGT 1777. Paratype: UF/FSM 21556, 11 km N, 2 km SE Cabo Rojo, Pedernales Province, Reptblica Dominicana, 23 March 1974, F. G. Thompson. Diagnosis: A large species of Sphaerodactylus (snout-vent length 33 mm) with large keeled boss-like dorsal scales (17-18 between axilla and groin); small cycloid smooth ventral scales (32-33 between axilla and groin); no middorsal rows of small granular scales; females (only sex known) pale drab grayish above with complex head pattern (see illustrations ), without dark collar and included ocelli or dark body bands with included ocelli, but with (in juveniles and very vaguely indicated in adults) three fine transverse black lines between axilla and groin, 370 Proceedings of the Biological Society of Washington Fic. 1. Sphaerodactylus thompsoni, holotype, UF/FSM 21555, dorsal view. anteriormost of these broadest and more or less attached to complex drak cephalic pattern. Description of holotype: Dorsum (as preserved; not strongly dif- ferent from life condition) very pale grayish or tan with dark brown head and body markings as follows. Head with median dark attenuate triangle on snout, its base anterior to eyes; dark canthal line extending through eye and auricular opening, with dark spot following it; dark marbling on upper lip, including dark spot below auricular opening; a transversely elongate hollow rectangle, its anterior side between pos- terior portions of eyes, its posterior margin at about level of mid-temple; another vague dark line connecting anterior corners of eyes; upper sur- face of neck with five brown spots (two lateral pairs and a median elongate blotch) followed by two transversely elongate markings, these in turn by an incomplete dark line across neck (the last three features all more or less united to and by dark dorso-lateral line which is the continuation of canthal-postocular line and blotch); very irregular and broken transverse line above forelimb insertions; remainder of dorsum with scattered dark scales that do not form a recognizable pattern; upper surfaces of all limbs and tail dotted with dark brown scales, those of tail forming a series of vaguely delimited rings which continue irreg- ularly on underside of tail and are more complete distally; throat vaguely stippled with dark gray. Measurements and counts from holotype are: snout-vent length 33 mm; 17 large keeled boss-like dorsal scales between axilla and groin, 38 scales around body at midbody, 11 fourth toe la- mellae, 2 internasal scales; 4/4 supralabial scales to mid-eye. Variation: The paratype is a juvenile with snout-vent length of 19 mm, 17 dorsal and 38 ventral scales between axilla and groin, 38 scales around body at midbody, 10 fourth toe lamellae, 1 internasal, and 4/4 supralabials to mid-eye. Head pattern much as described for holotype except pre- and post-forelimb dorsal crossbands more clearly delimited, less fragmented and darker than all other markings on body, which in holotype speckled with dark scattered scales. Distinct dark crossband above hindlimb insertion. Tail has six pale grayish rings (becoming white distally) alternating with dark gray rings (becoming more com- plete ventrally distally); these dark rings somewhat paler centrally to Gecko from Hispaniola 371 Fic. 2. Sphaerodactylus thompsoni, holotype, UF/FSM 21555, lat- eral view of head. give series of dark gray edges bordering pale gray to white rings; tail tip black; some fine dark stippling on throat. Comparisons: Comparisons of S. thompsoni wih any species other than S. copei are unnecessary; the large boss-like and keeled dorsal scales immediately distinguish S. thompsoni from all other species in Hispaniola. From S. copei, the absence of a middorsal zone of small scales or gran- ules distinguishes S. thompsoni. It is in pattern (and probably in col- oration in life) also that the two species are radically different. Thomas (1968:fig. 2d) showed a dorsal view of S. c. enochrus. The head in that subspecies includes three pale lines on an otherwise dark brown ground, followed by two rows of subcircular or oval pale dots, followed by a dark collar with included pale ocelli. The body has two or three (usually two) dark crossbands with included pale ocelli in juveniles, but this pattern becomes, in adult females, a random series of dark and pale scattered scales on a medium dark (brown) ground. Considering the illustrations of the other subspecies (Thomas, 1968) as well as the verbal descriptions of subspecies subsequently named (Schwartz, 1975), one can see that no other population of S. copei approaches S. thomp- soni in complete reduction of dark collar and included ocelli (although the ocelli may be present but not in a dark collar); in addition, the fine dark isolated juvenile crossbands in S. thompsoni differ strongly from the broad crossbands with their included ocelli in S. copei. None of the meristic characteristics of S. thompsoni differentiates the species from S. copei enochrus. In 33 specimens of S. c. enochrus examined, the largest female has a snout-vent length of 39 mm, larger than the single adult female S. thompsoni. Dorsals in axilla to groin distance are 13-18 in S. c. enochrus, 17-18 in S. thompsoni; ventrals in the same distance are 23-31 in S. c. enochrus, 32-33 in S. thompsoni; 372 Proceedings of the Biological Society of Washington midbody scales in S. c. enochrus are 38-45 and 38 in both specimens of S. thompsoni. No S. c. enochrus has 2 internasals (although this con- dition occurs in two subspecies (picturatus Garman, websteri Schwartz), whereas one of two S. thompsoni has 2 internasals. Both taxa normally have 4 supralabials to the center of the eye. Remarks: The holotype was taken under rocks near the top of a small talus cone, about 12 feet (3.7 m) above the base. The cone consisted of loosely packed limestone rocks up to 2 feet (0.6 m) in diameter and covered lightly with dry leaves. The cone was located against one wall of a rather steep-sided ravine or wash, the floor of which was made up of fine sediments and some cobble. The opposite wall consisted of ledges of limestone with shallow dry caves occupied by Cyclura. The vegetational cover was sub-mesic in the ravine with xeric conditions (Agave, Acacia, cacti) on the upper slopes and surrounding hillsides. There are no further data on the collection site of the paratype. We assume that S. thompsoni is widely distributed on the Peninsula de Barahona. The habitat noted above is widespread and rocky and talus cover are commonplace throughout the area. It is truly remarkable, de- spite intensive collecting in this region since 1963, that the species has not been previously collected. Since Dr. Thompson was collecting living snails in this xeric region, he made a point of disturbing and tearing apart talus and rubble piles; this activity may have been responsible for taking of the two specimens, since mere rock turning has repeatedly yielded no sphaerodactylus on the peninsula. In those areas where other species of Sphaerodactylus have been locally taken (Oviedo; Pedernales; and their vicinities) the ecology is regularly less harsh, more shaded, and somewhat more mesic than in those rigorous localities whence S. thompsoni has been collected. ACKNOWLEDGMENTS We wish to thank Dr. Fred G. Thompson for allowing us to describe this distinctive new species, and Nancy Halliday, the staff artist at the Florida State Museum, for her beautiful illustrative work. Comparative specimens of S. c. enochrus in the collection of the senior author were taken through the efforts of James W. Norton, William W. Sommer and Richard Thomas. LITERATURE CITED Tuomas, R. 1968. Notes on Antillean geckos (Sphaerodactylus). Herpetologica 24(1):46-60, Figs. 1-4. Scuwartz, A. 1975. New subspecies of Sphaerodactylus copei Stein- dachner (Sauria, Gekkonidae) from Hispaniola. Herpeto- logica 31(1):1-18. Rion * ‘ at ined | inde’ Cos + 10) i bg oar 38, No. 35, pp. 373-382 22 January 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON TWO NEW DIPLOPODS OF THE GENUS // POLYZONIUM FROM NORTH CAROLINA, //_ WITH RECORDS OF ESTABLISHED SPECIES | = (POLYZONIIDA: POLYZONIIDAE) By RowLanp M. SHELLEY i = North Carolina State Museum, P. O. Box 27647, \¢ f Raleigh, North Carolina 27611 \ Millipeds of the genus Polyzonium are common in the litter fauna of eastern North America. The small, yellowish animals are easily recognized as Polyzonium, but until the recent summary by Loomis (1971), assignment to species had been on an indiscriminate basis. Loomis recognized eight species in North America, two of which, P. bivirgatum and P. rosalbum, have been reported from several North Carolina localities (Causey, 1940; Wray, 1967; Loomis, 1944, 1971). Unfortunately, these records are rather confused, especially for the piedmont, and in a few cases both names appear to have been assigned to the same material. In addition, the polyzoniid fauna of North Carolina is considerably more com- plex than previously thought, with at least two additional species, both undescribed, occuring in the state. Reviewing the published North Carolina records in light of the criteria established by Loomis (1971), it becomes readily apparent that a number must be discounted until con- firmed by fresh material. Some listings are based on rep- resentatives of the two new species, while others cannot be verified due to absence of the specimens. For all practical purposes, it is necessary to discard the previous North Carolina records of Polyzonium and start over. Furthermore, an ad- ditional barrier exists to resolution of this problem—the iden- tity of Buotus carolinus, proposed by Chamberlin (1940) for an immature female polyzoniid from Duke Forest. The type 35—Proc. Bion. Soc. WAsH., VoL. 88, 1976 (373) 374 Proceedings of the Biological Society of Washington and only known specimen was deposited in the Chamberlin collection and is currently unavailable. The description is vague and unillustrated but suggests that the specimen might be of the genus Polyzonium. Establishing the specific identity of B. carolinus will be difficult because the type specimen lacks gonopods, but it may eventually be shown to be an undescribed species of Polyzonium in which the distal podomere* of the anterior gonopods is straight and not apically curved. For this form, the dominant polyzoniid in the eastern piedmont, I propose the name Polyzonium strictum. If B. carolinus should prove to be this form, its specific name will have priority and that of the former will become a junior synonym. Description of this form has been deferred because of the Buotus situa- tion, but a name must now be established for use in future studies on North Carolina diplopods. A series of papers is eventually planned on the milliped genera of North Carolina. Many regions remain to be col- lected, however, and these studies are several years from com- pletion. In view of the current confusion on Polyzonium, it seems that an interim report on the genus is desirable, in order to identify the species known from the state and to re- port authentic localities. This paper is submitted for that purpose. Descriptions and illustrations of the two new species are included along with figures of the anterior gonopods of P. bivirgatum and P. rosalbum. Individuals are most accurately identified by comparing the anterior gonopods of males with these drawings. Hopefully, this paper will clarify the status of North Carolina polyzoniids until a more thorough study can be prepared. Lines of affinity in the genus Polyzonium are virtually im- possible to determine due to our poor knowledge of most of the species. Loomis (1971) did not discuss relationships in his paper, which is the best available on the American forms. Thus, only very tentative conclusions can be drawn about the relationships of the two new species described herein. A real monograph of the genus, addressing problems of phylogeny, is clearly to be desired. * The term “distal podomere” is used in this paper in the same sense as “seminal joint’”’ by Loomis (1971). Diplopods from North Carolina 375 I am grateful to H. F. Loomis for his valuable advice and to Richard L. Hoffman for the loan of his Polyzonium col- lection. Material from Hanging Rock, Morrow Mountain, and William B. Umstead State Parks was collected with permis- sion of the North Carolina Department of Natural and Eco- nomic Resources, Division of State Parks. The specimens listed below are deposited in either the personal collection of Richard L. Hoffman (RLH) or the invertebrate collection of the North Carolina State Museum (NCSM). Polyzonium strictum, new species Figures 1-4 Type-specimens: Male holotype (NCSM 1881) and 7 male and 6 female paratypes collected by the author, 25 July 1973, from Wilkes County, North Carolina, 4.4 miles NE McGrady, along county road 1730, 0.2 miles N junction county road 1728. Paratypes deposited in the U.S. National Museum; the Florida Collection of Arthropods, Gaines- ville; and the Zoologische Museum Amsterdam. Diagnosis: Distal podomere of anterior gonopods apically straight, blunt, not curved mediad or laterad; sternal lobes low, moderately sep- arated, base and tip subequal in width. Similar in size, general appear- ance, and sternal lobes of anterior gonopods to P. rosalbum, with which it may occur sympatrically (probably not syntopically), but differing in configuration of distal podomere of anterior gonopods and in much thinner first two legs. Holotype: Length 9.5 mm, width 2.0 mm, 46 segments. Color yel- lowish, lighter on edges, posterior segments, and venter; dorsum with two rows of slightly darker, paramedian spots, one pair per segment, running entire length of specimen. Collum large, hood-like, completely covering ocelli and first two antennomeres. Sides of head slightly con- cave, meeting in front in angle of about 75°; antennae extending well beyond sides of body when held laterally; ocelli 3-3, equidistant, sec- -ond ocelli larger than others, anterior interocular space subequal to diameter of ocellus, posterior interocular space 4—5 times as wide. Pores of segment 5 (Fig. 2) lower than those that follow, sides raised, in- terzonal stria only slightly bulging behind pores. Ultimate segment evenly rounded. Anterior gonopods (Fig. 1) with sternal lobes short and broad, mod- erately separated; coxal lobe broad, slightly curved mediad at midlength but not apically, broad and rounded at tip; distal podomere slightly bulging at base, slightly curved mediad at midlength, but apex straight and blunt, not curved mediad or laterad. Legs 1 and 2 (Figs. 3, 4) smaller and thinner than succeeding ones, leg 3 slightly thickened. Sterna on legs 1 and 2 low and narrow, coxae almost touching; coxae of leg 3 more separated; sterna on other pregonopodal legs broad and low, 376 Proceedings of the Biological Society of Washington coxae well separated; sterna on postgonopodal legs low and narrow, coxae almost touching. Variation: The male paratypes agree with the holotype but have fewer segments. The female paratypes are generally darker, with brown speckles on the anterior segments. The collum of the females appears smaller than that of the males, and all sterna of the females are low and narrow, with the coxae almost touching, On the other males, the size of the collum varies, and on some individuals it does not completely cover the ocelli. All mature individuals have three pairs of ocelli; one immature male from Durham Co. has two pairs. The gonopods are quite consistent, with the distal podomere always being straight, never apically curved. The size of the bulge at the base of the distal podomere varies in relation to the angle at which the distal podomere is directed. This structure is not always perpendicular but may lean slightly mediad or laterad, and those individuals in which the distal podomere leans laterad have a greater basal bulge. Ecology: Polyzonium strictum is usually encountered in moist humus under or near rocks or logs in predominantly hardwood regions. It may occasionally be found under the bark of decaying pine logs. Distribution: The species is known from Virginia and North Carolina, where it ranges from the mountains to the inner coastal plain. It has not yet been collected in Tennessee, South Carolina, or most of the coastal plain of North Carolina, but this absence is undoubtedly due to lack of investigations in these areas. Specimens have been examined as follows: Virginia: Pulaski Co., Draper Mtn. above Pulaski, 13, 29, 4 Octo- ber 1959, R. L. Hoffman and R. E. Crabill (RLH). North Carolina: Madison Co., ravine along Southern railroad be- tween Hot Springs and Paint Rock, 14, 2 August 1962, R. L. Hoffman (RLH). Ashe Co., 7.2 miles SE Jefferson, 16, 692, 21 June 1972, R. M. Shelley (NCSM 1200). Wilkes Co., 4.4 miles NE McGrady along co. rd. 1730, 0.2 miles N jet. co. rd. 1728, 84, 69, 25 July 1973, R. M. Shelley (NCSM 1881) TYPE LOCALITY; and 9 miles SE Wilkesboro, 23,29, 26 July 1973, R. M. Shelley (NCSM 1827). Stokes Co., Hang- ing Rock State Park, 26, 49, 30 May 1973, R. M. Shelley (NCSM 1800). Stanly Co., 3 miles SE Albemarle, 44, 19, 19 October 1952, L. Hubricht (RLH); and Morrow Mtn. State Park, 16, 42, 9 August 1973, R. M. Shelley (NCSM 1901). Montgomery Co., 12 miles NW Troy, 2, 82, 19 July 1952, R. M. Shelley (NCSM 1175). Randolph Co., 7.2 miles NW Asheboro, 34, 15 juvs., 11 May 1972, R. M. Shelley (NCSM 1047). Caswell Co., 2.4 miles SE Yanceyville, 1¢, 23 May 1973, R. M. Shelley (NCSM 1779). Vance Co., 4.6 miles N Henderson, 16, 22, 25 August 1972, R. M. Shelley (NCSM 1500). Durham Co., 14.1 miles N Durham, 54, 129, 11 August 1972, R. M. Shelley (NCSM 1470). Orange Co., 1.5 miles SSW Hillsborough, 26, 29, 2 August 1973, R. M. Shelley (NCSM 1886). Wake Co., William B. Umstead State Park, 4¢, 99, 11 October 1971, R. M. Shelley (NCSM 2263); Diplopods from North Carolina 377 and 4 miles SW Cary, Hemlock Bluffs, 2¢, 29, 17 June 1975, J. C. Clamp (NCSM Al) and 124, 49, 16 June 1975, J. C. Clamp (NCSM A2). Lee Co., 10 miles NE Sanford, 54, 109, 28 May 1974, R. M. Shelley (NCSM 1516). Moore Co., High Falls, 1¢, 28 August 1974, R. M. Shelley (NCSM 2501); and 8.6 miles W Carthage, 16, 29, 5 July 1974, R. M. Shelley (NCSM 2355). Cumberland Co., Fayetteville, 26, 2 juvs., 22 November 1948, D. L. Wray (RLH); and 2¢, 39, 24 October 1974, P. D. Kinser (NCSM 2774). Remarks: Polyzonium strictum is the dominant polyzoniid diplopod in the eastern piedmont of North Carolina. It has been found sympatri- cally with P. rosalbum in Pulaski Co., Virginia, and Madison and Moore (High Falls) Cos., North Carolina, but the two species are probably not syntopic. In view of the morphological similarities between these two species, a close phylogenetic relationship is suspected. Previous records of Polyzonium from piedmont North Carolina in- clude P. bivirgatum from Durham (Loomis, 1944; Wray, 1967); Duke Forest (Wray, 1967); and Fayetteville, Cumberland Co. (Wray, 1967). Polyzonium rosalbum has been reported from Duke Forest (Causey, 1940); Fayetteville, Cumberland Co. (Loomis, 1971); and 3 miles SE Albemarle, Stanly Co. (Loomis, 1971). The Duke Forest and Fayette- ville records appear to be examples in which both names were assigned to the same specimens. The Fayetteville and Stanly Co. specimens are actually P. strictum and not P. bivirgatum or P. rosalbum, as previously reported. The location of the Durham and Duke Forest specimens is unknown, and the records cannot be confirmed. In all likelihood they too refer to P. strictum, which is abundant in the “Triangle” (Raleigh- Durham-Chapel Hill) region. It is virtually certain that the record of P. bivirgatum from Durham (Loomis, 1944) is incorrect, as the species is authentically known in North Carolina from montane localities only. Except for the collection of P. rosalbum in High Falls, Moore Co., P. strictum is the only polyzoniid authentically known from the piedmont, and the other records are therefore discounted pending discovery of another individual. Polyzonium laterale, new species Figures 5-8 Type-specimens: Male holotype and one male and one female para- type collected by L. S. Knight, 18 October 1969, from Linville Gorge near Table Rock Mtn., in Burke Co., North Carolina (RLH). Diagnosis: Distal podomere of anterior gonopods bent laterad apically, tip pointed; coxal lobe narrow, bent sharply mediad distally; sternal lobes high, broader at base than tip, narrowly separated. Holotype: Length 8.2 mm, width 1.8 mm, 41 segments. Color uni- formly light yellow dorsally and ventrally, becoming paler caudally. Collum large, hood-like, completely covering ocelli and first antenno- meres. Head short and very broad, sides meeting in almost a right angle, apex blunt; antennae club shaped, distal segments thicker, nearly 378 Proceedings of the Biological Society of Washington seis NID 8 Fics. 1-4. Polyzonium strictum new species (holotype, NCSM 1881). 1, Anterior gonopods, anterior view; 2, Segments 5 and 6, lateral view; 3, Leg 1 and sternum, caudal view; 4, Leg 2 and sternum, anterior view. Fics. 5-8. Polyzonium laterale new species (holotype, RLH). 5, An- terior gonopods, anterior view; 6, Segments 5 and 6, lateral view; 7, Sternum and basal joints of leg 1, caudal view; 8, Leg 2 and stemum, anterior view. Fic. 9. Polyzonium bivirgatum (Wood, 1864), specimen from Polk Co., North Carolina, 5 miles NE Saluda (NCSM 2036). An- terior gonopods, anterior view. Fic. 10. Polyzonium rosalbum (Cope, 1870), specimen from Madison Co., North Carolina, along U.S. hwy. 70, 0.8 miles W jct. N.C. hwy. 22 (RLH). Anterior gonopods, anterior view. Diplopods from North Carolina 379 equalling sides of midbody segments when held laterally; ocelli 3-3, equidistant, first ocelli slightly larger than others, anterior interocular space subequal to diameter of ocellus, posterior interocular space 4—5 times as wide. Pores of segment 5 (Fig. 6) lower than those that fol- low, sides raised, interzonal stria only slightly bulging behind pores. Ultimate segment evenly rounded. Anterior gonopods (Fig. 5) with sternal lobes high, subtriangular, wider at base than tip, bases almost contiguous; coxal lobe narrow, sharply bent mediad distally, sides gradually converging to narrow but rounded tip; distal podomere straight, bent sharply laterad apically, tip pointed. Legs 1 and 2 (Figs. 7, 8) crassate, leg 3 thinner. Sterna of legs 1 and 2 low and narrow, coxae almost touching; coxae of leg 3 more separated; sterna on other pregonopodal legs broad and low, coxae well separated; sterna on postgonopodal legs low and narrow, coxae almost touching. Variation: The male paratype agrees essentially with the holotype, except that the head is more pointed, with its sides meeting in an angle of about 70°. The female paratype is much larger than either of the males (length 15 mm, width 2.3 mm, 58 segments) and has a slightly smaller collum and a pointed head (angle approximately 75°). All sterna of the female are low and narrow, with the coxae almost touch- ing. The male from Tennessee differs considerably from those in the type series. It is brown along the caudal edge of the first 18 segments and the antennae are purplish (color pattern similar to that of P. bi- virgatum (NCSM 2036)). The collum of the Tennessee male has a distinct reflected lip and does not completely cover the ocelli or first antennomeres; the head is broad but pointed (angle approximately 90°). On the gonopods, the coxal lobe bends less sharply distally, and the sternal lobes are more widely separated. Distribution: The species is known only from the mountains of west- ern North Carolina and extreme eastern Tennessee. Specimens have been examined as follows: Tennessee: Carter Co., along U.S. Forest Service Rd., 4.5 miles N Shell Creek, 3500’, 16, 17 July 1965, R. L. Hoffman and K. P. Brown- ell (RLH). North Carolina: Burke Co., Linville Gorge near Table Rock Min., 26, 192, 18 October 1969, L. S. Knight (RLH) TYPE LOCALITY. Remarks: Polyzonium laterale was previously reported from North Carolina as P. bivirgatum by Loomis (1971). It is unique in the config- urations of the distal podomere and coxal lobe and does not appear to be closely related to any known species. Polyzonium bivirgatum (Wood, 1864) Figure 9 Octaglena bivirgata Wood, 1864, Proc. Acad. Natl. Sci. Phila., 16: 186. Polyzonium bivirgatum Cook and Loomis, 1928, Proc. U.S. Natl. Mus., 380 Proceedings of the Biological Society of Washington 72: 18—Chamberlin & Hoffman, 1958, Bull. No. 212, U.S. Natl. Mus., p. 188.—Loomis, 1971, Fla. Ent., 54: 155-156, figs. 6-10. The first North Carolina record of P. bivirgatum was from Durham, in the piedmont (Loomis, 1944). It has since been reported from Durham; Duke Forest; and Fayetteville, Cumberland Co., by Wray (1967). The last record refers to P. strictum, as stated earlier, and the others probably do too. The material from Durham and Duke Forest is unavailable, however, and these records of P. bivirgatum are therefore discounted. The species has also been reported from montane localities—Zionville, Watauga Co., (Wray, 1967); and Linville Gorge, Burke Co.; Madison Co. (two locations); and Mitchell Co. (Loomis, 1971). The Burke Co. record (Linville Gorge) is now assigned to P. laterale, as stated earlier; the Zionville material is unavailable for confirmation, although this is a plausible location for the species. Of the two Madison Co. records, one was based on a female, and I hesitate to report localities based solely on female specimens. The other, between Marshall and Hot Springs, was based on males and females unavailable to me but is also a plausible record for P. bivirgatum. Current evidence (from North Carolina and neighboring states) suggests that the species is restricted to montane localities in the state. The following are the two North Carolina localities which I know to be correct for P. bivirgatum: Mitchell Co., Roan Mtn. below Carver’s Gap, 5000’ elevation, 124, 49, 23 September 1950, L. Hubricht (RLH). Polk Co., 5 miles NE Saluda, 26, 16 October 1973, R. M. Shelley (NCSM 2036). Polyzonium rosalbum (Cope, 1870) Figure 10 Petaserpes rosalbus Cope, 1870, Trans. Amer. Ent. Soc. 3: 65.—Boll- mann, 1888, Proc. U.S. Natl. Mus., 11: 348. Hexaglena cryptocephala McNeill, 1887, Proc. U.S. Natl. Mus., 10: 328. Polyzonium rosalbum Williams and Hefner, 1928, Bull. Ohio Biol. Surv., 18: 104, figs. 7~8—Loomis, 1971, Fla. Ent., 54: 161-162, figs. 30-34. The original North Carolina record of this species was also from Durham (Causey 1940). Loomis (1971) reported it from Fayetteville, Cumberland Co.; 3 mi SE Albemarle, Stanly Co.; and four montane localities in Madison Co. The Fayetteville and Stanly Co. records both refer to P. strictum, as stated earlier, and the material from Durham is unavailable. In four years of collecting in the eastern piedmont, the only polyzoniid I have encountered around Durham is P. strictum. It is conceivable that Causey’s specimens were actually P. rosalbum, since I have collected this species in Moore Co. (High Falls), only 50 miles SW Durham. Without the specimens, however, the record must be Diplopods from North Carolina 381 deleted until new individuals are found. Current evidence indicates that P. rosalbum is widely distributed in the mountains and central pied- mont of North Carolina. The following are North Carolina localities which I know to be correct for P. rosalbum: Madison Co., along U.S. hwy. 70, 0.8 miles W jct. N.C. hwy. 22, 33, 23 July 1961, R. L. Hoffman (RLH); ravine at Paint Rock, 4¢, 292, 13 September 1952, L. Hubricht (RLH); Hot Springs, Silver Mine Recreation Area, 1 ¢, 2 August 1962, R. L. Hoffman (RLH); and ravine along Southern Railroad between Hot Springs and Paint Rock, 3¢, 39, 2 August 1962, R. L. Hoffman (RLH). Moore Co., south bank of Deep R. at High Falls, 24, 28 August 1974, R. M. Shelley (NCSM 2695 ). ADDENDUM As the paper was going to press, I received notice from Dr. Hoffman, who had just viewed the type of Buotus carolinus, that the milliped was not a polyzoniid but a chordeumid of the family Tingupidae, perhaps conspecific with the immature tingupids reported by Shear (1972) from Montgomery County, Virginia. LITERATURE CITED Causey, Nett B. 1940. Ecological and systematic studies on North Carolina myriapods. Unpublished Ph.D. Thesis, Zoology De- partment, Duke University, Durham, North Carolina. CHAMBERLIN, RatpH V. 1940. On some chilopods and diplopods from North Carolina. Can. Ent. 72:56—59. Loomis, H. F. 1944. Méillipeds principally collected by Professor V. E. Shelford in the eastern and southeastern states. Psyche 51: 166-177. 1971. The diplopod genus Polyzonium in North America. Fla. Ent. 71:179-184. SHEAR, WiLLIAM A. 1972. Studies in the milliped order Chordeumida (Diplopoda): A revision of the family Cleidogonidae and a reclassification of the order Chordeumida in the new world. Bull. Mus. Comp. Zool. 144:151-352. Wray, Davip, L. 1967. Insects of North Carolina, Third Supplement. North Carolina Department of Agriculture, Entomology Di- vision, Raleigh, North Carolina. 382 Proceedings of the Biological Society of Washington AN ausis Preece alps stant aplbonh Shain Oe , . 88, No. 36, pp. 383-304 Doljannany amos j f PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON i | . TWO NEW GENERA OF EUDRILID EARTHWORMS y FROM NIGERIA \ By A. O. SEcUN Museum of Natural History, University of Ife, Ile-Ife, Nigeria Few earthworms have previously been described from Ni- geria and as such undescribed species are sometimes found when new collections are made. Among the eudrilid earth- worms, Beddard (189la & 1891b) described four genera and species—Hyperiodrilus africanus, Heliodrilus lagosensis, Liby- odrilus violaceus and Iridodrilus roseus—from collections made in Lagos. Michaelsen (1913) recorded Eudrilus eugeniae from Old Calabar, Southern Nigeria and in 1915, he described Bed- dardiella dalzieli and Eutoreutus abinsianus from Northern Nigeria. Taylor (1949) described a new species, Hippopera nigeriae from Lagos and Ibadan. Clausen (1963) described two new species belonging to a new genus, Keffia variabilis and K. nigeriensis from Keffi, Jos and Kano, Northern Nigeria and Libyodrilus mekoensis from Meko on the Nigeria—Da- homey boundary in 1965. In a collection from Ibadan, Sims (1971) described a new genus and species, Ephyriodrilus afroccidentalis and recorded occurrence of Eminoscolex stein- dachneri. Segun described two new species of Hyperiodrilus from western and mid-western states of Nigeria (in press). The present paper deals with the description of two new genera and species of earthworms collected from the garden soil of catering Rest houses in Vom and Jos, Benue-Plateau State, and from a farm land of the University of Nigeria, Nsukka, East Central State. Other eudrilid earthworms iden- tified during these collections were species of Eudrilus, Hype- riodrilus, Iridodrilus and Keffia. 36—Proc. Bio. Soc. WasH., VoL. 88, 1976 (383 ) 384 Proceedings of the Biological Society of Washington COLLECTING TECHNIQUES Earthworms are collected in terrestrial habitats by digging or by direct application of dilute solution of formalin (18 ml. of 40% formaldehyde in 4.5 liters of water) or about 25 ml. of concentrated formaldehyde to an area of 1.5 square meters. A dilute solution of potassium permanganate can also be used. The worms obtained when using these solutions are washed in water for a few minutes as soon as they appear on the surface. Vomia, new genus Description: Eudrilinae with ab setae fairly widely, cd setae more closely paired. Male pores closely applied midventrally on segment XVII; female pores paired laterally on segment XIV; spermathecal pores paired midventrally in XIII, anterior to female pores. Oesophageal giz- zard present in segment VI, intestinal gizzards and supraintestinal glands absent. Unpaired ventral oeosphageal sacs present in segments X and XI, paired dorsolateral oesophageal glands in segment XII. Holandric, 2 pairs of testes enclosed in testis sacs forming sperm reservoirs in X and XI. Paired ovo-spermathecal apparatus completely separate but connected by a short duct. Ovaries paired and enclosed in ovarian sacs in segment XIII, each ovary leading by a narrow tube into an ovisac which communicates with a fertilisation chamber at the inner (ental) end of the oviduct. Each oviduct opens out ventrolaterally in segment XIV near 13/14 furrow. Spermathecal receptacula seminis paired, left and right parts completely separate in their whole length. Small, paired, muscular spermathecal ampullae present. Paired penial setae present in segment XVII. Distribution: Benue-Plateau State, Northern Nigeria. Type-species: Vomia prima new species. Remarks: The genus belongs to the Eudrilus-Metascolex-Parascolex- Hippopera group of West African Eudrilidae by possessing paired oesoph- ageal glands in segment XII and ventral, unpaired, oesophageal sacs in X and XI. It resembles Parascolex, Hippopera and Metascolex in having separate openings for spermatheca and oviduct, but differs from Meta- scolex which possesses only one pair of testes and vesiculae seminales, and also from Parascolex by the single nature and the positions of the male and spermathecal pores. The spermatheca is also single in Para- scolex for the greater part of its length. Vomia resembles Eudrilus and Hippopera in possessing paired male and spermathecal pores, but differs chiefly from Eudrilus by having separate openings for the oviducts and the spermathecae. Penial setae present in Vomia are completely absent in Eudrilus. Hippopera differs from Vomia in possessing a pair of pros- tatic pores (anterior male pores, Taylor, 1949) on segment XIV and clitellar pouch which covers most of the floor of the clitellum. Earthworms from Nigeria 385 Vomia prima, new species Figures 1-2, 4 26 clitellate specimens collected from the type-locality are now kept in the Natural History Museum, University of Ife, Ile-Ife, Nigeria. Reg. No. 1973.1.1—26. 3 clitellate specimens. British Museum (Natural History) Reg. No. 1973.27.46. External characters: The lengths of the earthworms vary from 120 mm and the diameters in the clitellar region from 6 to 9 mm. The number of segments varies from 130 to 190. The dorsal side of the body is grey with bluish tinge, whilst the ventral side is light grey; the clitellum is light orange. The prostomium is epilobous. The clitellum is annular, occupying segments XIII to XVII. Dorsal pores are absent. The setae have an eudriline arrangement with ab > cd; the setal formula (aa:ab:bc:cd) at segment X is 6:1-2:4:1, at segment XX = 16:2:6:1 and at segment XXX = 15:2:4:1. dd = % circumference. The male pores are paired between aa on segment XVII, very near 17/18 intersegmental furrow where they are situated on the apices of two small protuberances. These, in turn, are found in a shallow, oval depression occupying segments XVII and XVIII (Fig. 1). Penial setae are often visible inside the pores. The female pores are paired on small rounded papillae on XIV out- side setal line d, within dd. Spermathecal pores are closely paired and situated in two small de- pressions within setal lines aa on furrow 13/14 (Fig. 1). A pair of porophores occupying the positions of ab on both sides are found on segment XV. No other genital papillae and markings are present. The nephridiopores are paired in the posterior wall of each furrow within setal line cd, near c. Internal characters: The first septum is 4/5. Septum 5/6 is thin; septa 6/7 to 9/10 are conical being distended by the muscular pharynx; 10/11 and 11/12 are muscular and 12/13 is less muscular. The pharynx extends to segment IV where it leads into the oesophagus. In segment VI, the oesophagus dilates to form a large, very muscular gizzard which extends backwards to the level of segment IX. From seg- ment VII to XV, the oesophagus is undifferentiated apart from the un- paired, midventral, pear-shaped sacs in segments X and XI, and the paired, dorsolateral glands in segment XII. These stalked oesophageal glands are closely applied to septum 12/13 and lie directly beneath the large, posterior vesiculae seminales. Each of these glands opens by a short duct into the oesophagus. The gut dilates to form the intestine in segment XV and this continues to the posterior end of the worm and terminates in an anus. The intestine is uniform throughout, intestinal gizzards, supra-intestinal glands and typhlosole being absent. 386 Proceedings of the Biological Society of Washington "MOIA [VIJWIA ‘soamnqeoz [euiezxe “ds ‘u “nduohu snynasojousy ‘Gg ‘Bq “wtaysAs [eooyyeuutods-0AQ “GZ ‘AMoIA [eI]UOA ‘somjyeoy peutoyxg ‘T ‘ds ‘u ‘pwd pod “Z-T ‘SOI wniiy, aiod jpoayjowsads : ONpIA ad||idpg |pyua ay fonpiao aiod ajo ODSIAO, JequIDYD_UO!}psI1}39 tpoeulouneds Djindwy CTT TES WI Fa) {Doayjowsads 840d ajDW weld sas0ydosog 010, tong upion0 J [pDayjouuads LvTyiXe) Earthworms from Nigeria 387 The dorsal blood vessel passes anteriorly along the upper surface of the gut to segment VII where it breaks up into numerous tiny vessels to serve the gizzard. In segments X and XI, the dorsal vessel gives off branches to serve the oesophageal sacs. In the anterior region of seg- ment XII, it gives off a pair of blood vessels, the left branch serving the oesophageal glands and the right dipping to join the subneural blood vessel. A pair of vessels are also given off in segment XIII and these run forwards to serve the same oesophageal glands. In the in- testinal region, the dorsal vessel also gives off two pairs of parietal vessels in each segment. A subneural vessel runs on the left side of the ventral nerve cord from segment IX, then crosses to the right until it is joined in segment XII by the large right vessel emanating from the dorsal vessel. This subneural vessel then runs backwards directly undemeath the ventral nerve cord. Branches of subneural vessel serve the female reproductive organs in segments XIII to XVI. The ventral blood vessel passes anteriorly below the gut to the pharyngeal region. It also gives off branches to serve the oesophageal gizzard. Commissural blood vessels connecting the dorsal with the supra-oesophageal blood vessel before passing down to the ventral vessel, are present anteriorly with one pair in each of segments VIII to XI. The testes condition is holandric. Both anterior and posterior pairs are enclosed in pear-shaped sacs which lie closely against septa 10/11 and 11/12 respectively. Each of the sacs meets dorsally the opening of the corresponding vesicula seminalis with which it fuses. Ventrally, it leads into the vas deferens which pierces the septum immediately behind into the following segment. The vasa deferentia merge in seg- ment XII and continue posteriorly to segment XVII where both of them enter into the anterior (ental) region of the euprostatic gland of its side. Vesiculae seminales are paired in segment XI and XII, the posterior pair being much larger than the anterior pair. Each of the two euprostatic glands is in the form of an elongate fingerlike process which possesses a swollen ‘head’ followed by a ‘neck’ (Fig. 4). These glands bend laterally outwards, and their ectal end is more muscular than the ental part. The sacs with the penial setae join the euprostates at the ‘neck’ region just behind the swollen, anterior ends. The sacs are attached to the body wall in segment XVII. The penial setae are dagger-shaped and each is divided into two mirror halves by a median line (Fig. 4). The female generative organs are paired. Each ovary is enclosed within an ovarian capsule in XII where it lies by septum 12/13 on either side of the ventral nerve cord. An ovarian duct leads posteriorly from each capsule to an ovisac joined to the receptaculum ovorum (fertilisation chamber) (Figs. 2 & 4). The oviduct leads posterolaterally from the ectal end of each fertilisation chamber to the parietal wall in segment XIV traversing the 13/14 septum (Fig. 4). The spermathecal system is paired. Each spermathecal pore leads to a very muscular, tubular receptaculum seminis which lies alongside 388 Proceedings of the Biological Society of Washington Ovary per mothecal Ampulla Ovisac Atrium Vasa Deferentia Penial Seta Euprostate Gland Fic. 4. Vomia prima, n. sp., reproductive system. the ventral nerve cord. Sometimes, the posterior tips of the receptacula are tucked underneath the euprostate glands. Posteriorly, the receptac- ulum continues into segment XIV where it curves laterally and an- teriorly, it narrows and curls up before joining the ovarian duct. A spermathecal ampulla is present as a small ovoid, muscular body on each side. A communication exists between the two ovo-spermathecal systems of both sides. This is in the form of a short, narrow duct con- necting the two ovarian ducts (Fig. 2). The excretory system is meganephridial with the nephridia in the clitellar regions being larger than the others. The reservoir (vesicle) of each nephridium is thick-walled and the middle portion of the ne- phridial duct is convoluted. Earthworms from Nigeria 38S Agrotoreutus, new genus Description: Eudrilinae with ab fairly widely, and cd more closely paired. Male pore is single on raised papilla on 17/18 furrow; female pores paired laterally on 13/14 furrow; spermathecal pore single, mid- ventrally on papilla on 20/21 furrow, close to segment XXI, and pos- terior to male pore. Paired rounded genital papillae on segments XX to XXII and an almond-shaped papilla on XXIII-XXXI. Oecsophageal gizzard present in segment V, intestinal gizzards absent. Paired ventral oesophageal sacs present in segments IX, X and XI, paired kidney-shaped dorsolateral oesophageal glands in segment XIII. Typhlosole present. Holandric, two pairs of testes enclosed in testis sacs forming sperm reservoirs in segments X and XI. Euprostates usually long and tubular. Ovaries paired, enclosed in ovarian sacs in segment XIII. Each ovary, found usually attached to 12/13 septum near oesophageal glands, leads by a narrow short tube into an ovisac which communicates with a fertilisation chamber at the inner (ental) end of the oviduct. Each oviduct opens to the exterior ventro-laterally in 13/14 furrow near setal line d. Spermathecal atrium single, midventral dividing ectally into two diverticula (receptacula seminis). Two pairs of penial setae present in the form of rolled tubular setae in segment XVII. Distribution: East Central State, Eastern Nigeria. Type-species: Agrotoreutus nyongii, new species. Remarks: The genus belongs to the Polytoreutus-Eupolytoreutus- Eutoreutus group. It resembles Polytoreutus Michaelsen and Eupoly- toreutus Michaelsen in having an oesophageal gizzard in segment V, one pair of oesophageal glands in segment XIII, the male pore single on 17/18 furrow and the spermathecal pore single, midventral, lying behind the male pore. But, it differs from Polytoreutus which possesses one pair of testes, enclosed in sacs in segment XI. Paired penial setae which are present in the newly described genus are completely absent in Eupolytoreutus and Polytoreutus. It also resembles Keffia Clausen and Eutoreutus Michaelsen in the positioning of oesophageal gizzard in segment V and oesophageal glands in XIII, and also in the spermathecal pore being situated behind the male pore. But, the male and the spermathecal pores are paired in Keffia as opposed to their being single in this genus and Eutoreutus. The ovo-spermathecal system of the genus being described seems to be similar to the structures described for the genus Eutoreutus, but in the former genus, there is no “muscular copulatory pouch either situated behind and separate from the spermatheca, or opening in common with its ectal extremity” (Stephenson (1930) p. 879). Furthermore, the ventral oesophageal sacs in segments VIII to XI or segment IX to XI are unpaired in Eutoreutus whereas those found in the presently de- scribed genus are paired in segments IX, X and XI. This earthworm is named after Mr. E. T. Nyong, formerly of University of Nigeria, Nsukka, who first sent me a preserved specimen. 390 Proceedings of the Biological Society of Washington Agrotoreutus nyongii, new species Figures 3, 5 12 clitellate specimens collected from the type-locality are now kept in the Museum of Natural History, University of Ife, Ile-Ife, Nigeria Reg. No. 1974.1.1-2. 2 clitellate specimens. British Museum (Natural History) Reg. No. 1974.17.8-9. External characters: The lengths of the earthworms vary from 250 to 400 mm and the diameters in the widest region from 5 to 8 mm. The number of segments varies from 490 to 645, the clitellar segments are widest and the postclitellar segments are antero-posteriorly narrower than the preclitellar ones. The anterior portion of the body is pinkish whilst the posterior is bluish-purple. The body is not pigmented as the dorsal blood vessel, ventral vessel, the euprostate glands and parts of the spermathecal receptaculum show through it; the clitellum is dark yellow. The prostomium is prolobous. The clitellum is saddle-shaped, occupying segments XIV to XVII. Dorsal pores are absent. The setal arrangement is eudriline with ab > cd; the setal formula (aa:ab:be:cd) at segment X is 9:4:10:1, at segment XX = 9:3-6: 8:1 and at segment XXX = 9:4:8:1. aa < be on segment X whereas aa > be on XX and XXX. dd = %& circumference. The male pore is single between setal lines aa on 17/18 interseg- mental furrow where it is situated on a raised papilla. Two pairs of penial setae are often found protruding through the pore. The female pores are paired and they open out laterally in 13/14 intersegmental furrow, just ventral to the nephridiopores. The spermathecal pore is single being situated on a papilla on furrow 20/21 or segment XXI between setal lines aa (Fig. 3). A pair of rounded genital papillae (Fig. 3) are found occupying seg- ments XX to XXII and another almond-shaped papilla occupying seg- ments XXIII to XXXI midventrally. No other genital markings are present. \ The nephridiopores are paired in the posterior wall of each furrow from segment III within setal lines cd. Internal characters: The first septum, 4/5, is thin. Septum 5/6 is intermediate in thickness between the preceding septum and septa 6/7 to 10/11 which are thickened and conical. Septum 11/12 is thinner but conical. The buccal cavity opens into a large, muscular pharynx which extends to segment IV where it leads into the oesophagus. A number of dilator muscles attach the wall of the pharynx to the body wall. The oesopha- gus dilates in segment V to form the large, muscular and almond-shaped gizzard. The oesophagus then passes posteriorly and it is undifferen- tiated apart from the three paired glands which are present as ventral reddish pouches, a pair in each of segments IX, X and XI, and the paired bean-shaped oesophageal glands dorsolaterally in segment XIII. The Earthworms from Nigeria 391 intestine begins in segment XVII and runs uniformly throughout to the terminal anus. There are neither intestinal gizzards nor suprain- testinal glands, but a median typhlosole is present. The dorsal blood vessel passes anteriorly along the intestine and the oesophagus to the pharynx. It is very large in segment XIII where it gives off two pairs of vessels. While the posterior pair serve the dorso- lateral oesophageal glands, the left anterior vessel serves the sper- mathecal ducts of its side, crosses over underneath the oesophagus and joins the right anterior vessel to run alongside the ventral nerve cord. This blood vessel later becomes the subneural vessel in segment XIX. In segments VI to XI, paired commissural blood vessels encircle the oesophagus and interconnect the supraoesophageal with the dorsal ves- sel segmentally before passing down to the ventral blood vessel. They become progressively larger from VI to XI and function as lateral hearts since they are all contractile. A supraoesophageal blood vessel exists in VI to XI, and in each of IX, X and XI, it gives off paired vessels to the ventral oesophageal glands. A suboesophageal vessel gives rise to two rather large blood vessels in VI. These vessels run underneath the oesophageal gizzard to serve the pharyngeal region. The ventral blood vessel gives off four pairs of vessels to the ovospermathecal complex in segment XIII and then runs over the receptaculum seminis alongside the nerve cord. It gives off several branches in segments XVI to XXI and then dips downwards to lie above the nerve cord for the remaining length of the animal. The testes are holandric being paired in segments X and XI. Each is enclosed in a testis sac. A short duct leads posteriorly from the dorsal end of each testis sac through the septum into the seminal vesicle in the following segment. The testis sacs also contain the ciliated sperm fun- nels. The seminal vesicles are paired in segments XI and XII, the pos- terior pair being much larger. A vas deferens runs from the end of each testis sac and passes through the corresponding septum (10/11 and 11/12 respectively). The two vasa deferentia of each side unite in segment XIII and run posteriorly side by side over the parietal wall. From segment XIX, these ducts run along the edge of the euprostatic glands until they pass into these glands in segment XXII. The eupros- tates are paired, muscular, elongated and sausage-shaped occupying segments XV or XVI to XXXII. They are narrow anteriorly (entally) and the tapering anterior ends of both glands meet below both the nerve cord and spermathecal atrium (Fig. 5) before opening out at the single ventral male pore in furrow 17/18. The pairs of penial setae which are curved at the tips and which are in form of rolled tubes join the eupros- tates prior to their opening out. The ovaries are paired in segment XIII and each with the ovisac is enclosed in a pink, transparent and thin-walled ovarian vesicle. An ovarian duct leads posteriorly from each ovary to join an efferent duct which leads into the receptaculum ovorum (fertilisation chamber). From 392 Proceedings of the Biological Society of Washington Ovary Ovisac Oviduct Vasa deferentia ——YVentral nerve cord permothecal atrium Euprostate gland Fic. 5. Agrotoreutus nyongii, n. sp., reproductive system. here, the oviduct leads posterolaterally to open out in 13/14 furrow (Fig. 5). The spermathecal pore is single, midventral on 20/21 furrow and it leads into a muscular sac-like atrium which is situated below the ven- tral blood vessel and nerve cord. This immediately splits into two fingerlike, thin-walled appendages, the receptacula seminis, which occupy segments XXI to XXVIII, XXIX or XXX and in some cases to segment XXXIV on either side of the nerve cord. Anteriorly, the atrium main- tains its baggy shape until it narrows in segment XVII to form a sper- mathecal tube. In segment XV, this tube bends upon itself to become U-shaped on the left side (sometimes to the right). It then bifurcates, immediately after penetrating 13/14 intersegmental septum, into right Earthworms from Nigeria 393 and left narrow efferent ducts thereby becoming Y-shaped. This there- fore means that the spermathecal system is paired both anteriorly and posteriorly, but remains single medially (Fig. 5). The excretory system is meganephridial with one pair of nephridia to each of the segments from segment V backwards. Nephridia in the clitellar region are of the same size as the others. ACKNOWLEDGMENTS My thanks are due to Mr. R. W. Sims of Department of Zoology, British Museum (Natural History) for criticising the description of the newly described earthworms, and to Professor L. B. Halstead for his useful suggestions. LITERATURE CITED BEpDpDARD, F. E. 1891la. On the structure of two new genera of earth- worms belonging to the Eudrilidae, and some remarks on Nemertodrilus. Quart. Jour. Microscop. Sci. 32:235—278. 1891b. Preliminary account of an earthworm from West Africa referable to a new genus. Proc. Zool. Soc. London 1891:172-176. CiausEN, M. W. 1963. Two new genera and three new species of earthworms from West Africa. Vidensk. Medd. Dansk Naturhist. Foren. 126:1—15. ———. 1965. Description of two new species of Libyodrilus Beddard 1891, the remarks on the genus and on the female system in one family Eudrilidae. Vidensk. Medd. Dansk Naturhist. Foren. 128:273-292. ——. 1967. The African oligochaete genera Hyperiodrilus Beddard, 1891 and Legonea Clausen, 1963 (Oligochaeta). Vidensk. Medd. Dansk Naturhist. Foren. 130:179-207. MicHaAELsEN, W. 1913. Oligochaten vom tropischen und _ siidlich- subtropischen Africa II. Zoologica, Stuttgart, Heft 68:1-63, pls. 1-2. 1915. Oligochaten: Ergeb. 2. Zweit. Deutschen Zentral- Afrika-Exp. 1910-11. Bd. 1, Zool. Teil 1:185—-307. Leipzig. Sims, R. W. 1965. Acanthodrilidae and Eudrilidae (Oligochaeta) from Ghana. Bull. Brit. Mus. (Nat. Hist.) Zool. 12:283-311. 1967. Earthworms (Acanthodrilidae and Eudrilidae: Oligo- chaeta) from Gambia. Bull. Brit. Mus. (Nat. Hist.) Zool. 16: 1-34. 1971. Eudrilinae from Southern Nigeria and a taximetric appraisal of the family Eudrilidae (Oligochaeta). Jour. Zool. London 164:529-549. STEPHENSON, J. 1930. The Oligochaeta. Oxford, Clarendon Press, xiv + 978 pp. Taytor, A. G. 1949. A West African earthworm (Hippopera ni- geriae), belonging to a new family, Hippoperidae. Proc. Zool. Soc. London 119:703-710. 394 Proceedings of the Biological Society of Washington No. 37, pp. 395-398 22 January 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A NEW WATOPHILUS FROM UTAH, INCLUDING . A LIST OF ALL KNOWN SPECIES (CHILOPODA: /_ GEOPHILOMORPHA: CHILENOPHILIDAE) | By R. E. Cras, Jr. National Museum of Natural History, Smithsonian Institution, Washington, D.C. \ In his 1929 monograph of the Order, Attems was unable to present a key for Chamberlin’s species of Watophilus. I am unable to do so now, and for the same reasons: they are in- adequately described, and I have never seen a specimen other than the one presented here as new. An adequate discussion of the critical diagnostic characters is nowhere in the litera- ture. Neither have I been able to locate the typical specimens in the several Chamberlin Collections. In 1964 when I con- ferred last with Professor Chamberlin, he revealed that many, perhaps most, of his types described prior to 1913 could not be located; he presumed them to be lost. Most of the Cham- berlin forms presented here were described before 1913. I am therefore, at the moment, unable to do more than present a list of the known species; all are North American. Hopefully, additional Watophilus material will be collected in the future. The genus, whose members are apparently very rarely col- lected, is clearly most closely related to the ubiquitous Hol- arctic genus Arctogeophilus, whose many species are com- monly encountered in temperate to arctic North America. Both belong to a very small suite of chilenophilid genera one of whose notable features, in my opinion, is the absence of a ventral condyle on the second maxillary telopodite. So far as I am aware, this remarkable condition escaped the notice of Verhoeff and Attems. The presence in Watophilus of spinous processes on the second maxillary telopodite distinguishes it both from Arctogeophilus and all other North American chilenophilids. 37—Proc. Biot. Soc. WaAsH., VoL. 88, 1976 (395) 396 Proceedings of the Biological Society of Washington Although I am at the moment unable to distinguish all of the species of Watophilus from one another, it is possible to say, even from what little is known of them, that the species presented here as new cannot be referable to any other known member of the genus. All previously described species are said to possess a distinct tuberculate ultimate pedal pretarsus; knowltoni, new species, lacks an ultimate pedal pretarsus altogether. Possibly the new form is most like utus Chamberlin, known only from Utah, in which case they differ as follows. In utus: (1) pedal pairs, 65. (2) Ultimate pretarsus present and tu- berculate. (3) Prehensorial segments 2 and 3 each with a small denticle. In knowltoni: (1) pedal pairs, 49. (2) UI- timate pedal pretarsus absent. (3) Prehensorial segments 2 and 3 unarmed. Watophilus knowltoni, new species’ Holotype: female. Utah: Box Elder Co., Kelton Pass, in juniper duff; April 28, 1969; George F. Knowlton, leg. Deposited in the National Museum of Natural History. Description: GENERAL. Body length, 13 mm. Color, light yellow throughout. Body posteriorly gradually attenuate. Pedal segments, 49. ANTENNAE. Relatively long, ca. 2.8 longer than the head. Articles: filiform, each longer than wide; the more proximal articles with relatively few long setae, distally these increasing in number and becoming shorter. Ultimate article mesally and ectally at midlength and distal thereto with a patch of short robust hyaline modified setae. CzEPHALIC PLATE. Sub- stantially longer than wide (37:26). Shape: sides evenly slightly ex- curved; rear margin evenly rounded, excurved. Frontal suture absent. With two paramedian bands of coarse areolation extending from rear margin to labral level. Prebasal plate not evident. CLypEus. Antero- centrally with a small poorly-defined clypeal area (fenestra apud Cham- berlin) within which are 2 long setae; immediately anterior thereto on each side is a long seta. With a pair of prelabral setae; clypeus other- wise glabrous. Paraclypeal sutures slightly sinuous, complete. LABRUM. Sidepieces meeting centrally. Midpiece with long hyaline filaments. Clypeolabral suture distinct and complete. First MAxILLAE. Coxo- sternum: anteroposteriorly very deep, undivided, not suturate; lappets scrabrous, short, attaining only end of telopodite Ist article; medial lobes basally broad, narrowing rapidly distally. Telopodite: bipartite; lappet thick, robust, scabrous; extending slightly beyond end of telop- 1 Named in honor of Dr. George F. Knowlton, valued friend of many years, gen- erous and tireless collector, whose efforts for decades have contributed to our knowledge of Utah, his native state. Watophilus from Utah 397 odite. SECOND MaxitLaE. Isthmus: undivided and uniformly areolate, not hyaline; extending posteriorly beyond level of pore. Coxosternite: statumen well-developed, its posterior end curving anteroposteriorly; anteromesal projection absent; ventral condyle absent, dorsal condyle present. Telopodite: Ist article basally much narrower than the ex- panded distal half, anteromesal and anteroectal corners each with a small spinous process; 2nd article anteroectally with a similar though much smaller process (it is virtually abortive); ultimate article (homo- logue of a claw) is a short, blunt, minutely spinous tubercle. ForciePuLAR SEGMENT. Flexed prehensors exceeding anterior margin of head. Pre- hensor: Ist article very long, anteromesally with a blunt, unpigmented denticle; articles 2 and 3 unarmed. Tarsungula: with a large, pig- mented, pointed basal denticle; ungular blade edges smooth, not ser- rulate, the blade relatively short and robust. Poison calyx very small, suborbicular, its digitiform appendices extending only posteriorly; poison gland elongate, housed in articles 1, 2, and 3. PRostERNUM. Pleurograms absent. Pleuroprosternal sutures straight, parallel to prosternal margin, complete for ca. 95% of their length, anteriorly not quite attaining margin. Anteromedial denticles absent. Tercrres. Basal plate neither sututate nor sulcate, trapezoidal. Remaining tergites (except ultimate pedal) longitudinally bisulcate. Lzecs. Sparsely clothed with very long setae. Pretarsi: anterior parungues uncommonly long, ca % as long as their claws; posterior parungues present but minute, virtually abortive. STERNITEsS. Without glandular pores. Midlongitudinally weakly sulcate. Setae few, short. ULTIMATE PEDAL SEGMENT. Pretergite not bilaterally fissate. Tergite: sides notably excurved; greatest length exceeding great- est width; rear margin narrowly rounded. Sternite: shape trapezoidal; posteriorly truncate; sides straight, convergent; greatest length exceed- ing greatest width. Coxopleuron: only slightly inflated and virtually glabrous; pores anteroventrally and mostly exposed, there numbering 4-5, and anterodorsally and concealed beneath the pretergite, there numbering 2-3. Ultimate leg: % longer than the penult; articles slightly inflated, each much longer than wide; ventrally on trochanter through tibia with numerous tiny setae, in addition all articles sparsely clothed with very long setae. Tarsus: bipartite; first article slightly longer than second; article 2 subclavate, distally blunt, without a pre- tarsus of any sort; apically with a minute seta. PosTPEDAL SEGMENTS. Female gonopods flat, entirely fused medially without demarcation, unipartite. Anal pores prominent, lateral. The Known Species of Watophilus* Subgenus Watophilus Chamberlin, 1912A: 662 1. alabamae Chamberlin, 1912B: 421. —Type-locality: Anniston, Mapleville, Tallulah Falls, Ga. —Elsewhere recorded only from South Carolina. 2Tt was plainly Chamberlin’s intent to present Watophilus as new in the M.C.Z. Bulletin issued in April of 1912. Therein he stated that its type-species is alabamae, 398 Proceedings of the Biological Society of Washington 2. errans Chamberlin, 1912A: 662. —Type-locality: Berkeley, Cal. —No other record. 3. knowltoni Crabill, here described. —Type-locality: Kelton Pass, Box Elder Co., Utah. —No other record. 4, laetus Chamberlin, 1912A: 662. —Type-locality: Berkeley, Cal. —Elsewhere recorded from Stanford, Pacific Grove, Cal. 5. utus Chamberlin, 1928: 95. —Type-locality: La Sal Mountains, San Juan Co., Utah. —No other record. 6. dolichocephalus (Gunthorp, 1913: 169). —Type-locality: Cowley Co., Kans. —No other record. LITERATURE CITED ATreMs, Cart Grar. 1929. Geophilomorpha. Das Tierreich. Lief. 52:1-388. CraBILL, RALPH E., Jr. 1955. On the identity of the Gunthorp types. Can. Entomol. 87(5) :221-223. CHAMBERLIN, RALPH V. 1912a. The Chilopoda of California III. Pomona Journ. Entomol. 4(1):651-672 (February). 1912b. The Geophiloidea of the Southeasten States. Bull. Mus. Comp. Zool. Harvard 54(13):407-436 (April). 1928. On three chilopods from the La Sal Mountains of Utah. Entomol. News 39:93—96. Guntuorp, Horace. 1913. Annotated list of the Diplopoda and Chilopoda with a key to the Myriapoda of Kansas. Univ. Kansas Sci. Bull. 7(6):161—-182. the only species mentioned in that publication. However, the generic name first appeared in print in the February 1912 issue of the Pomona Journal of Entomology, at which time the genus, there noé presented as new, was said to contain only two species, laetus and errans: no type-species was mentioned. Therfore Chamberlin’s fixation of the type-species, alabamae, is neither through original designation nor monotypy, rather through subsequent designation. o. 38, pp. 399-428 22 January 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON A REPORT ON THE FIFTH INTERNATIONAL i SYMPOSIUM ON PLECOPTERA By Rictarp W. BAUMANN Smithsonian Institution, Washington, D.C. 20560 '\~ The first international Plecoptera symposium to be held in the new world took place September 3-6, 1974, in Wash- ington, D.C. The four previous meetings were all in Europe (Switzerland, 1956, Austria, 1960, Germany, 1963, and Sweden, 1968) therefore, this meeting provided an excellent oppor- tunity for American and European scientists to become bet- ter acquainted and exchange information on current research. The symposium was sponsored by the Smithsonian Institu- tion with the aid of the Office of Academic Affairs and the Department of Entomology. The international committee con- sisted of: R. W. Baumann, Smithsonian Institution, Washing- ton, D.C., USA, Organizing Chairman: J. Illies, Max-Planck Limnology Institute, Schlitz, West Germany, Honorary Chair- man; C. G. Froehlich, University of Sao Paulo, Sao Paulo, Brazil; P. P. Harper, University of Montreal, Montreal, Quebec, Canada; T. Kawai, Nara Women’s University, Nara, Japan: A. Lillehammer, University of Oslo, Oslo, Norway; I. D. Mc- Lellan, Westport, New Zealand: K. W. Stewart, North Texas State University, Denton, Texas, USA; P. Zwick, Max-Planck Limnology Institute, Schlitz, West Germany. Meetings were held during the days of September 3-5 in the Carmichael Auditorium of the Smithsonian Institution, National Museum of History and Technology. The opening session on the morning of September 3 included a welcome by the chairman and greetings by local dignitaries and was 38—Proc., Bron. Soc. Wasu., Vou. 88, 1976 (399 ) 400 Proceedings of the Biological Society of Washington - Baumann - S. W. Szezytkc 21. I. D. McLellan J. Spangler Ae 0b So wiln hy BRED 22. T. Kawai W. Stewart 3. Re 1) E 23. =. F. Benfield . Kohno V. Smolka 7. E. Ricker . R. Oblad it McElravy W. Woolman, IIT P. Stark 2. Tolbert Fic. 1. Group picture in front of Museum of History and Technology. highlighted by an invitational lecture by Professor Dr. Joachim Illies entitled, “The Status of Plecoptera Research Today.” Professor Ilies recounted the history of plecopterology and ended by noting that although much has been accomplished, there is still a lot to learn about all aspects of Plecoptera. Plecoptera Symposium A401 Fic. 2. Symposium participants, left to right: Peter Zwick, Joachim Illies and Dragica Kaéanski. Fic. 3. Portion of symposium audience during presentation of a paper. 402 Proceedings of the Biological Society of Washington Fic. 4. Field trip, left to right: Ian D. McLellan, Paul J. Spangler, Henry W. Woolman, III. The submitted papers were organized into five groups ac- cording to content and were presented as listed below: Tuesday, September 3, 1974 ( Afternoon ) Section leader: K. W. Stewart, Texas, USA. Basic Principles and Practices in Zoological Nomenclature, by C. W. Sabrosky, Washington, D.C., USA. The Phylogenetic System of the Order Plecoptera, by P. Zwick, Schlitz, Germany. Notes on Nomenclature and Taxonomic Growth of the Ple- coptera, by G. C. Steyskal, Washington, D.C., USA. Numerical Taxonomic Analysis of Relationships in Plecoptera. by C. H. Nelson, Tennessee, USA. Scientific Ilustration-Techniques and Media. by G. L. Venable and M. Druckenbrod, Washington, D.C., USA. Plecoptera Symposium 403 Fic. 5. Field trip, left to right: Rebecca F. Surdick, Oliver S. Flint, Jr., Briant R. Oblad. Wednesday, September 4, 1974 (Morning ) Section leader: P. P. Harper, Montreal, Canada. Notes on the Nearctic Genera of Perlidae. by B. P. Stark, Utah, USA. The Isoperla of Texas. by S. W. Szczytko, Texas, USA. Notonemouridae-Systematics and Distribution. by J. Illies, Schlitz, Germany. The Family Nemouridae. by R. W. Baumann, Washington, D.C., USA. African Species of the Genus Neoperla Needham (Perlidae). by P. Zwick, Schlitz, Germany. ( Afternoon ) Section leader: C. H. Nelson, Tennessee, USA. Mating Behavior of Paragnetina fumosa, Perlinella drymo and Hydroperla crosbyi; with Special Emphasis on External Sperm Transfer in H. crosbyi (film). by K. W. Stewart, Texas, USA. 404 Proceedings of the Biological Society of Washington An Emergence Sequence of Chloroperlidae in a Northeastern Ohio Stream. by M. A. Tkac, Jr., Ohio, USA. Emergence Patterns in Plecoptera. by P. P. Harper, Montreal, Canada. The Structure (Ultra) and Function of the Ventral Lobe and the Hammer of Plecoptera. by R. Rupprecht, Mainz, Germany. Thursday, September 5, 1974 (Morning ) Section leader: P. Zwick, Schlitz, Germany. The Stoneflies (Plecoptera) of the Rocky Mountains. by A. R. Gaufin, Utah, USA. Preliminary Studies on Pennsylvania Stoneflies. by R. F. Surdick, Pennsylvania, USA. Biogeography of New Zealand Plecoptera. by I. D. McLellan, Westport, New Zealand. A Preliminary Report of the Plecoptera Fauna in Bosnia and Herzegovina (Yugoslavia). by D. Kaéanski, Sarajevo, Yugloslavia.* ( Afternoon ) Section leader: I. D. McLellan, Westport, New Zealand. Autohemorrhage in Two Stoneflies and its Effectiveness as a Defense Mechanism. by E. F. Benfield, Virginia, USA. Some Morphological Features of Gills in Plecoptera. by N. N. Kapoor, Montreal, Canada. Notes on the Family Scopuridae. by T. Kawai, Nara, Japan. Distribution of Stoneflies (Plecoptera) Within the Hubbard Brook Watershed, New Hampshire. by S. M. Fiance, New York, USA. During the extra time made available, because some participants could not attend, some very interesting extemporaneous presentations were given. William E. Ricker explained the derivations of many of the names that he has proposed for Plecoptera genera and species. Joachim Illies showed films of the two most recent Plecoptera meetings * Paper not given at this time but abstract included. Plecoptera Symposium 405 in Germany and Sweden and presented the results of his emergence control studies on Breitenbach. Those in attendance on September 3 were recorded in the group picture. Everyone that attended the symposium and registered is listed at the end of this report along with their current address. The first issue of the plecopterists newsletter Perla was distributed and all felt that it was a positive step forward. Plans were made to continue the newsletter at approximately one-year intervals. A Plecopterists Luncheon was held in the Commons Room of the original Smithsonian Building on September 4 where all present were able to mingle and make new friends. The ladies in attendance at the Symposium were able to participate in an excellent program or- ganized by Carol M. Flint. The final day, September 6, was the field trip led by Dr. Oliver S. Flint, Jr., to the Bull Run Mountains in Northern Virginia. Although it rained most of the day and the interest of some was dampened, most everyone experienced a day to remember. One of the major objectives of this symposium was to bring together young plecopterists to meet and learn directly from the established ex- perts in the field. This was realized because approximately one-third of the participants were students and the open forum type of discussion allowed for free interchange of ideas. After hours contact was made difficult because of the housing arrangements available but hopefully this will be corrected in the next symposium tentatively scheduled to be held in Schlitz, West Germany. ABSTRACTS OF PRESENTED PAPERS BASIC PRINCIPLES AND PRACTICES IN ZOOLOGICAL NOMENCLATURE. By Curtis W. Saprosky, Systematic Entomology Laboratory, U.S. Department of Agriculture, Washington, D.C. The history of successive international codes of zoological nomen- clature was briefly reviewed, from the first in 1905 (then called the Régles, or Rules) to the 1961 and 1964 editions of the Code, to changes made at Monaco in 1972 and at Ustaoset in Norway in 1973. An Edi- torial Committee is now working on a 3rd edition that will, hopefully, be ready for final approval in late 1976. Basic principles were listed and discussed briefly: 1. The Priniciple of Availability 2. The Principle (“Law”) of Priority (with Conservation a Limi- tation on Priority) The Language of Scientific Names The Principle of Coordination The Principle (“Law”) of Homonymy The Principle of Typification ee ES £9 406 Proceedings of the Biological Society of Washington Terminology: Definitions or explanations were offered for certain nomenclatural jargon: nominal genus, junior and senior, objective and subjective, primary and secondary. A few “do’s” and “don’t’s” that would help in avoiding nomenclatural difficulties and expedite taxonomic work were suggested. THE PHYLOGENETIC SYSTEM OF THE ORDER PLECOPTERA. BY PETER Zwick, Limnologische Flufstation des Max-Planck-Instituts fiir Limnologie, Germany. Animals can be classified in many ways, to serve many purposes and these systems may be altered or abandoned, according to the needs of their users. However, taxonomy is then reduced to a subordinate mean, a file for other disciplines and such taxonomy cannot assist in scientific research. A more adequate way to treat taxonomy or systematics is as a science of its own, equal in rank to other sciences, with rules and procedures following scientific logic, and nothing else. Under no circumstances may these rules be bent or modified to suit other demands. The aim of such scientific taxonomy is establishment of the one system which evolved in time. Scientific taxonomy attempts to elucidate the genealogical re- lationships between species and supraspecific taxa of a given time, in the present or in the past. These relationships are represented in un- mistakable ways, either in written systems or in cladograms. Each of these two representations can be transformed into the other, without ambiguity or loss of information. Strict monophyly of supraspecific taxa is mandatory, paraphyletic or polyphyletic taxa are not admitted. Such a system has been termed phylogenetic by Hennig (1949). A brief account of the principles of phylogenetic systematics has been given in English by Hennig (1965). Brundin has also extensively dealt with it several times, e.g. 1966. For those who are critical of the theory of phylogenetic systematics it would be important to know its recent formulation by Schlee (1971), which is without some of the corollaries that have at times been taken to be essentials of the theory. Unfortunately, the former “evolutionary taxonomists” have started to use the term phylogenetic systematics for their classifications. They even try to refuse the use of the term according to its older definition by Hennig. This is unfortunate, because phylogenetic systems of evo- lutionary taxonomists lack a clear concept. They are mixtures of “science, most strictly speaking, and of an art’ (Simpson, 1961). Therefore, these systems are equivocal and are in fact little more than the usual intuitive taxonomy with evolutionary flavouring, unsuited for scientific work. The phylogenetic system of Plecoptera (in the sense of Hennig) shown below (Fig. 6), was briefly discussed (for details see Zwick, 1973). Plecoptera Symposium 407 o : : 4 : 0 0 cS cd) © Om) 6 sf cd) 6 -d is am © @) {I Oo dH oO oO dod “di “doy S| G-4 oO ee mM cl nal aa Se 4 4G di S€4d 860 Adv DH YO OO AS 0 @ po o 0 “4 AMY £08 dH ONO SG Aad a © fe yp p Ho HAO Ae HH HOH 040, Med Oo Oo 3 -d Oo, 308 YQ da OD OH GB wad od SS) OH ESmxe) Oe DMO Woo Cryo, S@ S dst} fond) sj 19) ee) Ged ae mS WO OO Of © Beh Oys poy oe O ® OH @O QH AYP HON EF ows) aA @ ns 6 oO G46 YH ON WAH OF WD aa Og) aS YP oO o 4 oo 00 G@S HY PO PAP tH PS G ¢ =a mo f£oa MAOHMm HA AKA AA NA DS i oo i p ia) i) % o o a) d) o © G © o a) w i O Le} D O iy © Lo} o df a) “di Dy “dl Le} ;S “di G S o © 3 o © 4 w qd 0) g AM @ “4d oe 6) D tg o d) Lo} 4 re ay “d a) 0 “UG cd E QQ cd w or - 0) 0 “ (| (oh, p poe 4 o Ce) 4 6s fe) S a (S| & ) fe) a, Od Sj cl Ss cd re fe) fo) fe) ay Ect 4 ) 3 s (e) fe) S & a Ce) 4 p re 5 am ay ) a, = p ) fe) re 4 4 qo cd) n a) , ote 4 ic) a) io) © id) ) fe} a) o p “dl 3 S| He senate iS) Z Z a on) a id) Ay A Ay fa) ea] < a » & j Subulipalpia Eusthe- Lepto- ie nioidea perloidea “es a ead . erlaria Arctoperlaria Antarctoperlar PLECOPTERA Fic. 6. The phylogenetic system of Plecoptera (from Zwick, 1969, modified ). LITERATURE CITED Brunpin, L. 1966. Transantarctic relationships and their significance, as evidenced by chironomid midges..... —Kungl. Svenska Vetenskapsakad. Handl., 4 Ser. 11(1):1-472, 30 pls. (Stock- holm). Hennic, W. 1949. Zur Klarung einiger Begriffe der Phylogenetischen Systematik. Forsch. Fortschr., 25:137-139 (Berlin). 1965. Phylogenetic Systematics. Ann. Rev. Ent. 10:97-116 (Palo Alto). ScHLEE, D. 1971. Die Rekonstruktion der Phylogenese mit Hennigs Prinzip. Aufs. Reden Senckenberg. Naturf. Ges. 20:1-62 (Frankfurt ). Smmpson, G. G. 1961. Principles of animal taxonomy. 247 pp. (New York). Zwick, P. 1969. Das Phylogenetische System der Plecopteren als Ergebnis Vergleichend-Anatomischer Untersuchungen. Thesis, Univ. Kiel, 291lpp; Abstract, 4 pp. 408 Proceedings of the Biological Society of Washington 1973. Insecta: Plecoptera. Phylogenetisches System und Katalog. Das Tierreich, 94:I-XXXII, 1-465 (Berlin, New York). NOTES ON THE NOMENCLATURE AND TAXONOMIC GROWTH OF THE PLECOPTERA. BY GEeorGE C. STEYSKAL, Systematic Entomology Laboratory, IIBIII, Agr. Res. Serv., USDA, c/o National Museum of Natural History, Washington, D.C. Students of the Plecoptera are fortunate in having up-to-date cata- logues and a systematic compendium of the order in the works of Illies (1966) and Zwick (1973). The literature of few orders of insects is so well indexed and the classification so well revised by the most mod- ern methods. Order and Suborder Group Superfamily Family Plecoptera Antarctoperlaria Eusthenioidea Eustheniidae Diamphipnoidae Gripopterygoidea Austroperlidae (Leptoperloidea ) Gripopterygidae Gripopteryginae Leptoperlinae Paragripopteryginae Antarctoperlinae Arctoperlaria Systellognatha ( Pteronarcyoidea ) Pteronarcyidae ( Peltoperloidea ) Peltoperlidae Perloidea Perlodidae (Subulipalpia ) Perlidae Chloroperlidae Euholognatha Scopuroidea Scopuridae Nemouroidea Taeniopterygidae Notonemouridae Nemouridae Capniidae Leuctridae The “Argumentationschema” on p. 2 and the dendrogram on p. 16 of the work by Zwick are to some extent at variance with the clas- sification followed in the catalogue and the systematic index, which is as shown above. The genus Crypturoperla is shown in the Schema as of equal rank with the Austroperlidae, but in the catalogue and the sys- tematic index the genus is treated as a member of that family. The Antarctoperlinae are treated in the Schema as of equal rank with the Plecoptera Symposium 409 1760 1600 1200 800 400 109) o -d 12) o jon 123) Ww (e} 109) 4 2 2) a ue) ) p © = =) = +) 10) 12) i ; 3 Vii “ : ' c 3 Lie \ : “ Pit ny) o. 39, pp. 429-432 22 January 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON PARAMETOPELLA INQUILINUS, NEW SPECIES FROM DELAWARE BAY OYSTER BEDS (AMPHIPODA: STENOTHOIDAE) By Les WATLING ie College of Marine Studies, University of Delaware, \\ Lewes, Delaware 19958 The genus Parametopella is, at present, characterized by the following features (Barnard, 1969): linear article 2 of pereo- pods 4 and 5; uniarticulate palp of maxilla 1; mandible without palp. To date, only three species in this genus are known: P. stelleri Gurjanova from the Russian Arctic, P. ninis Barnard from southern California, and P. cypris (Holmes) from north- eastern North America. This paper describes a new species of Parametopella from the polyhaline waters of Delaware Bay, US.A. STENOTHOIDAE Parametopella Gurjanova, 1938 Parametopella inquilinus, new species Figures 1-2 Diagnosis: Antenna 1, peduncle segments 1 and 2 subequal in length; gnathopod 2 powerful, palm with large teeth near hinge, excavate pos- teriorly with strong tooth on posterior corner; coxal plate 4 greatly ex- panded covering coxae 3 through 7. Description of male: length 3.0 mm. Head small, interantennal angle rounded. Antennae about % length of body. Antenna 1 peduncle article 1 stout, wider than article 2; articles 1 and 2 of subequal length; article 3 about 4% length of article 2. An- tenna 2 peduncle nearly as long as antenna 1; last 2 peduncle articles subequal in length; flagellum of 4 to 6 segments. Mouthparts typical for genus. Mandible without palp; maxilla 1 palp uniarticulate; maxilliped outer plate vestigial, inner plate small, sparsely setose, palp with few, strong setae. 39—Proc. Brox. Soc. WasH., VoL. 88, 1976 (429) 430 Proceedings of the Biological Society of Washington Fic. 1. Parametopella inquilinus, new species: a, Side view of body; b, Antenna 1; c, Antenna 2; d, Third maxilliped; e, Male gnathopod 1; f, Male gnathopod 2. Gnathopod 1 simple; dactyl finely pectinate; propodus with 3 setae on posterior margin; article 5 equal in length to article 6, posterior mar- gin with 3 groups of strong, pectinate setae; distal margin of article 4 with cluster of pectinate setae and group of short spines; article 2 with anterior row of long, simple setae. Gnathopod 2 strong; palm of propodus strongly excavate between distal group of teeth and large proximal tooth; article 5 with cluster of pectinate setae on postero-distal corner; article 3 with thin, transparent, anteriorly-directed projection; article 2 armed with anterior row of short, simple setae; coxal plate large, not covered distally by coxal plate 4, lower margin with several short spines. Amphipod from Delaware oyster beds 431 Fic. 2. Parametopella inquilinus, new species: a, Pereopod 1; b, Uro- some; c, Female gnathopod 1; d, Female gnathopod 2. Pereopod 3 coxal plate short, sharply tapered anteriorly, lower margin lightly armed along anterior portion; article 2 with 3 proximal and 2 distal spines along anterior margin; appendage notable for its lack of setation. Pereopod 4, coxal plate large, extending anteriorly to overlap slightly coxal plate 2 and posteriorly to cover coxal plate 7. Pereopods 5-7, article 2 slender. Urosome segments 2 and 3 fused; uropod 1 outer ramus with single spine, peduncle with 7 spines; uropod 2, peduncle with 2 dorsal and single distal spine, rami without spines; uropod 3 uniramous, unarmed, two segments of ramus equal in length to peduncle. Telson entire, with- out spines. ' Female: Similar to male; gnathopod 1 article 2 anterior margin more setose; palm of gnathopod 2 not as strongly excavate. Holotype: &, USNM No. 152671. Paratypes: 236 2, USNM No. 152672. Type-locality: Oyster beds in Delaware Bay, 75°22’W longitude, 39°12’N latitude; recorded salinity range 18-25%, temperature range 126°C, depth 8 m; substratum consists of hard, oyster shell reefs inter- calated with muddy shells and mud (Maurer and Watling, 1973). 432 Proceedings of the Biological Society of Washington Distribution: This species has thus far been found only in Delaware Bay. It occurs on hard substrata, usually in association with the hydroid Tubularia crocea. When the hydroid is found on fouling plates, Para- metopella inquilinus, Parapleustes aestuarius Watling and Maurer, and Stenothoe minuta Holmes are found in large numbers in the sediment trapped at the base of the colony. In the oyster community, Maurer and Watling (1973) incorrectly identified this species as P. cypris. Relationships: This species differs from P. stelleri (see Gurjanova, 1951) by its lack of an anteriorly-directed process on antenna 1 peduncle segment 1. It differs from P. ninis (see Barnard, 1962) by the large excavation of the palm of gnathopod 2 and by the short coxal plate 3. P. vulgaris can also be distinguished from P. cypris (see Holmes, 1905; Bousfield, 1973) by the form of gnathopod 2, and also by the antenna 1 peduncle segment 1 being shorter than segments 2 and 3 combined, the very long peduncle segments 4 and 5 of antenna 2, and by coxal plate 4 not covering the distal end of coxal plate 2. Etymology: The specific name, from the Latin noun inquilinus = so- journer, refers to its inquilinous life habit. Acknowledgments: The author would like to thank Mr. Tom Dean who brought the specimens to his attention and Dr. Don Maurer whose grants provided the funding under which the specimens were collected. LITERATURE CITED Barnarp, J. L. 1962. Benthic marine Amphipoda of southern Cali- fornia: families Amphilochidae, Leucothoidae, Stenothoidae, Argissidae, Hyalidae. Pacific Nat. 3:116—-163. ——. 1969. The families and genera of marine gammaridean Am- phipoda. U.S. Nat. Mus. Bull. 271:1-535. BousFIELD, E. L. 1973. Shallow-water gammaridean Amphipoda of New England. Cornell University Press, Ithaca. xiv + 312 pp. GuryaNnova, E. 1951. Bokoplavy morej SSSR i sporedel’nykh vod (Amphipoda-Gammaridea). Opred. po Faune SSSR, Akad. Nauk SSSR, 41:1—1029. Homes, S.J. 1905. The Amphipoda of southern New England. Bull. U.S. Bur. Fish. 24:459-529. Maurer, D., anpD Les WatTuLinc. 1973. Studies on the oyster com- munity in Delaware: the effects of the estuarine environment on the associated fauna. Int. Rev. Ges. Hydrobiol. 58:161—201. GH i MH No. 40, pp. 433-438 22 January 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON SHIINOA ELAGATA, A NEW SPECIES OF PARASITIC COPEPOD (CYCLOPOIDA) FROM ELAGATUS (CARANGIDAE) By Rocer F. Cressey Smithsonian Institution, Washington, D.C. 20560 I recently described Shiinoa inauris Cressey and the male » of S. occlusa Kabata from Atlantic and Pacific specimens of Scomberomorus respectively. Subsequent to the submission of that manuscript a third species was collected by Hillary Boyle from the nasal lamellae of the carangid Elagatus bipinnulatus (Quoy and Gaimard) from various Pacific localities. The hosts are in the collections of the California Academy of Sciences and I wish to thank Dr. William Eschmeyer for allowing Ms. Boyle to examine them for parasites. I examined a few speci- mens of Atlantic Elagatus in the Smithsonian collections but did not recover any further material. Examination of more Atlantic Elagatus might produce additional Shiinoa, as did the examination of Atlantic Scomberomorus cited above. All drawings were made with the aid of a Wild Drawing Tube. All material has been deposited in the collections of the Smithsonian Institution. Shiinoa elagata, new species Figures 1-12 Material studied: Holotype female (USNM 152651) from the nasal lamellae of Elagatus bipinnulatus and 4 paratype females (USNM 152652) from the same host collected from the Caroline Islands. One female from Elagatus sp. collected in the Gulf of Thailand (10°40’N, 99°32’E). Two females from Elagatus sp. from the same locality. One female from Elagatus sp. from Vanikoro Is. (11°41’S, 166°50’E). All additional material collected from the nasal lamellae of the host. 40—Proc. Brow. Soc. WasH., Vou. 88, 1976 (433 ) 434 Proceedings of the Biological Society of Washington Parasitic copepod from Elagatus 435 Description: Female.——Body form as in Fig. 1. Total length, 1.9 mm; greatest width, 1.25 mm, measured at widest part of genital segment. Body form more robust than S. inauris and S. occlusa. Rostral area pro- duced anteriorly, curved ventrally, covering recurved second antennae as in other Shiinoa species. Cephalon posterior to second antennae con- stricted, followed by laterally expanded area bearing mouthparts and first legs. Cephalon comprising about % body length. Thoracic seg- ments bearing second and third legs short, narrower than preceding and following segments with third leg appearing to be incorporated into genital segment ventrally. Genital segment expanded laterally (see Fig. 10), somewhat longer than wide (1.3 x 1.2 mm). Abdomen (Fig. 2) segmentation obscured. Caudal rami (Fig. 5) longer than wide (17lu x 112), bearing one lateral, one subterminal, and 4 terminal setae; all setae naked. First antenna (Fig. 3) 7-segmented with short naked setae as indi- cated in figure, aesthaete present on last segment (seen in young forms only). Second antenna (Fig. 4) recurved within rostrum, terminal half with hyaline sheath, basal segment with 2 short, blunt setae along inner edge as indicated in figure, terminal segment with single short seta. Mouthparts (Figure 5) somewhat removed from first and second an- tennae and situated on ventral protuberance (see Fig. 1). Mandible terminating in long flagellum bearing stout pyriform spines along outer edge and smaller spinules along inner; inner seta arising near base of flagellum bearing spines on outer and inner edges. First maxilla, small, with 3 terminal setae and spinules as indicated in figure. Second maxilla terminating as stout heavily spinose process with 2 setae near base of spinose area. Maxilliped absent. First leg (Fig. 6) with 3-segmented rami; exopod first and second segments with short spine on outer distal corner, last segment with 3 terminal spines and 3 short inner setae; endopod first segment with short seta on inner distal corner, second segment with prominent spinelike process on outer distal corner, last segment with 2 terminal prominent spinelike processes and 3 short inner setae; outer edges of both rami with spinules. Second leg (Fig. 7) as in first leg except exopod second seg- ment with inner seta and small patch of spinules on basipod. Leg 3 (Fig. 8) uniramose, bearing 2 terminal spines, 1 seta, and patches of spinules as indicated in figure. Legs 4 and 5 absent. Leg 6 (Fig. 9) represented by sclerotized area bearing 1 short seta and 1 short spine at area of egg sac attachment. Egg sacs broken in ovigerous females but appear to be of the usual cyclopoid type with many eggs. Male.—unknown. & Fics. 1-5. Shiinoa elagata n. sp., female: 1, Dorsal; 2, Abdomen and caudal rami; 3, First antenna; 4, Second antenna; 5, Mouthparts. 436 Proceedings of the Biological Society of Washington Parasitic copepod from Elagatus 437 Fics. 10-12. Shiinoa elagata n. sp., young female: 10, Ventral; 11, Abdomen, ventral; 12, Caudal ramus, ventral. Young female.—a few specimens of smaller non-ovigerous females are present in the material examined and these forms exhibit a few differ- ences from the mature females as follows; the ventral aspect (Fig. 10) was drawn from a young female but its general body form does not differ from the mature specimens. The abdomen (Fig. 11) segmentation is more distinct. The caudal rami (Fig. 12) each bear a lateral seta and 2 additional terminal setae. The first antenna bears more setae terminally (at least 7) and an aesthaete. The knobs present along the outer margins of the rami of legs 1 and 2 are pointed spinules in the younger forms. < Fics. 6-9. Shiinoa elagata n. sp., female: 6, First leg; 7, Second leg; 8, Third leg; 9, Sixth leg and area of egg sac attachment. 438 Proceedings of the Biological Society of Washington Remarks: The new species can be easily separated from S. inauris since the rami of legs 1 and 2 of inauris are 2-segmented rather than 3. The new species seems more closely related to S. occlusa but can be separated from it by the nature of the body form, the greater number of setae on legs 1 and 2 of S. elagata and the patches of spinules on leg 3 of elagata. Etymology: elagata. Latin, feminine, referring to the generic name of the host from which it was collected. LITERATURE CITED Cressey, R.F. 1975. A new family of parasitic copepods (Cyclopoida: Shiinoidae). Crustaceana 28(2):211-219. Kaxpata, Z. 1968. Copepoda parasitic on Australian fishes VII. Shiinoa occlusa gen. et sp. nov. Jour. Nat. Hist. 2:497—-504. Ne tithe : Den eo fo LO eee ee 3, No. 41, pp. 439-446 22 January 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON y ORCONECTES SAXATILIS, A NEW SPECIES OF “ CRAYFISH FROM EASTERN OKLAHOMA = By RayMonp W. BoucHarD AND JupITH W. BoucHARD \c : National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560 A new species of crayfish from the Kiamichi River system (Red River basin) was discovered during a trip to western Arkansas and eastern Oklahoma. This new species brings to 10 the number of representatives of Orconectes in Oklahoma. Orconectes leptogonopodus Hobbs (1948:146), O. menae (Creaser 1933:5), O. acares Fitzpatrick (1965:87), Procam- barus (Tenuicambarus) tenuis Hobbs (1950:194) and the spe- cies described herein all appear to be endemic to the Ouachita Mountains section. We wish to thank Horton H. Hobbs, Jr., of the Smithsonian Institution who reviewed the manuscript. We also are in- debted to the Smithsonian Institution for providing a Smith- sonian Postdoctoral Fellowship to the senior author. Orconecies saxatilis, new species Figure la—k Diagnosis: Body and eyes with pigment. Rostrum with marginal spines. Areola 6.0 to 9.3 times longer than broad, constituting 28.7 to 29.9 percent of total length of carapace (37.8 to 38.8 percent of post- orbital carapace length) and with 3 punctations across narrowest part. No cervical spines or tubercles present. Hepatic spines absent; branchio- stegal spine small and acute; suborbital angle weakly developed; post- orbital ridges moderately well developed and terminating cephalically in acute, corneous-tipped spine. Antennal scale longer than broad, broadest slightly distal to midlength; lamellar portion with broadly an- gulate to slightly declivous margin. Chela with 2 subserrate rows of low tubercles along mesial margin of palm; scattered smaller tubercles over dorsomesial half of palm; small setal tufts over entire dorsal sur- face; moderately well developed longitudinal ridges on dorsal surface 41—Proc. Brow. Soc. WasH., Vou. 88, 1976 (439) 440 Proceedings of the Biological Society of Washington Fic. 1. Orconectes saxatilis, new species. a, Lateral view of carapace of holotype; b, Mesial view of first pleopod of holotype; c, Mesial view of first pleopod of morphotypic male; d, Basipodite and ischiopodite of third pereiopod of holotype; e, Antennal scale of morphotypic male; f, Lateral view of first pleopod of morphotypic male; g, Lateral view of first pleopod of holotype; h, Dorsal view of chela of holotype; i, Epistome of allotype; j7, Annulus ventralis of allotype; k, Dorsal view of carapace of holotype. Crayfish from eastern Oklahoma 44] of fingers. Hook on ischium of third pereiopod of first form male over- reaching basioischial articulation. First pleopod of first form male with central projection corneous, tapering and reaching basis of first pair of pereiopods when abdomen flexed; shoulder present on cephalic sur- face; mesial process slightly shorter, non-corneous, with distal end curved mesiad. Annulus ventralis symmetrical; cephalic half divided by medial trough and bearing 2 caudally directed protuberances overhanging cen- trally located fossa; sinuate sinus extending from fossa to caudal edge of sclerite; tongue-like projection extending into fossa. Holotypic male, form I: Body subovate and somewhat laterally com- pressed (Figs. la, k). Abdomen narrower than thorax (9.0 and 9.6 mm). Greatest width of carapace greater than depth at caudodorsal margin of cervical groove (9.6 and 8.9 mm). Areola 9.3 times longer than wide with 3 punctations across narrowest part; length of areola 29.3 percent of entire length of carapace (38.0 percent of postorbital carapace length). Rostrum with thickened, concave margins terminating in large, acute, corneous spines. Acumen terminating in small, upturned, corneous tu- bercle reaching nearly to end of antennular peduncle. Rostrum excavate dorsally with submarginal punctations and scattered additional ones. Postorbital ridge moderately developed, grooved dorsolaterally, terminat- ing in prominent, acute, corneous tubercle. Suborbital angle poorly de- veloped; branchiostegal spine large, acute and cormeous. No cervical spines present; hepatic area tuberculate; dorsal surface and branchio- stegal portions of carapace punctate. Abdomen longer than carapace (24.3 and 22.2 mm); pleura of mod- erate length with caudoventral extremity subangular. Cephalic section of telson with single movable and immovable spine in each caudolateral corner; partly separated from caudal section by 2 oblique incisions. Basal podomere of uropod with spine extending over mesial and lateral rami. Lateral ramus of uropod with median and submedian ridges, former terminating in acute spine at transverse flexure. Proximal part of lateral ramus with row of small spines distally and large movable spine sub- marginally at caudolateral corner. Mesial ramus of right uropod with median ridge terminating distally in premarginal acute spine, left miss- ing; lateral margin terminating distally in acute spine. Dorsal surface of telson and uropods lightly setiferous. Cephalic lobe of epistome (Figure li) triangular and narrow with small cephalomedian projection and thickened, upturned cephalolateral margins. Main body of epistome with very shallow median fovea and pair of obliquely disposed oval fossae immediately cephalic and sub- parallel to thickened, arched epistomal zygoma. Proximal segment of antennule with small spine on ventral surface near base of proximal third. Antennae broken. Antennal scale (Figure le) with broadly an- gulate lamellar margin, broadest slightly distal to midlength; thickened lateral part terminating in prominent, acute, corneous-tipped spine pro- jecting anteriorly beyond tip of rostrum. Left chela (Fig. 1h) (right chela missing) with mesial margin of palm 442 Proceedings of the Biological Society of Washington bearing primary subserrate row of 9 tubercles and secondary row of 8 slightly smaller tubercles on dorsal surface lateral to primary row; scattered small tubercles over dorsomesial half of palm; distoventral sur- face of palm with 2 corneous-tipped spines. Lateral surface of propodus weakly costate with row of punctations rendering proximolateral base slightly impressed dorsally; dorsal and ventral surfaces with submedian ridges flanked by setiferous punctations; opposable surface with row of 6 tubercles along proximal three-fifths of finger, fourth from base largest; additional acute tubercle present on lower level near base of distal fourth; single row of minute denticles extending proximally from tip to third tubercle from base, interrupted by each of fourth through sixth tubercles. Dorsal and ventral surfaces of dactyl with median longi- tudinal ridges flanked by setiferous punctations; opposable margin with row of 8 tubercles, fourth from base largest; row of denticles extending proximally from tip to third tubercle from base, interrupted by fourth through seventh tubercles; mesial surface with irregular rows of tu- bercles on approximately basal half with single row of progressively smaller tubercles extending distally. Fingers terminating in large, acute, comeous spines. Left carpus with deep oblique furrow dorsally; mesial surface with large, procurved, corneous-tipped spine near midlength and 8 smaller, scattered tubercles; distoventral margin with 2 large, corneous-tipped spines, medial one procurved; dorsal surface with small, acute, corneous- tipped spine distomesially and row of 4 small tubercles dorsomesially; podomere otherwise punctate. Upper surface of left merus with 2 prominent corneous-tipped spines and 3 similar but much smaller ones at distal margin; ventral surface with lateral row of 5 cornified tubercles, distal 2 largest and mesial row of 11 similar tubercles; large corneous-tipped tubercle at lateral artic- ulation. Left ischium with row of 3 small tubercles on ventromesial margin, first cornified, other 2 broken or abraded. Hook on ischium of third pereiopod only (Figure 1d); hook simple, overreaching basioischial articulation and not opposed by tubercle on basis. Coxae of fourth and fifth pereiopods without prominent caudo- mesial boss. (See Measurements. ) First pleopod (Figs. 1b, g) reaching basis of first pereiopod ie abdomen flexed. (See Diagnosis for description. ) Allotypic female: Similar to holotype in most respects but differing as follows: areola constituting 29.9 percent of entire length of carapace (38.8 percent postorbital carapace length), and width 7.2 times longer than broad. Postorbital ridges and rostrum with larger spines. Antennal scale with more declivous margin on lamellar portion. First chelipeds regenerated. Sternum between third and fourth pereiopods broadly U- shaped and shallow. Annulus ventralis (Figure 1j) firmly attached to sternum cephalically. See Diagnosis for description of annulus ventralis. Morphotypic male, form II: Differing from holotype in following re- spects: rostrum and postorbital ridges terminating in larger spines. Crayfish from eastern Oklahoma 443 Areola constituting 29.1 percent of entire length of carapace (38.5 percent of postorbital carapace length), and 7.8 times longer than broad. Mesial margin of palm of right chela (left regenerated) with primary subserrate row of 10 tubercles. Propodus of right chela with 7 tubercles along opposable margin; single row of denticles interrupted by tubercles 4 through 7. Opposable margin of dactyl with row of 7 tubercles, first and fourth from base largest. Carpus with 6 and 9 tubercles on mesial and dorsal surfaces respectively; 2 subequal, corneous-tipped spines and 3 rounded tubercles on ventral surface. Upper surface of merus with 3 corneous-tipped spines and 3 of 4 tubercles at distal margin cornified; ventromesial margin with 12 tubercles; ventrolateral margin with distal tubercle largest. Ischium of right cheliped with 4 acute, corneous-tipped tubercles. Hook on ischium of third pereiopod much reduced, not reach- ing basioischial articulation. First pleopods (Figs. lc, f) long and of uniform texture, neither element corneous; central projection longer and curved at tip; mesial process straight. Type-locality: Pigeon Creek at Oklahoma State Highway 63 (Red River basin via Kiamichi River system), LeFlore County, Oklahoma. The creek bottom is comprised of masses of cobbles and numerous rocks of various sizes. During the summer the flow is sluggish with visible water confined to isolated pools, the latter connected by subsurface flow. On 11 August 1974 the clear water was less than 2 feet deep with a temperature of 24°C. Pigeon Creek is approximately 15 feet wide with oc- casional concentrations of the emergent aquatic plant, Justicia americana. Disposition of types: The holotypic male, form I (USNM 146577), the allotype (USNM 146578) and the morphotypic male, form II (USNM 146579) are deposited in the National Museum of Natural History, Smith- sonian Institution. Paratypes consisting of 1¢II, 29, 2¢ juv. and 29 juv. are deposited also in the Smithsonian Institution, and 1 ¢1, 1 ¢ II and 2@ are in the collection of the senior author. Range and specimens examined: This crayfish is known only from the type-locality and the type-series as indicated above. Probably the spe- cies will prove to be more widespread in the upper Kiamichi River system. Variation: Other than the usual range in meristics, morphometrics and setation, there is little variation in the available material. (See Diagnosis.) Abrasion of spines or tubercles in late intermolt stages are noticeable. Size: The largest specimen available is a female with a carapace length of 26.1 mm (postorbital carapace length 20.2 mm). The smallest first form male has corresponding lengths of 22.0 and 17.0 mm. Color notes: Cephalothorax and abdomen concolorous sienna. Bran- chiostegites and hepatic region lighter. Caudal edge of cephalothorax with prominent dark brown band. Caudal edges of abdominal terga with narrow, dark brown bands; pleura with red crescentic markings. Tail fan with narrow red margination. Pair of dark vermiculated blotches anterior to cervical groove marking attachment of mandibular muscles. Ventral aspects of cephalothorax and abdomen white. 444 Proceedings of the Biological Society of Washington TABLE 1. Measurements (mm) of Orconectes saxatilis Holotype Allotype Morphotype Carapace Height 8.9 10.1 8.5 Width 9.6 eat 9.5 Total length BNO) 26.1 21.3 Postorbital length 17.1 20.1 16.1 Areola Width 0.7 11 0.8 Length 6.5 7.9 6.2 Rostrum Width 2.6 2.9 2.4 Length 5.1 6.0 5.2 Chela Length, mesial margin palm 5.3 —* 5.2 Width, palm 7.5 — 6.3 Length, lateral margin 17.9 — 15.5 Length, dactyl 10.9 — 9.4 * Chelae regenerated Chelae sienna dorsally, white ventrally. Distal ends of fingers red. Pereiopods lighter sienna dorsolaterally, cream to white ventrolaterally; distal podomeres darker dorsally than proximal ones. Rostral margins dark brown; antennae sienna and tubercles on op- posable margins of fingers yellow. Life history notes: First form males have been collected only during the month of August. A large percentage of the adult male population is probably in reproductive form during late summer through early spring. No ovigerous females were collected but egg laying probably occurs during late winter to early spring. Ecological notes: Orconectes saxatilis was collected under rocks in the pool areas. Taken with this new species at the type-locality were Orconectes palmeri longimanus (Faxon 1898) and Procambarus (Tenui- cambarus) tenuis Hobbs (1950). Orconectes palmeri was the dominant crayfish at the type-locality. Unlike the usual habitat of the adults of its close ally, O. menae, which prefers the deeper, faster riffles and runs, the habitat of O. saxatilis had little current. The above mentioned as- sociates are also more common in quieter waters. Relationships: Orconectes saxatilis has its closest affinities with Or- conectes menae (Creaser 1933) which is known from the Ouachita and Red River systems of the Ouachita province. It differs from O. menae primarily in possessing longer gonopods with a distinct cephalic shoulder Crayfish from eastern Oklahoma 445 in the first form male (not reaching forward to the bases of the first pair of pereiopods in O. menae) and a more uniform color pattern (dark stippling or rosette patterns are obvious on O. menae). Orconectes saxa- tilis and O. menae share a unique combination of morphological char- acters such as a narrow, triangular cephalic lobe on the epistome; the absence of cervical spines; and lack of the caudally directed expansion of the annulus ventralis. Etymology: saxatilis, L., found among rocks; so named because of the habitat at the type-locality which is dominated by numerous rocks. LITERATURE CITED CREASER, Epwin P. 1933. Descriptions of some new and poorly known species of North American crayfishes. Occ. Pap. Mus. Zool., Univ. Michigan 275:21 pp. Faxon, WALTER. 1898. Observations on the Astacidae in the United States National Museum and in the Museum of Comparative Zoology, with descriptions of new species. Proc. U.S. Nat. Mus. 20( 1136) :643-694. FITZPATRICK, JosEPH F., Jk. 1965. A new subspecies of the crawfish Orconectes leptogonopodus from the Ouachita River drainage in Arkansas. Tulane Stud. Zool. 12(3):87-91. Hosss, Horton H., Jr. 1948. Two new crayfishes of the genus Or- conectes from Arkansas, with a key to the species of the Hylas Group (Decapoda, Astacidae). Amer. Midl. Natur. 39(1): 139-150. —. 1950. A new crayfish of the genus Procambarus from Okla- homa and Arkansas (Decapoda, Astacidae). Jour. Washington Acad. Sci. 40(6): 194-198. 446 Proceedings of the Biological Society of Washington GH | nk BAx MH 88, No. 42, pp. 447-458 22 January 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ye TWO NEW SPECIES OF OCTOPODS OF THE GENUS/ GRANELEDONE (MOLLUSCA: CEPHALOPODA) //—~ FROM THE SOUTHERN OCEAN! ox By Gupert L. Voss Rosenstiel School of Marine and Atmospheric Science, University of Miami, Miami, Florida 33149 The cephalopod collections taken from Antarctic seas by the USNS ELranin are rich in benthic octopods. These are being worked upon by the writer and the final results will form a monographic study of the octopods of the Southern Ocean. Because of the time involved in working up the col- lections and the complexities of the taxonomic problems, the descriptions of the new species are being published separately in a series of papers in order to make them immediately avail- able to other students of the group. ‘The genus Graneledone is poorly known. Most of the spe- cies descriptions are inadequate for identification and com- parisons, and, due to the poor state of preservation of most of the material, little is known concerning the internal anatomy of the component species. The present paper helps to remedy this situation and is preliminary to a more detailed and com- prehensive study of the genus. I wish to thank those responsible for the collection and preservation of the specimens, for their care in handling them, and for their excellent state of preservation. I also wish to thank Dr. George Llano, head of the Biology Program of the NSF Office of Antarctic Programs, for making this work pos- sible. This research was supported by National Science Foun- 1 Scientific Contribution from the Rosenstiel School of Marine and Atmospheric Science, University of Miami. This paper constitutes a scientific report to the Na- tional Science Foundation. 42—Proc. Bion. Soc. WAsH., Vou. 88, 1976 (447) 448 Proceedings of the Biological Society of Washington dation grants GA 0253, GA 1493 and BMS 70-00851 A04 for which I am grateful. The illustrations of Graneledone macrotyla are by Constance Stolen McSweeny; the drawings of G. antarctica are by the writer; Roger Hanlon took the photographs. The measure- ments and indices used are those defined by Voss (1963:11). The types are deposited in the U.S. National Museum of Nat- ural History. Graneledone antarctica, new species Figures la-i, 2 Material studied: Holotype—male, mantle length 41 mm, from Ross Sea, Eltanin Sta. 2110, 74°05.6’S, 175°05.2’W in 2341 m with 10-foot Blake trawl, February 8, 1968, USNM 729679. Paratypes—male, mantle length 39 mm, female, mantle length 45 mm, UMML 1667. Male, mantle length 38 mm, 3 females, mantle length 25-41 mm, USNM 729680. (All paratypes with same data as holotype. ) Description: The mantle is short, broadly rounded posteriorly, and very wide. It is distinctly flattened dorso-ventrally. The head is set off from the mantle by a slight constriction. The head is nearly as wide as the mantle, flattened, and bears large conspicuous eyes. The funnel is of moderate size; it is free for its anterior half which is tubular and tapered. The funnel organ is VV-shaped but shows con- siderable variation (Figs. la, b). It is composed of 2 elongate oval pads slightly to moderately split anteriorly with pointed to blunt tips. The arms are long and rather stout; the arm order is 1.2.3.4, either I or II always being the longest and IV always the shortest. The web is moderately deep with the formula C=B.D.A=E but showing some individual variation. The web extends only a short distance up the dorsal side of each arm after which there is no trace. On the ventral side of I, IJ, and III the web extends to the tip of the arm. It is low in the basal half but at about the midpoint of the arm it broadens, becoming widest at about the distal 15 of the arm. Only the extreme tip of the arm is free. On preservation, the contraction of the web curls the arm tip in a pronounced fashion. The arm web attains its greatest develop- ment on I and II. The suckers are small and in a single row on each arm. They are largest near the base of the arm and regularly decrease in size toward the arm tip. The third right arm is hectocotylized in the male. It is bordered ventrally by a membrane or web, its outer margin thickened, cream- colored, and rolled outward. The contraction of this thickened border curls the arm tip downward resembling a hook or a slight S-curve. This shape is probably not found in the living animal but is due to the action of the preservative. The tip of the arm bears a small ligula (Fig. Ic). It is spoon-shaped, pointed distally, and has thickened margins. The Two octopods from Southern Ocean 449 Fic. 1. Graneledone antarctica, new species. a—b, Funnel organs; c, Hectocotylus of holotype; d, Digestive tract of female; e—f, Mandibles; g, Radula of 39 mm specimen; h, Radula of 38 mm specimen. oral face is crossed by about 12-13 low fleshy folds. The calamus is projecting, low, and blunt. The gills are short, stout, and contain 6 lamellae on the outer demi- branch. None of the specimens appear to be sexually mature. The holotype had a well developed hectocotylus and penial apparatus but there were no spermatophores in Needham’s sac nor were any found in the other males. 450 Proceedings of the Biological Society of Washington Fic. 2. Graneledone antarctica, new species. Upper figures of holo- type; lower figure of 38 mm paratype. None of the females contained developed eggs. The digestive tract was dissected from one of the large females (Fig. 1d). The buccal mass is large and contains large anterior salivary glands. The paired posterior salivary glands are very small and poorly developed. The esophagus leads from the buccal mass posteriorly to the crop which it enters dorsally. The latter has no diverticulum. Posteriorly the esopha- gus and crop lead into a stout portion which connects with a moderately large two-parted stomach, one part thick-walled and muscular, the other thin-walled. The spiral caecum is small, stout, smooth externally, and united with the small heart-shaped liver by paired hepatopancreatic ducts. The intestine is stout, thin-walled, and leads anteriorly to the anus which does not appear to have anal flaps. The crop was opened Two octopods from Southern Ocean 451 in one specimen. It contained an amorphous mass of animal tissue in which numerous polychaete bristles were embedded. The beaks show no unusual features (Figs. le, f). Radulas were removed from two specimens. While there is no ques- tion that all of the specimens are conspecific and show a remarkably close conformity in other characters, the radular teeth vary greatly. The radula from specimen No. 2 (Fig. 1g) has only 5 teeth in each trans- verse row, none of them showing any great dissimilarity. The rachidians are undifferentiated but somewhat larger than the second laterals and without cusps. The admedians or first laterals are missing. The second laterals are only slightly curved with small bases; the third laterals are largest, tallest, and broadest, with conspicuous bases. There is no trace of marginal plates. The radula of specimen No. 3 (Fig. lh) has rather large, broad rachid- ians with broad bases, perhaps somewhat distorted in the figure by a sidewise orientation in the mount. The admedians are a little shorter than the second laterals, narrow and sharp. The second laterals are larger and more curved. The third laterals are very broad, almost flat- tened at their tips and show a slightly irregular outline. All of these teeth stand erect, are thin and transparent, and in No. 3 have very ir- regular almost root-like bases. All of the specimens were fixed in 10 percent buffered formalin after brief emersion in fresh water to kill them. They were thus fixed without undue distortion of the arms and were later transferred to 70 percent alcohol. They are in remarkably fine condition for examination. The general consistency is muscular with a thick overlying layer of almost gelatinous material, particularly posteriorly on the mantle and slightly less so dorsally, around the head and eyes, and surrounding the brachial crown. The surface is covered by a thin, very tough skin which is covered dorsally on the mantle, head and arms by numerous, evenly spaced, close-set warts (Fig. 2). These consist of a raised mound bear- ing from one to over a dozen small, cone-shaped papillae. They are very regularly distributed. There are none on arms IV; arms II and III are liberally covered dorsally but there are none ventrally. There is a dis- tinct line of fine warts on the periphery of the mantle arranged like a keel running from the corners of the mantle aperture around the mantle posteriorly. Below the keel there are no warts and the surface is smooth. There are no warts beneath the head except for a few bordering the lower eyelid and immediately adjacent to it. There is a distinct circlet of warts on the eyelid (Fig. li). These are somewhat larger than the other warts and two or three of them over each eye are greatly enlarged and, although not erected, probably represent ocular cirri. The color of specimens in alcohol is a pale yellowish brown dorsally, suffused with purplish hues at the base of the brachial crown and on and between the arm bases. On this ground color the mantle, head, funnel, and the base of the arms are reddish brown suffused with purple. The spermatophoral groove on the third right arm is yellowish. 452 Proceedings of the Biological Society of Washington TasLEe 1. Measurements (in mm) of 3 male specimens of Graneledone antarctica, new species, from ELTANIN Sta. 2110. Holotype No. a 2 3 Mantle length Al 39 38 Mantle width 44 37 40 Head width 40 34 38 Arm length I 164 165 138 138 138 136 II 164 164 123 130 127 136 III 150 39145 1 105) 119 +117 IV 3 Silos 103 98 1S Teel Os Total length 208 179 178 Arm width 8.0 7.0 7.0 Hect. arm length 145 105 117 Ligula length 45 4.0 3.8 Calamus length 2.0 2.0 15) Sucker diameter 3.3 2.3 DD Web depth A 28 30 25 B 46 47 35 38 33 A? € A5 52 35 35 38 38 D 33 ON 30 Gills 6 6 6 TaBLE 2. Indices of bodily proportions of three males of Graneledone antarctica, new species, from ELTANIN Sta. 2110. No. 1 2 3 iN Range S.D. ML A. 38) 2 8 BOs 28] 4 1.52 MWI 107.3 94.9 105.3 94.9 — 102.5 — 107.3. 6.65 HWI1 97.6 872 100.0 87.2 — 94.9 — 100.0 6.80 MAI 249 283 275 24.9 = 26:9 = 28: sueanlean ALI OS THO Wes Mu Wio= TOS i120 AWI 195 180 184 18.0 = 18:6 = WOON WDI 315 27.5 304 27.5 = 29.8 = Sie ao0e SIn 30 39 5.8 58— 6.6 8.0 1.24 HceAI MS Gil G48 76.1 = 82:9 = ‘S7SanGlon LLI Ball 28 3.2 Mi = BO 3.2 0.20 Cit 444 50.0 39.5 39.5 — 446 -— 50.0 5.25 PLI Arm formula 1.2.3.4 1.2.3.4 1.2.3.4 Web formula CBDEA BCDAE CBDE Gills 6 6 6 A Two octopods from Southern Ocean 453 TABLE 3. Measurements (in mm) of four females of Graneledone ant- arctica, new species, from ELTANIN Sta. 2110. No. 4 5 6 W Mantle length 45.0 4] 39 25 Mantle width 43 37 36 23 Head width 40 32 34 23 Arm length I 131 134 99 102 116 115 60 62 II 134 134 98 99 108 114 6 il Ill 122 118 97 93 96 102 58 60 IV 198+ 114 86 Sie GSCI 5) 5 Total length Wars 143 157 88 Arm width 8.0 Bl 6.0 6.0 Web depth A 32 25 28 17 B Al 42 30 32 32 37 18 — C 42, 43 35 33 32 37 21 20 D OM 38 32 32 32 Sil Ig ily E Bill 20 22 17 Gills 6 6 6 6 -+ = tip broken Holotype: U.S. National Museum of Natural History 729679. Type-locality: Ross Sea, Antarctica, ELTANIN Sta. 2110, 74°05.6’S, 175°05.2’W in 2341 m. Discussion: Two other species assigned to Graneledone must be con- sidered: G. challengeri (Berry, 1916) and ? G. setebos Robson, 1932. G. antarctica superficially resembles G. challengeri in its bodily pro- portions. It differs from it in the structure of the funnel organ, hecto- cotylus, radula, and sculpture. The funnel organ in antarctica is formed of double oval pads slightly to moderately spilt anteriorly; in challengeri these organs are typically VV-shaped with narrow, pointed, anterior TasLE 4. Indices of bodily proportions and counts of four females of Graneledone antarctica, new species from ELrantn Sta. 2110. No. 4 5 6 8 Range S.D. ML 45 Al 39 25 25 — 37.5 — 45 8.69 MWI 5 C0 O23 CeO 90.2 — 92.5 — 95.6 DNB) HWI SoS) —WSpl Moll 92.0 78.1 — 80.5 — 92.0 3.97 MAI 33.6 40.2 33.6 40.3 33.6 — 36.9 — 40.3 3.83 ALI 1.0 —s 8) 7 70.5 — 72.8 — 75.7 2.39 AWI WES 13.4 aoe 2420 13.4 — 17.6 — 24.0 4.61 WDI Soll 34.3 Sil) BB) 31.9 — 33.1 — 34.3 122, 454 Proceedings of the Biological Society of Washington limbs. In antarctica the ligula is small and moderately well differentiated; in challengeri the ligula is well formed, deeply excavated, with sharply differentiated calamus. The radula of antarctica is highly variable but shows a general uniformity of teeth; in challengeri the radula is more typically octopodan. The sculpture of antarctica varies from challengeri in the more scattered, more widely separated tubercles, and the presence of 2 to 3 much larger tubercles over the eyes. The wide membrane on the ventral side of the arms in antarctica also seems distinctive. There is no point of comparison with ? G. setebos for the reasons given in the general discussion. The etymology of the name is self-evident. Graneledone macrotyla, new species Figures 3a—g Material studied: Holotype, a female, mantle length 34.5 mm, EL- TANIN Sta. 1592, 54°43’S, 55°30’W in 1647-2044 m with 10-foot Blake trawl, March 14, 1966. USNM 729678. Description: Only a single specimen of this species was found in the collections. It is, however, so different from any known species of Graneledone that I do not hesitate to describe it as new. The mantle wall is thick and muscular with some semigelatinous ma- terial forming an outer layer. The mantle is large, round, but somewhat dorsoventrally flattened, and is very wide. There is no noticeable con- striction between the mantle and the head (Fig. 3a, b). The mantle aperture is small and the funnel-mantle locking apparatus is weakly developed. The funnel is stout and tubular but it is united to the ventral surface of the head for most of its length. The funnel is VV-shaped and stout with broad lateral limbs slightly stouter than the median limbs (Fig. 3c). The head is broad with medium sized eyes which do not project. There is a slight constriction between the head and the bases of the arms. The brachial crown is well developed. The arms are moderately long, stout, and taper gradually to slender points. The arm formula is 2.3.1.4. The suckers are small and crowded together but arranged regularly in a single row. The web is moderately deep. The web formula is C.D.B = A.E. The web extends as a broad membrane up the ventral side of each arm nearly to the tip. The mantle had been opened in search of mesozoan parasites and some of the internal organs were damaged. The gills are small, about equally developed in each demibranch, with 7 lamellae, including the terminal ones, on the outer demibranch. The digestive tract was dissected (Fig. 3d). The beaks offer no par- ticular differences but are well developed (Fig. 3e). The radula con- sists of a rachidian with asymmetrically arranged cusps as shown in the Two Octopods from Southern Ocean 455 Fic. 3. Graneledone macrotyla, new species. Holotype: a—b, Dorsal and ventral views; c, Funnel organ; d, Digestive tract; e, Mandibles; f, Radula; g, Details of tuberculations. figure (Fig. 3f). The admedian is small and somewhat bicuspid with the outer cusp more pronounced. The second lateral has a broad inner cusp with a short base. The third lateral is short, stout, and broad. The marginal plates are roughly triangular and rugose, the ridges or plica- tions extending longitudinally of the plate. 456 Proceedings of the Biological Society of Washington TaBLE 5. Measurements (in mm) and indices of the female holotype of Graneledone microtyla, new species. Mantle length 34.5 Mantle width 38.0 MWI 110.0 Head width 36.0 HWI 104.0 Arm length I 81.0 ALI 70.0 II 87.0 MAI 39.7 Ill 83.0 IV 80.0 Sucker diameter 2.5 Sl, U2 Arm width 7.5 AWI DLT Length of gills 8.0 Gills W Total length 124.0 Web depth A 24.0 WDI Ono B 24.0 C 28.0 D 25.5 E 20.0 Arm formula 2.3.1.4 Web formula CDB = AE The anterior salivary glands are small; the posterior salivary glands are triangular and small. The esophagus is stout and along much of its length is slightly dilated forming an indistinct crop of the type shown by Robson (1932, Fig. 30b) for Bathypolypus and which in fact is not a crop. This dilated portion of the esophagus leads into a strongly dif- ferentiated two-parted stomach and a distinct large, spiral caecum. The intestine is large, slightly dilated, and bent almost double upon itself. The anal pore is round, without the usual anterior and posterior folds or lateral flaps. There is no trace of an ink sac. The crop and stomach were opened but contained little food remains. None was identifiable. The genitalia unfortunately were removed at the time of capture while searching for mesozoans. The sculpture is distinctive (Fig. 3a, b, g). There is a low, thin, pe- ripheral keel or raised line around the mantle. Above the keel the sur- face is covered with well separated, distinct rugosities consisting of simple, sharp, cartilaginous tubercles, slightly larger groups of simple tubercles, and large areas 3-7 mm in diameter consisting of a pale raised area with small, simple tubercles surrounding an erect, central, sharp spine. There is a row of tuberculous papillae around each eye and a gigantic tuberculate cirrus about 10 mm in diameter above each eye. A color photograph of the living animal taken aboard the ship shows the two cirri erected to a height of about 15 mm. The tuberculations extend Two octopods from Southern Ocean 457 onto the web and bases of the arms nearly to the level of the web mar- gin. There are no tubercles on arms III and IV nor on the ventral sur- face of the head and mantle. The color is a brownish red with deeper purple in the area of the brachial crown. The raised pale areas are creamy brown. Holotype: U.S. National Museum of Natural History 729678. Type-locality: Near the Falkland Islands, Eltanin Sta. 1592, 54°43’S, 55°30’W in 1647-2044 m. Discussion: G. macrotyla resembles no other member of the genus in its superficial appearance. The few tubercles varying in size from minute to extraordinarily large ones, with a single enormous tubercle over each eye, immediately separates this species from all others in the genus. The radula is even more octopodan in general shape than that of G. challengeri which it most closely resembles in this feature. The name macrotyla is derived from the Greek meaning large knobs and refers to the large tuberculations characteristic of the species. GENERAL DISCUSSION The genus Graneledone Joubin, 1918 may be diagnosed as follows: suckers uniserial; ink sac absent; funnel organ VV-shaped; crop reduced or absent; gills small; hectocotylus small; mantle and arms covered with small to large cartilaginous spiny warts. Robson (1932), listed four taxa: G. verrucosa (Verrill, 1881), G. verrucosa media (Joubin, 1918), G. challengeri (Berry, 1916), and ? G. setebos Robson, 1932. G. verrucosa, and its possible subspecies media, is a North Atlantic species about which surprisingly little is known. From the literature, however, it differs considerably from the new species under considera- tion. At present, I am inclined to believe that the two forms do not war- rant separation and represent a northern hemisphere species. G. challengeri, of which the type is from off the Kermadec Islands, is confused because Robson (1932:311) included in his description Hoyle’s (1904:21) Moschites verrucosa from the Gulf of Panama. This latter ‘specimen needs reexamination, including study of other now available specimens from the same region. ? G. setebos was first described by Massy as Moschites sp. (Massy, 1916:159). Robson (1932:313) placed the species in Graneledone with- out giving a single character to support his decision. None of the char- acters given either by Massy or Robson indicates that the specimen was a Graneledone. The specimen was taken dead in a rock pool and was badly decomposed so that no observations were possible concerning skin and sculpture, color, funnel, funnel organ, ink sac, etc. In my opinion G. setebos is a species dubia and should be dropped from further con- sideration. LITERATURE CITED Berry, S. StinLMAN. 1916. Cephalopoda of the Kermadec Islands. Proc. Acad. Nat. Sci., Philadelphia 68:45-66, 4 pls. 458 Proceedings of the Biological Society of Washington Hoyie, W. FE. 1904. Reports on the Cephalopods, in Reports on the scientific results of the expedition to the eastern tropical Pa- cific... ALBATROss. Bull. Mus. Comp. Zool. 43(1):1-71, 10 pls. Joupin, L. 1918 Céphalopodes recueillis au cours des croisiéres de S.A.S. le Prince de Monaco, 5° Note: Moschites verrucosa (Verrill). Bull. Inst. Oceanogr. Monaco No. 339:1-9. Massy, A. L. 1916. Mollusca. Part II Cephalopoda. Brit. Antarct. (“Terra Nova”) Exped. 1910. Zool. 2(7):141-175. Rosson, G. C. 1932. A monograph of the recent Cephalopoda. Vol. 2. Part II. The Octopoda (Excluding the Octopodinae). Brit. Mus. (Nat. Hist.), London, 359 pp., 6 pls. VeERRILL, A. E. 1881. No. 5. Reports on the results of dredging.... “Blake” ...X. -Report on the cephalopods, and on some ad- ditional species dredged by the U.S. Fish Commission steamer “Fish Hawk,” during the season of 1880. Bull. Mus. Comp. Zool. 8(5):99-116, 8 pls. Voss, GiuBerT L. 1963. Cephalopods of the Philippine Islands. U.S. Nat. Mus. Bull. 234:1-180, 4 pls. 3, No. 43, pp. 459-468 22 January 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON ORCONECTES ETNIERI, A NEW SPECIES OF CRAYFISH FROM WESTERN TENNESSEE AND NORTHERN - MISSISSIPPI WITH NOTES ON PROCAMBARUS, ~ ABLUSUS AND ORCONECTES WRIGHTI | =: By RayMonp W. BoucHARD AND JuDITH W. BoucHARD\ ~~ National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560 This new species of crayfish is known from the Mississippi Embayment province of Tennessee in the Tennessee, Forked Deer, Hatchie and Loosahatchie River systems where it is one of the more common species. In many western Tennessee col- lections from fluvial environments, it was the only species represented. Two of its associates, Procambarus (Pennides) ablusus Penn (1963:121) and Orconectes wrighti Hobbs (1948: 85), are known from only a few specimens. Procambarus ab- lusus was described on the basis of 7 specimens (4 adults) from 5 localities in Mississippi and Tennessee. Until recently, O. wrighti was known from only 11 specimens collected at the type-locality. Large populations of P. ablusus are more com- mon in the Forked Deer River system, while O. wrighti occurs in an additional larger downstream tributary (see below) of the Tennessee River. Orconectes etnieri, new species Figure la—k Orconectes immunis.—Penn, 1963:125. Diagnosis: Body and eyes with pigment. Rostrum without marginal tubercles (may be present on juveniles). Areola 4.2 to 7.2 times longer than broad, constituting 26.3 to 30.3 percent of total length of carapace (35.4 to 39.4 percent of postorbital carapace length) with 2 to 4 punc- tations across narrowest part. Single cervical spine present on each side of carapace. Hepatic spines absent; branchiostegal spine small and 43—Proc. Brow. Soc. WasH., Vou. 88, 1976 (459) 460 Proceedings of the Biological Society of Washington Fic. 1. Orconectes etnieri, new species. a, Lateral view of carapace of holotype; b, Mesial view of first pleopod of holotype; c, Mesial view of first pleopod of morphotypic male; d, Basipodite and ischiopodite of third pereiopod of holotype; e, Antennal scale of paratopotypic male, form I; f, Lateral view of first pleopod of morphotypic male; g, Lateral view of first pleopod of holotype; h, Dorsal view of chela of holotype; i, Epistome of holotype; j, Annulus ventralis of allotype; k, Dorsal view of carapace of holotype. Crayfish from Tennessee-Mississippi 461 acute. Suborbital angle poorly developed, terminating cephalically in acute, corneous-tipped spine. Antennal scale longer than broad, broad- est slightly distal to midlength; lamellar portion with 3 sided margin. Chela with two subserrate rows of tubercles along mesial margin of palm; scattered smaller tubercles over dorsomesial half of palm; small setal tufts over entire dorsal surface; lateral base of fixed finger impressed; well defined longitudinal ridges on dorsal surfaces of fingers. Hook on ischium of third pereiopod of first form male overreaching basioischial articulation. First pleopod of first form male with central projection corneous, tapering and reaching basis of second pair of pereiopods when abdomen flexed; mesial process shorter, non-corneous with distal end flared into shallow trough; distal ends of both elements curved caudally at same angle to shaft of appendage (see Fig. 1b, g). Holotypic male, form I; Body subovate, somewhat depressed (Fig. la, k). Abdomen narrower than thorax (14.0 and 17.0 mm). Greatest width of carapace greater than depth at caudodorsal margin of cervical groove (17.0 and 14.0 mm). Areola 6.0 times longer than broad with 3 punctations across narrowest part; length of areola 28.8 percent of entire length of carapace (37.8 percent of postorbital carapace length). Rostrum with thickened, straight, subparallel margins devoid of mar- ginal spines or tubercles. Acumen with concave, slender, margins, ter- minating in very small upturned, corneous tubercle, latter reaching end of antennular peduncle. Rostrum comparatively flat dorsally with slight impression basally; low, broad carina evident along flattened portion. Rostrum with submarginal punctations dorsally, others scattered between. Postorbital ridge moderately developed, shallowly grooved dorsolaterally and terminating cephalically in acute, corneous tubercle. Suborbital angle poorly developed and obtuse; branchiostegal spine small and acute. Single acute cervical spine on each side of carapace. Hepatic area and lateral portion of branchiostegite tuberculate; dorsal surface of cara- pace punctate. Abdomen longer than carapace (35.2 and 31.3 mm); pleura of mod- erate length with caudoventral extremities subangular. Cephalic section of telson with marginal immovable spine and submarginal movable one in each caudolateral corner; partially separated from caudal section by 2 oblique incisions. Basal podomere of uropod with spine extending over lateral and mesial rami. Lateral ramus of uropod with median and submesial ridges, former terminating in acute spine at transverse flexure. ’ Proximal portion of lateral ramus with row of spines distally and large movable spine submarginally at caudolateral corner (broken on right side). Mesial ramus of uropod with median ridge terminating distally in acute premarginal spine (missing on right side). Dorsal surface of telson and uropods lightly setiferous. Cephalic lobe of epistome (Fig. 1i) spatulate with thickened, irregular cephalolateral margins, and lacking cephalomedian projection. Main body of epistome with very shallow median fovea and pair of obliquely disposed slitlike fossae immediately cephalic and subparallel to thickened, 462 Proceedings of the Biological Society of Washington arched epistomal zygoma. Proximal segment of antennule with small spine on ventral surface near midlength. Antennae broken. Antennal scale (Fig. le) broadest at at midlength with 3 sided margin on lamellar portion; thickened lateral portion terminating in prominent, acute, cor- neous-tipped spine reaching distal end of antennular peduncle. Left chela (Fig. 1h) with mesial margin of palm possessing primary subserrate row of 8 tubercles and secondary row of 6 smaller ones on dorsal surface lateral to primary row; scattered tubercles over dorso- mesial half of palm; distoventral surface of palm with 2 tubercles at base of dactyl; dorsal surface covered with numerous small setal tufts. Lateral surface of propodus costate with row of punctations rendering proximolateral base of finger impressed dorsally and less so ventrally; dorsal and ventral surfaces with distinct submedian ridges flanked by setiferous punctations; opposable surface with row of 10 tubercles along proximal two-thirds of finger, fourth from base largest; additional tu- bercle present on lower level near base of distal third; double row of minute denticles extending proximally from distal end of third tubercle, interrupted by fourth through tenth tubercles. Dorsal and ventral sur- faces of dactyl with median longitudinal ridges flanked by setiferous punctations; opposable margin with row of 16 tubercles, fifth largest; double row of denticles extending proximally from distal end to fifth tubercle, interrupted by each of sixth through sixteenth tubercles; mesial surface with irregular rows of tubercles on proximal two-thirds, reduced to single row distally. Fingers with distal tubercles very small and in- conspicuous; terminating in large, acute, corneous tips. Carpus with deep oblique furrow dorsally; mesial surface with large procurved spine near midlength, 9 and 10 scattered tubercles on left and right cheliped, respectively; dorsal surface with 9 tubercles (right with 12) and additional acute one at distomesial margin; distoventral margin with 2 acute tubercles (median one on left broken); podomere otherwise punctate. Dorsodistal surface of left merus with 2 acute to subacute tubercles (right with 3); ventral surface with lateral row of 2 corneous-tipped spines, proximal one smaller, and mesial row of 9 spines, some corneous, decreasing in size proximally. Ischium with single small corneous-tipped tubercle on ventromesial margin. Hook on ischium of third pereiopod only (Fig. 1d); hook simple, overreaching basioischial articulation and not opposed by tubercle on basis. Coxae of fourth and fifth pereiopods without caudomesial boss. (See Measurements. ) First pleopods (Fig. 1b, g) reaching bases of second pair of pereio- pods when abdomen flexed. (See Diagnosis for description. ) Allotypic female: Differing from holotype in following respects: areola constituting 28.7 percent of entire length of carapace (37.8 per- cent of postorbital carapace length) and 5.7 times longer than broad. Postorbital ridges terminating in more acute spines. Left chela (right missing) distinctly narrower and shorter with mesial surface of palm Crayfish from Tennessee-Mississippi 463 bearing 7 tubercles in primary and 4 in secondary rows. Propodus with opposable margin bearing row of 9 tubercles, fifth from base largest; additional tubercle present on lower level; double row of denticles ex- tending proximally from distal tip to fifth tubercle, interrupted by sixth through eleventh. Dactyl with opposable margin bearing 11 tubercles, first and fourth from base largest; double row of denticles extending proximally from tip to fifth tubercle, interrupted by sixth through elev- enth. Carpus of left cheliped with mesial surface possessing corneous- tipped, procurved spine; distoventral margin with 2 corneous-tipped spines. Left merus with ventral surface bearing mesial row of 10 spines, some corneous-tipped. Sternum between third and fourth pereiopods broadly V-shaped. An- nulus ventralis (Fig. 1j) symmetrical and firmly fused cephalically to sternum; cephalic portion elevated with shallow, longitudinal, median trough flanked by low ridges, latter produced laterally at midlength as transverse, elevated ridges; fossa centrally located; caudal half bearing P-shaped sinistral sinus. Postannular sclerite about two-thirds as wide as annulus. First pleopods of female uniramous and reaching midlength of annulus when abdomen flexed. Morphotypic male, form II: Differing from holotype in following re- spects: areola constituting 28.9 percent of entire length of carapace (39.7 percent of postorbital carapace length), and 7.2 times longer than broad with 2 punctations across narrowest part. Postorbital ridge with more acute spine cephalically. Mesial surface of palm of right chela (left regenerated) with 7 tubercles in primary and 4 in secondary rows; propodus bearing 11 tubercles on opposable margin, additional one on lower level. Dactyl with opposable margin possessing 15 tubercles; double row of denticles extending proximally from tip to third tubercle, interrupted by fourth through fifteenth. Carpus of right cheliped with mesial surface bearing corneous-tipped, procurved spine and 12 scattered tubercles. Merus with single corneous-tipped tubercle representing usual ventrolateral row. Right ischium with single non-cornified tubercle on ventromesial margin. Hook on ischium of third pereiopod much reduced, not overreaching basioischial articulation. First pleopod (Fig. lc, f) of uniform texture, subequal and neither element corneous. Distal ends of both elements recurved at angle of approximately 40 degrees. Type-locality: Robinson Creek at Tennessee State Highway 57 (Ten- nessee River system), Hardin County, Tennessee. The creek varies from 6 to 15 feet in width and as much as 3 feet in depth in pools and less than 1 foot at the riffles. During several visits to the creek, the water ranged from clear to slightly turbid. The substrate is mostly sand with gravel in the riffles, several small to moderately sized rocks below the bridge site and numerous pockets of leaf litter in the pools and around obstructions in the runs. Mixed deciduous trees line the banks of this small creek. Robinson Creek is also the type-locality for Orconectes wrighti Hobbs (1948). 464 Proceedings of the Biological Society of Washington Disposition of types: The holotypic male, form I (USNM 146583), the allotype (USNM 146584) and the morphotypic male, form IJ (USNM 146585) are deposited in the National Museum of Natural History, Smithsonian Institution. Paratypes consisting of 8 ¢1,6 ¢II,9 2,3 ¢ juv. and 11 @ juv. are in the Smithsonian Institution; 2 ¢I and 2 9? are deposited in the Tulane University Museum; and 8 ¢1, 16 ¢II, 19 9, 2 9 eggs, 1 ¢ juv. and 4 @ juv. are in the collection of the senior author. Range and specimens examined: This new species of crayfish is known from Tennessee and Mississippi. In Tennessee Orconectes etnieri has been collected from tributaries draining the west bank of the Tennessee River from Robinson Creek northward to Snake Creek in Hardin and MeNairy counties. In waters flowing westward into the Mississippi River, it has been collected from the Forked Deer, Hatchie and Loosa- hatchie River systems in Tennessee and Mississippi. A crayfish of the Mississippi Embayment province, it is absent from the sluggish waters that characterize the Mississippi River Floodplain section. The following material is designated as paratypes: TENNESSEE. HARDEMAN COUNTY. Dry Branch at Tennessee State Highway 125, 1.3 miles south of intersection with U.S. Highway 64 (Hatchie River system via Spring Creek). C. E. Comiskey, D. A. Etnier and M. Reese. 18 October 1968. 6 ¢I, 3 9. HARDIN COUNTY. Robinson Creek at Tennessee State Highway 57 and two additional localities approximately 0.7 and 1.2 air miles upstream (Tennessee River system). P. Yokley, Jr., and R. W. B. 15 March 1971. 5 61,6 ¢II,6 9,1 2 eggs, 1 ¢ juv. and 4 @ juv.; Robinson Creek at Tennessee State Highway 57 and single additional locality approximately 0.7 air miles upstream. J. D. Way and R. W. B. 14 March 1972. 2 61,7 411, 10 9,2 @ juv. and 6 @ juv:; Snake Creek at Tennessee State Highway 22 (Tennessee River system). C. E. Comiskey, D. A. Etnier and M. Reese. 19 October 1968. 2 61 and 2 9. MCNAIRY COUNTY. Snake Creek, at County Road 8235, 2.0 miles west of intersection with County Road 8083. T. Carson, B. Clark, M. Hughes and G. A. Schuster. 1 March 1974. 3 ¢I1, 10 ¢4II, 13 92 and 1 @ eggs. Variation: The most significant variation within the species occurs in the annulus ventralis. The anterior ridges bordering the median trough of the annulus may be lacking, and in some populations the trough is broader and more deeply excavated. Size: The largest specimen available is a second form male with a total carapace length of 32.5 mm (postorbital carapace length 24.0 mm). The smallest first form male has corresponding lengths of 18.4 and 13.9 mm. The smaller of the two females with eggs has a total carapace length of 24.6 mm. (postorbital carapace length 18.6 mm). Color notes: Cephalothorax and abdomen mottled with browns and a pair of dorsolateral and submedian broken, dark brown stripes on ab- domen extending onto cephalothorax as darker mottled dorsolateral areas. Stripes and dark dorsolateral mottling may be less obvious and faded in some individuals. Branchiostegites and hepatic region otherwise light Crayfish from Tennessee-Mississippi 465 TABLE 1. Measurements (mm) of Orconectes etnieri Holotype Allotype Morphotype Carapace Height 14.0 TPA 13.4 Width 17.0 13.9 15.6 Total length of carapace 31.3 29.6 32.5 Postorbital carapace length 23.8 22.5 24.0 Areola Width 1.5 15 1.3 Length 9.0 8.5 9.4 Rostrum Width 45 ow 4.3 Length U5) Toll 8.5 Chela Length, mesial margin palm 9.2 6.0 8.6 Width, palm 12.6 7.2 10.1 Length, lateral margin 30.2 18.8 26.5 Length, dactyl 18.7 11.0 15.9 brown to white. Rostral margins, postorbital ridges and lateral margins of antennal scales darker brown than basic color of body. Ventral aspects of cephalothorax and abdomen white. Chelae mottled to concolorous brown dorsally with lighter proximo- lateral area; white to cream ventrally. Tubercles along opposable mar- gins of fingers yellow to cream. Distal ends of fingers red in young, often orange or yellow in adults. Pereiopods mottled brown dorsally; fading to cream or white ventrally. Distal podomeres darker dorsally than proxi- mal ones. Life history notes: First form males have been collected during the months of March, July and October. Two females bearing eggs were collected, one each on 15 March 1971 and 1 March 1974. Ecological notes: Orconectes etnieri commonly inhabits the leaf litter which has collected in pools or above obstructions, such as fallen branches, in the runs. At the type-locality O. etnieri shared its preferred habitat with Procambarus acutus acutus (Girard 1852), P. ablusus, Orconectes validus (Faxon 1914), O. wrighti, Cambarus striatus Hay (1902) and occasional individuals of the primary burrowing crayfish C. diogenes Girard (1852). Orconectes wrighti was more common under rocks that littered the riffles. Cambarus striatus and Cambarus diogenes were com- mon burrowers in the stream bank. Collected along with Orconectes etnieri in one or more localities were Procambarus ablusus (Hatchie, Forked Deer and Tennessee River sys- 466 Proceedings of the Biological Society of Washington Fic. 2. Lateral view of left first pleopods of two species of Orconectes. a, First form male, O. etnieri; b, Second form male, O. etnieri; c, First form male, O. validus; d, Second form male, O. validus. tems), P. a. acutus (Tennessee River system), Cambarus striatus ( Hatchie, Forked Deer, Loosahatchie and Tennessee River systems), Cambarus diogenes (Hatchie and Tennessee River systems), Orconectes p. palmeri (Faxon 1884) (Loosahatchie River system), O. placidus (Hagen 1870), O. validus and O. wrighti (Tennessee River system). When Procambarus ablusus was described from the Hatchie River system in Mississippi and Tennessee, it was regarded by Penn as an un- common species and remains so in museum collections. Procambarus ablusus does appear to be rare in the Hatchie River system but in some localities in the Forked Deer River basin to the north it is very common. Orconectes wrighti is also a little known crayfish with an apparently restricted range. In addition to its presence in the type-locality, a pop- ulation of this rare species occurs in a stream flowing through Hardin and McNairy counties, Snake Creek, a tributary of the Tennessee River. Its range probably does not exceed the area bordered by these two lo- calities and it may not occur outside of the two creek systems. Crayfish from Tennessee-Mississippi 467 Relationships: Orconectes etnieri has its closest affinities with O. validus from which it differs primarily in the morphology of the male gonopod. In O. etnieri the terminal elements are recurved at the same angle (Fig. 2a, b) while in O. validus the angles differ, with the mesial process recurved at approximately 90 degrees (Fig. 2c, d). The rostrum of O. etnieri is much flatter in comparison to the deeply excavate one of O. validus. For the most part the two species occupy mutually ex- clusive ranges. Orconectes validus occurs in the Tennessee River system in streams draining the Highland Rim section downstream to Benton County, Tennessee. A single female from Montgomery County, Ten- nessee, (Cumberland River system) may also be a member of this spe- cies possibly indicating a larger range. Orconectes etnieri is known from only a small part of the Tennessee River system in Hardin and McNairy counties, Tennessee, (Robinson Creek and Snake Creek) and very pos- sibly occurs in nearby tributaries of the Tennessee River in Mississippi. Most of the range of O. etnieri, however, embraces westward flowing tributaries of the Mississippi River from which O. validus is absent. Both species were collected syntopically at the type-locality. Etymology: We take great pleasure in naming this new species of crayfish in honor of David A. Etnier, University of Tennessee, Knoxville, for his many contributions to our knowledge of Tennessee crayfishes and especially his enthusiam in collecting primary burrowing species. Acknowledgments: We are indebted to Joseph F. Fitzpatrick, Jr., University of South Alabama, Mobile, who with Horton H. Hobbs, Jr., originally noted the uniqueness of this new species and permitted us the honor of naming it after our former major professor. To Horton H. Hobbs, Jr., we are as always indebted for his kindness and valuable assistance and for reviewing the manuscript. To David A. Etnier and his students who helped collect much of the material upon which this description is based we owe thanks, especially to those who surveyed the West Tennessee fish fauna: George Boronow (Forked Deer River system), Byron Clark (Tennessee River system), William Dickinson (Obion River system) and Wayne Starnes (Hatchie River system). Ap- preciation is extended to the Smithsonian Institution for providing the senior author with a Smithsonian Postdoctoral Fellowship permitting the opportunity to describe this new species. LITERATURE CITED Faxon, WALTER. 1884. Descriptions of new species of Cambarus, to which is added a synonymical list of the known species of Cam- barus and Astacus. Proc. Amer. Acad. Arts and Sci. 20:107- 158. —. 1914. Notes on the crayfishes in the United States National Museum and the Museum of Comparative Zoology with de- scriptions of new species and subspecies to which is appended a catalogue of the known species and subspecies. Mem. Mus. Comp. Zool., Harvard Coll. 40(8) :351—427. 468 Proceedings of the Biological Society of Washington Grrarp, CHartEs. 1852. A revision of the North American Astaci, with observations on their habits and geographical distribution. Proc. Acad. Natur. Sci., Philadelphia 6:87-91. Hacen, HERMANN A. 1870. Monograph of the North American As- tacidae. Illus. Cat. Mus. Comp. Zool., Harvard Coll. 3:viii + 109 pp., 11 pls. Hay, Witu1aM Perry. 1902. Observations on the crustacean fauna of Nickajack Cave, Tennessee, and vicinity. Proc. U.S. Nat. Mus. 25( 1292) :417—439. Hosss, Horton H., Jr. 1948. A new crayfish of the genus Orconectes from southern Tennessee (Decapoda, Astacidae). Proc. Biol. Soc. Washington 61(15):85-91. PENN, GrorcGE Henry. 1963. A new crawfish from the Hatchie River in Mississippi and Tennessee (Decapoda, Astacidae). Proc. Biol. Soc. Washington 76(15):121—125. GH a BAX NH 3, No. 44, pp. 469-488 22 January 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON PERCINA (IMOSTOMA) TANASI, A NEW PERCID FISH / FROM THE LITTLE TENNESSEE RIVER, TENNESSEE/ By Davip A. ETNIER Department of Zoology, University of Tennessee, Knoxville, Tenn. 37916 | = a Witte iay | 1} f Wi \ % \ \ The Little Tennessee River begins in Macon County, North \ Carolina, and flows northwest to enter the Tennessee River near Lenoir City, Loudon County, Tennessee. The flow of the Little Tennessee River is interrupted by Fontana Reservoir, a large warmwater impoundment in Swain and Graham coun- ties, North Carolina. Tailwaters of Fontana Reservoir are consistently cold and well suited to a recreational trout fishery. The cold water habitat continues through Cheoah, Calder- wood, and Chilhowee reservoirs. Below Chilhowee Reservoir a trout fishery persists for about 21 river km, ending approxi- mately 2.5 river km above the U.S. Highway 411 bridge, Mon- roe County, Tennessee. Although trout occur sporadically in the lower 32 km of river, warmwater fishes are predominant. Records indicate that the original fish fauna of the Little Tennessee River system was rich. Among the more interest- ing species known from this system are an undescribed darter of the subgenus Catonotus (R. E. Jenkins, in lit.), Hybopsis monacha (Cope), Cycleptus elongatus (Lesueur), and Percina burtoni Fowler. Noturus baileyi Taylor, a madtom that is likely extinct, is known only from a rough fish removal opera- tion in Abrams Creek, tributary to the Little Tennessee River (Taylor, 1969). A number of reports of lake sturgeon ( Aci- penser fulvescens Rafinesque) from recent sightings and cap- tures in the river suggest the continued existence of what is possibly a reproducing population of this vanishing species in the lower Little Tennessee River. A reasonable number of fish 44_Proc. Brot. Soc. WAsH., Vou. 88, 1976 (469 ) 470 Proceedings of the Biological Society of Washington collection records are available from the tributaries and head- water portions of the system, and from the 21 km of river utilized as a trout fishery below Chilhowee Dam. The lower 32 km of the Little Tennessee River remained uncollected until recently. It seemed possible that (despite environmental changes that have taken place over the years) some elements of the original main channel fish fauna might persist in the lower portion of the river. An additional impoundment, Tel- lico Reservoir, is under construction by the Tennessee Valley Authority, and would innundate virtually all of the river re- maining below Chilhowee Dam. With this in mind, Robert A. Stiles and I visited the river at Coytee Spring, River Mile 7, Loudon County, Tennessee, on 12 August 1973, in an attempt to survey the fish fauna with the aid of face masks and snorkels. Although very few fishes were seen, a single specimen of a very unusual darter was observed and “scooped up” by hand by the author. Subsequent examination of this specimen in- dicated that it was a new species of the genus Percina, sub- genus [mostoma. In the genus Percina, the subgenus Imostoma had until re- cently been envisioned as containing two species, P. shumardi (Girard) and P. uranidea (Jordan and Gilbert). Three speci- mens collected in Shoal Creek, tributary to the Etowah River (Mobile Bay drainage), Cherokee County, Georgia, by Don Scott and O. Tyson, were recognized by J. D. Williams in 1966 as representing a distinct species allied with P. uranidea. Impoundment of Alatoona Reservoir apparently eliminated the population. Stiles and I collected a specimen of this spe- cies in the Conasauga River, Polk County, Tennessee, in 1968. Approximately 60 additional specimens have now been col- lected from the latter stream, and the description of this spe- cies by Williams and Etnier is in preparation. In 1970 I examined some specimens at Tulane University from the Ozark region collected by R. C. Cashner and B. A. Thompson. These distinctive specimens, collected with what was then known as P. uranidea, appeared to be a new species. Subsequent examination of the types of P. uranidea and its supposed junior synonym P. ouachitae (Jordan and Gilbert) by Thompson revealed that their “new” species was identical Percid fish from Tennessee AT] with the types of P. uranidea, and the types of P. ouachitae were identical with specimens then being referred to as P. uranidea (Thompson, in lit.). A paper concerning the status of P. uranidea and P. ouachitae is being prepared by Thompson and Cashner. Examination of the specimen from the Little Tennessee River suggested that its affinities were with true P. uranidea rather than with P. ouachitae, P. shumardi, or the undescribed species from the Conasauga River. Further stud- ies confirmed this view and indicated that the Little Ten- nessee River population represented a species distinct from P. uranidea. THe SusBGENUS I[MOSTOMA The subgenus Imostoma currently includes the following five species: P. shumardi (Girard), which lacks distinct dor- sal saddles; and four species with saddles—P. uranidea (Jor- dan and Gilbert), restricted now to the White and Saline river systems, Arkansas, but formerly occurring in the Wabash River, Indiana; P. ouachitae (Jordan and Gilbert), widespread in the lower Mississippi River valley and in the Gulf Coastal drainages; the undescribed species from the Conasauga River; and the new species from the Little Tennessee River, described herein. Page (1974) presented a brief diagnosis of the subgenus Imostoma based on P. shumardi, the undescribed species from the Conasauga River, and P. uranidea from a “ditch” in Mis- souri. His P. uranidea specimens were likely P. ouachitae, since P. uranidea is extremely rare in Missouri, and appears to be restricted to swift water in moderate size rivers. P. ouachitae does occur in ditch-type habitats. The inclusion of two additional species in the subgenus (P. uranidea and P. tanasi), plus examination of extensive material of the three species considered by Page, necessitates modification of the diagnosis, as follows. A “midbelly row of modified (enlarged and strongly toothed) scales” develops on individuals of all five species. These modified midventral scales are most con- spicuous in large adults, and may be lacking in juveniles and small adults. In nuptial males of P. uranidea there is a more or less well developed band of exposed but scarcely modified 472, Proceedings of the Biological Society of Washington ctenoid scales extending from the ventral prepectoral area transversely across the middle of the breast. In the other four species embedded cycloid and occasionally a few exposed ctenoid scales are present on the breast in many individuals. These are most conspicuous in nuptial males, and are usually confined to the posterior half of the breast. Key TO THE SPECIES OF THE SUBGENUS I[MOSTOMA 1. Dorsal saddles absent Percina shumardi (Girard) Four or five distinct dorsal saddles present — 2 2. Anteriormost dorsal saddle entirely in front of dorsal CENT y oes tee SUNN nro eet nls Noone Percina species Anteriormost dorsal saddle beneath dorsal fin ___ 3 3. Posterior margin of fourth dorsal saddle over dorsal in- sertion of caudal fin; fifth dorsal saddle absent; anal sott mays LS l2s(extremess0=13)) 4 Posterior margin of fourth dorsal saddle anterior to dor- sal insertion of caudal fin; fifth saddle adjacent to dor- sal insertion of caudal fin, moderately developed; anal soft-rays l0=1] (9-12) 2 Eee i See ae Percina ouachitae (Jordan and Gilbert) 4. Anal rays modally 11; pectoral fins long and pointed, overlying lateral-line scale 19-24 (males) or 17-23 (females and juveniles); pelvic fins 22% of standard length or longer in males, 20% of standard length or longer in females and juveniles; nuptial males with tubercles on pelvic spine and adjacent four soft rays eee SD Percina uranidea (Jordan and Gilbert) Anal rays modally 12; pectoral fins shorter and more rounded, overlying lateral-line scale 17-19 (males) or 15-18 (females and juveniles); pelvic fins less than 22% of standard length in males, less than 20% of standard length in females and juveniles; nuptial males with pelvic fin tubercles confined to the four median TAY Sj RAI suet sare lace ts ee tt Percina tanasi new species METHODS Measurements and counts (Tables 1-8) follow the methods of Hubbs and Lagler (1958) unless defined below. Measure- Percid fish from Tennessee 473 ments were made with a needle-point divider and read to the nearest 0.1 mm. Head length was measured from the tip of the snout to the tip of the membranous extension of the operculum. Orbital diameter and interorbital width were measured between the fleshy coverings of the bones delimit- ing these areas. Trans-pelvic width was the measurement be- tween the outer edges of the parallel pelvic spines. Saddle widths were measured along the dorsal midline and the saddle position from the occiput. The lateral-line scale under the tip of the pectoral fin (Table 3) was obtained by counting from the head to the fin tip when pressed against the lateral midline. MATERIAL STUDIED The number of specimens appears in parentheses after the catalog number. The following abbreviations have been used: SL (Standard length), TU (Tulane University), UT (Uni- versity of Tennessee), USNM (National Museum of Natural History, Smithsonian Institution), CU (Cornell University ), ‘FSM (Florida State Museum), INHS (Illinois Natural History Survey), NLU (Northeastern Louisiana State University), UAIC (University of Alabama Ichthyological Collection), UMMZ (Museum of Zoology, University of Michigan), OAM (Oklahoma State University), HWR (Henry W. Robison collection ). Percina tanasi, new species Snail Darter Figure 1 Type-specimens: Holotype, TU 90858, an adult male, 59.1 mm SL, Little Tennessee River at Coytee Spring, River Mile 7, Loudon County, Tennessee, 6 Nov. 1973, G. F. Boronow, B. F. Clark, Jr., D. A. Etnier, R. L. Hensen, W. C. Starnes, B. A. Thompson. Allotype female, TU 90859, 54.6 mm SL, same data as holotype. Paratypes taken with the holotype are TU 83994 (2) and USNM 214697 (2). Additional para- types taken at the type-locality are CU 64754 (5), FSM 20852 (5), INHS 75000 (5), NLU 32057 (5), UAIC 5020.01 (5), and UMMZ 197413 (5), all 16 Aug. 1973; USNM 214698 (5), 4 Oct. 1974; UT 91.803 (1), 12 Aug. 1973; UT 91.1074 (10), 23 April 1975. An ad- ditional paratype, OAM 9536 (1), is from River Mile 15, main channel side of Davis Island, Little Tennessee River, Loudon County, Tenn., 4 Oct. 1974. 474 Proceedings of the Biological Society of Washington OL sss O9-Ilh O8'8F OL 609 SP-0€ O69 61 FI8 6S-Fo Ic9r f G vF-96 O00SE OL 689 Is-ve O8 IP OL 982 €F-1G §=O1L'8G Il L€8 O9-$E OEr FA e[Pppes L 664 €F-¥G PIES OL 688 PS-8G OLE Ot cO9 €&-€L OF SG €I 09°8 09-0€ O0GF W YANoF fo YIPIM OL 87CI 624-66 OFFS Or Lg°¢ Ss-0OrF OT LP 6I GI8 6L-Gh FLe9 f G 8F-8E O00 EF OL 9TOL L9-9€ O16? Or 199 SE-8I OF 66 Il 896 O8-0S SFPD A s[ppes L £98 €9-LE FIL 6P OL G9ET LI-FE OL 6F OL O19 T?-61I OS 08 €I GYVOL I8-9F ST'39 W Pury? FO YIPIM OI ss°9 8LZ-9S 0619 OL S79 09-Gh OTIS 6I 19°9 r9-0F pg9g f G cs-ty O0LF Or 99°9 €9-SF 00S (0) Ge OS-PE OG CF Il €€6 SL-Lp 1619 A s[Ppes puo L 179 8&-6€ 98 °8P OL 89TLT TL-9€ O8 ES OL 092 Ig-GG OL'9E €I O0O0GI 6L4-0F 6995S W 7998 FO YIPIAA OL 888 G8-1S OLT9 OL &a7 8S-SP OSS 6I 9F6 9L-IF 1356S f I OO'TE OL 802 g9-9F OT19S 6 GYOLT I9-T€ 68'°8F OL S8 IL G8-IF 0069 A s[Ppes 9 LY9 8S-8E LOOP OT 896 IL-ch 00248 OL 608 LS-VE OF ST OL 766 V8-GS OCr9 W dsarF JO PIAA OL 88S GEC-LIG O08'S6G VG OF OL YGC-LET 8S'OIZ 6I 062 *F61-091 sGrost f G LIG-SIG O009IG Il 8€6 ¥FEC-80G 8I'Ecs 8I 886 GrE-LOG 68616 Il SLGI G66I-8ST LOF8I A Yasue] L 098 GESG-90G O09 IGG LI 9OIT 8FC-LIG CITES 6I GOL FPPS-8IS 68'86G €I 89ST €6¢6-S9I 00°00 W Ulf OIATod OL Gl9 €SC-OES 06'0FG VG SOOL PLC-ECG SL 8ts 61 S88 FEC-Z6I G6L'L0G f G Ssc-ISc 00°€S¢ Il €871I PLC-€GG EL'SSs €6 GIGLI I6G-FEG 6E'99G Il 6PETL S8EC-S6I GS8'8IG A Y}su9]| 9 OF 6 GLG-LSG FE69G SI FOIL FOE-F9S OS L8G 8G IG8I 9GE-GSC OF SSG €I SG9T 9LG-SIG 69'CFG W Ulf [B1OJ9qF OL F8€ COI-68 OT S6 OL 18% GOI-06 OS'S6 61 O€ F G6OI-E6 LE TOT f € 9eppes G s8-08 09°¢8 OL 7SS OOI-08 00°68 OL 684 OOI-I8 08°28 Il cO9 E€II-€6 1666 A JO ursIr0 ZL 60S 88-SL L9°¢8 OL cg7 96-78 0£ 06 OL 90S 88-FL 02408 €I O08 60I-9L 69°56 W MOT YIPIM OL 8L4Y 66I-SII O00'TGT OL 88F GEI-SIL O0'FGI 61 80S ShI-FEI sosET f Ulf [vs1op G €CI-GIT OG LIT OL 909 G6GI-€IL OL ISI OL 868 LEI-801I Od FEI Il POL G6FI-LOL IGSEL A OS FO ULSIIO 9 GIF YGI-FILT LT 6IT OL €rS €EI-FIL OL ESI OL 119 9GI-LOT OT 9IT €l L448 LPI-6IL 9FCGEL W AOTEq YIPIM Ot I”s OS-PE OG HF OL 8G sc-SP OL'6F 6I €1€ cc-cr LYePr G €F-OF OS IP 6 8sT1é 6F-OF_ IIL'FF OL 90°E OS-Irh OLSP Il 8¢7 19-9F 8TIS A UPI [eITqro 9 S8'€ G6F-GE O00'SF OL C66 6F-OF OS EF OL 866 GS-Pr OE LP €L O8'€ 8S-hr PS TS NW ~teyUr UNTO N *S AA x N “Ss M x NS AA x Noe as A x stoJOVILYO (‘YW YyseqeM ) vaprupin -g (CY eurles ) vapwpin «g (YH SHUM ) vaprupin +g ISDUD} * J (syMpeqns pue sopfuean{ = f[ ‘soyeuroz = J ‘sopeur = yy ‘ozis ofdures = NY ‘UOTVIAep prepurys = *g ‘osuex =M ‘UrstI =x) “suse, plepuejs Jo syypuLsnoy} ul posseidxe vaprupin ‘gq PUR IspUD} DULOIag Jo SyUSUIAINSeIW{ “[ ATAVY, A795 Percid fish from Tennessee Sail G I i wiayshs IOAN yseqeAA V6I OV © § £6i 6 L V W9}shs IOAN oUTTeS 861 09 cot &$ #® @ @O hi oo 6 I WI9}SAS IOATY OFA soyfuaan{ pue soreurey ‘paprupin -g GOL 9€ G II 6 6 6 SoftueAn! pur saves “spud, DUIAIAag C'6I 9 € ¢ wiayshs IOAN Yseqvy L0G I€ G11, 8S 48) 46 S I wiayshs IOATY OUTTeS L0G ial vpe6G¢ € I IT wiaysks IOATY OYA SajpUuL “paplupin *g Ost 61 G 9 8 G I so[eul “IspUD, DUIIIAag xe N SG 6 £6 CG I6 O06 61 SI LI 9T se VL SIL papwupin “q pue IspUud} DUIDIag Ul UF [eIoJed yoy Jo dy Yeeueq s]eos oUT-[e1o}eT “EC AIAV], LOS 6 Il G I I I © il wiayshs IOAN YseqeAr EG i I gS vy @6 ith Gl Sl 6 Qe vw 6 I WI9}sAS IOATY OUTTRS erg 88 y 8 ol 6 OE SS CU SOUL WI9}SAS IOATY OYA papwuvin *q Gs SES; I = Os Si ae 9 (c Olas 8 G&G ISDUD} DUIDIAd xs N oo ~6©66S0—CsS ZG gs SS PS eG GS Ig os 6h Shr LE 9F SQ[VOS OUI]-[e19}e[ JO Iaquinyy ‘DapwudIn “q Pue wDUDZ DULDIag UI Soy[eos OUI[-[B19}e] JO Aouonbal yy 'G ATIAV], 476 Proceedings of the Biological Society of Washington OFT 6 9 G I 9°01 6 9 G I WI94sAS IOATY YsequAA OS ci 8 96 ~=@OT SO ir 6G We Te Wi9}shS IOATY OUT[TeS 671 19 9 eV GI VOL L9 G LE GG € wiayshs IOANY FY AA papwupin “g GGT GS G LT 6G OL I COT VS V 6G LG I ISDUD} DULDIag os N 8I LT 9T ST FI €1 oe N IIx Ix xX XI skey IOS [esto sourds [es10oq ‘Dapwvin “gq PUe IsPUD} DUOlag JO SUIF [eSIOp UL skvI }Jos pue soulds Jo Aouonbery ‘ec ATAV, S6r 6 V V I Sit I G g I Wa3SAS IOATY YsequA, O0G SP Gl &6 6 I vyOL Gir @ el Bl ® T WoeSAS IOATY OUI[eS 6W6 Gl G ¥ cé 8 9 GOT 8& it Sh 86 We WI9}SAS IOATY SPY AA Dapiupin ‘d Oe SE 6 GE GT JO sg if 8 Bl CG ® IspUD] DULIIag s NM @6 i O06 Bl I es N @6¢ Sh Zl Of Sr pi gpunpeg [epney punory salvos SMOY 9[VOS ssIOAsURIy, ‘DapIudDIN *q Pure IspUD} DUlOlag 1O0F 9[oUNped [epnvo punore saror sfvos WNUTUIUE pue (UTZ [es1op 0} pxeArtoF Afonbi~qo ourds [eue jsilz WoIJ poyUNOD) sMor Q[Bos asiloAsueI}] JO AQuonbaly “Pp aATAV, ATT Percid fish from Tennessee Sel 9 g I Sl 8 V V u19}SAS IOATY YyseqeA\ OVI S&P Cum eSee ES OTL Sh 9 PS & wioyshs IOAN OUT[TeS Scr S84 fF © 8 BI SSl 78 6 Gr 6 wi93shs IOATY OFA Dapwpin “gq Sel Og 6 8 Ol Cat i ? ML Of IsDUD] DUIOIAd 8 N 9T ST PI €1 3 N ST VI €1 sey [elopog IWSIYy SAVY [eIO~og WoT Dapupin “gq pup spun}, pdulolag Ul SAvI UlF [eI0JOed Jo AOUONbealy “) AIAV], SO. 6 I G & wiayshs IOATY YseqeAr Sol Sir C6 WG 6 Gil ef Tf 6 OF € WI9}SAS IOATY SUITES SOL Lv Pl Me © GIT ag cI ce & woysks IOAN SFT AA papwpin “J Sc 1 € ry ® kit 7 Tt 8 Sl ®B IsDUD} DUIDIAg g NM AL 91 STI % N 1 GI Il O1 sAey [epneD sAey JOS [euy ‘paplupin “gq pue ispun} Dulolag Ul SXe1 [epneo yediourid pue sAvi [eure iJos Jo Aouonbely “9g AIAV], 478 Proceedings of the Biological Society of Washington TasBLeE 8. Measurements of Percina tanasi expressed in thousandths of standard length. (x = mean, W = range, Sx = standard deviation, N = sample size, M = males, F = females, J = juveniles and subadults) Characters x WwW S. N Head length 288.00 272-304 9.40 13 286.91 282-293 4.23 11 278-307 7.18 19 71.92 66-80 3.62 13 (2.0 68-76 2.21 JL 74.32 69-78 2.52 19 Orbit diameter bo (ee) © (ee) i Fics. 1-9. Lirceolus smithii. 1, 6 body, dorsal; 2, Antenna 1, 1.8 mm 9, dorsal; 3, Antenna 2, 3.3 mm 9, dorsal; 4, Left pereopod 1; 5, Dactyl of same; 6, Left pereopod 2; 7, Telson and uropod, dorsal, of 1.8 mm @; 8, Right uropod, dorsal, of 3.7 mm 9; 9, Penes. 491 Texas troglobitic water slater —— Nw ny [A a) laa SS f by fZl 492 Proceedings of the Biological Society of Washington Texas troglobitic water slater 493 Description: Length up to at least 3.7 mm. Body slender, about 314 times as long as wide; width increasing gradually posteriorly to pereonite 6, then decreasing. Coxae all visible in dorsal view. Margins of head, pereonites, and telson moderately setose. Head nearly twice as wide as long; anterior margin nearly straight, without rostrum; postmandibular lobes weakly developed. Telson unusually small, orbicular; width and length subequal; caudomedial lobe barely evident. Antenna 1 slightly longer than head; flagellum of 2—4 segments, last 2 segments at each bearing esthete. Antenna 2 about as long as body; flagellum of about 30 segments. Mandible with 3-cuspate incisors and 2-cuspate lacinia mobilis; spine row with 13 spines on each mandible; palp with unusually long Ist seg- ment and compact spination on 3rd segment. Maxilla 1 and maxilliped as in generic diagnosis and illustrations. Pereopod 1 propus not enlarged; palm without processes; flexor mar- gin of dactyl with 5-6 long slender spines. Pereopods 2-7 similar; basis with distinctive bulbous-based setae on margin; carpus and propus with long slender spines on 1 or both margins; dactyl with 2 spines on flexor margin. 6 pleopod 1 larger than pleopod 2; protopod slightly shorter than exopod, with 5 retinaculae; exopod oval, slightly less than twice as long as wide, with a few short naked setae on distal margin. ¢ pleopod 2 exopod, proximal segment produced distomedially over anterior surface of distal segment; distal segment with longitudinal furrow on posterior surface. Endopod slender, about 3.6 times as long as wide; margins parallel distal to short medial and lateral spurs at base; fissure oblique; medial process running obliquely laterad and ending in knob extending slightly beyond broadly rounded caudal process. 2 pleopod 2 suboval, slightly more than half as wide as long; margins without setae. Pleopod 3 exopod with margins unarmed except for 2 setae at distomedial corner. Pleopods 4—5 uniramous; groove encircling margin indicates line of fusion of rami. Uropod with narrow linear rami bearing long setae; endopod about 1.8 times as long as exopod. Relationship: The oblique suture on the exopod of pleopod 3 is a distinctive character which Lirceolus shares only with Lirceus, and sug- gests that Lirceus may be ancestral to Lirceolus or that the 2 genera share a common ancestor. The structure of maxilla 1 is unique; all known Asellidae have a maxilla 1 inner lobe with either 4 apical setae (Asellus (Asellus)) or 5 apical setae (all other Asellidae). The finding of 8 setae on L. smithii was so unexpected that the maxilla 1 of several speci- é Fics. 10-17. Lirceolus smithii. 10, Right mandible, gnathal margin; 11, Incisor of right mandible; 12, Incisor and lacinia of left mandible; 13, Mandibular palp; 14, Right maxilla 1; 15, Right maxilliped; 16, 2 pereo- pod 4; 17, 2 pereopod 6. 494 Proceedings of the Biological Society of Washington Fics. 18-26. Lirceolus smithii. 18, ¢ pleopod 1, anterior; 19, ¢ pleopod 2, anterior; 20, ¢ pleopod 2 exopod, posterior; 21, ¢ pleopod 2, endopod tip, posterior; 22, @ pleopod 2; 23, ¢ pleopod 3, anterior; 24, 2 pleopod 4; 25, 2 pleopod 5; 26, 2 pleopod 5, tilted to show line of fusion between rami. mens was examined in order to be sure that this number is constant. The position of the spines of the outer lobe of maxilla 1 is also unusual; in other Asellidae they are terminal rather than on an oblique subterminal shoulder. The extreme reduction of pleopods 4 and 5 is known in no other Asellidae, but these pleopods have not been described in most species. The small size of L. smithii, together with the well known low metabolism of troglobites, suggests that this species requires minimal respiratory surface. LITERATURE CITED Banta, ArrHuR M. 1907. The fauna of Mayfield’s Cave. Carnegie Inst. Washington Publ. No. 67:1-114. BistTEIn, J. A. 1951. Freshwater isopods (Asellota). Fauna SSSR, Crustacea 7 (5):1-140 [in Russian; English translation by Israel Program for Scientific Translation, 148 pp. 1964]. Texas troglobitic water slater 495 CHACE, FENNER A., Jr., J. G. Mackin, LEsLiIzE Hupricut, ALBERT H. BANNER, AND Horton H. Hosss, Jr. 1959. Chap. 31, Mala- costraca. Pp. 869-901 in W. T. Edmondson, ed., Fresh-water Biology, Second edition, New York, xx + 1248 pp. Cuappuis, P. A. 1927. Die Tierwelt der unterirdischen Gewésser. Die Binnengewasser 3:1—-175. CrEASER, Epwin P. 1931. A new blind isopod of the genus Caecidotea from a Missouri cave. Occ. Pap. Mus. Zool. Univ. Michigan, No. 222:1-7, pls. 1-2. DEAROLF, KENNETH. 1953. The invertebrates of 75 caves in the United States. Proc. Pennsylvania Acad. Sci. 27:225-241. EIGENMANN, Cart H. 1900. A contribution to the fauna of the caves of Texas. Science 12:301-302. FLEMING, LAURENCE FE. 1973. The evolution of the eastern North American isopods of the genus Asellus (Crustacea: Asellidae). Part II. Intern. Jour. Speleol. 5:283-310. Henry, JEAN-PAUL, AND Guy Macniez. 1970. Contribution 4 la sys- tématique des Asellides (Crustacea Isopoda). Ann. Spéléol. 25 (2) :335-367. JEANNEL, R. G. 1943. Les fossiles vivant des cavernes. Gallimard, 321 pp. Miter, Mitton A. 1933. A new blind isopod, Asellus californicus, and a revision of the subterranean asellids. Univ. California Publ. Zool. 39(4) :97—110. MITCHELL, ROBERT W., AND JAMES R. REDDELL. 1971. The inverte- brate fauna of Texas caves. Pp. 35-91 in Ernest L. Lundelius and Bob H. Slaughter, eds., Natural History of Texas Caves. Gulf Natural History, Dallas. Nicuo.tas, Bro. G. 1960. Checklist of macroscopic troglobitic or- ganisms of the United States. Amer. Midl. Nat. 64(1):123- 160. RAFINESQUE, C. S. 1820. Annals of nature or annual synopsis of new genera and species of animals, plants, etc., discovered in North America by C. S. Rafinesque. First annual number for 1820. Thomas Smith, Lexington, Kentucky, 16 pp. REDDELL, JaMEs R. 1965. A checklist of the cave fauna of Texas. I. The Invertebrata (exclusive of Insecta). Texas Jour. Sci. 17 (2) 143-187. —. 1970. A checklist of the cave fauna of Texas. IV. Additional records of Invertebrata (exclusive of Insecta). Texas Jour. Sci. 21(4) :389-415. REDDELL, JAMEs R., AND Ropert W. MircHetu. 1969. A checklist and annotated bibliography of the subterranean aquatic fauna of Texas. Texas Technological College, Water Resources Center, Spec. Rep. no. 24:1-48 (mimeographed). RicHarpson, Harriet R. 1905. A monograph on the isopods of North America. Bull. U.S. Nat. Mus. 54:LIII + 727 pp. 496 Proceedings of the Biological Society of Washington STEEvVES, Harrison R. III. 1968. Three new species of troglobitic asellids from Texas. Amer. Midl. Nat. 79(1):183-188. Uxricu, Cart J. 1902. A contribution to the subterranean fauna of Texas. Trans. Amer. Microsc. Soc. 23:83-101, pls. 14-18. Van NamMeE, Wittarp G. 1936. The American land and fresh-water isopod Crustacea. Bull. Amer. Mus. Nat. Hist. 71:1-535. NH 3, No. 46, pp. 497-502 22, January 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON SARSIELLA OZOTOTHRIX, A NEW SPECIES OF MARINE OSTRACODA (MYODOCOPINA) FROM THE // ATLANTIC AND GULF COASTS OF NORTH AMERICA| By Louis S. Kornicker AND Marcia BOWEN Department of Invertebrate Zoology, Smithsonian Institution, lee Washington, D.C. 20560, and New York Ocean Science Laboratory Montauk, New York 11954 Specimens of Sarsiella ozotothrix, the new species described below, were collected at three localities: 1, Stonewall Pond, Chilmark, Martha’s Vineyard, Massachusetts; 2, Long Island Sound between Eatons’s Neck, Long Island and Darien, Con- necticut; and 3, Alligator Harbor, Florida. The species was not reported from the Sapelo Island area of Georgia by Darby (1965), nor is it present in extensive collections from the Fort Pierce area of Florida examined by the senior author. This might indicate that the Atlantic and Gulf coasts populations are disjunct. The known distribution suggests that the species may have inhabited the southeastern coast of North America in the past, possibly during colder climates of the Pleistocene. A similiar explanation was proposed for the disjunct Atlantic- Gulf coast distribution of Sarsiella zostericola Cushman 1906, by Kornicker (1975:130). SUBORDER MYODOCOPINA SARSIELLIDAE Brady and Norman, 1896 Sarseilla Norman, 1869 Sarsiella ozotothrix, new species Figures 1-3 Etymology: The specific name is from the Greek ozotos = branching, branched, combined with thrix = hair and refers to the branching hairs or bristles on the carapace. 46—Proc. Bion. Soc. WAsH., Vou. 88, 1976 (497) 498 Proceedings of the Biological Society of Washington Fic. 1. Sarsiella ozotothrix new species, paratype, left lateral view of ovigerous female from Alligator Harbor, Florida, USNM 152441, length 0.95 mm. Material examined:—Alligator Harbor, Florida, Sta. I, 5-2, depth and precise locality unknown; Darrell K. Jones, collector, exact date unknown, about 1957; 1 ovigerous 2 paratype, USNM 152441. —Long Island Sound, sta. EB8-2, dump site between Eaton’s Neck, Long Island, and Darien, Connecticut, 41°00’N, 73°26’W, depth 23.5 m; sandy bottom, 93% sand, 3% silt, 4% clay; substrate and water surface temperature 3.0°C (substrate temperature taken aboard ship when grab was opened), Marcia Bowen collector, February 1975; 9 paratypes (re- turned to collector). —Martha’s Vineyard, Massachusetts, Stonewall Pond (between Gay Head and Chilmark) near ocean side, separated from ocean by bar; Mrs. C. R. Stoertz, collector, July 1965: Sta. 1, water depth 2 m (type- locality), 1 ovigerous @ holotype, USNM 152455 (in alcohol and on slides); 4 paratypes, USNM 152454. Sta. 6, water depth 3 m, 5 para- types including 1 ovigerous @ and 1 adult 2, USNM 152847, and 3 adult 6, USNM 153929 and 153920. Distribution: Martha’s Vineyard, Massachusetts; Long Island Sound; Alligator Harbor, Florida. Description of Adult Female: Carapace oval in lateral view with prominant caudal process and without rostrum or incisur (Figs. 1, 2a); carapace in posterior half widest dorsally, in anterior half widest ventrally. Ornamentation: Continuous oval ridge composed of bristles present on each valve approximately parallel to outer margin of valve except in vicinity of caudal process (Figs. 1, 2a); anteroventral segment of ridge less convex than anteroventral margin of valve; a rib composed of bristles present just within dorsal margin of valve; surface of valve with small flat-bottomed fossae; radial ribs composed of bristles per- pendicular to oval ridge present along anteroventral, ventral, and postero- ventral margins. Ostracode from eastern North America 499 AIL ut ugh wana ap 1) tile 4 Ww MAMAN Vlad ay MN LAC tit! NY) We he WA wn ws wy SS = = sted aye \ hh \ onan mM ly) LA . iN uit Fic. 2. Sarsiella ozotothrix new species, holotype, USNM 152455, ovigerous female from Martha’s Vineyard, Massachusetts: a, Left lateral view of left valve, open circles represent fossae which completely cover surface, black dots represent central adductor muscle endings; b, Detail of anterior infold of left valve near middle showing minute bristle on infold and 3 marginal bristles, medial view; c, Caudal process of left valve and 2 flagellate posterior bristles, medial view; d, Sketch of short and long bristles with branching tips that are abundant on outer sur- faces of valves. Bristles: Long bristles with pointed tips and broad ringed bases pres- ent along anterior and ventral valve margins and scattered over valve surface (Fig. 2a—c); long stout bristles with branching tips form ridges and ribs on valves (Fig. 2d); shorter bristles with branching tips pres- ent on bottoms of fossae and between fossae (Fig. 2d); transparent or- ganic substance present between bristles on most specimens. Infold: Anterior infold with minute bristle near middle (Fig. 2b); infold in front of caudal processes with 4—6 bristles (Fig. 2c); 2 fla- gellate bristles present on posterior infold dorsal to caudal process (Fig. 2c); a few minute bristles present along inner margin of infold in vicinity of caudal process. Selvage: Wide lamellar prolongation with smooth outer edge present along anterior, ventral, and posterior margins of both valves. Size: USNM 152441, length 0.96 mm, height 0.72 mm with caudal process; USNM 152454-A, length 0.99 mm, height 0.84 mm with caudal process, 0.83 mm without; USNM 152454-C, length 1.05 mm, height 0.86 mm with caudal process; USNM 152455-A, length 1.05 mm, height 0.93 mm with caudal process; USNM 152455-B, length 1.00 mm, height 0.85 mm with caudal process; adult female from Long Island Sound, length 0.95 mm, height excluding caudal process 0.70 mm. First antenna (Fig. 3a): Ist joint bare; 2nd joint with short bristle and long hairs on dorsal margin; 3rd joint with 2 short bristles, 1 ventral, 500 Proceedings of the Biological Society of Washington Fic. 3. Sarsiella ozotothrix new species, aj, Paratype, USNM 152441: a, Right Ist antenna, medial view; b, c, Endopodites of right and left 2nd antennae, medial views; d, Right mandible, medial view; e, Right maxilla, lateral view; f, Tip of 5th limb; g, 6th limb; h, 7th limb; i, Anterior of body showing lateral eye, medial eye and rod-shaped organ, anterior triangular process, and upper lip at bottom of figure; j, Pos- terior of body showing right lamella of furca, right genitalia and brush- like organ, and right Y-sclerite; k, USNM 152454-A, 6th limb. 1 dorsal; 4th joint with very short dorsal bristle, 2 long ventral bristles, and spines and hairs on medial surface and along ventral and dorsal margins; sensory bristle of 5th joint with 2 minute filaments; 6th joint with short medial bristle. Seventh joint: a-bristle about twice length of bristle of 6th joint; b-bristle about 4 times length of a-bristle, bare; Ostracode from eastern North America 501 c-bristle long with 2 minute filaments. Eighth joint: c- and d-bristles long, bare; e-bristle long, bare; f-bristle long with 3 minute filaments. Second antenna: Protopodite bare; endopodite with 1 short bare proximal bristle and 1 longer spinous terminal bristle (Fig. 3b, c). Exopodite: Ist joint with long proximal hairs along ventral margin and small medial spine on distal margin; bristle of 2nd joint long, with short slender spines along middle of ventral margin and long natatory hairs along proximal dorsal margin and distal part of ventral margin; bristles of joints 2-8 with natatory hairs; 9th joint with 3 bristles, 2 long with natatory hairs, 1 short with short slender marginal spines; spines ob- served along distal margins of joints 2-4; few small medial spines form- ing short row on joint 2 proximal to row of spines along distal margin. Mandible (Fig. 3d): Coxale endite consisting of stout spine; ventral margin of coxale with long hairs. Basale: dorsal margin with 1 minute subterminal spine; ventral margin with 5 minute spines: 3 proximal and I distal on medial surface; 1 proximal with base on lateral side. Endop- odite: 1st joint with distal medial spines and minute terminal spine on dorsal margin (not seen with certainty); no spines observed at base of main claw; 2nd joint with subterminal spine on dorsal margin; no spines observed at base of main claw; 3rd joint with minute dorsal spine at base of main claw; no teeth observed on claws of endopodite. Exop- odite absent. Maxilla (Fig. 3e): Endite I with 4 bristles; endite II with 3 bristles; endite III with 7 bristles; coxale with 1 anterior bristle; basale with 1 short bristle near exopodite; exopodite with 2 bristles, 1 short, 1 long; Ist endopodite joint with spinous alpha- and beta-bristles, and a few short spines distally on anterior margin; 2nd endopodite joint with 2 lateral a-bristles, 1 medial c-bristle, and usual 5 pectinate terminal bristles. Fifth limb: Epipodial appendage with 25-27 bristles; single endite with 1 short bristle. Exopodite (Fig. 3f): joints fused; Ist joint with 2 bristles; remaining joints with total of 5 bristles; surface of exopodite hirsute. Sixth limb (Fig. 3g, k): Endite with 3 bristles, end joint with 11 bristles. Seventh limb (Fig. 3h): Proximal group with 2 bristles, 1 on each side; distal group with 6 bristles, 3 on each side; each bristle with 2-5 bells; all bristles without marginal spines; terminus with opposing combs, each with about 6 or 7 opposing teeth. Furca (Fig. 3j): Each lamella with 5 pointed claws; claw 1 separated from lamella by suture; claws 2—5 joined to lamella; claws 1-3 with few stout and many slender teeth along posterior margins; 2 minute spines and many long hairs on lamella following claw 5. Eyes (Fig. 3i): Lateral eyes with 4-5 ommatidia; medial eye pig- mented, about same size as lateral eye. Rod-shaped organ (Fig. 3i): Elongate with proximal wrinkles and rounded tip. 502 Proceedings of the Biological Society of Washington Genitalia (Fig. 3j): Oval amber-colored process present on each side of body anterior tc furca. Brushlike organ (Fig. 3j): Consisting of 3 minute ringed bristles above genitalia. Posterior of body: Bare. Upper lip: Simple with rounded anterior process at tip. (Fig. 3i). Y-sclerite: Typical for family (Fig. 3j). Eggs: USNM 152441 with 6 eggs in marsupium, USNM 152454-A with 5 eggs. Comparisons: The new species, S. ozotothrix, differs from other spe- cies of Sarsiella in having on each valve a continuous well-developed oval ridge formed of bristles with branching tips. The morphology of the endopodite of the 2nd antenna differs from many species of Sarsiella in having 1 proximal and 1 terminal bristle. LITERATURE CITED Darsy, D.G. 1965. Ecology and taxonomy of Ostracoda in the vicinity of Sapelo Island, Georgia, pp. iii-vi, 1-76, text-figs. 1-89. Re- port No. 2 in Kesling, Darby, Smith, and Hall, Four Reports of Ostracod Investigations Conducted under National Science Foundation Project GB-26. University of Michigan, Ann Arbor. Kornicker, L. S. 1975. Spread of ostracodes to exotic environs on transplanted oysters, pp. 129-139, text-figs. 1-3. In F. M. Swain, L. S. Kornicker, and R. F. Lundin (eds.), Biology and Paleobiology of Ostracoda. Bull. Amer. Paleontol. 65( 282) :1— 687. Paleontological Research Institution, Ithaca. NH No. 47, pp. 503-508 22 January 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON TWO NEW INDO-WEST PACIFIC CARDINALFISHES // OF THE GENUS APOGON I] By Tuomas H. FRASER AND JOHN E. RANDALL Environmental Quality Laboratory, | 2 Port Charlotte, Florida 33952 and ile Bernice P. Bishop Museum, Honolulu, Hawaii 96818 —\ Recent deep-SCUBA collections of coral reef fishes in the Solomon, Palau and Cocos-Keeling Islands have yielded two closely related species of Apogon. These are described herein and their subgeneric relationships discussed. All measurements were made with dial calipers. Gillrakers counts include all rudiments. The raker at the angle of the arch is included in the lower arch count. The last dorsal and anal rays are a composite of two elements divided to the base; they are counted as one. Radiographs have been examined for all specimens. Type-material is deposited at the following museums: Acad- emy of Natural Sciences, Philadelphia (ANSP); Australian Museum, Sydney (AMS); Bernice P. Bishop Museum, Hono- lulu (BPBM); California Academy of Sciences, San Fran- cisco (CAS); Rijksmuseum van Natuurlijke Historie, Leiden (RMNH); National Museum of Natural History, Washington, D.C. (USNM). The senior author gratefully acknowledges support from the South African Council for Scientific and Industrial Re- search, Rhodes University and the Smithsonian Institution. The junior author received support for field work from the National Science Foundation and the National Geographic Society. We thank James F. McKinney and Janet Gomon for taking radiographs of the specimens. E. A. Lachner and M. M. Smith are thanked for reading draft portions of the manuscript. 47—Proc. Bion. Soc. WAsH., Vou. 88, 1976 (503 ) 504 Proceedings of the Biological Society of Washington Apogon dispar, new species Figure | Diagnosis: A species of Apogon with a narrow, red band from the tip of the snout to the base of the caudal fin, ending in a squarish red spot; two small red spots behind and slightly above the eye; and caudal fin lobes broadly tipped with red. Description: For general shape and pigment pattern see Figure 1. Proportions for nine of the 14 specimens 37-48 mm SL (as percent of standard length with holotype in parentheses): body depth (32) 28— 32; head length (37) 36-40; eye length (12) 11-12; snout length (10) 9-12; bony interorbital width (7) 7; upper jaw length (18) 17-18; caudal peduncle depth (10) 10-13; caudal peduncle length (28) 28— 31; dorsal spine length—first (1) 1-2, second (11) 10-11, third (17) 15-17, fourth (15) 14-15, last (14) 13-14 (spine is in second dorsal); anal spine lengths—first (3) 2-3, second (14) 14-15; pectoral fin length (20) 20-21; pelvic fin length (18) 18-21. Dorsal fin VII—I,9 except one with VII—I,8; anal fin II,8 except one 11,9; pectoral fin (both sides counted) 13(11) or 14(5), frequencies in parentheses; pelvic fin 1,5; principal caudal rays 9+8; secondary caudal rays 6-10 above and 7-10 below; well developed gillrakers 21-25 (4-5+17-20), including rudiments 23-26 (0-3-+-4-5; 17-20+0-1), the holotype 0+-5-18+1; pored lateral line scales 25; transverse scale rows above lateral line 2; median predorsal scales 4 or 5; circumpeduncular scales 12-13 (5-2-5 or 6); other scale counts could not be taken because of lost scales. Premaxilla with two rows of teeth anteriorly becoming a narrow band on the side; all teeth villiform with the outer teeth larger than those of the inner row. Dentary with two rows of teeth anteriorly becoming one row on the side, all villiform and about the same size; vomer with one row of three small teeth on each side; none to three teeth on the palatine, no teeth on ectopterygoid or endopterygoid, occasionally teeth on base of basihyal. Vertebrae 10+14; basisphenoid present; supramaxilla absent; 2 pre- dorsals; 2 dorsal spines on the first pterygiophore; 7 epipleural ribs; 5 free hypurals; 3 epurals; 1 pair of uroneurals; suspensory pharyngeal present; 4 upper pharyngeal tooth patches; 7 branchiostegal rays; no hidden eighth dorsal spine; posttemporal serrated; preopercular ridge smooth, ventral and posterior edges serrated; infraorbitals smooth; shelf on third infraorbital. Swim bladder with a single rete bundle and gas gland; the presence of an oval could not be established. Coloration: From an Ektachrome transparency taken soon after death: Most of body translucent with orange-red and white markings; an orange- red stripe extending from tip of lower jaw to anterior margin of eye, thense continuing as a narrow post-ocular stripe, ending at base of New Indo-West Pacific cardinalfishes 505 Fic. 1. Apogon dispar, holotype, BPBM 9378, 44.1 mm SL, Palau Islands, photographed freshly fixed in formalin. caudal fin in a squarish spot; a small white spot above this orange-red spot; two small reddish spots behind and slightly above eye; caudal fin lobes broadly tipped with red; cheek and abdomen silvery. Color in 55% isopropyl alcohol: All fins pale and translucent; a faint dark band along midside becoming a diffuse basicaudal mark; scales above lateral stripe outlined with black chromatophores; a small patch of dark chromatophores on nape behind and slightly above eye; dusky stripe on snout to tip of lower jaw; peritoneum white with many small dark chromatophores; intestine white with few dark chromatophores, stomach becoming blackish with numerous chromatophores. Coloration of the Cocos-Keeling Island specimens similar in 70% ethyl alcohol, ex- cept lobes of caudal fin dusky. Distribution: Apogon dispar is known from 14 specimens taken at the Palau and Solomon Islands in the Pacific Ocean and from Cocos- Keeling Islands in the Indian Ocean, at depths of 18-58 m. Holotype: BPBM 9378 (44.1 mm SL), Palau Islands, Augulpelu Reef at southwest end, small cave at base of drop-off, depth 36 m, 13 April 1970, J. E. Randall. Paratypes: BPBM 17017 (2 specimens, 44.3-44.8 mm), USNM 212134 (2, 45.8-48.2), USNM 212135 (1, 44.3, cleared and stained), CAS 30648 (1, 48.1), Solomon Islands, Florida Island, south side of Tanavula Point, 09°02’44”S, 160°04’07”E, vertical reef wall with caves, 18-37, 30 July 1973, Randall and G. R. Allen. ANSP 128364 (6, 21.7-38.1) Cocos-Keeling Islands, Turk Reef, vertical face, 51-58 m, 25 March 1974, W. F. Smith-Vaniz and P. Colin. ANSP 128365 (1, 38.6) Cocos- Keeling Islands, Turk Reef, drop-off reef face and cave, 45-49 m, 8 March 1974, Smith-Vaniz and Colin. Etymology: The latin adjective dispar, meaning unlike or different, is used because of the unusual body shape and coloration of this Apogon. 506 Proceedings of the Biological Society of Washington Taste 1. Frequency distributions of selected characters of Apogon dispar and Apogon melanoproctus. Upper arch Total Second dorsal gillrakers gillrakers spine in % of SL 5 G6 7F G3} Be Ol bs 3 Or @® )«li) 1ilg A. dispar 6* 6 1 = 1 S25 2. =) eo A. melanoproctus = 2 9* 1 = = 2 8 2 = ieee Caudal peduncle Body depth length in % of SL in % of SL 26 27 28 29 30 31 28 29 30 31 32 33 B84 A. dispar = = 2 4 2 OO 93 A. melanoproctus Io @ = = ea = = 8 4 3 = Upper jaw Pectoral fin length in % of SL in % of SL 15 16 17 18 20) 2 2223 A. dispar - 9 3 4 Be B= = A. melanoproctus G@ @ = = = a) SESS * Includes holotype Apogon melanoproctus, new species Figure 2 Diagnosis: A species of Apogon with a faint brown-orange caudal base with a white spot above; a white spot just behind the base of the soft dorsal fin; and a black area around the anus. Description: For general shape and pigment pattern see Fig. 2. Pro- portions for the 12 specimens (as percent of standard length with holo- type in parentheses): body depth (32) 31-34; head length (39) 36— 39; eye length (12) 12-13; snout length (10) 8-10; bony interorbital width (7) 7-8; upper jaw length (16) 15-16; caudal peduncle depth (13) 12-14; caudal peduncle length (27) 26-28; dorsal spine lengths— first (1) 1-3, second (11) 11-13, third (17) 15-18, fourth (16) 15-16, last (13) 13-15 (spine is in second dorsal); anal spine lengths—first (3) 2-3, second (15) 14-15; pectoral fin length (22) 22-23; pelvic fin length (21) 19-22. Dorsal fin VII—1,9; anal fin II,8 except one with II,9; pectoral fin (both sides counted) 14 (11), 13 (1), frequencies given in parentheses; pelvic fin 1,5; principal caudal rays 9-++-8; secondary caudal rays 8-9 above and 8-9 below; well-developed gillrakers 22-24 (4-5+18-19), including rudiments 25-27 (2-3-++-4-5; 18-19+0-1), holotype 2+5-19 +0; pored lateral line scales 24; longitudinal row above lateral line 24; New Indo-West Pacific cardinalfishes 507 Fic. 2. Apogon melanoproctus, holotype, BPBM 15474, 40.0 mm SL, Solomon Islands, photographed freshly fixed in formalin. transverse rows above lateral line 2; transverse rows below lateral line 6; median predorsal row 5; circumpeduncular scales 12 (5-2-5). One row of teeth on dentary; two or three rows on premaxilla be- coming one row on side; teeth on palatine few or absent; no teeth on vomer, ectopterygoid or endopterygoid; occasionally teeth on base of basihyal. Vertebrae 10+14; basisphenoid present; supramaxilla absent; 2 pre- dorsals; 2 dorsal spines on the first pterygiophore; 7 epipleural ribs; 5 free hypurals; 3 epurals; 1 pair of uroneurals; suspensory pharyngeal present; 4 upper pharyngeal tooth patches; 7 branchiostegal rays; no hidden eighth dorsal spine; posttemporal serrated; preopercular ridge smooth, ventral, and posterior edges serrated; infraorbitals smooth; shelf on third infraorbital. Swim bladder with a single bundle and gas gland; the presence of an oval could not be established. Coloration: From an Ektachrome transparency taken soon after death: Most of body translucent with brownish-orange, black and white mark- ings; tip of lower jaw brownish-orange; snout and post-ocular region without a well defined stripe, but with some orange pigment; base of caudal with a brownish-orange bar and a smaller adjacent white spot positioned anterodorsally; another white spot at base of last dorsal ray; a black area around anus; cheek and abdomen silvery. Coloration in 55% isopropyl alcohol: All fins pale and translucent; a faint basicaudal mark displaced slightly dorsal to the pored lateral line scales; a dusky stripe on snout to tip of lower jaw; only nape scales out- lined with dark chromatophores; anal and genital openings surrounded by circular black spot; peritoneum white with many small dark chromato- phores; intestine and stomach black. 508 Proceedings of the Biological Society of Washington Distribution: Apogon melanoproctus is known from one collection of 12 specimens from the Solomon Islands. Holotype: BPBM 15474 (40.0 mm SL), Solomon Islands, Florida, Island, south side of Tanavula Point, 09°02’44”S, 160°04’07”E, vertical reef wall with caves, 18-37 m, 30 July 1973, J. E. Randall and G. R. Allen. Paratypes: (All same data as holotype) USNM 212133 (1 specimen, 38.8 mm, cleared and stained), USNM 212132 (2, 36.0-38.1), RMNH 27056 (2, 36.5-37.5), BPBM 15475 (2, 37.2-38.0), CAS 30649 (2, 30.1- 37.3), AMS I. 17888-001 (2, 34.5-37.2). Etymology: The name melanoproctus is a Greek noun from melanos, black, and proktos, anus. It refers to the black pigment that completely surrounds the anus. Discussion: These two species are closely related to each other, based on the combination of seven visible first dorsal spines, two predorsal bones, a pair of uroneurals and similar body shapes and dentition. The combination of two predorsal bones and ‘seven spines in the first dorsal is unusual in seven-spined species of Apogon and is known only to oc- cur in Apogon queketti Gilchrist, out of the approximately 49 seven- spined species examined by Fraser (1972) in the subgenera Nectamia and Pristiapogon. The new species do not belong to the group of species that includes A. queketti nor do they belong in Pristiapogon. We ten- tatively assign the new cardinalfishes to the subgenus Nectamia. The subgeneric classification of Apogon proposed by Fraser (1972) was based on the osteological study of about 80 of the approximately 130 valid species. Future evaluation (now in progress) of the subgenera and species groups is necessary before stable and, hopefully, definitive understanding of this large genus is attained. Apogon dispar and melano- proctus may eventually warrant subgeneric designation. Like other deeper-dwelling reef apogonids, we expect these species to have wider distributions than are presently recorded. LITERATURE CITED Fraser, T. H. 1972. Comparative osteology of the shallow-water cardinal fishes (Perciformes: Apogonidae) with reference to the systematics and evolution of the family. Ichthyol. Bull. Rhodes Univ. 34:i-v, 1-105, 43 pls., 6 tabs. GH ut BAX NH No. 48, pp. 509-514 22 January 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON KEY TO SUBFAMILIES OF ROBBER FLIES BASED ON PUPAL CASES, WITH A DESCRIPTION OF THE | - PUPAL CASE OF DORYCLUS DISTENDENS : (ASILIDAE: MEGAPODINAE) By Lioyp KNuTSsoN nee Systematic Entomology Laboratory, IIBIII, Agr. Res. Serv), USDA, Beltsville, Maryland 20705 The immature stages of Asilidae are poorly known, as are those of most other families of Diptera. In my review (Knut- son, 1972), I noted that only about 2% of the approximately 5,000 world species of Asilidae are known in any immature stage. Only four of the five generally recognized subfamilies are included in this number. Most of the relatively few genera and species for which descriptions of immature stages are available are members of the three largest subfamilies (Asi- linae, Dasypogoninae, and Laphriinae). Limited data are available for the Leptogastrinae. Until now, nothing has been published on the immature stages of the remaining subfamily, the Megapodinae. Doryclus distendens (Wiedemann) Megapodinae Pupal case (Figures 1-3) Length (including anterior antennal processes), 10.8-12.8 mm; great- est width of thorax, 2.3-3.0 mm; greatest width of abdomen, 2.4—3.2 mm, tapering to 1.5-2.0 mm at greatest width of last abdominal seg- ment. Subshining straw yellow, integumentary processes glistening reddish brown. Head with a pair of large, acuminate, straight anterior antennal processes (aap) not joined at base, a group of 3—4 basally con- fluent posterior antennal processes (pap), and one truncate dorsal an- tennal process (dap) at base of each anterior antennal process. Facial area with a pair of small medial spines (fs). Labral sheath (Is) and hypopharyngeal sheath (hs) elongate and smooth. Maxillary sheath 48—Proc. Biou. Soc. WAsH., Vou. 88, 1976 (509 ) 510 Proceedings of the Biological Society of Washington pmc ———}._ i ih ead = i j\= ' WMV ANY —, => ie HSMN viny Fics. 1-3. Doryclus distendens (Wiedemann), pupal case. 1, lateral view; 2, ventral view; 3, dorsal view. aap, anterior antennal process; abs, abdominal spiracle; acs, anterior coxal sheath; dap, dorsal antennal process; fs, facial spines; hs, hypopharyngeal sheath; ls 1, 2, 3, leg sheaths; Is, labral sheath; ms, maxillary sheath; pap, posterior antennal process; pmc, posterior mesothoracic callosity; sos, suborbital spine; ths, thoracic spiracle; tpr, terminal process; ws, wing sheath. (ms) with pair of elongate areas extending below labral sheath, almost to apex of anterior coxal sheath (acs). Two or three suborbital spines (sos) on either side of head ventrolaterally above base of anterior coxal sheath. Paired thoracic spiracles (ths) elongate oval, surrounded by a C-shaped ring of thickened cuticle, situated dorsally at anterior margin of thorax on rugulose area. Without anterior or posterior mesothoracic spines, but with a well developed posterior mesothoracic callosity (pmc) at base of wing. Sheaths of third pair of legs reaching anterior margin of abdominal segment 4. First abdominal segment with dorsal trans- verse row of about 30 erect, subequal, more or less uniformly shaped, straight, spinelike, yellowish processes; about 8 similar processes in Pupal cases of robber flies 511 transverse row behind abdominal spiracle (abs); venter obscured. Sec- ond segment with median transverse row of about 20 short, stout, dull, dark, posteriorly directed processes, and with elongate, straight, yellowish spinelike processes interspersed between every 3rd stout process dorsally; about 11 such spinelike processes laterally on each side behind spiracle, and a mesally interrupted row of such spines ventrally. Third through seventh abdominal segments similar to each other, each with stout pro- cesses and spinelike processes dorsally as on second segment, with stout processes becoming recurved to erect, and sharper on posterior segment; 10-16 spinelike processes laterally behind each abdominal spiracle, and complete transverse row of spinelike processes ventrally. Abdominal segment 8 (terminal segment) composed of a ringlike anterior portion with a transverse circlet of processes, interrupted midventrally or ab- sent ventrally, and a tapered posterior portion with a pair of strong, dark, dorsolateral terminal processes and a silghtly smaller pair of ven- tral terminal processes (tpr). Martin and Papavero (1970) gave the distribution of D. distendens as, “Brasil: Santa Catarina. Para, Santarem.” Six pupal cases of D. distendens are in the collection of the U.S. National Museum, Wash- ington. Three specimens are labeled only, “Z-3224”, “A. E. Pritchard Collection, 1962.” Three other specimens are labeled, “Parita, Panama, Zetek #1689;” one of the latter pupal cases contains a well-formed, partially emerged adult. Bromley (1951) was apparently referring to these six specimens when he noted “The Megapodinae are wasp-like in appearance and are undoubtedly mimetic of certain aculeate Hyme- noptera with which they are associated in a Mullerian association. Their affinities with the other Laphriinae are indicated by the fact that their larval and pupal stages take place in decaying wood. Dr. Zetek has reared certain species of this group from hollow citrus trees in Panama.” THE SUBFAMILY CLASSIFICATION OF ASILIDAE Various authors have used two, three, four, five, or eight subfamilies for the classification of the approximately 400 world genera of Asilidae. Bromley (1946) recognized four subfamilies: Asilinae, Dasypogoninae, Laphriinae, and Leptogastrinae. Hull (1962) recognized five subfamilies: Asilinae, Dasypogoninae, Laphriinae, Leptogastrinae, and Megapodinae. Martin (1968) raised the Leptogastrinae to family status. Martin and Papavero (1970) recognized only three subfamilies, considering the Laphriinae to represent tribes in the Dasypogoninae, and the Lepto- gastrinae as a separate family. Papavero (1973a) recognized eight sub- families of Asilidae (Apocleinae, Asilinae, Dasypogoninae, Laphriinae, Laphystiinae, Ommatiinae, Stenopogoninae, and Trigonomiminae). Carrera (1949) proposed the tribe Megapodini for five Neotropical genera (Doryclus Jaennicke, Megapoda Macquart, Pronomopsis Hermann, Pseudorus Walker, and Senobasis Macquart). The Megapodini were raised to subfamily rank by Hull (1962). Forty-three species of Mega- 512 Proceedings of the Biological Society of Washington podinae in six Neotropical genera (the above five genera plus Pseu- doryclus Carrera) were recognized by Martin and Papavero (1970). Oldroyd (1974) p. 17, commented, “Hull is, therefore, justified in re- taining this tribe [Megapodini], though perhaps not in raising it to the status of a subfamily.” Papavero (1973b) included Lagodias Loew (Ethiopian, Palearctic, and Oriental Regions) and Pegasimallus Loew (Ethiopian Region) with the six Neotropical genera mentioned above in the tribe Megapodini, subfamily Dasypogoninae. Papavero’s tentative classification of the subfamilies is based on phylo- genetic considerations, is worldwide in scope, and is probably a better arrangement than the classifications of previous authors. However, to facilitate identification of asilid pupae and pupal skins and for the purposes of the key presented below, I am recognizing the five groups accorded subfamily status by Hull (1962). Kry To SUBFAMILIES OF ASILIDAE BASED ON PUPAL CASES 1. Anterior antennal processes small, truncate; posterior antennal pro- cesses absent; last abdominal segment with only 1 pair of ter- minal processes; abdomen laterally with elongate (longer than length of segment), hairlike processes Leptogastrinae Anterior antennal processes large, acuminate; posterior antennal processes present; last abdominal segment with 2-6 pairs of terminal processes; abdomen laterally with short (shorter than length of segment), spinelike processes 2 2. Dorsal antennal processes present; anterior antennal processes al- most straight; without anterior or posterior mesothoracic spines, but with posterior mesothoracic callosity; with suborbital spines and) medial ttaciallspines! esse a Megapodinae Dorsal antennal processes absent; anterior antennal processes some- what curved ventrad; with anterior and/or posterior mesothoracic spines, without suborbital spines; with or without medial or lat- eral facial spines23 vie sath ee be Sy ee 3 3. Lower face with small medial or lateral spines; posterior antennal processes usually with 4—5 confluent hooks each _______ Laphriinae Lower face without medial or lateral spines; posterior antennal pro- cesses usually with 3 confluent hooks each 4 4, Abdomen laterally behind spiracles usually with only 3 spinelike processes on each side; last abdominal segment with 2 or more Pairs) Of terminal sprocessesy =e ta wae ene Dasypogoninae Abdomen laterally behind spiracles usually with more than 5 spine- like processes on each side; last abdominal segment with 4 pairs of terminal processes, ee Asilinae ACKNOWLEDGMENTS I thank Miss Pamela Hollyoak for preparing the illustrations and Drs. D. Steve Dennis, Robert J. Lavigne, Nelson Papavero, and Aubrey G. Scarbrough for reviewing the manuscript. Pupal cases of robber flies 513 LITERATURE CITED BroMLey, S. W. 1946. Third fascicle. Asilidae. In Guide to the insects of Connecticut. Pt. VI. The Diptera or true flies of Connecticut. Conn. Geol. Nat. Hist. Surv. Bull. 69. viii + 51 pp. 1951. Asilid notes (Diptera), with descriptions of thirty- two new species. Amer. Mus. Novitates No. 1532:1-36. Carrera, M. 1949. Contribuicao ao conhecimenta dos Asilidae Neo- tropicais (Diptera). Arq. Zool., Sio Paulo 7(1):1-148. Hutz, F. M. 1962. Robber Flies of the World. The genera of the family Asilidae. U.S. Nat. Mus. Bull. 224, pt. 1:i-x, 1-430; pt. 2:433-907. Knutson, L V. 1972. Pupa of Neomochtherus angustipennis (Hine), with notes on feeding habits of robber flies and a review of publications on morphology of immature stages (Diptera: Asilidae). Proc. Biol. Soc. Wash. 85:163-178. Martin, C. H. 1968. The new family Leptogastridae (the Grass Flies) compared with the Asilidae (Robber Flies) (Diptera). Jour. Kans. Entomol. Soc. 41:70—100. Martin, C. H., anp N. Papavero. 1970. Family Asilidae. Fascicle 35b:1-139. In a Catalogue of the Diptera of the Americas South of the United States. Mus. Zool., Univ. Séo Paulo, Sao Paulo, Brazil. Oxtproyp, H. 1974. Some comments on the tribal classification of Asilidae (Diptera). Israel Jour. Entomol. 9:5-21. Papavero, N. 1973a. Studies of Asilidae (Diptera) systematics and evolution. I. A preliminary classification in subfamilies. Arq. Zool., SAo Paulo 23(3):217-274. 1973b. Studies of Asilidae (Diptera) systematics and evolu- tion. II. The tribes of Dasypogoninae. Arq. Zool., Séo Paulo 23(4):275-294. 514 Proceedings of the Biological Society of Washington , No. 49, pp. 515-548 22 January 1976 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON CLARIFICATION OF FIVE GENERA OF PHOXOCEPHALIDAE (MARINE AMPHIPODA)? By J. LAurRENs BARNARD AND MarcareT M. DRUMMOND Department of Invertebrate Zoology, Smithsonian Institution, Washington, D.C. 20560, and Fisheries and Wildlife Division, Marine Pollution Studies Group, 605 Flinders St. Extension, Melbourne, Victoria 3000, Australia The following four genera were treated as subjective syn- onyms of Paraphoxus Sars, 1895, by J. L. Barnard (1960): Pontharpinia Stebbing, 1897, Parharpinia Stebbing, 1899, Pro- tophoxus K. H. Barnard, 1930, and Trichophoxus K. H. Bar- nard, 1930. We present new evidence to effect their revival and to suggest that a fifth genus, Metharpinia Schellenberg, 1931, is also valid. We have examined the type-species of all genera except Metharpinia and have found minute char- acters, heretofore unreported, that confirm the distinction of the genera. This study is an outgrowth of a larger work, in preparation, concerning 86 species in 23 genera of Australian Phoxocephalidae. All but three of the Australian genera are new and 80 of the 86 species are new. This pool of taxa ap- proximately doubles the number of species and genera in this family and has given.us new information on the generic clas- sification of Phoxocephalidae. We believe that the Australian fauna of Phoxocephalidae contains the most primitive living members of the family but also contains an array of highly advanced genera so as to present a full display of systematic modifications in generic characters. The Australian fauna contains one new endemic genus with 38 species, several new genera with 4-8 species and many others with fewer species. The tightly knit generic characterization of the largest genus, 1 Publ. 35 in the Ministry of Conservation, Victoria, Environmental Studies Series. 49—Proc. Bio. Soc. WasuH., VoL. 88, 1976 (515) 516 Proceedings of the Biological Society of Washington allied with the demonstrated cohesiveness of characters in the genera comprising 4-8 species, provides a background out of which the generic validity of Pontharpinia, Parharpinia, Proto- phoxus and Trichophoxus can be established. The large genus, Birubius, is established with its type-species but the other 37 species are reserved for description in another place (Barnard and Drummond, in prep.). MATERIALS AND METHODS Specimens of the type-species of the 4 genera (2 of these Australian) to be reestablished plus 84 other Australian species in 21 genera and 20 additional species formerly assigned to Paraphoxus have been examined. Approximately 250 char- acters have been analyzed in each species. Many of these are characters not heretofore considered as important for record- ing in Phoxocephalidae and might be termed “minute” or “trivial” characters, for example, the presence, number, size and groupings of spines on the medial surfaces of appendages. Among many other “trivial” characters, the fine structure of right lacinia mobili and of cuticular setules has also been examined. Many of the characters actually are class-char- acters, those in which numerous alternatives of condition have evolved. Many new kinds of attributes have been discovered, mainly those occuring in Australian species but rarely else- where as far as we can determine from the phoxocephalid literature. For example, several genera, including Tricho- phoxus, to be revived, bear a spike on uropod 1, composed of a large interramal spine on a ventral protrusion of the peduncle. Characters such as “non-skid” cuticle, heretofore found only in genera of nonphoxocephalid families are found to occur in Australian species. A list of these characters will be presented by Barnard and Drummond (in prep.). The 86 Australian species were arranged in species clusters according to subjective appraisal of their common characters and the clusters then ordered in numerous trials based on models concerning evolutionary flow in morphocharacters. The main precept of each model was a manifold series of de- cisions as to the primitive or advanced state of each of the characters, these being juxtaposed and evaluated simulta- Clarification of five amphipod genera 517 neously in each model. The principle of parsimony was ap- plied to the system and the tenet of irreversibility was adopted, so that once an attribute was lost entirely it could not be re- gained in a later step. The large genus, Birubius, originally comprising 40 species assembled by subjective taxonomic means, was found to have 9 clusters of species. One of these clusters, of 2 species, was found to have qualitative characters separating it from the other 8 clusters. The characters were seen to have apparent generic significance when compared with characters in other taxa. Those other genera had been separated by ordinary taxonomic evaluations and not by clustering methods. The distinctions, comprising fully quali- tative characters, found in the separation of 2 species from 38 others, suggested the need to search for similar kinds of dis- tinctions in an overall generic analysis of 86 species. The 8 clusters remaining in the large genus were found to be dif- ferentiated only by quantitative characters. We define “quali- tative” as meaning present or absent; we define “quantitative” either as being slightly different in shape, proportion, degree and count, or subject to evidence of intergradation from one extreme to another, such as an accessory nail being represented in another species by a vestigial setule. These manipulations were extended to the remaining Australian phoxocephalids so as to result in 86 species being divided among 23 genera. The nomenclature of the genera was then rectified by examining the type-species of named genera and fitting those species to the generic pool. In this process we determined that the 4 genera to be reestablished in this paper are highly distinctive and not synonymous with Paraphoxus. Pereopods are counted as gnathopods 1-2 and pereopods 3-7. The term “mark” (M.) refers to a stated distance from a basal point on a 0 to 100 scale. The term “ordinary” is used in the descriptions to avoid lengthy explanation for charac- teristics of common occurrence in phoxocephalids. An or- dinary rostrum is elongate and unconstricted from dorsal aspect. Ordinary setation on maxilla 2 is a contrast apposed to “weak” setation and is a subjective value judgment. Or- dinary inner plates of the maxilliped are broad, elongate, and unfused medially. Ordinary pereopod 7 refers to the great size 518 Proceedings of the Biological Society of Washington of this appendage and its lack of facial setae. The contrast to ordinary refers to the highly miniaturized pereopod 5 of Trichophoxus (see illustrations). The term “mitelloid” refers to the shape of the hands on gnathopods 1-2 in Trichophoxus which resemble goose-neck barnacles. An epimeron 3 of “rounded-classification” bears a convex posterior margin, no posteroventral tooth and, in addition, has fewer than 3 long setae anywhere on the epimeron. Occasional examples of this kind of epimeron may have 3 extremely small setules pos- teroventrally. The “special spine” on the peduncle of uropod 1 refers to its displacement on the apex in a ventral direction away from the dorsal margin of the peduncle. Broad and narrow forms of article 2 on pereopod 5 refer to vastly dis- tinctive alternatives in width of that article. Facial counts of spines on article 4 of antenna 1 commence at the distal end. Spine and seta counts on article 1 of the outer ramus on uro- pod 3 commence proximally; the lateral margin bears ac- clivities each with one or more spines and setae, thus the formula, acclivities = 8, spines = 0 X 5, 1 X 4, setae = 1 X 5, 2 X 2 and 2 X 3 translates to “spine absent on first 5 acclivities, one seta present on each of next 3 acclivities plus apex of article (counted as ninth armament position), one seta on each of first 5 acclivities, 2 each on next 2 acclivities, 3 each on next acclivity plus apex.” HIisTorRY OF THE PARAPHOXUS PROBLEM Paraphoxus Sars, 1895, based on P. oculatus (Sars), was established to separate that species from the familial type- genus, Phoxocephalus Stebbing, on the equality of gnathopods and the nontriturative molar. Paraphoxus oculatus is pri- marily a northern coolwater species widely distributed in the Atlantic and Pacific Oceans. In the description of the genus Sars erred in stating that the first maxillary palp is uniartic- ulate; instead, it is biarticulate. This unfortunate error in an otherwise superlative description commenced a series of mis- steps culminating in the gradual amalgamation of several phoxocephalid genera described in later years. The final re- sult was the synonymy promulgated by J. L. Barnard (1960) in which Pontharpinia, Parharpinia, Protophoxus and Tricho- Clarification of five amphipod genera 519 phoxus were treated as subjective junior synonyms of Para- phoxus. Many good taxonomic characters had been over- looked by that time, those characters having been considered to be valuable only at specific level. Stebbing (1897) established Pontharpinia for Urothoe pin- guis Haswell, 1879b. Stebbing did not mention Paraphoxus in his comparisons, and unfortunately his diagnosis is erroneous because he did not have in hand Urothoe pinguis Haswell but another species of phoxocephalid now known as Paraphoxus stebbingi J. L. Barnard, 1958. We have learned that pinguis and stebbingi are not congeneric, indeed they are vastly dif- ferent and probably should be relegated to distinct subfamilies (to be considered elsewhere) within the Phoxocephalidae. Although Pontharpinia is a confounded genus we believe that it should be retained, with Urothoe pinguis as type-species; a new diagnosis of the genus is presented below. Paraphoxus stebbingi belongs to a new genus with affinities to Tricho- phoxus. That genus will be described elsewhere but Stebbing’s species will not be designated as the type-species. Although we recognize the genus to which P. stebbingi belongs, we have been unable to locate Stebbing’s specimen either in the British Museum of Natural History, the Universitets Zoologiske Mu- seum in Copenhagen or The Australian Museum. In the Australian study we have not been able to identify stebbingi among the 5 species in the genus. The specimen Stebbing had in hand was one set to him by Haswell long after the establishment of U. pinguis and from a different site than the type-locality. Stebbing’s Pontharpinia was included in a paper otherwise concentrated on collections of the Copenhagen Museum but apparently the specimen was not given to the Copenhagen Museum nor returned to Haswell and deposited in the Australian Museum, nor was it given to Stebbing’s na- tional museum, the British Museum of Natural History. To trace the exact species Stebbing had in hand may require in- vestigation of the kinds of amphipods in Jervis Bay, New South Wales, the original site of Stebbing’s specimen. The 5 species of the new genus, to be described, that will contain Paraphoxus stebbingi, are characterized interspe- cifically by spination patterns on uropods 1-2 and the pres- 520 Proceedings of the Biological Society of Washington ence, absence, or degree of development of combs on those uropods. The crucial Beseupuen of these parts was not given by Stebbing. Pontharpinia is highly distinc! from Paraphoxus. J. L. Bar- nard (1960) made a partial evaluation of the type-species pinguis based on notes made by K. Sheard and concluded that the genus was synonymous with Paraphoxus. Unfortunately the mandible of the so-called type-specimen was missing; that mandibular molar is triturative, thus demonstrating a strong distinction. We have now identified numerous speci- mens of pinguis by comparison with that so-called type-speci- men and with Haswell’s description and find numerous dis- tinctions from Paraphoxus as outlined in the generic diagnosis to follow. Stebbing (1899) established Parharpinia in a short para- graph, with Phoxus villosus Haswell, 1879a, as type-species. Again, this is a confounded genus because Stebbing had in hand a specimen from New Zealand loaned to him by G. M. Thomson and identified as Phoxus batei. Stebbing was under the impression that this entity was synonymous with Phoxus villosus from Australia. Hurley (1954) demonstrated that the specimen identified by Thomson as P. batei is in fact Ponthar- pinia australis (K. H. Barnard), originally described as the type-species of Protophoxus and now to be returned to that appellation. Phoxus batei Haswell 1879a, itself is a good spe- cies to be relegated to Birubius, new genus. Stebbing (1906) relegated P. batei to P. rostratus (Dana, 1853), but that synonymy cannot be supported. Parenthetically, the so-called original material of Phoxus batei is confounded; the specimens labeled Phoxus batei residing in the Australian Museum are clearly not what Haswell described; we can identify what Haswell described with fair precision. J. L. Barnard (1974) discussed this recurrent problem that most, if not all, of Has- well’s amphipod types were labeled “Type” or perhaps re- identified at a much later date, between 24 and 74 years after the fact, some apparently in the early 1900’s by unknown hands but some as late as the 1950’s by the late Dr. Sheard. In most cases the later selection was correct, but, the specimens se- lected for P. batei belong to Birubius. We believe that in most Clarification of five amphipod genera 521 cases Haswell did not participate in the selection of specimens labeled “Type” and that therefore they would qualify, at best, as lectotypes, once codified by publication as such. We in- tend to ignore publishing the specimen of P. batei now called “Type so as to conserve Haswell’s name for an easily rec- ognized species. We now know that Parharpinia differs from Paraphoxus in numerous characters as seen in the diagnoses to follow but more importantly, Parharpinia, with 2 species in Australia, forms a genus with intergradations between Birubius, the basic Australian genus (containing 38 species), and the two groups of genera to which Trichophoxus and Protophoxus belong. Birubius carries only apicoposterior telsonic spines, whereas Parharpinia carries dorsal telsonic spines or setae and further- more bears a special spine on uropod 1. Parharpinia also ap- pears to be near the ancestral pool for several other new Aus- tralian genera by virtue of posterior setation on epimera ]-2. Protophoxus K. H. Barnard, 1930, for P. australis K. H. Bar- nard, 1930, was described sketchily to distinguish a species thought by Barnard to bear relationship to Phoxocephalus and therefore was distinguished from that genus but not from Paraphoxus nor Pontharpinia. The latter genus had by that time become the repository for the bulk of paraphoxinlike species because those species fitted the diagnosis of Ponthar- pinia best in light of its known biarticulate palp of the first maxilla and the broad pereopods 5-6; other species had, how- ever, been placed in Parharpinia to fit the concept of thin pereopods 5-6. K. H. Barmard’s diagnosis of Protophoxus con- tained only 2 characters. the biarticulate palp of maxilla 1 and the elongate wrists of the gnathopods. Five figures, i.e., gnathopods 1 and 2, epimeron 3, telson and a sketch of an- tenna 2 plus a medium-sized description established the type- species. We now recognize that the presence of dorsal spines on the telson is the best clue to suggest that Protophoxus bears no relationship to Paraphoxus. Hurley (1954) redescribed “Pontharpinia” australis from a specimen collected a few hun- dred miles from the type-locality and which differed, as far as Hurley could distinguish from the meager first description, only in the reduced number of dorsal telsonic spines. We have 522 Proceedings of the Biological Society of Washington now reexamined some of the syntypes of P. australis and find Hurley’s excellent portrayal of the species to be exact and the distinction he noted to be minor but perhaps indicative of infraspeciation, perhaps a reflecton of clinal phenotypy. Proto- phoxus bears a close resemblance to Parharpinia, especially in the presence of dorsal spination or setation on the telson. But several good generic distinctions separate the 2 taxa as stated in the diagnoses to follow. Each genus has a special spine on uropod | but on opposite sides from each other. Trichophoxus was established by K. H. Barnard, 1930, for T. capillatus K. H. Barnard, 1930, again in comparison to Phoxocephalus and with a short diagnosis confined to “man- dible with molar conical, lacking the triturating surface, and tipped with 4 strong spines; and 5th joint of both gnathopods elongate.” Illustrations of head, gnathopod 1, antenna 1, telson, epimeron 3 and pereopod 7 accompanied a medium- sized description. We now know that various characters on pereopod 7, gnathopod 1 and the telson have generic value. The written description of the mandibular molar weakly fits the general situation in Paraphoxus and in the supposed but confounded descriptions cited for Parharpinia and Ponthar- pinia. We have examined and established a lectotype for T. capillatus and have determined that this molar is actually distinct in high degree from the molar attributed to various species formerly placed in Paraphoxus, Pontharpinia and Par- harpinia. The molar is large, subconical, but truncate, has a definite semiarticulate base and bears 4 very large spines only partly articulate and consistently fixed in a pattern of size and shape (see illustrations). We confirm the consistency of this molar in 5 species of a related new genus from Aus- tralia and in Paraphoxus rakiura Cooper and Fincham, 1974, a species from New Zealand which should be transferred to a new and distinctive genus based on patterns of uropodal spination and setation. Metharpinia Schellenberg, 1931, was described for 2 spe- cies, M. longirostris Schellenberg, 1931, and M. cornuta Schel- lenberg, 1931, neither of which, as yet, has been designated as type-species of the genus. Although J. L. Barnard (1960) reexamined these species, few new observations were pub- Clarification of five amphipod genera 523 lished. The lateral setation of the telson suggests generic dif- ferentiation from those genera diagnosed herein but numerous other minute details must be determined before a type-species is selected and the genus revived. During the years between 1906 and 1957 three species of phoxocephalid were described in Paraphoxus, 9 in Pontharpinia (with 3 additional synonyms) and 7 species in Parharpinia. Apparently these species were assigned to genera mainly on the basis of Stebbing’s (1906) key to the genera of Phoxo- cephalidae, in which Paraphoxus was erroneously attributed a uniarticulate palp on maxilla 1, Pontharpinia was erroneously attributed a nontriturative mandibular molar and Pontharpinia and Parharpinia were distinguished on the stoutness of pereo- pods 5-6. Apparently the species assigned to Paraphoxus were placed there mainly on the overall facies rather than on the condition of maxilla 1. J. L. Barnard (1960) in assembling a pool of 40 species, synonymized all of these genera on the thesis that stoutness of pereopods 5-6 was variable in the pool of species being treated, that all were known or suspected to have a biarticu- late palp on maxilla 1, that the elongation of article 5 on gnathopods was variable and that widths of rostra were vari- able. All were presumed to have a simple mandibular molar bearing a few semiarticulate, tightly clumped spines, a fact we now know to be erroneous in Trichophoxus and Ponthar- pinia; and we now know that the kind of spine clumping on the molar of Paraphoxus is a good generic character. Our study of Australian phoxocephalids demonstrates a secondary importance of rostral dimensions as a generic char- acter and an imperfect association between pereopodal widths and generic or subfamilial categorizations, but many of the following characters should be brought into the pool of im- portant characters: fine structure of molars, proportions of and setal distribution on article 2 of antenna | and on article 5 of antenna 2, facial spine and setal distributions on articles 3-4 of antenna 2, shapes and dominance of epistome and upper lip but not necessarily the degree of cuspidation, size of palpar hump on mandible, presence or absence of cones on the lower lip, setation counts on the inner plate of maxilla 1, 524 Proceedings of the Biological Society of Washington degree of cleft between inner lobes of the maxillipeds, linear or nonlinear arrangement of facial setae on article 3 of the maxillipedal palp, length and shape of the dactyl on the max- ilipeds and the condition of the apical nail and attendant ac- cessory setae, shapes and setosity of gnathopods in terms of shortened and cryptic article 5, anterior setal density on hands, degree of palmar slope, enlargement of gnathopod 2 and eusirid-like attachments of articles, presence or absence of posteroproximal setae on article 5 of pereopods 3-4, presence and size of midapical subdactylar spine on article 6 of pereo- pods 3-4, apical ornamentation of dactyls on pereopods 3-4, size and shape of pereopod 7 articles 2-3, presence or absence of dactyl on pereopod 7, presence or absence of combs in various places on pereopods and uropods including medial surfaces and rami, presence or absence of long flexible setae in various places on epimera and uropods, numbers and shape of apical setae on article 2 of outer ramus on uropod 3, pres- ence or absence or degree of development of thick spines on uropods and urosomal surfaces, lengths of rami of uropods 1-2 with special attention to inner ramus of uropod 2, degree of fusion of inner ramus of uropod 2 to peduncle, size and presence of glandular tissue in urosome, fusion, setation and presence of spines on urosomites, presence and number of apical, dorsal and lateral spines on telson, and, finally, gill formulas. Paraphoxus Sars Paraphoxus Sars, 1895: 148-149. Diagnosis of male and female: Eyes present; flagella of antennae 1-2 unreduced, article 1 of antenna 1 bearing medial fuzz in male, article 2 short, setae confined distally, primary flagellum with calceoli in male; antenna 2 lacking ensiform process, article 3 with only 2 facial setules, apicalmost elongate, article 5 somewhat shortened, bearing dorsal cal- ceoli in male, flagellum with calceoli in male; upper lip and epistome distinct, epistome dominant; right mandibular incisor with 3 teeth, molar not triturative, small, forming weakly conical boss, bearing clump of 3 articulate spines on basal plate, no fuzz, right lacinia mobilis bifid, rami subequal in length, lacking facial denticles, palper hump of mandible small; lower lip bearing cones; palp of maxilla 1 biarticulate, inner plate with 2 setae; inner plates of maxilliped ordinary, poorly setose, palp article 3 scarcely protuberant, dactyl elongate, apical nail elongate, not Clarification of five amphipod genera 525 immersed; gnathopods small but hands weakly enlarged, almost identical in shape, article 5 shortened, hands ovatorectangular, poorly setose an- teriorly, palms oblique; article 5 of pereopods 3-4 bearing setae pos- teroproximally but not fully to base, article 6 bearing normal 2 rows of spines, middistal armament absent, dactyls lacking inner tooth, bearing subapical slit, apical nail distinct but partly immersed; article 2 of pereopod 5 of broad form, untapering, of pereopod 7 of ordinary phoxo- cephalid kind, article 3 of pereopod 7 small (ordinary), articles 4-5 of pereopods 5-6 narrow, pereopod 7 of normal size, article 2 ventrally naked (lacking long setae); peduncular apices of uropods 1-2 minutely combed, peduncle of uropod 1 normally elongate, lacking ventral spike or special spines, medial spines widely spread, dorsal spines confined apically (not shown in Sars, 1895), inner ramus of uropod 1 bearing only 1 row of spines, peduncle of uropod 2 with 1 short medial spine set apically plus 1 midmedial seta, inner ramus spinose, as long as outer ramus, uropod 3 especially elongate, article 2 of outer ramus with only 2 apical setae; telson naked laterally and dorsally except for normal pair of setules set about mark 50 (not shown in Sars, 1895); urosomite 1 devoid of major lateral spines and ventral setae; epimera 1-2 lacking posterior setae, with facial setae located ventrally (not shown in Sars, 1895), none in midface, epimeron 3 lacking posterior setae, of special “rounded-classification” form, lacking midfacial setae or spines. Description: Rostrum large and broad; facial spines on article 4 of antenna 2 primarily in one row, but second row of one spine and setule present, article 5 thin; article 1 of mandibular palp weakly elongate, apex of article 3 truncate but scarcely oblique, palp thin, distal branch of right lacinia mobilis broad, serrate minutely, proximal branch like raker spine; outer plate of maxilla 1 with 11 spines plus inner articulate apical plumose seta (not shown in Sars, 1895), one spine especially thickened; setation of maxilla 2 ordinary; inner and outer plates of max- illiped ordinary but inner plate poorly setose, lacking thick apical spine(s) [this character not confirmed]; coxae 2—4 lacking anterodorsal humps; article 5 of gnathopod 2 free (not cryptic); article 2 of pereopod 7 with small posterior teeth, lacking facial setae; gills present on pereo- nites 2-7. ! Type-species: Phoxus oculatus Sars, 1879 (monotypy). Composition: The type-species and possibly Parharpinia simplex Gur- janova, 1938. All other species ever assigned to Paraphoxus are to be removed to other genera, mostly new. See “Allocation in Paraphoxus” to follow. Relationship: Our study of Australian phoxocephalids has revealed a group of 5 new genera containing 9 species in which Paraphoxus fits as a highly specialized and advanced taxon from cool waters of the northern hemisphere. This group of genera is distinguished by the fol- lowing combination of characters: (1) the sole presence on the man- dibular molar of a group of 3 tightly packed and partly fused (basal) spines (inadequately shown in Sars, 1895); (2) article 2 of antenna 1 526 Proceedings of the Biological Society of Washington short; (3) all thick facial spines on article 4 of antenna 2 peduncle aligned in a single row; (4) setae on inner plate of maxilla 1 reduced in number from the usual 4 found in primitive Australian genera; (5) article 2 of pereopods 6-7 with only one main facial ridge; (6) inner plate of maxilliped lacking stout apical spines; (7) epimeron 3 of “rounded-classification”, lacking any elongate setae, protruding pos- teriorly, and, at most, bearing 3 setule notches. The following char- acters are often useful in recognizing members of this generic group but the characters are not universally present: (1) apical combs on the peduncles of uropods 1-2; (2) articles 4-5 of pereopods 5-6 very thin (with one weak exception); (3) article 5 of antenna 2 shortened in females, this observation based on primitive phoxocephalids bearing a more elongate article 5; (4) article 6 of pereopods 3-4 with some com- ponent spines elongate and thin, almost classifiable as flexible setae (with 2 generic exceptions). These genera also share numerous other sim- ilarities most of which in the broader context are useful in genera out- side this system at specific level. One example is the absence of accessory apical nails on the rami of uropods 1-2; in certain new Australian genera the presence or absence of these accessory nails is useful taxonomically only at specific level; another example is the presence of a special cusp or articulate setule medially on the outer plate of maxilla 1. The Australian genera of this complex can be arranged in an evolu- tionary sequence to demonstrate that in the primitive state taxa are characterized by an enlarged gnathopod 2 with cryptic article 5, whereas in the advanced or specialized state gnathopod 2 is reduced to a size similar to gnathopod 1, with the posterior margin of article 5 free from envelopment. Paraphoxus represents the advanced state of this condition. The characters of the generic diagnosis have been confirmed from material of Paraphoxus oculatus (Sars) deposited in the U.S. National Museum of Natural History and identified by comparison to pl. 51 of Sars, 1895. ALLOCATION OF SPECIES IN PARAPHOXUS Those species formerly assigned to Paraphoxus not now confined to the genera herein restricted must be removed from Paraphoxus and al- located to newly created genera. This process is difficult because many of the species are poorly described and cannot be so allocated until they have been reexamined minutely. We detect a pattern of classifi- cation in the species adequately described as outlined below. New genera for these taxa will be described elsewhere. I. Peduncle of uropod 1 with special medial spine. A. Pereopods 5-6 with stout articles 4-5; wrists of gnathopod 2 slightly elongate. Pontharpinia epistoma Shoemaker, 1938: 326; J. L. Barnard, 1960: 205 (as Paraphoxus); Bousfield, 1973: 126 (as Tricho- phoxus ) Clarification of five amphipod genera 527 Pontharpinia floridana Shoemaker, 1933: 5; J. L. Barnard, 1960: 226 (as Paraphoxus ) Paraphoxus gemmatus J. L. Barnard, 1969: 222 Paraphoxus jonesi J. L. Barnard, 1963: 463 Paraphoxus lucubrans J. L. Barnard, 1960: 212 B. Pereopods 5-6 with thin articles 4—5. 1. Wrist of gnathopod 2 shortened; setae on article 2 of an- tenna 1 widely spread. Paraphoxus cognatus J. L. Bamard, 1960: 233 PParharpinia fuegiensis Schellenberg, 1931: 78; J. L. Bar- nard, 1960: 271 (as Paraphoxus); pereopod 3 not thin (to a new genus to be described ) Paraphoxus similis J. L. Barnard, 1960: 230 Paraphoxus spinosus Holmes, 1905: 477; J. L. Barnard, 1960: 243 2. Wrist of gnathopod 2 not shortened; setae on article 2 of antenna | set apically (and telson with lateral setae). Par- harpinia sinuata K. H. Barnard, 1932: 103 II. Peduncle of uropod 1 lacking special spine. A. Articles 4—5 of pereopods 5-6 stout, facial spines on article 4 of antenna 2 in 2 or more rows. 1. Epimera 1-2 with posterior setae, telson with supernumerary dorsal setation. Pontharpinia milleri Thorsteinson, 1941: 82; J. L. Barnard, 1960: 266 (as Paraphoxus); Pontharpinia robusta Gur- janova, 1938: 262 Pontharpinia obtusidens Alderman, 1936: 54; J. L. Barnard, 1960: 249 (as Paraphoxus ); Pararpinia [sic] pontarpioides Gurjanova, 1953: 229 2. Epimera 1-2 lacking long posterior setae, telson lacking supernumerary dorsal setation. a. Gnathopods with mitelloid hands, thin and/or highly setose anteriorly. Paraphoxus abronius J. L. Barnard, 1960: 203 Paraphoxus bicuspidatus J. LL. Barnard, 1960: 218 Paraphoxus fatigans J. LL. Barnard, 1960: 209 Pontharpinia tridentata J. L. Barnard, 1954: 4; 1960: 261 (as Paraphoxus ) Paraphoxus variatus J. L. Barnard, 1960: 198 Paraphoxus vigitegus J. L. Barnard 1971: 70 (but uro- some with spike ) b. Gnathopods ordinary, hands ovatorectangular, poorly setose anteriorly. Paraphoxus daboius J. L. Barnard, 1960: 210 Paraphoxus heterocuspidatus J. L.. Barnard, 1960: 224 Pontharpinia rostrata Dana of Pirlot, 1932: 62 Paraphoxus stenodes J. .. Barnard, 1960: 211 528 Proceedings of the Biological Society of Washington B. Articles 4-5 of pereopods 5-6 thin, facial spines on article 4 of antenna 2 mostly confined to one row. 1. Wrists of gnathopods elongate, ventral setae on article 2 of antenna 1 widely spread. Parharpinia calcarata Gurjanova, 1938: 272; J. L. Barnard, 1960: 238 (as Paraphoxus ) Paraphoxus robustus Holmes, 1908: 618; J. L. Barnard, 1960: 235 Paraphoxus subuncigerus Kudrjaschov, 1965: 1776 2. Wrists of gnathopods short, ventral setae on article 2 of antenna 1 confined apically. Pontharpinia uncinata Chevreux, 1912: 100 List oF SpEcIES REJECTED FROM PARAPHOXUS Protophoxus australis K. H. Barnard, 1930: 335 (to Protophoxus) Phoxus batei Haswell, 1879: 259 (to Birubius) Trichophoxus capillatus K. H. Barnard, 1930: 336 (to Trichophoxus) Metaharpinia cornuta Schellenberg, 1931: 68 (to Metharpinia) Metharpinia longirostris Schellenberg, 1931: 65 (to Metharpinia) Urothoe pinguis Haswell, 1879a: 325 (to Pontharpinia) Paraphoxus rakiura Cooper and Fincham, 1974 (should be established in a new genus) Paraphoxus stebbingi J. L. Barnard, 1958: 148 (to a new genus to be described ) Paraphoxus tattersalli J. L. Barnard, 1958: 148; 1960: 282 (to a new genus to be described ) Phoxus villosus Haswell, 1879: 258 (to Parharpinia) Paraphoxus waipiro J. L. Barnard, 1972: 143 (to a new genus to be de- scribed ) Paraphoxus sp. Australia, J. L. Barnard, 1960: 285 (to a new genus to be described ) List oF UNCLEAR SPECIES OF “PARAPHOXUS” Pontharpinia barnardi Pirlot, 1932: 62; for P. rostrata of K. H. Barnard, 1931:119 [not Dana, 1853]; pereopod 5 and epimeron 3 unusual Pontharpinia centralis Schellenberg, 1938: 15; pereopods 5-6 thin Pontharpinia maxima Stephensen, 1947: 42; possibly close to P. pyripes; uropod 1] with ventral peduncular setae Pontarpinia [sic] nasuta Gurjanova, 1936: 249; 1951: 382 Parharpinia obliqua K. H. Barmard, 1932: 101; probably similar to new genus of paraphoxin from Australia to include P. tattersalli, new genus, to be established by Barnard and Drummond (in prep.) Pararpinia [sic] ochotica Gurjanova, 1953: 225 Paraphoxus pyripes K. H. Barnard, 1930: 332; perhaps similar to P. maxima but gnathopod 2 probably enlarged Clarification of five amphipod genera 529 Parharpinia rotundifrons K. H. Barnard, 1932: 104; PStephensen, 1947: 44; gnathopods probably like basic Australian genus as cited above under P. obliqua Parharpinia uncigera Gurjanova, 1938: 267; 1951: 388; pereopods 5-6 thin, article 4 of antenna 2 stunted, possibly article 6 of pereopods 3-4 lacking posterior spines. Pontharpinia Stebbing Pontharpinia Stebbing, 1897: 32-33. Diagnosis of female [adult male unkown]: Eyes present; flagella of antennae 1-2 unreduced, fuzz on article 1 of antenna 1 unknown in male, article 2 of antenna 1 short, setae widely spread on ventral sur- face, calceoli on primary flagellum of male antennae unknown; antenna 2 with weak ensiform process, article 3 with groups of 5 + 1 facial setae; article 5 shortened; upper lip and epistome not strongly distinct from each other; right mandibular incisor with 3 teeth, molar fully triturative, small, cylindrical, right lacinia mobilis bifid, rami equal, bearing weak facial denticles, palpar hump of mandible of medium size; lower lip of mandible lacking cones; palp of maxilla 1 biarticulate, inner plate with 5 or more setae; inner plates of maxilliped ordinary, palp article 3 weakly protuberant, dactyl elongate, apical nail small, mostly immersed; gnatho- pod 1 small, gnathopod 2 enlarged, of diverse shapes, article 5 of gnathopod 2 greatly shortened, hands expanded, rectangular, poorly setose anteriorly, palms oblique; article 5 of pereopods 3—4 lacking posteroproximal setae, article 6 bearing normal 2 rows of spines, mid- distal armament composed of medium to small spine, dactyls bearing 1 inner tooth; article 2 of pereopod 5 of broad form and expanding dis- tally, of pereopod 7 of ordinary phoxocephalid kind, article 3 of pereo- pod 7 small (ordinary), articles 4-5 of pereopods 5-6 broad, pereopod 7 of normal large size, article 2 ventrally setose; peduncular apices of uropods 1-2 not combed, peduncle of uropod 1 normally elongate, lack- ing ventral spike but bearing special large apicomedial spine, other me- dial armaments widely spread, dorsal spines widely spread, inner ramus of uropod 1 bearing only one row of spines, peduncle of uropod 2 with medial spines widely spread, inner ramus spinose, as long as outer ramus, uropod 3 ordinary, article 2 of outer ramus with 3 apical setae and sub- apical lateral spine-seta; telson with only apical spines but each lobe with huge crescentic brush of setae midlaterally, other setules abnormal, one each confined dorsobasally on each lobe; urosomite 1 with ventral crescentic brush of setae; epimera 1-2 naked posteriorly, with all facial setae ventrally placed, epimeron 3 with midfacial setae. Description: Rostrum large and ordinary; facial spines on article 4 of antenna 2 in 2 or more rows; article 5 thin; article 1 of mandibular palp short, palp of medium thickness but huge in comparison to man- dibular body; outer plate of maxilla 1 with 11 spines, one spine especially thickened; setation of maxilla 2 ordinary; coxae 2-4 lacking antero- 530 Proceedings of the Biological Society of Washington dorsal humps; article 5 of gnathopod 2 cryptic (posterior margin con- cealed by articles 4 and 6); article 2 of pereopod 7 with small posterior teeth, bearing numerous groups of facial setae; gills present on pereo- nites 2-7. Type-species: Urothoe pinguis Haswell, 1879a (monotypy). Relationship: The genus and its unique species may be the most primitive living phoxocephalid in the present context of this family (as- suming full rostral development as a prerequisite or barring other dis- coveries). Like Phoxocephalus and Leptophoxoides it bears a fully triturative mandibular molar, a character assumed to be primitive, short article 2 of antenna 1, and an enlarged gnathopod 2, but unlike the more advanced Phoxocephalus and Leptophoxoides it has a higher number of setae on the inner plate of maxilla 1, continuously spinose rami on uropods 1-2 and facial setae on article 2 of pereopod 7, all characters which might be marks of relationship to gammaroid-haustorioid an- cestors. Nevertheless Pontharpinia obviously is a very advanced taxon of amphipod in that it has specializations that must remove it from the direct ancestral pool of other phoxocephalids. Its primitiveness is simply a hypothetical assessment; it has never lost certain marks of its ancestors which we consider are weak character linkages to an ancestral stream of species. Pontharpinia pinguis retains many of the characters lost in various degrees in other genera of phoxocephalids, such as: special spine on peduncle of uropod 1, enlarged gnathopod 2, short and cryptic article 5 of gnathopod 2, dispersed facial setae on article 3 of palp on maxilliped, full rostrum, facial setae on pereopod 7, triturative molar, ensiform an- tenna 2, fully setose maxilla 2, widely setose inner plate of maxilla 1, inner tooth on dactyls of pereopods, broad articles on pereopods 5-6, posterior setae on article 2 of pereopods 5-6, posterior setae on coxa 4, anterior setae on coxa 1, setal brushes on telson, fully flabellate right lacinia mobilis, setose dactyl of pereopod 7, ventrally spread setae on article 2 of antenna 1, continuously spinose rami of uropods 1-2, widely spread ventral setation on epimera 1-2, mainfold setae on article 2 of outer ramus on uropod 3, facial setae on epimeron 3, ventral setae on urosomite 1, subapical setae on peduncle of uropod 3, and lateral setosity on palp article 2 of mandible. Pontharpinia pinguis, however, has the following characters about which we are unable to confirm the degree of primitiveness; absence of basofacial setae on uropod 1, absence of double spine rows on the inner ramus of uropod 1, absence of lateral spination on the urosome, *short and immersed apical nail on palp article 4 of the maxilliped, unequal flagella on antenna 1, absence of thick apical spines on the inner plates of the maxillipeds, presence of only one (not 2) dorsolateral setule on each lobe of the telson, absence of anterodorsal humps on coxae 1-4, absence of facial setae on coxae 6-7, absence of posteroproximal setae on article 5 of pereopods 3—4, *absence of cones on the lower lip, almost fully amalgamated upper lip and epistome and multiple sets of spines Clarification of five amphipod genera 531 on article 4 of antenna 2. Except for the two characters marked by asterisks we believe all others are advancements in phoxocephalid evolution. Parharpinia Stebbing Parharpinia Stebbing, 1899: 207; 1906: 147. Diagnosis of male and female: Eyes present; flagella of antennae 1-2 unreduced, article 1 of antenna 1 bearing medial fuzz in male, article 2 short, setae widely spread on ventral surface, primary flagellum lacking calceoli in male (as far as known); antenna 2 lacking ensiform process, article 3 with only 2 facial setules, article 5 not shortened, lack- ing calceoli in male (as far as known), flagellum also lacking calceoli; upper lip and epistome not strongly distinct from each other; right man- dibular incisor with 3 teeth, molar not triturative, small, forming pillow- shaped boss, bearing small articulate spines, fuzz weak to absent, right lacinia mobilis bifid, rami unequal, bearing facial denticles, palpar hump of mandible of medium size; lower lip bearing cones; palp of maxilla 1 biarticulate, inner plate with 4 setae; inner plates of maxilliped ordinary, palp article 3 weakly protuberant, dactyl elongate, apical nail of medium size, mostly immersed; gnathopods small, almost identical in shape but article 5 of gnathopod 2 slightly shortened, hands ovate-rectangular, poorly setose anteriorly, palms oblique; article 5 of pereopods 3—4 bear- ing posteroproximal setae, article 6 bearing normal 2 rows of spines, middistal armament composed of small to medium spine or thick short seta, dactyls bearing one inner tooth; article 2 of pereopod 5 of broad form but tapering distally, of pereopod 7 of ordinary phoxocephalid kind, article 3 of pereopod 7 small (ordinary), article 4-5 of pereopods 5-6 narrow, pereopod 7 of normal large size, article 2 ventrally setose; peduncular apices of uropods 1-2 not combed, peduncle of uropod 1 normally elongate, lacking ventral spike but bearing special large apico- medial spine, other medial spines widely spread, dorsal spines mainly apical, inner ramus of uropod 1 bearing only 1 row of spines, peduncle of uropod 2 with only 1 medial spine (spinule) set apically, inner ramus naked, as long as outer ramus, uropod 3 ordinary, article 2 of outer ramus with 2-4 setae (variable infraspecifically); telson variable in spination, with either or both lateral and dorsal spines, midlateral setules normal, placed at about mark 33, comprising a pair on each side; uro- somite 1 devoid of major lateral spines and ventral setae except for distal setae at base of uropod 1; epimera 1-3 with posterior setae, epimera 1-2 with facial setae all ventral, epimeron 3 with few midfacial setae or spines. Description: Rostrum large and ordinary; facial spines on article 4 of antenna 2 primarily in one row scarcely divided into sections, article 5 thick; article 1 of mandibular palp short, palp thin; outer plate of maxilla 1 with 11 spines, one spine especially thickened; setation of maxilla 2 ordinary; inner and outer plates of maxilliped ordinary; coxae 532 Proceedings of the Biological Society of Washington 2-4 with very weak anterodorsal humps; article 5 of gnathopods free (not cryptic); article 2 of pereopod 7 with small posterior teeth, lack- ing facial setae; gills present on pereonites 2-7. Type-species: Phoxus villosus Haswell, 1879 (original designation). Remarks on type-species: Stebbing (1899) designated Phoxus vil- losus as type of his Parharpinia but confounded the generic establish- ment because he did not have Phoxus villosus in hand. He probably had Protophoxus australis K. H. Barnard in hand as evidenced by his statement that he had a specimen of Phoxus batei Haswell as identified by G. M. Thomson, an entity confirmed by Hurley (1954) to be Proto- phoxus australis. The diagnosis was not affected by this confusion be- cause the diagnosis was so brief that it did not touch upon differences between Protophoxus and Parharpinia. Composition: The type-species and one new species to be described by Barnard and Drummond (in prep.) from Australia. Relationship: Parharpinia differs from Paraphoxus Sars in the pres- ence of dorsal-lateral spines on the telson; in Paraphoxus any telsonic spines are always fully apical. Parharpinia also bears a special, en- larged medial spine on the peduncle of uropod 1 and has posteriorly setose epimera 1-2. The apical nail on palp article 4 of the maxilliped is more strongly immersed than in species of Birubius. Protophoxus K. H. Barmard resembles Paraphoxus in the absence of posterior setae on epimera 1-2 but the special spine on the peduncle of uropod 1 is fully lateral instead of being fully medial as in Parharpinia. Both Proto- phoxus and Parharpinia have article 3 of pereopod 5 tapering distally. The telsonic spination in Protophoxus is more consistently dorsal than in Parharpinia, the latter having a highly variable placement of dorsal and lateral elements often so flexible as to be termed setae; this vari- ability is infraspecific. Protophoxus resembles Paraphoxus in the ab- sence of long ventral setae on article 2 of pereopod 7 but Parharpinia has those setae fully developed. Parharpinia differs from Birwbius in the dorsal telsonic setospination, in the presence of the special spine on uropod 1, in the presence of pos- terior setation on epimera 1-2, and in the taper of article 2 on pereopod 5. Protophoxus K. H. Barnard Protophoxus K. H. Barnard, 1930: 335. Diagnosis of male: Eyes present; flagella of antennae 1-2 unreduced (in male), article 1 of antenna 1 bearing medial fuzz, article 2 short, setae widely spread on ventral surface, primary flagellum with calceoli; antenna 2 lacking ensiform process, article 3 with only 2 facial setules, article 5 not shortened, bearing dorsal calceoli, flagellum with calceoli; upper lip and epistome not strongly distinct from each other; right man- dibular incisor with 3 teeth, molar not triturative, small, forming pillow- shaped boss, bearing small articulate spines and fuzz, right lacinia mobilis Clarification of five amphipod genera 533 bifid, branches unequal, bearing facial denticles, palpar hump of man- dible of medium size; lower lip bearing cones; palp of maxilla 1 bi- articulate, inner plate with 4 setae; inner plates of maxilliped ordinary, palp article 3 unprotuberant, dactyl elongate, apical nail of medium size, distinct, not deeply immersed; gnathopods small, almost identical in shape but article 5 of gnathopod 2 slightly shortened, hand ordinary ovatorectangular, poorly setose anteriorly, palms oblique; article 5 of pereopods 3-4 bearing posteroproximal setae, article 6 bearing normal 2 rows of spines, middistal armament composed of medium spine, dactyls bearing one inner tooth; article 2 of pereopod 5 of broad form but taper- ing distally, of pereopod 7 of ordinary phoxocephalid kind, article 3 of pereopod 7 small (ordinary), articles 4-5 of pereopods 5-6 narrow, pereopod 7 of normal large size, article 2 ventrally naked (lacking long setae); peduncular apices of uropods 1-2 naked and weakly combed respectively, peduncle of uropod 1 normally elongate, lacking ventral spike but bearing special large apicolateral spine, medial spines widely spread, dorsal spines widely spread, inner ramus of uropod 1 bearing only 1 row of spines, peduncle of uropod 2 with only 1 medial spine set apically, inner ramus spinose, as long as outer ramus, uropod 3 especially elongate, article 2 of outer ramus with only 2 apical setae; telson naked laterally, bearing dorsal spines, midlateral setules normal, placed at about mark 33, comprised of a pair on each side; urosomite 1 devoid of major lateral spines and ventral setae; epimera 1-2 lacking posterior setae, with facial setae located ventrally, none in midface, epimeron 3 lacking pos- terior setae, with midfacial setae or spines. Description: Rostrum large and ordinary; facial spines on article 4 of antenna 2 primarily in 1 row scarcely divided into sections, article 5 thick; article 1 of mandibular palp short, apex of article 3 obliquely truncate, palp thin; outer plate of maxilla 1 with 11 spines, 1 spine es- pecially thickened; setation of maxilla 2 ordinary; inner and outer plates of maxilliped ordinary; coxae 2—4 lacking anterodorsal humps; article 5 of gnathopods free (not cryptic); article 2 of pereopod 7 with small posterior teeth, lacking facial setae; gills present on pereonites 2—7. Type-species: Protophoxus australis K. H. Barnard, 1930 (monotypy). Relationship: This genus is now known to resemble Parharpinia Stebbing and to be distinguished from Paraphoxus Sars and Birubius in the following characters: (1) presence of dorsal spination on the telson; (2) presence of a special spine on the peduncle of uropod 1 and (3) strong taper on article 2 of pereopod 5. Protophoxus, however, differs from Parharpinia in: (1) the shift of the special spine on uropod 1 onto the full lateral side of the peduncle, whereas in Parharpinia that spine is fully medial; (2) the absence of posterior setation on epimera 1-3; (3) the absence of strong ventral setation on article 2 of pereopod 7. Protophoxus and Parharpinia also have thin articles 4—5 on pereopods 5-6 but this character is almost evenly intergraded with some of the thin-membered species of Birubius. 534 Proceedings of the Biological Society of Washington Protophoxus australis K. H. Barnard Figure 4 (part) Phoxus batei—Thomson, 1882: 232-233, pl. 17, fig. 2 (not Haswell). Protophoxus australis K. H. Barnard, 1930: 335-336, fig. 12. Pontharpinia australis—Hurley, 1954: 581-587, figs. 1-28. Material: Terra Nova 135, British Museum (Natural History) No. 1930.8.1:133-137, 8 specimens, 7 in alcohol, 1 mounted on 3 slides; 2 of these are juveniles or very young males as stated by K. H. Barnard, the other 6 are males, none is a female as stated by K. H. Barnard; 1 of the specimens in alcohol resembles female in some respects but de- tached head is male-like and carcass bears penial processes. Remarks: The depiction of this species by Hurley (1954) is excellent; all of the details of the generic diagnosis and description given above have been checked and confirmed on the material in hand. Hurley pointed out that his specimen from Cook Strait had only 1 dorsal spine on each lobe of the telson whereas K. H. Barnard showed 5. K. H. Barnard, however, noted specimens from Terra Nova 106 that lacked these spines. We have not seen these but would suspect their identity on that basis. The smallest specimens, 3.5—-4.0 mm, in the sample on hand, have 2 spines on each lobe; the largest whole specimen, about 6.5 mm, has 3 on each lobe; K. H. Barnard shows the apical telsonic spination to be composed of 2 short spines separated by a setule; the specimen in hand mounted on slides bears 3 spines on each lobe, 2 outer spines separated by the setule from the inner third spine. We estimate such variables to be normal from our study of Australian phoxo- cephalids. Hurley's specimen may, however, represent a Cook Strait race of the species. We append figures of the right lacinia mobilis and uropod 1 from the slide-mounted specimen (size unknown, carcass missing). Inner plate of maxilla 1 of medium size, bearing long middle subapical pluseta, 1 shorter medioapical pluseta and 2 even smaller, almost naked, weakly lateral setae. Trichophoxus K. H. Barnard Trichophoxus K. H. Barnard, 1930: 336. Diagnosis of male: Eyes present; flagella of antennae 1-2 unreduced (in male), article 1 of peduncle on antenna 1 lacking medial male fuzz, article 2 elongate, setae spread widely on ventral surface, primary fla- gellum with calceoli; antenna 2 with weak ensiform process, article 3 with numerous facial setae, upwards of 15, article 5 shortened, bearing dorsal calceoli, flagellum with calceoli; upper lip and epistome strongly distinct; right mandibular incisor with 4 teeth, molar not triturative, long, massive, conical, bearing 4 enlarged but strongly fixed spines, right lacinia mobilis bifid, rami subequal, bearing facial denticles, palpar hump of mandible small; lower lip lacking cones; palp of maxilla 1 biarticulate, Clarification of five amphipod genera 535 inner plate with 4 setae; inner plates of maxilliped ordinary, palp article 3 unprotuberant, dactyl elongate, apical nail short and fully immersed or fused; gnathopods small, gnathopod 2 slightly smaller than gnathopod 1, almost identical in shape, wrists elongate, hands thin, mitelloid, heavily setose anteriorly, palms transverse; article 5 of pereopods 3—4 lacking posteroproximal setae, article 6 bearing normal 2 rows of spines, mid- distal armament comprising huge spine, dactyls bearing 1 inner tooth; article 2 of pereopod 5 of broad form, nontapering, of pereopod 7 ex- tended-nasiform, article 3 of pereopod 7 greatly enlarged, articles 4-5 of pereopods 5-6 very broad, pereopod 7 miniaturized; article 2 ventrally setose; peduncular apices of uropods 1-2 weakly combed, peduncle of uropod 1 normally elongate, bearing weak ventral spike, medial spine(s) confined to apical end, dorsal spines fully spread, inner ramus of uropod 1 bearing only 1 row of spines, peduncle of uropod 2 with more than 1 medial spine, inner ramus naked but as long as outer ramus, uropod 3 especially elongate and article 2 of outer ramus probably with 3 apical setae; telson with lateral spines, lateral setules reduced to 1 on each side and shifted highly distad; urosomite 1 with 2 rows of lateral spines, 1 ventral crescent of setae; epimera 1-2 with both posterior and midfacial setae. Description: Rostrum very small; facial spines on article 4 of antenna 2 in 2 or more main rows, article 5 thin; article 1 of mandibular palp short, apex of article 3 obliquely truncate, palp of medium thickness; outer plate of maxilla 1 with 11 spines, 1 spine especially thickened; setation of maxilla 2 ordinary; inner and outer plates of maxilliped or- dinary; coxae 2 and 4 with anterodorsal humps; article 5 of gnathopods free (not cryptic); article 2 of pereopod 7 with small posterior teeth, bearing facial setae; gills present only on pereonites 2-6. Type-species: Trichophoxus capillatus K. H. Barnard, 1930 (mono- typy). Relationship: This genus is now known to be distinguished by nu- merous characters separating it from Paraphoxus Sars to which it was relegated by J. L. Barnard (1960). We have compared it with 38 spe- cies of Birubius, the basic genus from Australia, and 20 other species of various genera from the northern Pacific Ocean. Trichophoxus has its closest relationships to a new genus (genus of 5 species, Barnard and Drummond, in prep.) from Australia. These 2 genera are characterized by the following attributes: (1) the semitriturative molar, enlarged, bearing 4 enlarged poorly articulate spines similar to teeth; (2) the miniaturized pereopod 7 with enlarged article 3; (3) the weakly ensiform process on antenna 2; (4) manifold setation of article 3 on antenna 2; (5) naked posteroproximal margin on article 5 of pereopods 3-4; (6) absence of medial fuzz in males on article 1 of antenna 1; (7) strong distinction between upper lip and epistome; (8) absence of cones on lower lip; (9) poorly developed apical nail on palp article 4 of maxil- liped; (10) anterodorsal humps on coxae 2—4; (11) ventral spike on peduncle of uropod 1; (12) presence of 3 (not 2) apical setae on article 536 Proceedings of the Biological Society of Washington 2 of outer ramus on uropod 3; and (13) patchwork (in contrast to single row) of denticles on each lobe of male telson. Trichophoxus is distinguished from the new Australian genus compris- ing 5 species in numerous characters to be discussed by Barnard and Drummond (in prep.), but a few are given here so as to justify the nec- essarily long generic diagnosis: (1) lateral spination on urosome; (2) lateral spination on telson; (3) far more numerous setae on article 3 of antenna 2; (4) presence of 4 setae (not 3) on inner plate of maxilla 1; (5) immensity of apical midspine on article 6 of pereopods 3-4; (6) elongation and bareness of inner ramus on uropod 2; (7) normally elongate peduncle of uropod 1; (8) proximal spination on peduncle of uropod 1; (9) presence of only 1 row of spines on inner ramus of uro- pod 1; (10) presence of only 1 ventral crescent of setae on urosomite 1; (11) strong distal shift of and reduction in number of dorsal setules on telson; (12) absence of any combs on pereopods 5-7; (13) presence of facial humps on right lacinia mobilis; (14) distad shift in primary tooth on dactyls of pereopods 3-4; (15) presence of more than 1 medial spine on peduncle of uropod 2. In many ways Trichophoxus is more primitive than the diverse Aus- tralian genus in such characters as peduncles of uropods 1—2 and lack of pereopodal combs but is far more specialized than the Australian genus in telsonic and urosomal ornamentation. Trichophoxus, our unnamed new Australian genus, and the genus to which Paraphoxus rakiura Cooper and Fincham, 1974, should be rele- gated (new), have numerous affinities as marked by mandibular molar, coxal humps, article 3 of antenna 2, ensiform process of antenna 2, gnathopod form, setation pattern on article 5 of pereopods 3-4, pre- buccal parts and the other characters noted above. Paraphoxus rakiura should be given a genus of its own, to differ from Trichophoxus in the presence of setae on the peduncle of uropod 1, the pattern of the various ornaments on uropods 1-2, and the giant tooth of epimeron 3 among others. Trichophoxus capillatus K. H. Barnard Figures 1-3, 4 (in part) Trichophoxus capillatus K. H. Barnard, 1930: 336-337, fig. 13. Description of female: Head (broken) about 18+ percent of total body length, greatest width about 110 percent of length, rostrum con- stricted, narrow, short, exceeding middle of article 1 on antenna 1; eyes > Fic. 1. Trichophoxus capillatus K. H. Bamard, lectotype, male, 7.15 mm. A, antenna; B, prebuccal lateral; C, coxa; D, dactyl of pereopod; E, epimeron; F, accessory flagellum; G, gnathopod; H, head; I, inner ramus, J, molar; L, lower lip; M, mandible; O, outer ramus; P, pereopod; Q, cuticle; R, uropod, S, maxilliped; T, telson; U, upper lip; V, pleon; Clarification of five amphipod genera 537 W, palp; X, maxilla; Y, right lacinia mobilis; a, peduncle; b, broken; d, dorsal; e, enlarged; f, flagellum; i, medial; 1, left; n, enlarged; r, right; s, setae removed; z = abnormal. 538 Proceedings of the Biological Society of Washington Sipe —— Spr lm me 2 ENSS [| a mi su AN he I) Bp // 4a 40 Ve Z fi U7, Fic. 2. Trichophoxus capillatus K. H. Barnard, lectotype, male, 7.15 mm. Symbols, see Fig. 1. Clarification of five amphipod genera 539 large, clear of occluding pigment; article 1 of peduncle on antenna 1 about 1.3 times as long as wide, about 1.9 times as wide as article 2, ventral margin with about 20 setules, strongly produced dorsal apex with 3 setules, article 2 about 0.85 times as long as article 1, with ven- tral crescent of 16 setae, primary flagellum with 9 articles, about 0.7 times as long as peduncle, bearing 1 long aesthetasc on articles 1-8, accessory flagellum with 8 articles; article 3 of antenna 2 with 17 facial setae, spine formula of article 4 = 4-11-11-7 or 4-11-11-6, dorsomedial margin fuzzy, ventral margin with 15 groups of 1-3 long to short setae, 3 ventro- distal setae and 4 dorsodistal spines, article 5 about 0.77 times as long as article 4, facial spine formula = one plus setule, dorsal margin bearing 7 sets of male setae and 1-2 calceoli, ventral margin with 5 sets of short setae, 4 ventrodistal setules, flagellum elongate, with 33 articles, one calceolus each on articles 4, 6, 8,...28 (one side broken); mandibles with weak palpar hump, right incisor with 4 teeth, left incisor with 7 teeth in 2 groups, right lacinia mobilis bifid, distal branch little shorter than proximal, narrow, with 3 facial humps, proximal branch simple, pointed, left lacinia mobilis with 4 teeth, middle teeth weakly shortened, right rakers 12 plus 1 rudimentary, left rakers 11 plus 1 rudimentary, molar composed of elongate soft cones, each molar with 4 large spines, one spine enlarged, each molar with bifid setule, palp article 1 short, article 2 with 6 medium to short inner apical setae and 2 other shorter inner setae and 1 long outer seta, article 3 about as long as article 2, oblique apex with 10 spine-setae, 4 inner setae, basofacial formula = 3 or 2; inner plate of maxilla 1 ordinary, bearing 1 long apical pluseta, and 3 other much shorter setae, palp article 2 with 5 apical spines and 4 submarginal setae; inner plate of maxilla 2 shorter than outer, outer much broader than inner, without apicolateral setae, inner with 7 medial setae; inner plate of maxilliped with 2 large, thick apical spines, 6 apicofacial setae, 4 medial setae, outer plate with 11 medial and apical spines (not all in illustration), 1 apicolateral seta, palp articles 1-2 lacking apicolateral setae, medial margin of article 2 moderately setose, article 3 with 17 facial setae, 8 lateral setae, nail of article 4 fully immersed, with 3 accessory setules; coxa 1 scarcely expanded apically, anterior margin weakly convex, main ventral setae of coxae 1-4 = 11-15-15-32, posteriormost seta of coxae 1-3 shortened, anterior and posterior margins of coxa 4 weakly divergent, posterior margin oblique, convex, posterodorsal corner rounded, posterodorsal margin of ordinary length, almost straight, width-length ratio of coxa 4 = 4:5; long posterior setae on article 2 of pereopods J-4 = 6-5-5-6, short posteriors = 0-2-0-0, long anteriors = (10—14)-12- 3-27, short anteriors = 2-3-4-0, gnathopod 1 also with 4 facial setae; gnathopods with apically narrowed hands, especially on gnathopod 2, width ratios on articles 5-6 of gnathopods 1-2 = 30:26 and 27:25, length ratios = 61:46 and 64:46, palmar humps ordinary to small, re- spectively, palms transverse, article 5 of gnathopods elongate, ovate, posterior margin rounded-flat; pereopod 4 stouter than pereopod 3 es- pecially on article 4, no margins with combs, facial setae formula on 540 Proceedings of the Biological Society of Washington Fic. 3. Trichophoxus capillatus K. H. Barnard, lectotype, male, 7.15 mm. Symbols, see Fig. 1. Clarification of five amphipod genera 541 Fic. 4. Trichophoxus capillatus K. H. Barnard, lectotype, male 7.15 mm; lower box, Protophoxus australis K. H. Barnard, syntype, size un- known. Symbols, see Fig. 1. article 4 = 5 and 6, on article 5 = 3 and 6, main spine of article 5 ex- tending to M. 85 on article 6, spine formula of article 6 = 7 + 9 and 7 + 8 plus huge middistal spine, acclivity on inner margin of dactyls of pereopods 3—4 highly apicad, sharp, produced as tooth, emergent setule short, midfacial pluseta absent; coxae 6-7 with both facial and marginal setae; articles 4-5 of pereopods 5-6 broad, facial spine rows dense, facial ridge formulas of article 2 on pereopods 5-7 = 0-1-1, width ratios of articles 2, 4, 5, 6 of pereopod 5 = 48:44:38:10, of pereopod 6 = 58: 542 Proceedings of the Biological Society of Washington 55:34:13, of pereopod 7 = 53:13:10:3, length ratios of pereopod 5 = 74:39:45:45, of pereopod 6 = 83:60:51:52, of pereopod 7 = 74: 15:27:19, article 2 of pereopod 7 reaching middle of article 5, with facial and ventral setae, no combs on pereopods 5—7; posteroventral cor- ner of epimeron 1 rounded, posterior margin straight, setose, antero- ventral margin with 12 medium setae, face with 16 short to medium setae, posteroventral comer of epimeron 2 rounded, weakly protuberant, posterior margin weakly convex, setose, ventral facial setae = 7, none set vertically, midfacial setae = 5, posteroventral corner of epimeron 3 rounded, weakly protuberant, with sinus, posterior margin short, weakly convex, serrate, setose, ventral margin with 9 setae, mainly in posterior half, face with several posterodorsal setae; rami of uropods 1-2 with fused apical nails, no ramus with accessory nails, outer ramus of uropod 1 with 4 proximal dorsal spines, apical margin naked, inner with 5 in single row only, outer ramus of uropod 2 with 5 dorsal spines, apical margin naked, inner with none, inner ramus as long as outer, peduncle of uropod 1 with 6 dorsolateral spines, medially with 6 marginal setae and spines, apicalmost an ordinary spine, ventrally with weak spike, peduncle of uropod 2 with 7 dorsal spines, medially with 4 spines, apico- lateral corners of peduncles on uropods 1-2 with weak comb, none on rami; peduncle of uropod 3 with 12 ventral spines, dorsally with 1 lat- eral spinule, 2 medial spines, rami masculine, inner extending to M. 100+ on article 1 of outer ramus, apex with 2-3 (3 probably normal), setae, medial and lateral margins setose, article 2 of outer ramus short, 0.10 as long as article 1, bearing 2 long setae, medial margin of article 1 setose, lateral margin with 11-12 acclivities, spine formula = 0 x 6, 1 X 3, OX 2 or Ox 577, O10 1 O51 0) 0; setal formula = 2 aloe Soule 1, 2 x 7, 3, 2, 3, 3; telson weakly elongate, length-width ratio = 33:29, not fully cleft, each apex of medium width, undulate, lateral acclivity deep, with lateral and medial spines separated by setules, each lobe with 3 other lateral spines, midlateral setules shifted distad, one on each side, each lobe with irregular mediobasal denticle patch; urosomite 1 with ventral crescent of 10 setae, 2 lateral facial rows of 12 and 5 spines, articulation line complete, urosomite 1 weakly protuberant dorsally; cuticle with enlarged bulbar setules mixed with pipes, setules surrounded by clear spaces in midst of fine striations in form of linear fingerprint pattern, emergent setules long, branched. Observations and illustrations: Rostrum broken; eyes slightly dis- associated; mandibular molar pointing towards observer but flattened in whole view; all long setae of uropod 3 plumose, article 3 of antenna 2 also with 1 disjunct setule (ordinary phoxocephalid with only 2 facial setules highly disjunct, this species with groups of 17 and 1); telsonic denticles intermediate between linear row and broad patch; right uropod 3 of lectotype with only 2 apical setae, left with additional setule (we presume 3 is normal from our study of related new genera); maxilla 1 with only 11 spines on outer plate, outer spine with facial hump showing as twelfth projection. Clarification of five amphipod genera 543 Lectotype: British Museum (Natural History) No. 1930.8.1: 142-144, male, 7.15 mm, from syntypes of K. H. Barnard (1930). Type-locality: Terra Nova Station 185, New Zealand, North Cape, L.ix.11, 3 meters, night. Birubius Barnard and Drummond, new genus Diagnosis of male and female: Eyes present; flagella of antennae 1-2 unreduced, article 1 of antenna 1 bearing medial fuzz in male, article 2 elongate, setae widely spread ventrally or fully ventral, primary flagellum with calceoli in male; antenna 2 lacking ensiform process, article 3 with only 2 facial setules, apicalmost usually elongate, article 5 not shortened (in standards based on Australian genera), bearing dorsal calceoli in male, flagellum with calceoli in male; upper lip and epistome scarcely distinct, often fully fused but recognizably with upper lip dominant; right mandibular incisor with 3 teeth, molar not triturative, medium to small, pillow-shaped, bearing 4 or more partially articulate spines, usually with fuzz (usually contained on a distinct plusetule), right lacinia mobilis bifid or simple, flabellate or thin, bearing or lacking denticulations and facial humps, palpar hump small to medium; lower lip bearing cones; palp of maxilla 1 biarticulate, inner plate with 4 setae (rarely 3); inner plates of maxilliped ordinary, strongly setose, palp article 3 scarcely or not protuberant, dactyl elongate, apical nail distinct, medium to elongate, not immersed; gnathopods small, similar in shape but article 5 of gnatho- pod 2 often shorter than on gnathopod 1, never cryptic, hands ovato- rectangular, poorly setose anteriorly, palms oblique; article 5 of pereo- pods 3-4 bearing setae posteroproximally, article 6 bearing normal 2 rows of spines, middistal spine or seta present, dactyls bearing inner tooth, apical nail weakly distinct or partly fused to body of dactyl; article 2 of pereopod 5 of broad form, untapering, of pereopod 7 of ordinary phoxocephalid kind, article 3 small (ordinary), articles 4—5 of pereopods 5-6 broad to narrow, pereopod 7 of normal size, article 2 ventrally naked or setose; peduncular apices of uropods 1-2 not combed, peduncle of uropod 1 normally elongate, lacking ventral spike or special spines, medial spines widely spread, dorsal spines confined apically, inner ramus of uropod 1 bearing only 1 row of spines, peduncle of uropod 2 with spines either widely spread or confined apically, inner ramus free and elongate; uropod 3 ordinary, article 2 of outer ramus with only 2 apical setae; telson naked laterally and dorsally except for normal pair of setules set about mark 33-45; urosomite 1 usually naked laterally (but occa- sionally with lateral setae); epimera 1-2 lacking posterior setae, with facial setae located ventrally, none in midface, epimeron 3 ordinary, vari- ous, often with facial, posterior and ventral setae, and teeth. Description: Rostrum various, from ordinary to severely reduced and often narrowed from dorsal view; facial spines on article 4 of antenna 2 in 2 or more rows, article 5 broad; article 1 of mandibular palp either short or weakly elongate, apex of article 3 oblique, palp thin, distal 544 Proceedings of the Biological Society of Washington branch of right lacinia mobilis either flabellate or thin; outer plate of maxilla 1 with 9-11 spines, no inner seta, one spine especially thickened; setation of maxilla 2 ordinary; inner and outer plates of maxilliped or- dinary; coxae 1—4 lacking anterodorsal humps; article 5 of gnathopod 2 free (not cryptic); article 2 of pereopod 7 with small posterior teeth, lacking facial setae; gills present on pereonites 2—7. Type-species: Birubius panamunus Barnard and Drummond, new species. Composition: The type-species and 37 other species to be described by Barnard and Drummond (in prep.). Relationship: Externally, Birubius is difficult to separate from the Paraphoxus group of genera but all species have 4 or more spines on each molar of the mandible whereas the Paraphoxus group has only 3 spines characteristically clumped basally and rather elongate. The inner plate of maxilla 1 in Birubius bears 4 setae, rarely 3, whereas all members of the Paraphoxus group bear only 2 setae on the inner plate. Birubius has an elongate article 2 of antenna 1 and normal sized article 5 of antenna 2, whereas the Paraphoxus group has a short article 2 of antenna 1 and a reduced size of article 5 on antenna 2 (males and females differing in the overall size but each with the article reduced in comparison to each sex of Birubius). Birubius differs from Parharpinia and Protophoxus in the absence of a special spine on the peduncle of uropod 1, in the absence of super- numerary dorsal spination and setation on the telson and in the elongate article 2 of antenna 1. New American genera of phoxocephalids will be distinguished from Birubius (in prep. ) Birubius panamunus Barnard and Drummond, new species Diagnosis: Birubius with fully broadened rostrum; right lacinia mo- bilis bifid, apical branch flabellate, apically denticulate, proximal branch simple, pointed; coxa 4 with long ventral setae, anterior and posterior margins parallel; epimeron 3 with large posteroventral tooth, bearing oblique facial row of setae, bearing row of ventral setae, posterior mar- gin naked; uropod 1 with basofacial setae, inner ramus with accessory spine adjacent to apical nail, each lobe of telson with 1 stout apical spine and attendant setule. Holotype: National Museum of Victoria, female “h’, 6.5 mm. Type-locality: Port Phillip Bay Environmental Study Station 1266, 11 March 1971, Port Phillip Bay, Victoria, Australia, 8 m, fine sand. Remarks: Only 3 of the 38 species in this genus, to be described by Barnard and Drummond (in prep.), have the combination of long setae on coxa 4 and accessory apical nail on the inner ramus of uropod 1. This species differs from the other 2 species in the parallel margins of coxa 4, the fully broadened rostrum and in the presence of a fully de- veloped oblique row of setae on the lateral face of epimeron 3. Material: 30 samples from 10 stations in Port Phillip Bay, 139 speci- mens. Clarification of five amphipod genera 545 Distribution: Australia, Victoria, Port Phillip Bay, 6-22 m, sand, silty sand, clay. Etymology: Birubius, Latinized, masculine, from an Aboriginal word meaning “Southern Cross”; panamunus, Latinized, from an Aboriginal word meaning “ocean”. ACKNOWLEDGMENTS We thank Dr. Roger J. Lincoln of the British Museum of Natural History, Dr. D. J. G. Griffin of the Australian Museum and Dr. Torben Wolff of the Universitets Naturhistorisk Museum for their assis- tance in attempting to locate specimens. Dr. Thomas E. Bowman of the Smithonsian Institution and Dr. E. L. Bousfield of the National Museum of Canada kindly read the manuscript and offered helpful suggestions. Miss Carolyn L. Cox of Smithsonian Institution inked our drawings for publication. Charline M. Barnard checked references and supplied the Literature Cited. LITERATURE CITED ALDERMAN, A. L. 1936. Some new and little known amphipods of California. Univ. California Publ. Zool. 41:53-74, figs. 1-51. BaRNarD, J. L. 1954. Marine Amphipoda of Oregon. Oregon State Monog., Stud. Zool. 8:1—103, pls. 1-33. ———. 1958. Index to the families, genera, and species of gammari- dean Amphipoda (Crustacea). Allan Hancock Found. Pub. Oce. Pap. 19:1—145. ——. 1960. The amphipod family Phoxocephalidae in the eastern Pacific Ocean, with analyses of other species and notes for a revision of the family. Allan Hancock Pacific Expeds. 18:175- 368, pls. 1-75. ——. 1963. Relationship of benthic Amphipoda to invertebrate communities of inshore sublittoral sands of southern California. Pacific Naturalist 3:437—467, figs. 1-7. ——. 1969. A biological survey of Bahia de Los Angeles, Gulf of California, Mexico. IV. Benthic Amphipoda. Trans. San Diego Soc. Nat. Hist. 15:175-228, figs. 1-30. — — 1971. Gammaridean Amphipoda from a deep-sea _ transect off Oregon. Smithsonian Contr. Zool. 61:1-86, figs. 1-48. —. 1972. Gammaridean Amphipoda of Australia, Part I. Smith- sonian Contr. Zool. 103:1-333, figs. 1-194. —. 1974. Gammaridean Amphipoda of Australia, Part II. Smith- sonian Contr. Zool. 139:1—148, figs 1-83. Barnarp, K. H. 1930. Amphipoda. British Antarctic (“Terra Nova’ ) Exped., 1910, Nat. Hist. Rep. 8:307—454, figs. 1-63. 1931. Amphipoda. Great Barrier Reef Exped. 1928-1929, British Museum (Nat. Hist.), Sci. Rep. 4:111-135, figs. 1-4. 1932. Amphipoda. Discovery Rep. 5:1—326, figs 1-174, pl. 1. 546 Proceedings of the Biological Society of Washington CHEvREUX, E. 1912. Amphipodes. Deuxiéme Expédition Antarctique Francaise (1908-1910) comandée par le Dr. Jean Charcot. Sci. Nat.: Doc. Sci.:79-186, figs. 1-62. Cooper, R. D. anp A. A. FincHaM. 1974. New species of Haustoriidae, Phoxocephalidae, and Oedicerotidae (Crustacea: Amphipoda) from northern and southern New Zealand. Rec. Dominion Museum 8:159-179, figs. 1-13, tables 1 and 2. Dana, J. D. 1853. Crustacea. Part II. U. S. Expl. Exped. 14:689- 1618; atlas of 96 pls. [published in 1855]. Guryanova, E. 1936. Neue Beitrige zur Fauna der Crustacea-Mala- costraca des arktischen Gebietes. Zool. Anz. 113:245-255, figs. 155. 1938. Amphipoda, Gammaroidea of Siaukhu Bay and Sud- zukhe Bay (Japan Sea). Rep. Japan Sea Hydrobiol. Exped. Zool. Inst. Acad. Sci. USSR. in 1934, part 1; 241-404, figs. 1- 59 (in Russian with English summary ). —. 1951. Bokoplavy morej SSSR i sopredel’nykh vod (Amphip- oda-Gammaridea). Akad. Nauk SSSR, Opredel. po Faune SSSR, 41:1-1029, figs. 1-705. 1953. Novye dopolnenijka k dal’nevostochnoi fauna morskik bokoplavov. Trudy Zool. Inst., Akad. Nauk SSSR 13:216—241, figs. 1-19 (in Russian). Haswe.i, W. A. 1879a. On Australian Amphipoda. Proc. Linn. Soc. New South Wales 4:245-279, pls. 7-12. 1879b. On some additional new genera and species of amphip- odous crustaceans. Proc. Linn. Soc. New South Wales 4:319- 350, pls. 18-24. Houmes, S.J. 1905. The Amphipoda of southern New England. U.S. Bureau Fisheries, Bull. 24:459-529, pls. 1-13. 1908. The amphipods collected by the U.S. Bureau of Fish- eries Steamer, “Albatross,” off the west coast of North America, in 1903 and 1904, with descriptions of a new family and sev- eral new genera and species. U.S. Nat. Mus. Proc. 35:489-543, figs. 1-46. Hurry, D. E. 1954. Studies on the New Zealand amphipodan fauna no. 3. The family Phoxocephalidae. Trans. Roy. Soc. New Zealand 81:579-599, figs. 1-5. KupryascHov, V. A. 1965. Novye vidy bokoplavov (Amphipoda, Gammaridea) iz vostochnoi chasti Oxotskogo Morja. Zool. Zhur., Akad. Nauk SSSR 44:1776-1789, figs. 1-10. Prrauot, J. M. 1932. Les amphipodes de l’expédition du Siboga. Deux- iéme partie. Les amphipodes gammarides. I.-Les amphipodes fouisseurs. Phoxocephalidae, Oecdicerotidae. Siboga-Exped. Mon. 33b:57-113, figs. 12-34. Sars, G. O. 1879. Crustacea et Pycnogonida nova in itinere 2do et 3tio Expeditionis Norvegicae anno 1877 & 78 Collecta. (Pro- Clarification of five amphipod genera 547 dromus descriptionis). Archiv Mathematik og Naturvidenskab, Kristiania 4:427—476. ——. 1895. Amphipoda. An Account of the Crustacea of Norway with Short Descriptions and Figures of All the Species 1:i—viii, 1-711, pls. 1-240, suppl. pls. 1-8. SCHELLENBERG, A. 1931. Gammariden und Caprelliden des Magel- langebietes, Sidgeorgiens und der Westantarktis. Further Zool. Res. Swedish Antarctic Exped. 1901-1903, 2(6):1-290, figs. 1-136, pl. 1. 1938. Litorale Amphipoden des tropischen Pazifiks. Kungl. Svenska Vetenskapsakad. Handl. 16(3):1-105, figs. 1-48. SHOEMAKER, C. R. 1933. Amphipoda from Florida and the West Indies. American Mus. Novitates 598:1-24, figs. 1-13. STEBBING, T. R. R. 1897. Amphipoda from the Copenhagen Museum and other sources. Trans. Linn. Soc. London, Zool. (2)7:23- 45, pls 6-14. —. 1899. Revision of Amphipoda (continued). Ann. Mag. Nat. Hist. (7)4:205-211. 1906. Amphipoda, I: Gammaridea. Das Tierreich 21:1-806, figs. 1-127. STEPHENSEN, K. 1947. Tanaidacea, Isopoda, Amphipoda, and Pycno- gonida. Norske Videnskaps-Akad. Oslo, Sci. Res. Norwegian Antarctic Exped. 1927-28, 27:1-90, figs. 1-26. THomson, G. M. 1882. Additions to the crustacean fauna of New Zealand. Trans. Proc. New Zealand Inst. 14:230-238, pls. 17, 18. THoRSTEINSON, E. D. 1941. New or noteworthy amphipods from the north Pacific coast. Univ. Washington Publ. Oceanogr. 4:50- 96, pls. 1-8. 548 Proceedings of the Biological Society of Washington INDEX TO NEW TAXA VOLUME 88 (New taxa indicated in boldface; n. c. = new combination) PHANEROGAMAE GOODENIACEAE Seaevolandaccada. Bryan iit (00 ike Al 2 Wc dy dla 75 thevgceiaky ieioiatel te (ny ee eee 74 PLATYHELMINTHES TURBELLARIA DisesiawlaniZONensis) jue ees Ae 113 MOLLUSCA CEPHALOPODA Graneledone antarctica _____-----------------an enna 448 UTTER CHOLY, De foe ee A ORS A NN I UL 454 ANNELIDA OLIGOCHAETA AVES Ts EOE UEUIS seo Se 2 UU a Cea Ne Lee ee 389 UY OTD NH er a I cA Lee ZOE RAEN LID a ee et HANA 390 ES BES URNA Aa | ETCLY) «ls yy se Ase el ASML La la nae Sa 171 Stephensomiara, terrae yey ava ree ten i \ YETI EDN SECA TE ee na Teer oN weet a AUN A ees MIM MODUS AN Ve LNG MUR ot SE 8 384 10) 011 Fee aa =e ed NE ANAC Ef UPN DPD ce De J I 385 POLYCHAETA Hermenia neoverruculosa ______._..----------------------2- 222 QA2, ARTHROPODA PYCNOGONIDA Anioplodactylus: compackus) mice, 28a oe ) aaeranil ds 191 CRUSTACEA Ascetocythere bouchardi eee a pseudolita:, 1.0020 ee 10 triangulata 20.000 ee ll Veriita. 922.020 NR a 12 Asellus (Asellus) alaskensis 60 Atya. dressleri jo8: 00 51 Bathyconchoecia diacantha 150 hardingae: 22230 2 ee 144 Birwbisgse 9 22a AO eee nc, 543 bate 6) 23. ee 520 panamunus 2222-00 oe 544 Calliasmata srimolii: 222.00 oo eee 37 Chorisquilla, -andamaniea 02 2 ee eee 258 EXCaVata Ti @) ao a A _ 254 pococki 2000. ee eee 256 Dactylocythere erena 200. ee 14 Scissuta: 22-00 oe Oe ae 15 Eurysquilla, pacifiea, 2. eee eee 249 Bireceolis)) es. eee ee a ec eee 490 Srinith if men ee 490 Macrobrachium erybelum ___...- 2-22 30 Orconectes etniert 2.2. ee eee 459 Saxatilisy 20500) Usaha ee 439 Parametopella inquilinus 220 eee 429 Psittocythere® ee 18 Dsittay See ee es ee 18 Sarsiella ezotothrix 6.0000 ee eee 497 Shiinoa: elapata. <0. 9 A ao), 433 DIPLOPODA Exallostethius) 2280000 ee) ee 212 Chrinax: 22.2 212 Nannaria domestica) 2.00 eee 186 rutherfordensis: -........... ee 184 PolyzoniumijJlaterale, 0.) ee 377 Stich) 22 et 375 INSECTA Watophilus knowltoni) 2000. elon eee ee 396 ECHINODERMATA OPHIUROIDEA Amphilimna valida tn: (c)..2 a ee eee 134 CHORDATA PIscEs PAU COTIMCLES PUea Eye ee ae ne oe eR eS 504 ARE | QT Go PCC TES ea NB 506 Gorythoichthys) amplexus) 225.2020 271 Bets Oo bry,co raise setae tal cau emai Ree tena SLE UG oe ee ee ee 332 Deh y e003 oer LN Se ee nec Leen) Sass ele sl 334 Glyphidodontops springeri 2.000 347 mAb OCMTOMIShtE EAL GS): 2 etese sy cee I otek Fs Pe 2 ee 320 RETO In aR EATa 2 S| ieee ete OW ee nae oN ue te i a ee 473 Synonathus” darrosanus)) 20.22 o 269 AMPHIBIA BNE TIONS HCW UAE ZH, he eee eae Ne ee eee See ena AEC en ee 315 Eleutherodactylus: leucopus 2.022 351 REPTILIA pxmistellipergy Mee Heil (2. eto bes Vie vue wean. heat aA MAL OL see 308 Sphaerodactylus thompsoni 202 369 Trophidophis nigriventris hardyi — 86 pilsbryi galacelidus 2) 5-22 81 MAMMALIA plkeina asa Gl Wr apTTTE Xi CAT) Aled Cops tae eaten epee ean seats As 104 Fossiz AVES Milvavomalexan drives) De Dinos ie Wk See 356 i Ril) i f SAN RU A NOMI vir My i : ose enn ah Sa ei TA Mera ae oink HIER MACrD Neenah NAHE ISH ‘iS ROR RNIN eRe ATU MN DN eA KATA Se rt : a a arn ae 3) 7 Shy NeW, ; Mitt Men beainter) AAS ib hie Fa) ae \ of Da ait “8 INSTITUT S31uvUs INSTITUT s?saiuvy: INSTITUT. INSTITUTION NOILALILSNINVINOSHLIWS ‘, -e NVINOSHLIWS saluvuaiy LIBRARIES NVINOSHLINS S31YV4U: SMITHSONIAN ty NWINOSHLIWS LIBRARI ES SMITHSONIAN SMITHSONIAN: N¥INOSHLIW INSTITUTION NOM. s " i a | a 2 z 2 cs = oc zd it ee = Ws ~ re) a = a e = 9 fe) = i zi = Zz 2 a SMITHSONIAN {NSTITUTION NOILNLILSNI NVINOSHLINS S3IYVYUGIT LIB cS ie Zz tas = r a z c z r -_ Paes) — ie) wo 5 of ae 5 - = 2 GM > e - : = te Ae ys re rex a = 2 Wi REM ae Me m 2 o | | NVINOSHLINS S3IYVYSIT LIBRARIES SMITHSONIAN INSTITUTION NOI rs 2] Pa ve 72) Pac Mo w S x Bh 5 WN = 5 Qe a Z SG 4e WE Z 2K By = = “yy EE NS 2 = 2G = > - = ee = = i < a Sa ee a ees SMITHSONIAN INSTITUTION NOILONLILSN!I NVINOSHLINS S3IYWVYEIT LIB NS 4 SMITHSONIAN INSTITUTION NOILALILSNI NVINOSHLINS S3I1YVuaIT LIB =H ms ” 3 f Fl ae H Z = AE 5 eo Xe # & foes _ ox at \Y we [eg — = = 3 = RQ = “ ae A: (e} — = ie) ay ; z aay = Oe eared a ; NVINOSHLINS S3!1uYvVusIT LIBRARIES SMITHSONIAN NOI mS z c z Cie = z 4 w - — w == Ve, ow — 2 \ = Be) = Gly x = > SQ : > Vie SV IW = E Ye Ee m SQ wo Ba n* = on a pi = a = wn = uo = = Be w e) = > = Z w z 7) % z 5 i Fhiy 3 OUY:. 5 r % Up 7 2 O lA Gi 2 a E z Gi = 24 7 ee = >" = > = 7) Zz 7) 2 | a NVINOSHLIWS LIBRARIES SMITHSONIAN INSTITUTION A NOI. = z ” a ud uj ; LJ 2 : g = Uy" a a AO’ ss = SMe sas 0 2) AN = fe) ay, re) oa | = a) Zz iy 2 J ; . SMITHSONIAN INSTITUTION NOILOLILSNI NVINOSHLINS S3IYVUSIT LIB = FE = ig z Bi : o So ae ders VSL OS Lew. a oO Ak lee A = a = _ V/s NA 2 4 : ‘e > KAY = = i= a r X& S\S wn = Canin = w ¢ MITHSONIAN INSTITUTION NOILMLILSNI NVINOSHLINS Dail: < = = 1) 5 \ = x 5 Ww) ‘3 ‘N . x (op) (82) op} = WA" 2 — = ba A \\ eZ Z 2 IWINOSHLINS SA3IYVYEIT LIBRARIES SMITHSONIAN INSTITUTION NOI LIBRARIES SMITHSONIAN Tp) = KS : Wl a ri, ul Ws oc = yo . S| XS Ba 4 4 FANG ARN = fe) O eas I Fa 2 Hh SMITHSONIAN INSTITUTION NOILALILSNI NVINOSHLINS SAIYVUSIT Lie an FS a oe z iG Paes) = o c ° ow yes 2 5 = KS 5 2 Shey a E > WQS = = Gy F 2 NY = 2 YP m ”) m N ig) m w & er) ~ = ”) IWINOSHLINS S3SIYVYEIT LIBRARIES SMITHSONIAN INSTITUTION NOI 7) z as 7) > ‘i a) = ye = < NY = ; a Oo, Y 3E fe) x Le 2 2 = 2% = B ne = ae MITHSONIAN _INSTITUTION NOILALILSNI NVINOSHLINS S3ZIYVYUSIT LIE ie a We ne FAvig ie @ = ra NX as, o os Wy ed oc a oc a iy Ho co@ - a =| f aA a a = a Yi fe) ee oO we re) z my z2 De a 2 VINOSHLIWS (S43 1uVvug less LIBRARI ES SMITHSONIAN TINS FITUTION NO! * rs Oo es & » i = = : = Gy ° E a 2 2 sb, > = = > = Wy <> i - “ - YF) ee = w wm” i (ep) z= oe Z o Zz MITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3I1uvUugi7 LIB Pad ; wn a 2) \ FE “ < = " = >" = 7) ES 2 wo Zz 7) WINOSHLINS SaluVdail LIBRARIES SMITHSONIAN INSTITUTION NOI us 5 i z i \ ox A ow pea o Sa. Se ‘ Pe = << ie a << WO" a a ae a o aS res (e) —_ (@) = —_ Fa, J z 4 FE THSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS SS3IYVWYUSGIT LIE i z i = iG Mee henry {ESN CD ih PGW Ta RR Oe 2 Hi ARIES | LIB | | 05 207 | | | Wilh INSTIT 3 9088 012 | | | | | | whowietg . Sioned seneteee Nee pera eta aejeadicramre ere er ee a ee Wand at eae peaene agatha n eran ame ae ON pele eerie pegeaarweerete oo se ut abe tt Sed rae tate te a ene ee eta Cue ge tae be ar Nw BE tote eans teva aietad rep tate neh ot 2294 eee pea waar late Taha epnineesi Mee teenie) reenenran ent Fase Sent et ars ates renner ae date pe aimee Ligaen aD lia at peeagngenee ne arse ne eke coeds Mos cone See eR Jot OAT ne kant aa seaea stew aser y= sasae ge syaaeee SON EA ENN ge ONDE ESAT OO ee ee a pc mahi etaees way Fee RS, Leet Rew erat ets cnetent seat eye ae iat aise Hak pita iy vers i atta ated eae rere eee sapoat aes Net ey ha Sag ae ET et saaerete eae Pras Dhar tetas sat ete naa ped eatin cinta aig sone uae we RS er fis wan yes prememnnrnr Tr Try) a Mase a ped age asi aneer ae Pre ee en Een rial he seg HAM ar eye ate freer ees “ aapnade stay tyr et aed grad Pape ID MDE Week pet gh Notas AGERE! (ERT SARIN IU RED EA MAE ES 12204 8S ekg AeA Ate hiaaee lV IA ote ADA INA ATT ie Ueia Spa ams Denne aide gece trian Rpeeer ern Per Dien Lan hone Re ae st peatig viata ganee (yp Uwe w CEA OT ARS COAT UM SITS A Facto deve a STEN, a frrorto ee oe Few kee re eee ett ate ede Perera poste pe gene wea wae Seine pea ANNs arg aynteg= 9a esa med ATs Mare. 46 8 ay we grat gen are 1d peas eT ee Ae et 4s Ot aden ant ew sp rei sania nena eat ae 6 wr 2 prone rsa sn 7 yas toeeg WE dane t ig aise ins aig iseabs aun + tee err eer eae cee lane area eta ete RE ae Beatin ee shana err hare tts i PO ere a rey sti el e a d Pererr avert tie le ee td tsa lee erenrr) aes eae 6 AUER RAND BEDE LED WM e gen sea ge nee ee Star eugiets ren WWAdAA UO amstFA was ey cas Tpteradye sip wei waeeeens etied Ord GS ENS: svene renter nt yrs r are rca PO I a sesy, ee yada cheney